PSI/AIDSMark Social Marketing Research Tool Kit 2004

6.2
5. Draft an interview guide.

IN-DEPTH INTERVIEWS

LEARNING OBJECTIVES
By the end of Chapter 6.2, the reader will be able to:
1. Understand when to use in-depth interviews. 2. Identify the advantages and disadvantages of using interviews. 3. Write an analysis plan, including study objectives. 4. Plan fieldwork, including informant recruitment.

6. Understand how to interview informants and promote discussion. 7. Understand appropriate formats for reporting interview results.
KIM LONGFIELD PSI 2004

PURPOSE OF IN-DEPTH INTERVIEWS

In-depth interviews are a method of gathering inforinformation from individuals about their behavior, opinions, feelings, and experiences. Although less often used than focus group discussions, in-depth interviews can help social marketing programs improve their understanding of target audiences and the issues under study; clarify a health phenomenon In-depth interviews usually last 1 to 2 hours, depending on the patience and availability of informants and their interest in the discussion.

6.2-1

In-Depth Interviews

Chapter 6.2

from the audiences perspective; and explore the why and how behind behavior. During an interview, a skilled researcher speaks with one informant.1 Informants can be members of the target audience or key informants, individuals who have special knowledge about the target audience, status among audience members, access to important information, or a willingness to share their knowledge and skills (Hogle and Sweat 1996). The interview format can be structured, unstructured, or semi-structured2. Advantages of In-Depth Interviews The major advantage of in-depth interviews is the confidential atmosphere in which informants can share sensitive information. Informants are able to provide detail about their perStructured interviews can help sonal experiences, views, and behavior. A private researchers explain a setting also means that peers do not influence inphenomenon whereas formants responses to study questions. Although unstructured, open-ended not as time- or cost-efficient as FGDs, in-depth ininterviews help them terviews allow researchers to collect a great deal of understand the phenomenon under study. information in a short period of time. Interviews permit access to audience members when groups (Hogle and Sweat, 1996). are difficult to coordinate or contact. They are also a mobile method that allows researchers to work with geographically dispersed populations. Finally, interviews are adaptable to different circumstances, permitting researchers to collect data in urban and rural areas. Disadvantages of In-Depth Interviews Despite the advantages of in-depth interviews, researchers must recognize their limitations and avoid using them inappropriately. The most common mistake is using interviews in lieu of surveys. In-depth interviews should not be used to measure program impact or to draw conclusions about the number of audience members reached by a
Throughout this chapter, the term informant is used to describe the individual who provides information during an interview. Other appropriate terms include interviewee, respondent, participant, and source (Hogle and Sweat 1996). 2 Because most interviews PSI researchers conduct are semi-structured, the majority of this chapter will focus on the appropriate format for semi-structured interviews rather than those that are structured or unstructured.
1

6.2-2

PSI/AIDSMark Social Marketing Research Tool Kit 2004 campaign or the extent of behavior change. In-depth interviews are less efficient in terms of FIGURE 6.2-1: BEHAVIOR CHANGE FRAMEWORK time and cost than FGDs are because researchers must Perception conduct more of Severity sessions to obtain as Social many different Affordability Support perspectives. Because in-depth Personal Risk interviews provide Assessment Availability Product detail about inAwareness of Use Transmission formants personal experiences, they do Solution not reveal social Brand Efficacy Appeal norms or trends and, Self Effilike other qualitative Awareness cacy methods, of Problem nonprobability sampling is used to recruit informants, meaning that the sample of informants is not representative of the larger population. As a result, study findings are only suggestive of trends among audiences and cannot be generalized to populations. Because interviews rely on self-reported data, researchers must be aware of the potential for information bias. Finally, thorough data analysis is time intensive, and its importance should not be undervalued.

WHEN TO USE IN-DEPTH INTERVIEWS

In-depth interviews are appropriate for use at all three project stages: planning (strategy development and marketing plan development), performance monitoring, and evaluation. As demonstrated in Figure 6.2-1, in-depth interviews are useful tools for gaining insight into determinants of behavior change and designing interventions. During the planning stage, in-depth interviews provide information about relatively unknown behavior or personal information, such as types of sexual activity or barriers to condom use.

6.2-3

In-Depth Interviews

Chapter 6.2

In-depth interviews are often conducted in conjunction with FGDs: researchers may conduct interviews with key informants to BOX 6.2-1: USE IN-DEPTH INTERVIEWS TO: obtain more detailed information Explore a relatively unknown behavior. than was obtained in FGDs or to Examine a sensitive study topic. solicit information from experts who know about the target audiObtain information from knowledgeable inforence. mants. Results from in-depth interviews are sometimes used to develop survey content and question structure. Researchers can learn what questions to include, in local languages, on a questionnaire (Bernard 1995; Werner and Schoepfle 1987) During marketing plan development, in-depth interviews identify issues important to target audience members, highlight areas for intervention, and allow managers to prioritize behavior change messages. Although FGDs are the preferred method for pretesting, in-depth interviews can also be used for this activity, enabling researchers to identify images and concepts that appeal to audiences. Learn the how and why behind behavior. Study complex behaviors and motivations. Uncover local terms related to a topic. Work with geographically dispersed informants. Obtain information that might be influenced by peer pressure during FGDs. Generate hypotheses for future research. Develop language and survey content. Generate new ideas for a program. Improve project implementation. Inform campaign/program development (pretesting). Reveal images, language, concepts, and packaging that appeal to audiences (concept testing). Clarify survey findings.

During performance monitoring, in-depth interviews provide programs with audience feedback about intervention efforts and identify areas for improvement. During evaluation, interviews are used to clarify survey findings and solicit additional feedback from target audience members, key informants, or project implementers. Interviews with project implementers can assess opinions about interventions effect on target audiences, successes and challenges associated with implementation, lessons

6.2-4

PSI/AIDSMark Social Marketing Research Tool Kit 2004 learned, and future directions interventions should take. Appropriate project implementers to interview include project managers at various levels, outreach workers, peer educators, trainers, decision makers, community leaders, and any other individuals connected with an intervention (Hogle and Sweat, 1996). TABLE 6.2-1: ADVANTAGES AND DISADVANTAGES OF IN-DEPTH INTERVIEWS Advantages Cost efficient Require little time for data collection Appropriate when access to groups is limited Key informant interviews can complement FGD data Provide detail about sensitive information, including personal experiences, views, and behavior Provide confidential atmosphere for informants Responses less influenced by peers than FGDs Mobile method Disadvantages Less cost-efficient than FGDs Require great deal of time for data analysis More interviews needed than FGDs to reach as many participants Require more skills than moderating FGDs Inappropriate for determining program impact, social norms and trends Not generalizable to larger audiences Quality of data depends on interviewer skills and quality of transcriptions Researcher has little control over environment since interviews may take place in variety of settings

TYPES OF INTERVIEWS
Several authors offer valuable guidance on types of interviews that are useful in social marketing research, among them Denzin and Lincoln 2000; Bernard 1998, 1995; Miller and Dingwall 1997; Hogle and Sweat 1996; and Weller 1979. Unstructured Interviews This format is flexible and relies only on a list of question topics. Unstructured interviews resemble a conversation more than an interview and are used to understand a target audiences complex behavior. No categories of responses are imposed on the interview. Unstructured interviews are most appropriate when administered by experienced researchers to examine study areas about which little is known. Anthropologists use unstructured interviews during the early stages of investigating a sociocultural phenomenon and during longterm fieldwork.

6.2-5

In-Depth Interviews Semi-Structured Interviews

Chapter 6.2

This is the most common interview format used for PSI studies. Semi-structured interviews facilitate the collection of new information with the flexibility to explore different topics in-depth with different informants. Their format is somewhere between unstructured and structured: they usually contain a one-page guide of open-ended questions that are asked of all informants in the same order. Semistructured interviews are best in situations where researchers will only have one opportunity to interview each informant. Structured Interviews Structured interviews use precoded questionnaires that are similar to surveys: they probe knowledge, attitudes, beliefs, and practices. Closed-ended questions are usually administered with a limited set of response categories and interviewers are instructed to never stray from the instrument. Responses may include multiple choice or Likert scales and, in some cases, open-ended questions. Structured interviews allow researchers to make comparisons among informants. Interviews by telephone, face-to-face interviews in informants households, intercept interviews, and interviews associated with survey research are the most likely to be included in the structured interview category (Denzin and Lincoln 2000).

BUDGET AND TIMELINE

Costs for conducting in-depth interviews depend on the settings BOX 6.2-2: IN-DEPTH INTERVIEW RULE: for the interviews and number of SET UP A REALISTIC SCHEDULE interviews to be conducted (Table 6.2-2). Costs increase with each Do not schedule more than two interviews area and type of informant included in per day for each interviewer. Too many the study. The least costly option is interviews in a short period of time fatigues conducting interviews with key inforresearchers and compromises data mants in urban areas (near program quality. Inexperienced interviewers should conduct only one interview per day. sites) or with a limited number of audience members. Sufficient time should be set aside for conducting interviews (Figure 6.2-2). Typically it takes 2 weeks to outline study objectives and draft an interview guide. Recruiting should start at least 3 weeks before

6.2-6

PSI/AIDSMark Social Marketing Research Tool Kit 2004 conducting interviews. Allow 1 week each for pretesting and refining the guide. For each interview to be conducted, allow half a day for preparation (contacting gatekeepers, arranging appropriate dates, determining incentives, etc.), half a day to conduct each interview, and 1 or 2 days to translate and transcribe results. Analysis and report writing can take anywhere from 2 to 8 weeks to complete, depending on the number of interviews and complexity of data to be analyzed.

TABLE 6.2-2: AN EXAMPLE BUDGET FOR INDEPTH INTERVIEWS IN ZAMBIA (N= 30) Recruiter honorarium Training for interviewers Interviewer honorarium Transportation Participant incentives Refreshments Supplies (microcassette recorders, cassettes, batteries, etc.) Translation, transcription and analysis Report writing Total (US dollars) 295.00 124.00 882.00 118.00 118.00 50.00 335.00 2,065.00 1,000.00 4,987.00

Note: This budget is from the HIV Perception amongst Outof-School Boys study. All costs are approximate and only include fees incurred by PSI staff. No research agency was hired to conduct this study, resulting in lower-than-normal costs. In addition, no translator was used: PSI staff conducted simultaneous translation and transcription.

Although less efficient than FGDs, interviews are a cost-effective means for gathering a great deal of qualitative information in a short period of time. In-depth interviews are generally inexpensive to design and conduct; however, transcription, translation, and analysis can be costly. During formative research, interviews uncover the how and the why associated with behavior to inform project design. They can assist programs with developing survey content and decrease the cost of subsequent quantitative data collection. During pretesting, interviews may uncover flaws in preliminary concepts and messages, saving programs money during production. In-depth interviews used during monitoring are an inexpensive means for uncovering programmatic shortcomings and can prevent additional spending on unsuccessful activities. Finally, during evaluation, interviews can clarify survey findings, providing programs an opportunity to adapt or change strategies, improve project proposals, and garner additional funds.

6.2-7

In-Depth Interviews

Chapter 6.2

There are several BOX 6.2-3: COST-EFFECTIVE IN-DEPTH INTERVIEWS key questions you can use to help evaluate the Adequate planning can save time and money during cost of interviews (adapted in-depth interviews (adapted from Morgan and from Morgan and Scannell Scannell 1998): 1998): Hire more than one interviewer to conduct siCan staff members plan, multaneous interviews. conduct, and analyze inAppoint a principal investigator to make major terviews or must you hire decisions. an outside agency? Hire recruiters, interviewers, transcribers, and Who should study inforother team members with prior experience. mants be and what will it Define study objectives and state research take to recruit them? questions up front to expedite planning. Are all of the informant Define the study population to expedite recruitcharacteristics youve outing. lined important or can you cut costs and conduct Use gatekeepers to identify informants. fewer interviews? Simplify analysis for pretesting results. How thorough must the Use computer software for 10 or more trananalysis be? Is a synthescripts. sis of findings sufficient or do you require a detailed Write detailed field notes in lieu of full tranreport for external distribuscriptions (keep in mind this will sacrifice tion? data).

DATA ANALYSIS PLAN

Use a template for report writing.

When developing the analysis plan, researchers should think Analysis of in-depth interviews is called contextual about how interview results will analysis, which highlights common and divergent be used to make programmatic themes in informant responses. Reports on formative decisions. If in-depth interviews research should tell a comprehensive story and proare used for formative research vide readers with a picture of the study population and and exploring a relatively unbehavior. known behavior, researchers should write a comprehensive report appropriate for external distribution.

6.2-8

PSI/AIDSMark Social Marketing Research Tool Kit 2004

FIGURE 6.2-2: SAMPLE GANTT CHART FOR CONDUCTING IN-DEPTH INTERVIEWS Tasks
1

May
2
x

June
4 1 2 3 4 1

July
2 3 4 1

August
2 3 4

Decide on Study Objectives and Research Questions to be Answered Determine Study Design Develop Data Analysis Plan Develop Interview Guide & Training Manual Prepare logistics

Recruit informants
x x

Train interviewers Pretest Interview Guide Revise Interview Guide Conduct Interviews in Urban Area Conduct Interviews in Rural Areas Translate and Transcribe Tapes Analyze Data* Prepare Report* Likewise, if the team plans to report findings at an important meeting or conference, ensure that the analysis is thorough and highlights unexpected findings. In general the report content and format should follow study objectives. Reports should provide a general description of the study sample, explain the behavior in question, highlight specific factors associated with behavior, present unexpected findings, and propose suggestions for targeting the behavior in question. Interviews conducted during pretesting do not require an extensive report. A synthesis of findings divided by topic area (e,g., message, actors, slogan) is sufficient to provide agencies with necessary information to improve campaign design.
x x x x

6.2-9

In-Depth Interviews

Chapter 6.2

Interviews conducted for program monitoring usually require a short report that can be distributed to decision makers. Reports should provide audience feedback about activities and outline suggestions for improving programs. Results from interviews conducted during pro- BOX 6.2-4: IN-DEPTH INTERVIEW RULE: STUDY OBJECTIVES ject evaluation should be woven into a larger end-of-project report that Always outline study objectives before starting reincludes findings from surveys and search. Remember that the study objectives are the other relevant data. most important part of any study and that they should For more detail, refer to this chapters sections on Data Analysis and Interpretation and Report Writing. be carefully identified during the planning stage (adapted from Oladosu 2000). The objectives should be clear and specific and should define the target population and its characteristics. Recall that objectives for qualitative research are expressed differently than those for quantitative research. Objectives for qualitative research usually include such words as: Explain Explore Qualify Define Identify Refine Discover Understand Describe Elicit Discern

SAMPLING METHODOLOGIES FOR INDEPTH INTERVIEWS

Informants for in-depth interviews are typically selected through convenience sampling.

Unlike most surveys, interviews do not require probability samples, but informants are selected purposively (i.e. with eligibility criteria in mind). For the most part PSIs qualitative studies use purposive sampling. Typical methods for selecting informants are snowball sampling, quota sampling, judgment sampling, typical cases, extreme cases, and panels (Figure 6.2-3). Snowball Sampling

Snowball sampling (also called network sampling) is the most common sampling strategy used for in-depth interviews. Snowballing refers to the process of asking people researchers have already contacted for the names of additional informants; the analogy refers to a small snowball growing as it rolls downhill.

6.2-10

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Researchers should use snowballing when they have limited access to the study population or when the behavior in question is stigmatized or illegal (e.g., drug use or commercial sex). Snowball sampling should also be used when gatekeepers are available to help with recruitment. An example would be asking club owners to recruit youth who are likely to know about recreational drug use. Initial informants could then suggest other youth that would be appropriate for BOX 6.2-5: EXAMPLE OF STUDY OBJECTIVES FOR IN-DEPTH INTERVIEWS inclusion in the study. Quota Sampling Researchers should use quota sampling when time and cost constraints are concerns and when the researchers can determine how many interviews are required to provide needed information. A fixed number of informants are recruited for the study. An example would be recruiting 20 men from a bar where commercial sex workers (CSWs) usually solicit clients. Convenience Sampling In-depth interview research is a linear process. Study objectives inform data collection, analysis, and report writing. For example, in a study of Zambian males misconceptions about HIV/AIDS: Study objective: Identify young Zambian males misconceptions about HIV/AIDS. Interview guide questions: How is HIV transmitted? and Who is most vulnerable to HIV/AIDS? Analysis & report: Overall, informants awareness of HIV/AIDS was high, but specific knowledge about transmission, prevention, symptoms, and risk factors was limited. Most informants identified promiscuous, careless, and unhygienic individuals, especially women, as likely transmitters of HIV. Misconceptions about HIV being transmitted through mosquito bites, kissing, and biting were also common.

During convenience sampling, researchers include indi(Longfield, Cramer, and Sachnigongu 2003) viduals in their study as they please. This method is often used to obtain exploratory information in a quick and inexpensive manner. An example would be to include any individuals found at an outdoor market in the study. The disadvantage of this strategy is that the only characteristic informants have in common is the location where they were recruited. As a result, the authors do not recommend using convenience sampling for PSI studies and urge researchers to use purposive sampling in order to select informants according to eligibility criteria.

6.2-11

In-Depth Interviews Judgment Sampling

Chapter 6.2

Researchers conduct judgment sampling when they know the population well enough to determine its members expertise and potential contribution to the study. For example, in an analysis of family planning procedures, researchers could recruit clinic staff that administers FIGURE 6.2-3: SAMPLING METHODS FOR IN-DEPTH INTERVIEWS family planNonprobability Sampling ning programs. Researchers would inNeed exploratory Do you want Time & cost information in a quick to track a constraints? clude only Limited access to Are there indiand inexpensive group over Can you deterviduals who manner? population? time? mine how many should be inthese indiAre there no eligibility Gatekeepers interviews are cluded in the criteria for informants? available? required? study because of viduals in the their expertise or Panel experience? study because they Quota Judgment are the most Snowball Typical Convenience informed and have the Extreme most to contribute. There is no need to ask anyone else. Typical Cases Researchers recruit typical cases when a behavior is relatively rare and they wish to describe behaviors that are distinct to a study population. An example would be recruiting men who have sex with men (MSM) and examining their risk perception for HIV/AIDS. Extreme Cases Researchers use extreme cases when they wish to describe behaviors that fall outside of the norm for a particular group. For example, if the purpose of a study is to examine household feeding practices and the average number of children per family in a country is 3, researchers would recruit mothers with very few children (one or two) and those with several children (five or more).

6.2-12

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Panels Panels are used when researchers wish to track attitudes and behavior among a particular group of individuals over time. An example would be conducting annual interviews with the same group of youth over 3 years. Informant Selection It is often difficult to find informants who fit the required study profile and feel comfortable discussing issues with researchers. Care should be taken when recruiting informants to ensure that they can contribute to a discussion, will not feel intimidated by an interviewer, and can provide responses that reflect experiences typical to the target audience. When using gatekeepers for recruitment, ensure that they provide you with a variety of informants who fit the study profile. Researchers should not include only the best, brightest, or most popular individuals in their studies. Researchers should also be aware of recruits who have participated in multiple studies and may be unable to provide new information. BOX 6.2-6: TYPICAL INFORMANT CHARACTERISTICS Sex Age Marital status Sexual activity Education Occupation Socioeconomic status Contraception status HIV/AIDS status Ethnicity Religion Location or residence

Informants should not know the specific subject of the interview in advance. If they do, they may formulate ideas before the discussion and lack spontaneity during the interview. Recruiters often share the general topic area with recruits, but do not disclose the specific subject. For example, if a discussion is about sexual behavior and risk assessment, recruiters may tell recruits that they will discuss relationships between men and women (AIDSCAP 1994).

COMPOSITION AND NUMBER OF IN-DEPTH INTERVIEWS

The number and composition of interviews required for a study rely upon the study objectives, characteristics of the target population, and study locations. Factors that can increase the required number

6.2-13

In-Depth Interviews

Chapter 6.2

of informants include, comparing different populations (single vs. married), including several population characteristics (sex, age, and ethnicity), and sampling multiple sites (urban and rural). As a general rule, researchers have found that between 6 and 20 interviews is appropriate for formative research. Fewer interviews are needed when they supplement other data (e.g., FGDs) or when research objectives are very limited (Bernard 1998). Informants Overall, an appropriate informant is someone who meets the study criteria and is a good source for information. Friends and relatives of interviewers are not eligible candidates since their familiarity with the topic and researcher may bias study results. According to Spradely (1979), a good informant: Knows the local culture Is involved with the study topic Can share firsthand experience with study topic Has adequate time to devote to an interview Is impartial; has not already analyzed study topic from an outsiders perspective. Number of Interviews BOX 6.2-7: IN-DEPTH INTERVIEW RULE: NUMBER OF STUDY INFORMANTS Conduct at least two interviews for each type of informant (member of target audience). Recruit the most important key informants to include in the study (there may not be two of each).

Over-recruit so that if some informants At a minimum, researchers should drop out of the study, there will still be conduct two interviews for each type enough for statistical validity. of informant in the target audience. This will ensure that if one interview does not go well, the research team still has another interview with the same type of informant from which to collect data. Conducting at least two interviews for each type of informant also permits researchers to confirm the reliability of study data. When interviewing key informants, it may be impossible to find two of each type. Researchers should be sure to include the most important key informants in their study and provide a variety of perspectives on the study topic.

6.2-14

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Figure 6.2.4 illustrates the process for segmenting a target population and determining the minimum number of interviews required for a study with the objective: Examine sexual risk-taking behavior for HIV/AIDS among CSW clients. The desired informant characteristics are: male, CSW client, 25-40 years of age, married/single, urban/rural. Population segmentation shows that there are 12 categories of informants. To interview two informants from each category will require 12 categories x 2 interviews, or a total of 24 in-depth inFIGURE 6.2-4: POPULATION SEGMENTATION FOR IN-DEPTH INTERVIEWS terviews.
Male CSW clients, 25-40 years

Urban

Rural

Married

Single

Married

Single

25-30

31-35

36-40

25-30

31-35

36-40

25-30

31-35

36-40

25-30

31-35

36-40

DEVELOPING IN-DEPTH INTERVIEW GUIDES

A comprehensive discussion guide is essential for conducting good interviews. Guides should include appropriate sections, contain questions that answer study objectives, and meet the needs of data users. A common mistake is to ask too many questions, resulting in an unwieldy guide, long interviews and compromised data. When guides are too long, interviewers must rush through questions, addressing them only superficially. Interviewers also find it difficult to probe interesting questions, and the session appears more like a survey than a dynamic discussion (Dawson, Manderson, and Tallo 1993).

6.2-15

In-Depth Interviews

Chapter 6.2

When developing guides, researchers must reflect on their analysis plan and the appropriate report format. They should share draft interview guides with others for review and feedback. Several model guides are available in the appendices that provide examples of question order, structure, and quantity. These models can be adapted for specific programs and different country settings. Interview Guide Format The following sections discuss the essential components of an interview guide. For examples of guides, refer to the appendices. Introduction BOX 6.2-8: IN-DEPTH INTERVIEW RULE: INTERVIEW GUIDES Restrict questions to answer study objectives. Avoid the temptation to ask too much. Ensure that guides are context specific. Ask open-ended questions. Use simple language. Keep questions short. Do not use questions that comprise several parts. Start general and get more specific. Place sensitive questions in the middle of the guide. Provide a complete wrap-up.

In many ways, the introduction is the most important component of an interview. It is the interviewers opportunity to make informants feel comfortable, explain the purpose of the study, and set the tone for the (Bernard 1998; and adapted from rest of the discussion. Although Dawson, Manderson, and Tallo 1993) an introduction should be thorough, it should not be intimidating. Interviewers should make informants feel valued and happy about their inclusion in the study. Interviewers should cover the following information during the introduction (adapted from Andrew Fletcher Consulting 2001): Introduce the interviewer and explain the reason for conducting interviews. Ask the informant to answer honestly. Explain that there are no right or wrong answers. Acknowledge the tape recorder. Reassure informant that his or her responses are confidential. Tell informant how long discussion will last.

6.2-16

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Ask informant to introduce himself or herself. The informant can use a pseudonym if he/she feels uncomfortable giving a real name. Obtain general information from the informant (e.g., name, age, occupation, education). When asking informants to introduce themselves, interviewers should ask them to share information about themselves. This strategy creates a casual atmosphere and shows that the interviewer is interested in getting to know the informant. Appropriate introductions include: Tell me how long youve been working in X, Tell me what you like to do in your free time, Tell me what you enjoy most about school/your work, and Tell me how long youve lived in this city. Icebreaker (Warm-up) A general question asked at the beginning of an interview is intended to break the ice and make informants feel comfortable. A good icebreaker should sound natural and have something to do with either the informant or the study topic (Bernard 1998). The warm-up question should never be too serious or intimidating. Example icebreakers include: When I say AIDS, whats the first thing that comes to mind? When did you first become interested in (health topic)? What do you like most about (health topic)? Key Questions Key questions are contained in the general discussion component of the BOX 6.2-9: IN-DEPTH INTERVIEW RULE: NUMBER OF QUESTIONS guide. This section follows the icebreaker and constitutes the heart of the guide. Key questions should be relevant to study obLimit interview questions to one page whenever possible. jectives and remain faithful to the topic at Informants may worry about hand. Questions in the beginning should be discussion length when they general and those toward the middle and end see interviewers flipping more specific. It is important for interviewers to through several pages. know which questions constitute key questions so they allow sufficient time to discuss these

6.2-17

In-Depth Interviews issues.

Chapter 6.2

During formative research, sensitive questions should be placed in the middle of the guide so that informants feel comfortable answering them, but the discussion does not end on a sensitive note. During interviews for pretesting, key questions address concept testing and allow informants to offer opinions about draft concepts. Transition Questions Transition questions should be used between key questions to move the conversation smoothly from topic to topic. Wrap-up (Ending) At the end of the discussion, interviewers provide a wrap-up for informants to bring closure to the discussion. That is, they briefly summarize the discussion, thank the informants for their participation, and ask if there is anything else the informants would like to add. A good wrap-up should make informants feel that their participation was important and a major contribution to the study. The wrap-up can contain three types of questions (Krueger 1998a): BOX 6.2-10: EXAMPLES OF ALL-THINGSCONSIDERED QUESTIONS All-thingsOf all the needs we discussed, which is most important to you? considered Suppose you had one minute to talk to the person in charge of questions: At this project, what would you tell him/her? the end of interviews, interGive me one sentence that best describes how you feel about the viewers allow topic we discussed today. informants to state their final position on important areas of concern. The allthings-considered question allows informants to express their opinion about what were the most important points brought up during the interview. This question can also help informants clarify their position on a topic if they have provided conflicting statements during the interview (Krueger 1998a). Summary questions: The summary question is asked after the interviewer has provided a brief summary of the interview and identified the major ideas that emerged during discussion. The

6.2-18

PSI/AIDSMark Social Marketing Research Tool Kit 2004 summary should last only 1 or 2 minutes. The interviewer then asks about the adequacy of the summary (Krueger 1998a).

BOX 6.2-11: EXAMPLES OF SUMMARY QUESTIONS

Is this an adequate summary of our conversation?

Final questions: The fiDid I correctly describe what you said? nal question during an How well does that capture what you said today? interview is asked to ensure that no important issues have been overlooked. The question begins with a short overview of the purpose of the study and followed by the interviewer asking if anything has been BOX 6.2-12: EXAMPLES OF missed (Krueger 1998a). FINAL QUESTIONS Have I missed anything? Is there anything else we should have covered but didnt? Guides for Evaluating Interventions Interview guides for evaluating interventions can contain as few as six or seven open-ended questions that take approximately 20 minutes to administer. A more complex guide would contain specific implementation-related questions BOX 6.2-13: IN-DEPTH INTERVIEW (Hogle and Sweat 1996). Table RULE: QUESTIONS TO INCLUDE 6.2-3 offers a list of esEvaluative interviews should include the following questions (Hogle and sential quesSweat 1996): tions to in1. How do you feel the target audience has responded to this intervenclude in tion? 2. What have been the strengths of this project from your perspective? evaluative 3. What have been the weaknesses of this project? interviews. 4. How could the weaknesses be addressed in future activities? 5. From your experience, what are the lessons to be learned from this project that could be applied to future interventions? 6. How would you do things differently in the future? 7. What is an appropriate follow-on project to this one? Where should we go from here?

6.2-19

In-Depth Interviews

Chapter 6.2 TABLE 6.2-3: EXAMPLES OF QUESTIONS FOR INTERVIEW GUIDES (KRUEGER 1998A)

Introduction Icebreaker Transition Key Key Transition Key Key Key Key Key Ending

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Tell me how long youve lived in this city What to you is good health? What to you is a healthy lifestyle? Think back to the last time you wanted to make a change relating to health. It may have been a change in what you eat or your exercise habits. What kind of barriers did you run into? What helped you or would have helped you the most in making the change? Suppose your doctor has told you that your cholesterol level is too high. What would you want to know? What kind of information would you want to get about foods and your diet to help you lower your cholesterol? There are a lot of different ways you could get the types of information you have been talking about. How would you like to get that information? Of all those ways of getting information, which do you feel is most important? Suppose that a workshop on cholesterol was held. What would get you to attend? I realize that its hard to stay motivated to learn about things like cholesterol, but what would entice you to come back for more? What would keep you interested? We are going to be putting together programs for residents on how to lower their cholesterol. Based on what you shared today, what advice would you offer the person in charge of this project?

Using Open-Ended Questions Qualitative data collection requires open-ended questions that elicit detail from informants. Unlike surveys, interview guides should not contain questions that restrict informants to categories of responses, such as yes/no or numbers. Instead, open-ended questions encourage informants to determine the direction of the response and encourage them to reply at length. The major advantage of using openended questions is that they reveal what is on informants minds, as opposed to what researchers suspect is on their minds (Krueger 1998). Although using closed-ended questions during in-depth interviews is discouraged, they are not totally off-limits. Toward the end

6.2-20

PSI/AIDSMark Social Marketing Research Tool Kit 2004 of discussions, it may be useful to narrow informants responses by asking some closed-ended questions. This is especially true during structured interviews and pretesting.

BOX 6.2-14: IN-DEPTH INTERVIEW RULE: QUESTION FORMAT

Some questions are deceptive and appear to open-ended when they are not. Questions that being with how much, how often, and to what extent imply a range of responses, such as very, somewhat, and not at all. Pretesting Interview Guides

Use open-ended questions that start with: For What How Describe Avoid closed-ended questions that start with: Do Is When Where

Once an interview guide is drafted, interviewers must conduct two or three preliminary interviews to pretest it with key informants or members of the target audience. Pretesting should determine the appropriateness of question format, content, language, and length. Researchers should revise the interview guide accordingly and then conduct study interviews. When pretesting interview guides, ask the following: Are the questions understandable? Do the questions sound awkward? Are any questions leading? Are any questions closed-ended? Do informants understand the questions? Do the questions promote discussion? Is the guide too long? Does the guide address only whats important to the study? What works? What doesnt? Are you obtaining the responses you need?

6.2-21

In-Depth Interviews Other Considerations for Interview Guides

Chapter 6.2

BOX 6.2-15: IN-DEPTH INTERVIEW Even with the same study topic, it may RULE: CONSISTENCY be necessary to prepare different guides for different types of inforOnce a final interview guide is developed for mants, especially if key informants structured or semi-structured interviews, do and members of the target audience not change questions. Maintain as much are included in the study design. For consistency as possible throughout the series example, a study examining contraof interviews since it is important to compare ceptive use will require a different and contrast themes that emerge during disinterview guide for current users of cussions. Analysis is reliant upon theoretical saturation, which is only possible with consiscontraceptives, non-users, and protency of questioning. viders (adapted from Debus 1995).

FIELD PREPARATION

(Krueger 1998).

Although field preparation for interviews is simpler than that required for surveys, a great deal of coordination can be required to ensure that data collection proceeds without any major complications. Interviewers must be adequately trained, past research studies examined, contact made with the gatekeepers and informants, supplies gathered, and arrangements made for study locations. The BOX 6.2-16: THE INTERVIEWER more preparation that takes place in advance, the better the Responsibilities: Interviews informants and eninterviews and ultimately, the gages them in study topic data collected. Staff Requirements and Functions For each interview, an interviewer is required. Unlike FGDs, note-takers and assistants do not participate in data collection. The interviewer must serve as all three. Requirements: Confident, objective, good listener, flexible, patient, organized, makes informant feel comfortable answering personal questions, informed about study topic, takes passive rather than assertive role, empathic, can take limited notes while listening, probes in timely fashion, ensures continuous rapport, discerns negative signs early and offers solution, displays seriousness combined with humor, informants can relate to him/her (e.g., not too young/old, speaks same language).

Without the group process to loosen inhibitions, interviewers must be more skilled than their FGD counterparts to gain informants trust (Baron et al. 1993)

6.2-22

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Interviewer behaviors influence the success of data collection. According to Debus (1995), interviewers must be capable of: Accurately receiving information: This capability may be inhibited by interviewer fatigue, boredom, bias, preoccupation with taking notes, and technical language that is foreign to the interviewer. Accurately recalling information: This capability may be inhibited by confusing content between interviews, retaining only select information, or attempting to retain too much information Critically evaluating information: The interviewer must be capable of filtering the information respondents provide. A good interviewer can steer informants away from irrelevant information and probe valuable responses to gain richer information. Acting upon the information received to obtain as much detail as possible within a given interview and across interviews: A good interviewer FIGURE 6.2-5: ENSURE THAT INTERVIEWERS probes for more ARE ADEQUATELY TRAINED AND FAMILIAR detail, keeps the inWITH STUDY OBJECTIVES formant on topic, and keeps study objectives in mind during an interview. Altering the process across interviews allows researchers to incorporate information obtained from initial interviews into those that occur later in the study, facilitating more in-depth data collection.

6.2-23

In-Depth Interviews Training

Chapter 6.2

Before selecting interviewers, ensure that they have received adequate training to complete their responsibilities (Figure 6.2-5). A typical training would include instruction on outlining study objectives, writing interview guides, pretesting, conducting interviews, preparing field notes, transcribing, and preparing for data analysis. Interviewers may require a great deal of training. In addition to the listed requirements, they should have experience working with sensitive topics and feel comfortable engaging informants in discussion. It takes time to refine interviewing skills and understand the appropriate approach to take with different informants. Interviewers must also be clear about the purpose behind all questions in the guide so they can elicit essential information. Advance Preparation 1. Conduct Preliminary Research Before Entering Field. This step is necessary for formative research, but not for pretesting or concept testing. Before collecting data, researchers should establish what has already been completed on the study topic. They should pull past studies from peerreviewed journals, read gray literature (internal documents) from other organizations, and request past PSI studies from headquarters. 2. Contact Gatekeepers if Necessary. In addition to reading past studies, researchers should collect information from gatekeepers about the target audience. Gatekeepers are valuable sources of information on behavioral issues, audience preferences, and appropriate interventions. In addition, gatekeepers can help researchers establish trust within the

BOX 6.2-17: IN-DEPTH INTERVIEW RULE: WHEN MAKING AN INITIAL VISIT TO THE INTERVIEW VENUE, ASK THE FOLLOWING: Is the site private? If the site is not private, will the interviewer be able to hear the informant? Will informants feel comfortable being interviewed at this site? Is the site non-threatening and neutral? Is the site accessible to informants? Are there doors and windows that shut to minimize outside noise? Will others have access to the site? Will you have access to the site before the start of the interview in order to set up? Is there sufficient light for note taking? Is the room quiet enough to make a clear audio recording? Will anyone need the site during or immediately following the interview?

6.2-24

PSI/AIDSMark Social Marketing Research Tool Kit 2004 study community. They can contact informants and encourage them to participate in the study (key for snowball sampling), assist with logistics, and make introductions before interviews begin. 3. Decide upon Appropriate Incentives. Before entering the field, researchers should discuss the appropriateness of providing incentives to informants. They should identify informant expectations, other organizations policies, reasonable values for incentives, and the local appropriateness of gift giving. Providing incentives may set a precedent, encouraging future informants to demand items in exchange for participation. Researchers must also consider the sustainability of providing incentives. Will the program always have the funds to BOX 6.2-18: IN-DEPTH INTERVIEW do so? Will providing incentives RULE: ENSURE INFORMANTS make it impossible for organizations COMFORT DURING THE INTERVIEWS with fewer means to conduct research? Can they get to the site? Incentives should not be costly or elaborate. In the past, PSI offices have provided informants with reimbursement for local transportation and PSI giveaways (such as T-shirts and key chains). Will they feel comfortable being seen at the site? Is the interview planned for an appropriate time or do informants have scheduling conflicts (e.g., school, work)? What language will be used during interview? Are informants most comfortable speaking a local dialect or the national language? Will you use colloquial terms during discussion? Are you familiar with terms informants use? How will you dress? Are you dressed as casually/formally as informants?

4. Conduct Mock Interviews. Interviewers should conduct mock interviews before entering the field. They should identify coworkers, friends, and family who can serve as informants and provide feedback about their performance. Programs can save time and money during pretesting by having interviewers practice with people they know. Practice sessions also help interviewers build confidence. Pretesting sessions should be regarded as additional practice, but not relied upon as the only opportunity to practice. Interviewers should be familiar with interview guides before working with the target audience. Ask the following questions during mock interviews (Krueger 1998a): Are the questions conversational?

6.2-25

In-Depth Interviews Do they sound good when said aloud? Are they easy to ask, or do you stumble over words? Do you feel comfortable asking the questions? Do informants understand the questions? Are you receiving the type of responses youd expect? 5. Select Appropriate Interview Sites, Obtain Permission, and Make Prior Visits. Under some circumstances, researchers may be unable to select interview sites. This is especially true when working with hard-toreach populations (e.g., those who are primarily found in bars or on the street) or when expert informants insist that interviews be conducted in their offices. When researchers can control the sites selected, they should ensure that the space is appropriate and minimize the possibility of unpleasant surprises, such as interruptions or a lack of privacy. If researchers are unsure of appropriate spaces, they can ask gatekeepers to identify sites where informants would feel comfortable being interviewed. 6. Accommodate Your Informants. Solid preparation means ensuring that informants feel comfortable during interviews. During planning, re-

Chapter 6.2

BOX 6.2-19: THE INTERVIEW CHECKLIST Name and contact information for informant Tape recorder Cassette tapes Batteries Pens Paper Incentives Copy of interview guide List to track money given to informants for transportation (some donors require informants signature)

BOX 6.2-20: IN-DEPTH INTERVIEW RULE: KNOW YOUR TAPE RECORDER

Do you know how to record, play, rewind, fast forward, pause, and eject cassette tapes? Does the recorder require batteries? If so, what type? Do you have spare batteries available? How long are cassette tapes? 60 minutes? 90 minutes? How many tapes are required for each interview? Will tapes need to be flipped during the interview? Is the voice activation option off on the recorder? At what speed are you recording? Do voices sound too slow or too fast? Where should the recorder be placed in order to pick up informants voices but not make them feel self-conscious about being taped?

6.2-26

PSI/AIDSMark Social Marketing Research Tool Kit 2004 searchers should consider informants lifestyle, access to transportation, schedule, and preferences. 7. Prepare a Checklist. Researchers should minimize the possibility of forgetting necessary items by making a checklist. All interviewers should have a copy and be responsible for taking items to interviews and returning them after fieldwork. 8. Ensure That All Equipment Works and Interviewers Are Familiar with Its Operation. Before entering the field, researchers should double-check to ascertain that tape recorders, televisions, VCRs, and other equipment work properly. All interviewers should be familiar with their operation to avoid interruption or confusion during interviews. 9. Arrive Early at Interview Site. It is important that interviewers arrive early at study sites to set up. They should contact officials in charge of the site (if necessary), load tape recorders, and make other necessary arrangements. Being prepared for informants arrival allows interviewers time to relax and make individuals feel welcome as they arrive.

CONDUCTING INTERVIEWS
Now that researchers have prepared the field, they are ready to conduct interviews. This process involves interviewing, taking notes, transcribing tapes, keeping records of interviews, and following up with gatekeepers and informants. Successful interviews are fun, informative, and leave interviewers and informants feeling good about their interaction. Refreshments Refreshments

Do not offer noisy snacks, In some cases it may be appropriate to offer inforsuch as chips or food in wrapmants refreshments. Offering food or drinks demonpers. The extra noise can instrates hospitality and can make informants feel terfere with audio recordings. more comfortable during discussion. If interviews take place at informal sites, such as restaurants, truck stops, or bars, researchers should offer informants a modest snack or refreshment. If interviews take place in more formal locations, such as informants offices, places of work, or clinics, offering refreshments is usually inappropriate.

6.2-27

In-Depth Interviews Introduction Rules and Ground

Chapter 6.2

FIGURE 6.2-6: FIELDWORK SHOULD BE FUN!

The discussion begins with a brief introduction of the study, the interviewer, and the informant. This component of the interview offers an opportunity to make informants feel welcome, thank them in advance for their participation, and lay the foundation for a successful interview. During the introduction, interviewers should cover ground rules to ensure that informants provide honest responses and know that their comments will be kept confidential. For more information, refer to the section entitled, Developing In-Depth Interview Guides. Encouraging Discussion There are several strategies for promoting discussion during interviews. The first strategy is to ask questions in a conversational manner. Interviewers should use the discussion guide as a tool for permitting dialogue with informants; it should not be read aloud as if it were a survey. Rather than conduct interviews as a series of questions and answers, interviewers should use the discussion as an opportunity to learn as much about the study topic as possible. The wording of questions should be direct, comfortable, and simple. Are the questions easy to ask, or does the interviewer stumble over words (Krueger 1998a)? Ensure that the language is appropriate for the target audience; it should not be too technical or contain acronyms and jargon. Colloquialisms should also be used sparingly.

6.2-28

PSI/AIDSMark Social Marketing Research Tool Kit 2004 When conducting discussions, interviewers should concentrate on one topic at a time and stick to the study objectives. They should avoid asking two questions at the same time or using long sentences or explanations that confuse informants. Interviewers should also keep in mind that informants have a tendency to stray from topics and talk about more than one issue at a time. It is the interviewers job to bring the dialogue back to the original question and address other topics at a more appropriate time during the discussion. Sometimes informants ask questions or give incorrect information during interviews. It is natural for researchers to want provide accurate information; however, this BOX 6.2-21: SPONTANEOUS PROBES should not be done during the interview. Interviewers should Ask for clarification assure informants that time will What do you mean by that? be reserved to discuss other is What about that? sues at the end of the interview I dont understand. (Zimmerman et al. 1996). What makes you feel that way? Encouraging dialogue with an informant is an art form. Interviewers must gauge informants for their ability to provide complete responses to questions and explain unclear statements. Probing is key to exploring questions more fully. Probes are short, follow-up questions that encourage informants to provide more detail. Im not sure I understand how youre using the word_____ Ask for more detail Could you be more specific? Could you elaborate on that point? Can you tell me more about that? What are some of your reasons for feeling as you do? Ask for examples Would you give an example? What other examples are there? Use cross-checks for consistency Earlier you said X, now youre saying Y.

Researchers prepare anticipated probes in advance of interviews and include them in interview guides to ensure that essential topics are covered during discussion. For example, if the question in the interview guide asks, In your opinion, which activities increase your personal risk for contracting HIV/AIDS? the researcher could anticipate using probes about multiple sex partners, inconsistent condom usage, no condom usage, drug use, selling sex, and so forth. It is up to interviewers to think of spontaneous probes and use them during discussion to elicit more detail from informants.

6.2-29

In-Depth Interviews

Chapter 6.2

Avoid the temptation to ask why as a probe. Asking why implies a rational answer and can put informants on the defensive. In real life, people make decisions based on impulse, habit, tradition, superstition, and other irrational processes. When asked why, informants provide answers that seem rational or appropriate, but may not reflect their true experience. Such responses can be unreliable (Krueger 1998a). Why questions can also make interviews appear to be interrogations rather than discussions. When interviewers are nervous, they tend to talk too much and give informants insufficient time to formulate BOX 6.2-22: IN-DEPTH INTERVIEW responses. They should fight the urge to RULE: NONVERBAL BEHAVIOR CAN ask a follow-up question or probe imINFLUENCE PARTICIPATION mediately after asking a question. Interviewers should practice the 5second rule: ask a question and count silently to himself or herself for 5 seconds. Informants usually respond within this amount of time. Although interviewers are familiar with study questions and have had an opportunity to practice, informants are hearing questions for the first time during discussions. Using the following techniques can promote discussion during interviews (Bernard 1998; adapted from Debus 1995, Zimmerman et al. 1996): Be aware of your physical distance from informants. You should not be too close or too far. Refrain from fidgeting, frequently shifting position, or making nervous gestures. Show that you are relaxed; your comfort will put informants at ease. Maintain appropriate eye contact to communicate interest, sincerity, confidence, and engagement. Use nonverbal encouragement by nodding, smiling, uttering uh-huh or mmmm. Smile and laugh when appropriate. Control the volume of your voice. Dont whisper or shout.

(Adapted from Andrew Fletcher Consulting 2001; Demonstrating sophisticated Bernard 1998; Spradley 1979) navet (incomplete understanding): The interviewer signals the need for more detailed information while not appearing to fake ignorance. Oh, I dont know about that. Can you tell me more about it? Employing the mirroring technique: The interviewer repeats what the informant has just said in order to solicit more detail. So you think its important.

6.2-30

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Repeating as a question: Interviewers sometimes find that repeating an informants comment is effective at obtaining more detail. Its good? Reweaving: Interviewers weave information provided at an earlier stage in the interview into the current discussion. Reweaving reassures informants that the interviewer is actively listening to their comments. Likewise, informants sometimes make critical comments at a premature stage in the discussion. The interviewer may let the comment drop until a more appropriate point and then ask the informant to expand TABLE 6.2-4: INTERVIEWER DOS AND DONTS upon earlier comDo Dont Interrupt informants ments. Rebecca, Give informants sufficient time to speak earlier you said X. I Accept the first response as think thats an impor- Practice the 5-second rule the complete response tant point. Can you Adopt a curious, sympathetic Express shock or surprise at a expand upon that for attitude response me? Use probes and follow-up ques- Accept the response I dont Summarizing infor- tions to solicit more detailed re- know sponses mant responses: Ask for clarification of colloquial Give the impression that reSome interviewers or unfamiliar terms sponses are right or wrong find it useful to sum- Ask for clarification if you dont Embarrass informants by inmarize informants understand a response sisting they respond to questions that make them feel uncontributions to keep comfortable the discussion on Use nonverbal cues (e.g., nod- Ask why topic and solicit more ding head) and silence as contributions. So far prompts youve told me that X, Y, and Z contribute to your drug use. What else might contribute? Information Bias Information bias is common among informants and can take several forms. Informants might withhold revealing information about themselves or conform to the opinion of others in order to fit the norm. Socially appropriate responses are common especially when informants do not want to deviate from their peers or cultural

6.2-31

In-Depth Interviews

Chapter 6.2

expectations (The Bible tells us that youth shouldnt be sexually active). Information bias can result when informants have difficulty remembering the details surrounding an event. Recall bias is the result of memory failure or informants misremembering events and reporting what they think happened rather than what actually happened (Denzin and Lincoln 2000; Bernard 1998; Spradley 1979; Weller 1979). The interview itself may affect study results. Contextual effects, such as the characteristics of the interviewer, informants comfort level with the interview format, or the study setting, may influence informant responses (Denzin and Lincoln 2000; Bernard 1998; Weller 1979). During formative research and pretesting, courtesy bias is common. Informants often share responses they think interviewers would like to hear (I have seen interventions in the neighborhoods and find them effective). During pretesting, informants may be reluctant to hurt researchers feelings and avoid expressing criticism of campaign ideas or materials (The ad seems fine as it is). Informants can also introduce bias into study results by providing what they consider to be impressive responses. Some individuals feel compelled to demonstrate competence in a topic area (Everyone knows that the ABCs of prevention are abstinence, being faithful, and condom use). Others participate in bravado, hoping the interviewer will be impressed with their statements (Guys like me, we like to have several girlfriends at the same time.). Information lost during translation and transcription can compromise the spirit and content of discussion. Transcribers should confer as much as possible with interviewers to ensure the quality of study transcripts. If colloquial phrases and context-specific vocabulary do not translate well to English (or another language used for reporting) keep original terms and provide definitions in transcripts and reports. There is also the possibility of researcher or observer bias where researchers misinterpret findings or have difficulty separating their personal views from informants contributions (Bernard 1998). Controlling such bias requires a systematic approach to data coding and communication between researchers to remain aware of personal biases.

6.2-32

PSI/AIDSMark Social Marketing Research Tool Kit 2004 The potential for information bias means that researchers must be alert to its presence during data analysis. They must also be discerning when reporting interview results and search for the intent behind informants responses. Verbal Informants Sometimes informants tell interviewers too much and provide longwinded responses to study questions. New interviewers are often reluctant to interrupt verbal informants, being afraid that doing so would be rude or compromise BOX 6.2-23: IN-DEPTH INTERVIEW the interviewing technique. RULE: HOW TO TAKE GOOD NOTES Learning to interrupt gracefully takes time and will come with Use the tape recorder to capture most diaexperience (Bernard 1995). logue, but note key words and important points to trigger your memory while reviewing Note-Taking recordings. Researchers should never unNote the time discussion starts and stops. derestimate the importance of Include a description of the informant (name, good note-taking; however, inage, occupation, education, etc.). terviewers should refrain from Use shorthand or paraphrase comments. taking too many notes during Use underlining or bold letters to indicate ininterviews. Writing may interterviewer questions. fere with paying attention to inTake additional notes immediately following formants and it can send mixed interview to avoid forgetting important informessages to informants, sugmation. gesting that certain comments Note informants reactions, body language, are more important than others and your impressions. (Bernard 1998). Extensive Discuss common trends and impressions of note-taking may also make ininterviews with other interviewers formants feel self-conscious and distracted. Once a discussion has ended and interviewers are alone, they should write their impressions of the interview, informants nonverbal behavior, and their rapport with the informants. Interviewers should also review tape recordings as soon as possible to refresh their memory and provide additional detail to notes.

6.2-33

In-Depth Interviews

Chapter 6.2

TABLE 6.2-5: TROUBLESHOOTING: PROBLEMS AND SOLUTIONS Problem Informants respond as if the discussion is a question-answer session. Informants comments appear rehearsed, or they only provide responses that are socially and politically appropriate. Informants appear uncomfortable answering certain questions. Solution The interviewer should repeat the ground rules and encourage informants to speak freely and elaborate responses. The interviewer should assure informants that their comments are confidential and they should express their opinions openly.

The interviewer should practice the 5-second rule. If the informant still appears uncomfortable, the interviewer can move to another question and suggest that more sensitive questions be addressed later or when the informant feels like discussing them. An informant asks an unrelated The interviewer should reserve time at the end question during the discussion. of the discussion to answer additional questions.

AFTER THE IN-DEPTH INTERVIEW

Transcribing Transcribing can be time-intensive, especially if discussions must be translated as well as typed. For every 1 hour of discussion, researchers should allow 6 to 8 hours for According to Krueger and King (1998), transcription. Transcribers are often unin1 hour of discussion requires 6 to 8 volved in data collection so they should hours to transcribe, assuming that the work closely with interviewers to ensure typist is experienced, familiar with the that transcriptions reflect the content and vocabulary, uses proper equipment, nature of discussions. and has no interruptions. A 1-hour FGD yields approximately 20 When transcribing in-depth interviews: to 25 pages of transcript (Knodel Use high-quality playback equipment. 1993). Avoid tape players with small speakers, and consider using earphones. Minimize distractions.

6.2-34

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Follow the interview guide, using questions as subheadings to organize transcriptions. Use bold print or underlining to identify interviewers statements. Identify informants by using codes or initials next to their comments. Double space between speaker comments. Dont correct informants grammar. Few people talk in complete sentences. If additional words are needed to complete a thought place these words in parentheses ( ). If words are unintelligible, use an ellipsis [...] to indicate that words are missing from the transcript. Explain colloquial phrases. Use parentheses ( ) to note informant reactions and nonverbal cues. Use brackets [ ] to indicate researchers impressions.

BOX 6.2-24: ZAMBIA STUDY ON HIV/AIDS RISK PERCEPTION AMONG YOUTH Below is an example of interview transcripts with identifiers for researcher and informant. The researchers questions are labeled with an R while the informants comments are labeled with an I. (Source: Transcripts from PSI/Zambia) R. What is HIV? What is AIDS? I. If I am to say what I know, I usually hear other people referring to HIV or AIDS. I however do not know what these really are. R. What is it that you have heard other people say? I. They say that HIV and AIDS are sexually transmitted diseases. R. Is there a difference between the two? I. I do not know if there is a difference or not. R. How could one cure these diseases? I. AIDS is not curable. This is particularly so for those who keep on changing sexual partners. Having multiple partners will only worsen the infection and that person will die more quickly.

Proofread transcriptions and ensure that words are not missing (e.g., the elimination of not from a phrase can mean the difference between an affirmative or negative comment). Refer to interviewers notes or ask for clarification when necessary. Verify transcripts with gatekeepers to ensure that language and context are correct.

6.2-35

In-Depth Interviews Maintaining Records of Interviews

Chapter 6.2

Including key information in interview records can expedite report writing and planning for additional studBOX 6.2-25: INFORMATION TO ies. Researchers should attach records INCLUDE IN INTERVIEW RECORDS to transcripts or include them in field notes for future reference. Follow-up Costs associated with study Gatekeeper names and contact information Locations where interviews took place Description of informants (e.g., age, sex) Time required to conduct each interview Major themes Informant concerns Unexpected findings Potential bias Researcher impressions Challenges encountered in the field Suggestions for improvement

Timely follow-up with informants and gatekeepers is important for maintaining a positive relationship. Small gestures, such as sending thank-you notes to individuals who assist with logistics, are important. If team members take photos during fieldwork, they should send copies to informants or gatekeepers and secure permission to include them in reports if necessary. Some individuals may also appreciate a final copy of the study report. Such efforts may facilitate access to the same target audience in the future. Data Analysis and Interpretation

Analysis of interview data is very different than the analysis used for surveys. Although survey data are analyzed with software that generates specific outputs (e.g., frequencies, significance levels), qualitative analysis is an iterative process that is Using a team approach for data time consuming and reliant upon researchers analysis improves study reliabilinterpretation of study data. ity, especially when researchers contribute to different analysis Analysis of interviews is a process that occurs steps and then compare their throughout the study and it should follow the results. set objectives. It is essential that interviewers discuss common trends and their impressions (Knodel 1993) of informants comments during data collection. Researchers continue analyzing data by reviewing tape recordings, notes, and completed transcripts.

6.2-36

PSI/AIDSMark Social Marketing Research Tool Kit 2004 For some researchers, analysis of qualitative data comes easily. Others become BOX 6.2-26: THOROUGH ANALYSIS GOES BEYOND WORDS distracted by trivial issues, forget the research objectives, or have difficulty separating Researchers sometimes assume their personal views from informant contributhat responses only consist of the tions. Although education helps, it is not the words informants use. The actions solution. Openness to new ideas, approaches, and behaviors of informants can and concepts is essential (Krueger 1998b). often speak louder than words. Analysis should observe all factors The process is called contextual analysis, in the communication: body lanwhich requires thorough reading and coding guage, gestures, and tones of of all study transcripts. This type of analysis is voice. like detective work: it requires researchers to (Krueger 1998b) search for trends and patterns (clues) that reappear among different interviews (Krueger 1998b). The contextual analysis process also requires researchers to search for explaFIGURE 6.2-7: EXAMPLE OF TRANSCRIPT CODED IN nations of behavETHNOGRAPH (SOURCE: LONGFIELD ET AL. 2002) ior and supporting quotes that will be integrated R. OkayWhat are the risks in these 582 kinds of relationships? 583 into the final report. I. To the man or the lady? 585 586 Although formative research R. Both. #-ECONRISK studies require contextual analysis, pretesting efforts do I. For the man, the risks areyou can 588 # have economic risks whereby he 589 | not. It is usually sufficient to might end up spending most of his 590 | consolidate findings from %-FAMILY pretesting interviews into a money on this lady.It could be so- 591 # % cial risks, where the wife could come 592 | simple matrix or list. | to know about it.Or the girls par- 593 Do researchers need softents or relatives from both sides 594 | ware to conduct analysis of $-DISEASE in-depth interviews? Most could find out. In terms of health, the 595 $ % man could contract sexual infectious 596 | PSI programs conduct qualiranging from STIs to HIV/AIDS 597 $ tative analysis by hand, us#-PREGNANCY ing grids or matrices to To the lady, she could become 598 # summarize themes and or%-VIOLENCE ganize findings; however pregnant. She could also be beaten 599 # % software programs are availby the man's wife. These are I think 600 | the major issues/risks involved in this 601 | able that save time and askind of a relationship 602 % sist with data interpretation.

6.2-37

In-Depth Interviews

Chapter 6.2

When working with a limited number of transcripts (fewer than 10), using software can prove more cumbersome than helpful. However, for larger studies, software packages can make data analysis more efficient, thorough, organBOX 6.2-27: CODING TRANSCRIPTS BY HAND ized, and less prone to bias. Among the most common software packages are: Assemble all materials: interview guide, tranEthnograph scripts, tapes, demographic information about informants, and field notes. Nu*Dist Read field notes at one sitting to refresh your Tally memory about the interview context and tone Atlas TI of discussion. Read each transcript completely. DT Search. Develop an initial set of codes corresponding Figure 6.2-7 depicts data coded in to major themes in interviews. Ethnograph. The researchers quesContinue to refine codes throughout analysis tion is labeled R, and the infor- create new codes for new themes and more mants comments are labeled with specific codes for complex themes. an I. Numbers indicate lines of text Mark sections that relate to study themes and the informants comments are Identify quotes to use in the report. coded with symbols #, %, and $) to Highlight words or phrases that represent indicate different themes. classification categories. Create an overview grid (matrix) to categorize Researchers must remember that informant comments and themes and simplify qualitative software packages do not data classification. generate definitive answers; they Examine one question at a time. After reading simply assist with data organization, all responses to a single question, prepare a enabling the team to better interpret summary statement that describes patterns study results. across interviews. Computer packages cannot identify: (AIDSCAP 1994; Knodel 1993) Important words or themes Manner in which something was said Informants reactions to statements. Additionally, researchers should note that qualitative data analysis is subjective: different researchers may lend different interpretations to study findings. However, approaching analysis systematically, com-

6.2-38

PSI/AIDSMark Social Marketing Research Tool Kit 2004 municating with other interviewers, and remaining aware of personal biases can minimize subjectivity. BOX 6.2-28: FACTORS TO CONSIDER DURING ANALYSIS

REPORT WRITING

In general, qualitative reports should be rich in detail and presented in narrative form. Researchers should cover recurring themes across interviews, patterns in target audiences attitudes and behavior, informants concerns, conflicting information, and unexpected findings. Format of Report In-depth interview reports should include the following: Cover Page Abstract (or Summary) Executive

Words used by informantsWhich words are new or unique to the target audience? Context of responsesHow was a response triggered? Internal consistencyDo informants contradict themselves or change their opinions during discussion? Frequency of commentsHow often did a concept or topic arise? Extensiveness of commentsHow many informants spoke about a particular issue? Intensity of commentsDid informants speak about certain issues with intensity, passion, or depth of feeling? Specificity of responsesDid informants provide specific examples from their experiences or did they provide vague and impersonal comments? Issues left unsaidWas a topic overlooked by informants? Perhaps it wasnt as relevant as researchers suspected. Big ideasWhich trends or ideas cut across all interviews? (Krueger 1998b)

Introduction (including a brief literature review); Methods Results (includes data interpretation and salient quotations) Conclusions (including study limitations) Programmatic Implications Appendices.

Good reports tell a story unknown to most readers and provoke thought about behavior and health-related issues. See the appended report Misconceptions, Folk Beliefs, and Denial among Zambian Youth as an example. The Results and Conclusions sections are the

6.2-39

In-Depth Interviews

Chapter 6.2

heart of a qualitative report. They should be clear, succinct, and entertaining for the reader. FIGURE 6.2-8: METHODS OF PRESENTING INTERVIEW DATA The following examples are taken from PSI studies and illustrate methods for presenting different types of interview data. Recurring Themes (Longfield et al. 2002) Most informants felt that men increase their social status among peers when they are seen with younger, attractive women. Consequently, they experience a sense of increased pride and continue pursuing cross-generational relationships. If youre seen with a young girl, your friends say, This guy, you know, he has such a good babe. Its for identification. Its for status (male supervisor, 38 years, Mombasa). Consensus Among Informants (Richter and Bobin 2002) Both men and women said that condoms are less likely to be used in affectionate relationships than in those that are primarily for sex. The word trust is commonly used when describing affectionate relationships, and the level of trust between partners is a major factor determining condom usage or non-usage. Informants Concerns (Sachingongu 2001) When asked what issues respondents talked about with their girlfriends, marriage and pregnancy were most often cited. Often, the girls initiated such discussions, in particular those concerning their boyfriends intentions with respect to marriage. Sometimes (my girlfriend) asks me if I would marry her. She also asks me what I would do if she becomes pregnant. I told her that I would marry her but not now (male, 19 years, Grade 7). Divergent Opinions (Longfield, Cramer, Sachingongu 2003) There was awareness among informants that HIV/AIDS is a very dangerous disease and most understood that it is incurable. However, a few informants argued that if infected individuals are treated early or with modern medicine, they can be cured of HIV/AIDS. Conflicting Information (Longfield, Cramer, and Sachnigongu 2003) Many informants stated that as long as they remain sexually faithful to their partners, they can avoid infection. As a result, informants viewed monogamous sexual relationships as risk free even though the majority of their relationships lasted no longer than 6 months. Unexpected Findings (Waithaka 2000) Certain myths and misconceptions about HIV/AIDS, STDs, and condoms were evident in the study. Some informants doubted the efficacy of condoms while in Meru, one man was quoted as saying that he swallows a tablet before sex to avoid getting infected with STDs. 6.2-40

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Common Pitfalls When Writing Reports When writing reports, it is important that researchers remain faithful to informant comments and prevent themselves from imposing assumptions on study data. They should keep in mind that interview settings and dynamics of discussions influence informants comments. Do not quantify findings from interviews. Unlike quantitative reports, qualitative reports should not present numbers or tallies of responses. Instead, descriptors such as most, some, a few, and several should be used. If comments are only relevant to one or two informants in a study, they are probably not worth reporting. Avoid making definitive statements about how informants think or feel. BOX 6.2-29: GUIDELINES FOR PREPARING Researchers can only report what REPORTS ON IN-DEPTH INTERVIEWS informants said they think or feel. They must keep in mind that inforBe objective and dont make assumptions. mants may be influenced by the Avoid quantifying qualitative findings. study setting and may not want to Use words like some, a few, most, sevshare their true thoughts or feeleral, and so forth. ings with the interviewer. Avoid definitive arguments. Reports should not present inforReport what informants said. mants comments at face value (litComplement text with salient quotations. erally), but should look beneath the Add quotations only when they enlighten the surface (Debus 1995). Another preceding statement. important intent of qualitative reProvide responses for each type of informant ports is to synthesize and concepwhen possible (males, females, etc.). tualize findings from different interIdentify the type of respondent attributed to views, not simply repeat what was quotes (Female, 15 years, Lusaka). said. Dont use too many quotations. Researchers should include salient quotations from informants that illustrate study findings; however, quotations should be used sparingly so they do not lose their effectiveness. If too many quotations are used, the report will read like a transcript rather than a narrative. Qualitative research reports are generally longer than those produced for quantitative studies. However, it is important to restrict reports for interviews to a reasonable length in order to maintain reader

6.2-41

In-Depth Interviews

Chapter 6.2

interest. Because researchers become so involved in data analysis and close to the topic at hand, there is a temptation to include everything in reports, making it difficult to limit their length. It is often useful to have colleagues read and edit researchers reports, allowing them to trim nonessential information and keep only that essential to telling a comprehensive story.

MODIFIED INTERVIEWS
Particular field situations may call for modified interviews that employ alternative strategies for collecting data from informants. Appropriate circumstances under which interviews should be modified are when researchers work in countries where individuals are distrustful or skeptical of the researchers intentions, when it is inappropriate for researchers to contact members of the target audience, or when researchers wish to explore an unknown behavior or cultural domain. If you are interested in using any of the following methodologies, please contact your Washington-based researcher or regional research coordinator. Paired Interviews Paired interviews are also called friendship paired interviews, dyads (two people), or triads (three people). These interviews combine the dynamics of a group discussion with the more personal atmosphere of an individual interview. When two friends or colleagues are interviewed together, they tend to support each other. They also challenge the accuracy of each others accounts and play off of one anothers comments (NOP 2003). This process (like FGDs) allows researchers to gain insight into the true meanings behind informants comments. In addition to collecting formative data, paired interviews are an effective method for pretesting materials. When it is too expensive or difficult to gather several participants for FGDs, paired interviews can be a cost- and timeefficient alternative. Peer Ethnographic Research This research method uses peers to conduct in-depth and unstructured interviews with individuals from their social networks. Peer researchers become key informants by virtue of their status within a

6.2-42

PSI/AIDSMark Social Marketing Research Tool Kit 2004 community, their local knowledge, and their ability to contact other members of the community (Price 2002). Rather than conducting several interviews with different informants, peer researchers conduct a series of interviews with the same small sample of informants. The advantage of the peer ethnographic method is that it allows researchers to conduct research with hardto-reach groups (e.g., MSM and CSWs) that are reluctant to participate in interviews with outsiders. A basic principle of the approach is that peers have established relationships of trust with the individuals they interview. Peer interviews are intended to uncover the meanings associated with behavior and reveal informants perception of others within their social network (Price 2002). Peer researchers are trained to conduct fieldwork and debrief PSI staff members. They also participate in follow-up workshops to analyze data and identify key issues that emerge from interviews. For more information on peer ethnographic research see Richter and Bobin, 2003. Pile Sorts Pile sorts are a type of classification study in which interviewers ask informants to sort cards that contain different terms into categories. Free-listing exercises precede pile sorts; target audience members brainstorm related words, concepts, or statements along a single study theme (i.e., cultural domain). Local terms solicited during free-listing exercises are entered into a program such as Anthropac or SPSS to reveal the most common elements in a domain. Terms are then transferred to 3 x 5 index cards. During pile sorts, informants are asked to sort cards, each containing the name of an item, into piles so that items in a pile are more similar to each other than they are to items in separate piles (Weller and Romney 1988). Cards can be sorted into a predetermined number of categories or informants can determine an appropriate number (Bernard 1998, 1995). Once informants have created their piles, interviewers ask them to explain the groupings. The analysis of pile sort data is called cluster analysis. As such, it uncovers informants own system of categorizing information and reveals the level of agreement that exists between informants perception of the same domain (Bernard 1998; Borgatti 1989; Spradley

6.2-43

In-Depth Interviews

Chapter 6.2

1979; Weller 1979). Pile sort exercises are useful for examining a variety of health topics, such as different types of sexual partners, classifications of illness, and methods of contraception. This methodology is recommended for larger samples (of at least 30 people) or for studies that contain several local terms (Weller 1979).

CHALLENGES
The process of planning, collecting, transcribing, analyzing, and reporting in-depth interviews poses many challenges. Listed below are some common challenges researchers face when conducting interviews: Logistic Confusion Insufficient communication among researchers, recruiters, interviewers, and other individuals responsible for interview logistics can result in compromised results. Common mistakes are recruiting too many informants (or not enough), double-booking interviews (requiring one informant to wait or reschedule), inadequate transportation for informants, informants who arrive late, sites that lack privacy, and forgotten equipment. To limit confusion, create checklists for all individuals involved that cover their responsibilities and your expectations. Follow up with meetings, visits, and phone calls to remind individuals of important dates, supply needs, and deadlines. Time and Resource Constraints Too few resources and too little time can compromise study results and make study implementation a frustrating process. Supervisors often ask researchers to stray from study objectives, collect additional data from informants, limit the required number of interviews, or disseminate results under impractical deadlines. Researchers should work with their supervisors to design the study, outline objectives, and consider the consequences of changing study protocol. Insufficient Time for Data Analysis and Report Writing Researchers should not underestimate the time required for thorough analysis and report writing. Coding data can be a tedious process. On average it takes researchers at PSI/Washington 2 hours to code transcripts that are 8 pages long. Additional time is required when working as a group: Researchers require time to discuss

6.2-44

PSI/AIDSMark Social Marketing Research Tool Kit 2004 codes, apply consistent codes to transcripts, and interpret results in the same manner. After analysis, researchers should be allowed at least 2 full weeks to write a preliminary report and additional time to incorporate reviewers edits. Generalizability Findings from interviews are not intended to be generalized to larger populations. The goal of qualitative research is to examine a topic in depth and provide explanations for behavior. Although interviews can indicate a range of views and opinions, they do not reveal the distribution of such views among a larger audience (Dawson, Manderson, and Tallo 1993). Other research methods (such as surveys) sacrifice depth for breadth and are used to make generalizations about groups. What interviews offer is transferability: Findings from one study may suggest trends in other environments and contexts. It is up to individuals reviewing the research to consider the methods, procedures, and audience included in a study and to decide upon the applicability of findings to other audiences (Krueger 1998b). Increasing the number of interviews to improve coverage and representativeness achieves neither and rarely improves the utility of data. Conducting too many interviews suggests a misunderstanding of the purpose of qualitative research. It also results in too much data, overwhelming researchers who lack time to sufficiently analyze results (Shedlin and Schreiber 1995).

6.2-45

In-Depth Interviews

Chapter 6.2

APPENDIX 6.2.1
INTERVIEW GUIDES
In-depth interview guide from Zimbabwe for VCT clients (English) INDEPTH INTERVIEW GUIDE FOR NEW START CENTRES Client code No.
Introduction: PSI wants to understand the reasons why people come for Voluntary Counseling and Testing at New Start Centres so that the organization can improve its voluntary counseling and testing services in order to attract more clients I would appreciate if you can spare some time to answer the questions below. Your responses will not be identified with you and the information that you will give me will be held in strictest confidence. Thank you for your cooperation.

Perceptions on HIV/AIDS severity 1. How severe do you think the problem of HIV/AIDS is in Zimbabwe and why? _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ ____________ New start 2. When did you know about New Start for the first time? _____________________________________________________________________ _____________________________________________________________________ ______ 3. How did you know about New Start?

_____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _________ 4. Do you think HIV testing is helping in mitigating the epidemic and how? _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _________

6.2-46

PSI/AIDSMark Social Marketing Research Tool Kit 2004

5. What do you think are the possible solutions to abating the epidemic and why? _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ ____________ Decision to take test 5. Why did you decide to take a test today? Probe on costs ($50) and promotions? _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ ____________ 6 When did you actually start to think about taking an HIV test? _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _________ 7. Why is it that you did not take the test earlier? _____________________________________________________________________ ___ _____________________________________________________________________ ___ _____________________________________________________________________ ___ _____________________________________________________________________ ___ 8. Is there anybody whom you told that you will be taking an HIV test today and if yes how are you related to that person? _____________________________________________________________________ _____________________________________________________________________ ______ Plan 9. What do you plan to do now that you know your HIV status? _________________________________________________________________ _________________________________________________________________ _________________________________________________________________ _________________________________________________________________ ________________

6.2-47

In-Depth Interviews

Chapter 6.2

Any other comments regarding New Start Centres and voluntary counseling and testing _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _______________ Thank you

10.

6.2-48

PSI/AIDSMark Social Marketing Research Tool Kit 2004

In-depth interview guide for men from Kenya cross-generational relationship study (English) Kenya Cross-Generational Relationship Study In-depth Interview Guide for Men

Good morning/afternoon/evening.Mr.__________ Thanks for accepting to meet with me for this discussion today. We shall be discussing how men and women socialize in Kenya today, given the prevailing economic, social and health conditions. 1. Why do older men have sex with women ten or more years younger than them?

2. What do they look for in the young women? (Details!) attributes education y traits n occupatio personalit level of physical

3. Where do the older men meet the young women? (Details!) Venue Time

4. When they meet at venue __________ at __________ time, how do they establish contact with the young woman? ( a venue at a time)

6.2-49

In-Depth Interviews what is said? interest in this relationship expressed? Give a description of the dialogue that would take place Repeat for different venues and times given in q3

Chapter 6.2 Exactly

How is

5. Once they have established the relationship, what happens? What takes place? (Details!)

6. Where do they have sex? (Details!) Place Time

7. Are there any risks in such a relationship?

8. How do men resolve the issue of(q6)?

9. How long does this kind of a relationship last?

10. Describe the kind of man that engages in this relationship? (Details!) status on 11. In your opinion, are these kinds of relationships common today? 6.2-50 Language Attire Age etc Occupati Social

PSI/AIDSMark Social Marketing Research Tool Kit 2004

12. Is this a problem? Why or why not? What can or should be done about it?

13. Have you ever engaged in such a relationship? (Details!) 14. Do the older men think of the HIV status of the young girls?

15. (If yes in Q14) What do they do about it?

6.2-51

In-Depth Interviews

Chapter 6.2

In-depth interview guide on Anemia & Vitamins (English)

Anemia & Vitamins In-Depth Interview Guides

Anemia In-depth Interview Guide
KAP Anemia, Part I 1 - Demedicalized Questions
1

These questions ask about anemia - tiredness for no reason that wont go away, or weakness - without using its medical name. The things people are doing now - coping strategies - are our competition. In Zambia, coping strategies were drinking black tea,

6.2-52

PSI/AIDSMark Social Marketing Research Tool Kit 2004

1. Preparatory question 2 a) Tell me about times this year when you were tired. What causes tiredness? 2. Probing for experience with anemia 3 a) Have you every felt tired for a long time, or do you know anybody whos felt tired for a long time, for example, for more than two days? b) Does it seem like tiredness wont go away, or weakness? c) Probe (e.g. tell me more about it). 3. Beliefs about causes 4 a) What do you think causes this lasting tiredness? 4. Coping strategies used a) What do people do to about this tiredness? b) What other things can people do about it? (probing further) 5. Local Lexicon a) What are other expressions that people use to say they feel this tiredness? (e.g. lazy, faint)

getting exercise, doing some work, sleeping and time. Notable by its absence was eating nourishing foods, much less taking vitamins.
2

This question will elicit responses about tiredness due to hard work, long hours etc - it is just preparation for the next question.

In Zambia, ALL focus group participants had, and spoke at length about it (e.g. Sometimes I wake up in the morning and Im so weak, I just dont want to do anything, not even wash a dish.
4

3 In countries with high anemia rates, you can be certain that the majority of your respondents will have had this experience.

Important to be open to other systems of thought. In Pakistan, women spoke about how a tense mind causes tiredness, and therefore it is important to have a relaxed mind, to have music and a beautiful atmosphere at home. This can help us in advertising (metamessages), and also tells us, in the consumers mind, what the competitors are. To prevent anemia, relaxing will only help so much!

6.2-53

In-Depth Interviews 6. Prevention a) What can people do to prevent feeling tired or weak? b) In general, what can people do to increase their energy? 7. Current health-care practices

Chapter 6.2

a) If a person felt tired or weak, who would he or she normally ask for advice? b) Would the person go to a traditional healer (use local term), a pharmacist, a clinic, a doctor? c) If the person began to feel very ill, where would he/she go? d) What are the advantages of the traditional healer? the pharmacist? the clinic? the doctor? e) What are the disadvantages? 8. Traditional Healers Products Expand questions as appropriate for traditional healing practices for particular country. a) What kinds of medicines do traditional healers provide? (Probe for details.) b) How does a person take them (how often, how much/many, what time of day)? KAP Anemia, Part II - Medicalized Questions 9. Awareness of and knowledge about anemia, (using medical terminology) a) Have you ever heard of anemia? b) What have you heard about it? 10. Symptoms a) Do you have any idea of how to tell if someone has it? What are the symptoms? 6.2-54

PSI/AIDSMark Social Marketing Research Tool Kit 2004

11. Causes of Anemia a) Do you have any idea how people get anemia? b) Is it contagious? 12. Treatments for anemia a) Can anything be done to cure anemia? b) Can it be prevented? 13. Personal risk a) Can children get anemia? Older people? Teenage girls? Young women? b) Could any woman, like us in this room, get it? c) Does everyone get it sooner or later? (like malaria) d) Does everyone get it during their life? or just some people? 14. Lexicon a) Have you heard of any other names for anemia?

6.2-55

In-Depth Interviews

Chapter 6.2

Vitamins
In-depth Interview Guide
1. Personal experience with vitamins5

a) Have you heard of vitamins? b) What are they for? Why do people take them? c) Have you ever used vitamins, or known people whove used vitamins? What was your experience with them, or what did you hear about their experience.

2. Vitamin category knowledge

a) What have you heard that vitamins are made from? (e.g. herbs, fruits, chemicals, medicines etc.) b) Are they natural things, like food, or chemical things, like medicine. d) Tell me something about the different kinds of vitamins e) Have you heard of multivitamins? f) What are the differences between multivitamins and other vitamins? 3. Consumption Practices a) When do people take vitamins? b) How many do they take at one time? c) How often do people take vitamins? d) For how many days or weeks do people take vitamins?

Bolivia research found that 50% of the lowest-income women had never heard of vitamins. Before asking a whole questionnaire about vitamins, it will necessary to screen out women whove never heard of them, and include only those respondents who have. 6.2-56

PSI/AIDSMark Social Marketing Research Tool Kit 2004 e) Are vitamins taken with food or alone? 4. Purchase experience and practices a) Where do people obtain vitamins? b) How do people decide which vitamin to use? c) Do they seek advice from anyone? d) Do they seek advice of a pharmacist, doctor or nurse? e) Do people find it confusing or easy to choose a type of vitamins? f) Why would someone choose a certain type of vitamins? g) How many tablets do people usually purchase at one time? h) How much do vitamins cost? 5. Package a) What does a package of vitamins look like? b) How many tablets are in a package? c) Is this too many or too few? d) Do they come in a bottle or in a blister pack? e) Which is better? 6. Organoleptic Characteristics a) Do people mind the size of vitamin tablets? b) What size would they prefer? c) Do people mind the taste of vitamins? d) What flavor would they prefer? e) Do people mind the color of vitamins?

6.2-57

In-Depth Interviews f) What color would they prefer? 7. Advertising

Chapter 6.2

a) Have you ever noticed a radio or television advertisement for vitamins? b) What did the advertisement say? c) Do you remember which brands you heard or saw advertised? d) Who were the ads directed to? 7. Fears a) In your opinion, are vitamins safe? b) Can they be taken everyday? c) Are they safe for children? d) Are they safe for women? e) Are they safe during pregnancy? f) Can they cause any problems? 8. Side effects a) Do you know anyone who has had a bad experience with vitamins? b) What type of problems did they experience?

6.2-58

PSI/AIDSMark Social Marketing Research Tool Kit 2004

Price Sensitivity In-Depth Interview Guide

1. Reference-Point Expenditures a) How many people are in a typical family? b) How much does a typical family spend on food weekly? c) In the past three months how much do you think the typical family spent on cosmetics/beauty products? d) How much on doctors? e) On medicines? 2. Purchasing Practices - Self Reported a) Do you know people who have taken vitamins? b) Where did they obtain them? c) How much did they cost? d) How many did they buy? e) Do you think they would buy them again? 3. Suggested Retail Price Range This is an example of a vitamin we are planning to introduce for women. a) What would be a good price for this box of vitamins? b) What would be the lowest and highest prices that would be good? c) Whats wrong with a price lower than xx ? A price higher than xx ?

6.2-59

In-Depth Interviews

Chapter 6.2

In-depth interview guide from Gates project in Cameroon on adolescent reproductive health (French)

Projet de sant Reproductive pour les Adolescents au Cameroun Guide de discussion pour entretiens individuels

Objectifs de ltude : Lobjectif de ces entretiens est dvaluer les ractions la campagne 100% jeune qui est actuellement diffuse la tlvision et la radio. Les entretiens se feront auprs de jeunes faisant partie du groupe cible de la campagne, cest dire garons et filles de 15 19 ans et jeunes hommes de 20 24 ans. Points aborder : 1. 2. 3. 4. La description des spots TV et radio dont la personne se souvient spontanment Les ractions gnrales de la personne lors de lexposition aux spots La description faite des personnages et le jugement port sur ces personnages Lidentification ou non de la personne avec les personnages/situations de la campagne 5. Les ractions quand aux instigateurs probables de la campagne et son bien-fond

Prslection : Il est conseill de mener entre 10 et 15 entretiens pour chaque groupe cible. Une prslection sera ncessaire qui se fera travers un questionnaire court : 1) Quel ge avez-vous ? (entre 15 et 20 ans, poursuivre linterview sinon arrter) 2) Au cours des 2 derniers mois, avez-vous vu la tlvision ou entendu la radio des annonces sur le thme du SIDA, des grossesses indsires ou de lutilisation du prservatif par les jeunes ? (si oui poursuivre, sinon arrter) 3) Nous aimerions vous poser quelques questions sur les annonces que vous avez vues, tes vous dispos participer un entretien ? 4) Commenons par quelques questions de base : - Quel est votre niveau de scolarit ? - Etes-vous actuellement scolaris ? - Est-ce que vous avez un emploi ? - Etes-vous mari ? (si oui arrter, sinon continuer)

6.2-60

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Quelle est votre religion ? Possdez-vous un poste de tlvision ? Un poste de radio ?

Entretien : Lenquteur devra laisser libre cours la conversation et permettre la personne consulte dapprofondir ses penses le plus possible sans suggrer de rponses. Le but de lentretien nest pas seulement dobtenir des ractions sur la campagne, mais galement dclaircir les raisons de ces ractions au fil de la conversation. Il est important pour lenquteur, dans la mesure du possible, dexplorer les ractions de la personne consulte trois niveaux : motionnel, cognitif et comportemental. Autrement dit : Est-il possible dexpliquer les ractions/commentaire par un aspect motionnel (peur, dgot, colre, ennui, honte, etc.) Sont-elles motives par des convictions/perceptions particulires (religion, morale, ducation) Sont-elles relies des comportements/ habitudes particulires (usage des prservatifs, partenaires multiples, etc.)

Tout ceci en influenant le moins possible la conversation !

Suggestions pour lenquteur :

Les questions suivantes peuvent tre utilises. Dautres questions sont galement suggres (en italiques) afin dexplorer les rponses plus en dtail. Elles restent la discrtion de lenquteur :

Vous avez mentionn avoir vu (entendu) des spots TV (radio) sur le thme du VIH/SIDA, lutilisation des prservatifs, etc. Pouvez-vous dcrire ces spots ? Quelle a t votre raction spontane aprs avoir vu (entendu) ces spots ?

Quelles motions (sensations) avez-vous ressenti ? Quelles penses vous sont venues lesprit ? Y avait-il dautres personnes prsentes ? Quelles ont t leurs ractions ? Avez-vous discut les annonces avec ces personnes ?

Quels aspects vous ont particulirement plu ou dplu dans cette campagne ? Pour quelles raisons ? Quels changements pourrait-on apporter aux annonces ? Que pensez-vous des personnages de ces spots ?

6.2-61

In-Depth Interviews

Chapter 6.2

Poser des questions afin de vrifier de quels personnages parle la personne Comment dcririez-vous ces personnes ? Est-ce que ces personnes vous rappellent des gens que vous connaissez ? Pouvez-vous les dcrire ?

Vous identifiez vous certaines de ces personnes ?

Approfondir (en quoi ?) Y a t-il un personnage auquel vous ne vous identifiez pas du tout ou qui vous parat invraisemblable ?

Quauriez-vous fait la place des personnages dont nous venons de parler ? A votre avis quelle organisation est lorigine de ces annonces ? Est-ce que vous approuvez ces annonces ? Pensez-vous quil faut poursuivre ce genre dinitiative ? Pensez-vous que ces annonces auront un impact sur les jeunes camerounais ? Lequel ?

Mthodes et Rsultats requis Il sera ncessaire de fournir une transcription intgrale (mot pour mot) des discussions obtenues. A cet effet lusage dun magntophone perfectionn est recommand. Les matriaux suivants devront tre soumis lissue de lenqute qualitative : 1) Transcription crite des discussions sur disquette 2) Rapport prliminaire 3) Rapport final prenant en compte les commentaires et questions soumises par PSI

6.2-62

PSI/AIDSMark Social Marketing Research Tool Kit 2004

In-depth interview guide from Togo on removing barriers to condom use (French) GRILLE DENTRETIENS DES INTERVIEWS DE GROUPES SUR LES IST/SIDA, LES GROSSESSES NON-DESIREES ET LUTILISATION DU PRESERVATIF CHEZ LES JEUNES ADOLESCENTS EN MILIEU SCOLAIRE AU TOGO

INTRODUCTION
Je suis . .. et je vous remercie trs sincrement pour avoir accept de participer cette interview de groupes que nous organisons dans le cadre du programme de sant de la reproduction des adolescents de PSI au Togo. En effet, PSI/Togo, conscient de limportance et du poids de la jeunesse pour toute socit et pour le Togo, choisi les jeunes comme une cible particulire pour ces activits de lutte contre les IST/SIDA et les grossesses non dsires. Les activits de PSI/Togo ciblant spcifiquement les jeunes sont, entre autres, lducation par les pairs et les campagnes par les mdias de masse. Dans la perspective de lamlioration et de lextension de ces activits, PSI/Togo souhaite avoir votre point de vue sur quelques sujets relatifs la lutte contre le IST/SIDA et les grossesses non dsires. Soyez donc libre de dire tout ce que vous pensez. Nayez aucune crainte car vous ne serez pas identifi. Vos noms ne seront pas demands et ne seront pas mentionns. Cest sur une base anonyme que se fait cette discussion. Avec votre permission, nous allons enregistrer les discussions avec cet appareil, dans la perspective de rendre compte avec le plus de fidlit possible les points de vue que vous allez exprimer. Nous sommes donc ici pour couter vos opinions, les opinions de chacun dentre vous. Merci davance pour votre comprhension et votre franche collaboration.

NOTE A LATTENTION DE LANIMATEUR

Il ny a pas de bon ou de mauvais point de vue. Chaque participant doit sexprimer librement, mme sil y a des points de vue diffrents. Cest la dynamique densemble qui importe. On ne cherche pas le consensus mais lexpression libre de chaque opinion.

6.2-63

In-Depth Interviews -

Chapter 6.2

Pour faciliter la discussion, il est important de veiller ce que tous ne parlent pas au mme moment. Donner donc la parole tout le monde mais, une personne la fois. Dans tout groupe, il y a des personnes qui parlent beaucoup et dautres qui sexpriment un peu moins. Il y a des personnes qui ne respectent pas les rgles implicites du groupe (qui dvient chaque fois la conversation) et dautres qui cherchent obtenir le consensus. Il est donc trs important que lanimateur repre toutes ces personnes pour ne pas perdre le contrle de la discussion. DU

- La discussion est prvue pour dure 1 heure 30 minutes. 1. IST/SIDA, GROSSESSES NON-DESIREES ET UTILISATION PRESERVATIF CHEZ LES JEUNES EN MILIEU SCOLAIRE

Nous allons discuter des IST/SIDA, des grossesses non-dsires et de lutilisation du prservatif.

CONNAISSANCE DES IST/SIDA (5 minutes)

Selon vous, le SIDA cest quoi ? Beaucoup de gens disent que le SIDA nest quune invention et quen ralit il nexiste pas. Quen pensez-vous ? Pensez-vous que le SIDA soit aujourdhui un srieux problme pour le monde ? Pensez-vous que le SIDA soit un problme pour votre pays ? Est-ce un problme srieux, moyen ou nest-ce pas un problme pour le Togo ? Pourquoi pensezvous cela ?

CONNAISANCE DES MODES DE TRANSMISSION ET MOYENS DE PREVENTION DES IST/SIDA (15 minutes)
Selon vous, comment une personne peut attraper le SIDA ? Connaissez-vous dautres maladies qui se transmettent de la mme faon que le SIDA ? Quelles sont ces maladies ? Certaines personnes disent quil ny a pas de diffrences entre le SIDA et les autres maladies sexuellement transmissibles que vous venez de citer ? Quen pensez-vous ? Il y a des gens qui disent que pour ne pas attraper le SIDA, il faut viter de saluer ou de manger avec une personne qui a le SIDA ? Quen pensez-vous ? Pouvez-vous reconnatre une personne qui est infecte par le virus du SIDA ? Par quels signes le reconnatrez-vous ? Quelle diffrence faites-vous entre une personne infecte par le VIH et une personne malade du SIDA ? Dautres personnes disent que la seule faon pour les jeunes comme vous dviter le SIDA, les IST et les grossesses non-dsires cest de ne pas avoir des rapports sexuels ? Quen dites-vous ? 6.2-64

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Pensez-vous quil y a dautres choses que vous pouvez faire pour viter dattraper le SIDA ou les autres IST ? Pensez quil est possible des jeunes comme vous de sabstenir compltement des rapports sexuels ? A partir de quel ge ou de quel moment pensez-vous quil soit normal ou bon de commencer avoir des rapports sexuels ? Pourquoi ?

PERCEPTION DU RISQUE PERSONNEL (10 minutes)

Daprs vous, quelles sont les personnes (hommes ou femmes) qui courent un grand risque dattraper le SIDA ou les autres IST ? Pourquoi ces gens courent-ils un grand risque ? Connaissez-vous des garons/des filles comme vous qui courent un grand risque dtre lauteur/davoir une grossesse non-dsires ? Pour quelles raisons pensezvous que ces personnes courent un grand risque ? Pensez-vous vous-mme courir un grand risque dattraper le SIDA, les autres IST, une grossesse non-dsires ? Pourquoi pensez-vous cela ? Comment pouvez-vous juger votre risque : pensez-vous quil soit lev, moyen ou faible ? Pourquoi pensez-vous cela ?

DISCUSSION SUR LES IST/SIDA ET LES GROSSESSES NON-DESIREES (15 minutes)

Certaines personnes pensent quil nest pas bon de parler des IST/SIDA ou des grossesses non-dsires car ce sont des sujets qui ont trait la sexualit ? Quel est votre avis ? Pensez-vous que la sexualit soit un sujet tabou ? Pourquoi ? Quelles sont les personnes avec lesquelles vous discuter le plus souvent de la sexualit, des IST/SIDA et des grossesses non-dsires ? Pourquoi discutez-vous le plus souvent avec ces personnes-l ? Lorsque vous discutez de la sexualit, quels sont les sujets que vous dbattez le plus souvent ? Pensez-vous avoir besoin de plus dinformations sur le SIDA, les autres IST et les grossesses non-dsires ? Quelles sont les informations supplmentaires que vous souhaiteriez avoir ?

6.2-65

In-Depth Interviews -

Chapter 6.2

Selon vous, quelle serait la meilleure personne ou la meilleure source pour vous procurer les informations dont vous avez besoin sur le SIDA, les autres IST et les grossesses non-dsires ? Pourquoi cette personne / source vous parat-elle tre la meilleure ?

ACCEPTABILITE ET ACCESSIBILITE DU PRESERVATIF (15 minutes)

Beaucoup de personnes disent que les prservatifs diminuent le plaisir sexuel et ne protgent pas contre les IST/SIDA ou les grossesses non-dsires ? Quel est votre avis ? Vous personnellement, pour quelles raisons utilisez-vous les prservatifs ? Est-ce que vous utilisez systmatiquement ( chaque rapport sexuel) les prservatifs ? Avec quelles personnes nutilisez-vous pas de prservatifs ? Pour quelles raisons ? Beaucoup dautres personnes disent quelles nutilisent pas des prservatifs parce quelles ont honte den acheter elles-mmes en pharmacie ou dans les centres de sant o ils sont vendus. Comprenez-vous ces personnes ? A votre avis, quel endroit ou auprs de quelle personne aimeriez-vous acheter vous-mme vos prservatifs sans aucune honte ? Pourquoi cet endroit ou cette personne ? Selon vous, quest-ce qui faciliterait lutilisation des prservatifs chez les jeunes comme vous ?

SOUTIEN SOCIAL (10 minutes)

Beaucoup de gens pensent que les filles qui disent leurs partenaires dutiliser le prservatif sont celles qui ne sont pas srieuses, celles qui savent quelles ont des multiples partenaires ? Quen dites-vous ? Pensez-vous quil soit appropri une fille de parler du prservatif avec son partenaire, quel que soit lge de celui-ci ? Pour beaucoup dautres personnes, il est normal un garon davoir plusieurs copines. De ce fait, cest seulement les garons qui doivent proposer le prservatif aux filles ? Selon vous, que pensez-vous que vos ami(es) pensent des jeunes qui utilisent les prservatifs ? Selon vous, que pensez-vous que vos parents pensent des jeunes qui utilisent les prservatifs ?

6.2-66

PSI/AIDSMark Social Marketing Research Tool Kit 2004 Vous sentez-vous capable de dcider seul(e) dutiliser le prservatif et de limposer votre partenaire, vos ami(es), vos parents.

FIN ANIMATEUR : Remercier les participants et annoncer que vous allez leur remettre, au nom de PSI/Togo et en signe de reconnaissance pour leur collaboration, une petite motivation de 1 000 fcfa et dun lot de prservatifs.

6.2-67

PSI/AIDSMark Social Marketing Research Tool Kit 2004

APPENDIX 6.2.2
IN-DEPTH INTERVIEW REPORTS
Misconceptions, Folk Beliefs, and Denial: Young Mens Risk for STIs and HIV/AIDS in Zambia. Misconceptions, Folk Beliefs, & Denial: Young Mens Risk for STIs & HIV/AIDS in Zambia Kim Longfield Rebecca Cramer Nkenda Sachingongu PSI Research Division Working Paper No. 53 2003 Dr. Kim Longfield is a Research Officer for AIDSMark/Population Services International, Washington, DC. Ms. Rebecca H. Cramer is a Research Assistant for Population Services International, Washington, DC. Mr. Nkenda Sachingongu is a Research Analyst for the Society for Family Health/Population Services International, Lusaka,Zambia. The opinions expressed in this report are those of the author(s) and do not necessarily reflect the views of PSI, SFH, USAID, or DFID. Acknowledgments Funding for this study was provided by USAID Zambia. Additional financial support was provided by PSI, which has core support from the British Department for International Development (DFID). The authors thank the communities in the following compounds for their cooperation and participation: Bauleni, Chainda, Chawama, Chazanga, Gardern, Jack, Kabanana, Kalingalinga, Misisi, and Soweto. The authors are indebted to Caroline Trigg, Shannon England, and Nchima Chulu Mwaba for their suggestions regarding the study design and preliminary report. The authors also thank Josh Barrett for his help with the literature review and editing, Rodrigo Boccanera for his assistance with the literature review, and Karen Eddleman for her editorial services. Finally, Sohail Agha, Kimberly Ashburn, Robert Kelly, and Andrea Plautz reviewed and provided useful comments on this paper.

Abstract

6.2-68

In-Depth Interviews

Chapter 6.2

Objectives: This study identifies young Zambian males misconceptions and folk explanations for sexually transmitted infections (STIs), as well as human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS). In addition, this study describes their sources of information and assesses the reliability of those sources. The authors distinguish denial about infection from misconceptions and folk beliefs and explain how each influences young mens personal risk perception and strategies for avoiding infection. Methods: Thirty in-depth interviews were completed with out-of-school males aged 1519 years in 10 compounds in Lusaka, Zambia. Participants discussed their knowledge of transmission and prevention of STIs and HIV/AIDS, sources of information, personal risk perception, and patterns of sexual behavior. Data analysis in Ethnograph 5.0 highlighted common beliefs about STIs/HIV and their role in risk perception and sexual behavior. Results: Overall, awareness of STIs and HIV/AIDS was high, but specific knowledge about transmission, prevention, symptoms, and risk factors was limited. Most participants identified promiscuous, careless, and unhygienic individuals, especially women, as likely transmitters of STIs/HIV. Misconceptions about HIV being transmitted through mosquito bites, kissing, and biting were common as were folk explanations linking infection to the strength of individuals blood, menstruation, or sorcery. Despite knowing that a healthy-looking person can be infected with STIs and HIV, most participants continue to rely upon outward appearance as a means of identifying individuals who are likely to be infected. Overall, risk perception among study participants was low, and several participants comments served as rationalizations for risky sexual behavior, suggesting that denial renders them unwilling to admit that current actions increase their risk for infection. Denial appears to be the result of the stigma attached to STIs and HIV/AIDS, behaviors associated with infection, and conflicts between local values and youths sexual behavior. Conclusions: Misconceptions, folk beliefs, and denial can impede personal risk perception for infection and interfere with the adoption of safer sexual behaviors. A common prevention strategy among participants was to avoid individuals thought to be at risk for STIs and HIV/AIDS rather than adopting more reliable methods, such as abstinence, condom use, or evaluation of risk through testing. The findings from this study suggest several programmatic implications for improving young mens knowledge about infection and their subsequent risk assessment, including developing more peer-based interventions and presenting science-based messages that work within a local health belief framework. In addition, programs should work with communities to decrease stigma associated with youths sexual activity and HIV/AIDS. 1

Introduction
Worldwide, an estimated 11.8 million young people aged 1524 are infected with human immunodeficiency virus (HIV), most of whom (77%) live in sub-Saharan Africa (Henry J. Kaiser Family Foundation 2002). Youth in this region are also

6.2-69

PSI/AIDSMark Social Marketing Research Tool Kit 2004 disproportionately affected by sexually transmitted infections (STIs), which increase their likelihood of acquiring and transmitting HIV (AGI 1998; UNAIDS and WHO 1997). In Zambiaone of nine African countries hardest hit by the epidemicHIV prevalence among 15- to 49-year-olds exceeds 21% (UNAIDS 2002a and 2002b). Several factors contribute to the rapid spread of HIV in Zambia, including high rates of STIs, a norm of multiple sexual partners, low rates of condom use, cultural beliefs, poverty, and poor health status (Weiss et al. 2001; USAID and Impact 1999). Zambias population is estimated to be 10.6 million, 21% aged 1524 (UNICEF, UNAIDS, and WHO 2002). Approximately 40% of the population is urban, and nearly two-thirds of Zambians live below the poverty line (CSO 2002). Although most Zambians receive some primary school education, few attend secondary school1 usually due to families inability to pay school-related costs (Kelly 2000; Feldman et al. 1997). A lack of access to education and economic opportunities can exacerbate youths vulnerability to HIV and STIs. Zambians with more years of schooling are less likely to have casual sexual partners and more likely to use condoms (CSO and ORC Macro 2002; Magnani et al. 2002; Agha 2000). Recent evidence from Zambia demonstrates a decline in HIV/AIDS prevalence, especially among youth and educated segments of the population. Decreases in seroprevalence have been attributed to delayed sexual debut, lower rates of sex with casual and multiple partners, and increased use of condoms (MAP 2002; CSO 2002; TVT Associates 2002). Nevertheless, only modest changes in seroprevalence have been detected among poor and less-educated groups (Fylkesnes et al. 2001). Zambian youth continue to participate in activities that put them at increased risk for STIs and HIV infection. Early sexual activity is common among youth, with a median age of 16.6 years for females and 16.0 years for males at sexual onset (UNICEF, UNAIDS, and WHO 2002). Many youth report having multiple sexual partners, and although rates of condom use are increasing, consistent use remains low (Fetters, Mupela, and Rutenberg 1998). Only 26% of 15- to 24-year-old males and 20% of females in the same cohort reported using a condom at last sex, and only 7% of youth report consistent condom use (UNAIDS 2002b, USAID and Impact 1999). Increasing rates of STIs are a problem among young Zambians, with rates of herpes simplex virus type 2 infection as high as 50% in some areas of the country (Weiss et al. 2001). Nearly a quarter of young men in a recent study reported a history of STIs, and 72% said that they knew someone who had had an STI (Ndubani and Hojer 2001). Several studies have demonstrated that despite high levels of awareness about HIV/AIDS, youth worldwide lack specific knowledge essential to protecting themselves from infection. Misconceptions persist about modes of STI/HIV transmission, methods of prevention, and forms (1 Primary education in Zambia is the first 7 years of schooling, with secondary being an additional 5 years (CSO 2002) of treatment (Summers, Kates, and Murphy 2002; Mshana et al. 2002; PRB 2001). 2 Although 94% of 15- to 24-year-olds in Zambia have heard of AIDS, only 75% know that HIV exposure can be avoided. Approximately one-fourth of youth aged 1524 believe that mosquitoes transmit HIV (CSO 2002). More than 80% of Zambian youth know that a healthy person can be infected with HIV, but youth continue to make assumptions about peoples serostatus based on appearance (CSO 2002; Trigg 2001). Knowledge about STIs is even lower. Three-quarters of youth have heard of 6.2-70

In-Depth Interviews

Chapter 6.2

STIs but only half are able to cite common symptoms (CSO 2002). Overall, researchers estimate that only 26% of Zambian youth have sufficient knowledge to protect themselves from STIs and HIV infection (UNICEF, UNAIDS, and WHO 2002). Youths knowledge about STIs and HIV/AIDS is complicated by local beliefs that overlap with scientific messages. Like most individuals, youth adhere to multiple theories of transmission and prevention in order to make sense of the epidemic (MacIntyre, Brown, and Sosler 2001; Warwick and Aggleton 2001). Their beliefs about infection are a result of their education, culture, socialization, and personal experience (Bernardi 2002; Astatke and Serpell 2000; Nzioka 1996; Prochaska et al. 1990). In addition, youth receive conflicting information about HIV/AIDS and sexuality from different sources (Dowsett and Aggleton 2000). For example, even though 78% of Zambian males aged 1524 state that having one faithful partner is a method of HIV prevention, many believe that multiple sexual relationships are essential for achieving manhood (CSO 2002; Ndubani and Hojer 2001). Local beliefs that HIV is transmitted through supernatural forces also persist: 16% of males and 22% of females aged 1524 stated that HIV is the result of witchcraft (CSO 2002). Zambian youths limited scientific knowledge and contradictory beliefs about STIs and HIV/AIDS may stem from their reliance upon interpersonal sources of information. Most youth rely upon peers for information, and others seek advice from grandparents or traditional healers (Fetters, Mupela, and Rutenberg 1998). These individuals may be ill informed about transmission and unprepared to suggest reliable methods of prevention. In Zambian culture, it is considered inappropriate to discuss sexual matters with ones parents, and evangelical churches label premarital sex and condom use as immoral, contributing to a stigmatization of STIs and HIV/AIDS (Agha 2001; Feldman et al. 1997). Stigma and a reluctance to discuss sexual matters openly with youth increase the likelihood of youth receiving incorrect information, denying their risk for infection, and failing to adopt protective behaviors (Population Council 2001; JHUCCP 2001; Kirby 1997). Researchers have identified several factors associated with risk perception for STIs and HIV/AIDS. Behavioral theories, such as the Health Belief Model, Theory of Reasoned Action, and Social Cognitive Theory, outline antecedents essential for the adoption of safer sexual behaviors (Kirby 1999; Sweat and Denison 1995; Rosenstock, Strecher, and Becker 1994; Fishbein, Middlestadt, and Hitchcock 1994; Bandura 1994). Although the role of knowledge in increasing risk perception and changing behavior is unclear, it is thought to be a prerequisite for adopting protective behaviors (Peruga and Celentano 1993). Improving youths knowledge about STIs and HIV/AIDS and the environment in which they learn about infection has the potential to change youths attitudes about risk behaviors, improve their ability to negotiate safer sexual practices, and decrease stigma associated with risk behaviors. 3 It is important that researchers identify misconceptions youth harbor about STIs and HIV/AIDS in order to improve behavior change communication (BCC) programs and social marketing efforts. This study identifies accurate beliefs, common misconceptions, and folk explanations for STIs and HIV/AIDS among young Zambian males, as well as their sources of information, and the reliability of those sources. Authors distinguish denial about infection from misconceptions and folk beliefs and 6.2-71

PSI/AIDSMark Social Marketing Research Tool Kit 2004 explain how each influences young mens personal risk perception and strategies for avoiding infection. Based on study findings, the authors recommend strategies for relaying accurate information and improving youths risk perception and adoption of protective strategies.

Methodology
Data were collected in May and June 2001 for two purposes: (1) to create a BCC strategy for Zambian youth encouraging consistent condom use; and (2) to develop a multi-round survey to track behavior change. Thirty in-depth interviews were conducted with 15- to 19-year-old out-of school males in 10 low-income neighborhoods (compounds) randomly selected from Lusaka. Although complementary interviews were conducted with females, the focus of this paper is on the data collected for males. Participants came from randomly selected households within each compound, and a trained interviewer conducted one-on-one interviews with participants. Most interviews were conducted in Chinyanja, the rest in Ichibemba, two of the most widely spoken languages in Zambia. An open-ended discussion guide was used that had been pretested with the same population and revised. Discussions covered the respondents social history and background, knowledge of transmission and prevention of STIs and HIV/AIDS, sources of information, personal risk perception, and patterns of sexual behavior. Interviews were audiotaped, translated, and transcribed into English. The authors completed data analysis in Ethnograph 5.0 to highlight common themes in participant comments about accurate knowledge and misconceptions about STIs and HIV/AIDS.

Results
Profile of study participants The majority of study participants were born in Lusaka and lived with both parents. Among those living with a single guardian, most lived with their mothers, usually because their fathers had died. Others lived with brothers, grandparents, uncles, or brothers-in-law. Many of the participants guardians were unemployed or had lowpaying jobs. The majority of participants left school during or at the end of grade 6. The lowest level of education attained was grade 3 and the highest, grade 9. Most participants could read, albeit with difficulty, both English and a vernacular language, but few could write. The main reason cited for dropping out of school was lack of money, and for several, this action had been prompted by the death of their guardian. Other participants attributed dropping out of school to their own irresponsibility. Most regretted the fact they were no longer in school and expressed eagerness to continue their education. Poverty was a central concern of all participants. The most common ambition respondents expressed was a desire to resume their education, which could enable them to secure gainfulemployment and contribute to family income. Few participants mentioned contracting STIs or HIV/AIDS as a concern. 4 Common beliefs, misconceptions, and folk explanations about STIs and HIV/AIDS All study participants had heard of STIs and HIV/AIDS; most, however, lacked specific

6.2-72

In-Depth Interviews

Chapter 6.2

knowledge about modes of transmission, methods of prevention, symptoms, and risk factors associated with infection. Most participants demonstrated partial knowledge about infection or harbored misconceptions about transmission and prevention. In addition, many shared folk explanations that linked infection to weak blood, menstruation, or sorcery. Only a few participants stated that everyone is at risk for contracting STIs and HIV/AIDS; most identified promiscuous, careless, and unhygienic individuals, especially women, to be at greatest risk. Although most participants understood that STIs and HIV/AIDS are sexually transmitted, they also described mosquito bites, kissing, and biting as likely modes of transmission. Likewise, several participants misunderstood methods of prevention and mentioned only condom use or abstinence as options rather than multiple forms of prevention. Finally, despite knowing that a healthy-looking person can be infected with HIV, most participants continued to rely on outward appearances to identify individuals who are likely to be infected. General STI and HIV/AIDS information Participants demonstrated a wide range of knowledge about HIV/AIDS and often mixed correct information with misconceptions. There was awareness among participants that HIV/AIDS is a very dangerous disease, and most understood that it is incurable. A few participants, however, argued that if infected individuals are treated early or with modern medicine, they can be cured of HIV/AIDS. Most participants believed that all STIs are curable and sometimes confused them with other diseases. A virus transmits HIV while STIs are diseases like cholera, which are transmitted by something else. STIs are curable unlike AIDS, which is curable only in a very few cases (16 years, Chainda Compound). Participants demonstrated mixed knowledge about STIs and their relationship to HIV/AIDS. Some knew that untreated STIs could result in infertility, but others believed that STIs could be deadly if left untreated. Most participants did not understand that STIs increase ones risk for HIV infection, and some believed that untreated STIs could develop into AIDS. As long as (STIs) arent treated in time or if one doesnt finish taking the prescribed medication, STIs are what eventually turn into AIDS. For example, when I suffered from bolabola2 (chancroid), if I hadnt finished taking the medicine, the disease wouldve turned into AIDS (18 years, Chainda Compound). Only a few participants were able to distinguish HIV from AIDS. The majority stated that HIV and AIDS are the same or said that they did not know if there is a difference between the two.2 Bolabola is a local term for a chancroid or an STI that
causes swelling of lymph nodes in the pelvic region. Literally translated, the term means ball ball, referring to the shape of swelling.

5 Some participants appeared to understand that HIV is a virus; however, they were unclear about the viral disease process and how HIV affects the immune system. A few participants believed that HIV becomes lodged in individuals bones and weakens their bodies as a result. HIV/AIDS isnt curable because its caused by a germ that lodges (itself) in the bones. Once in the bones, it isnt possible to dislodge this germ (17 years, Chawama Compound). STI and HIV/AIDS symptoms

6.2-73

PSI/AIDSMark Social Marketing Research Tool Kit 2004 When asked to cite symptoms associated with STIs and HIV/AIDS, nearly all participants described symptoms apparent in the later stages of AIDS. Only a few argued that it is impossible to know if individuals are infected by their outward appearance. You know, AIDS, just like syphilis, destroys many parts of a persons body including the brain even when a person may look physically healthy (19 years, Chainda Compound). Some participants noted common STI symptoms such as sores or pus around the genital region and blood during urination. Others provided anecdotal descriptions for infection such as walking with ones legs far apart, having flies hover around the genital region, or emitting a foul odor. Bolabola (chancroid) comes with sores around the private parts. Therefore, ones known by his manner of walking with legs wide apart (15 years, Jack Compound). Participants listed a wide range of symptoms for HIV/AIDS. Most mentioned symptoms of endstage AIDS and did not know that individuals infected with HIV could remain asymptomatic for years. Severe weight loss, hair loss, changes in skin tone, and changes in hair color were the most commonly cited symptoms. Some participants stated that people with HIV/AIDS develop shingles (locally known as Gods fire), and others believed that infected individuals crave meat or western food3. Other symptoms included a deepening of the voice, full body rash, diarrhea, ringworm, protruding shoulders, paleness, vomiting, and swollen legs. An infected person loses weight, the shoulders protrude, and the hair becomes light and falls off. When the infection becomes very acute, he begins to yearn for various foods, which he doesnt even eat. The person may ask for chicken and when you bring the chicken, he wont eat it (16 years, Soweto Compound).
3 Low-income

groups in Zambia typically eat inexpensive foods such as vegetables. Individuals who suspect that they are HIV positive may go to extremes to maintain a healthy diet and purchase nutritious foods that are considered western, especially meat, fruit, and milk.

6 Individuals at risk for STIs and HIV/AIDS Overall, participants believed that individuals who are sexually promiscuous are at greatest risk for STIs and HIV/AIDS. Nearly all agreed that females are more promiscuous than males and the most likely transmitters of STIs and HIV/AIDS.4 Several participants identified commercial sex workers (CSWs) and young women engaging in relationships with older men as especially promiscuous. Most participants believed that promiscuous females are easily identifiable by their manner of dress, propensity to frequent bars or nightclubs, and their participation in bad behavior. Those who are more vulnerable (to STIs and HIV/AIDS) are those who are involved in bad activities such as proposing love to boys and wearing short skirts in order that they may entice men to have sex with them (15 years, Chazanga Compound). Other individuals considered by participants to be at risk for STIs and HIV/AIDS were careless people who do not use condoms, unhygienic individuals, and boys who had had older girls as sexual partners. Only a few participants stated that everyone who is sexually active is vulnerable to contracting STIs and HIV/AIDS. Theres no one who can be said to be less vulnerable (to infection) because one never knows how he becomes infected (16 years, Chainda Compound). Participants appeared divided on the risk STIs and HIV/AIDS present to youth. Some believed that youth are at increased risk for infection because of high levels of sexual activity. Others, however, argued that youth are at low risk for infection and that adults are at greatest risk

6.2-74

In-Depth Interviews

Chapter 6.2

because they have been sexually active longer and have had more partners. Some participants stated that as long as youth avoid having sex with adults, they remain safe from infection. STI and HIV/AIDS transmission and prevention Although knowledge about modes of transmission for STIs was generally high among participants, correct information about HIV/AIDS transmission was low. Nearly all participants agreed that STIs and HIV/AIDS are sexually transmitted, but most believed that HIV infection is inevitable after one sexual encounter with an infected individual. Some also argued that HIV/AIDS is transmitted more easily through sexual contact than other modes of transmission. A few participants noted correctly that HIV/AIDS is present in blood, and they were able to identify bloody objects, such as razor blades and syringes, as infectious. (You get infected) by being cut by a razor blade contaminated by an infected person. In hospitals, a person can also be infected if a syringe previously used for an infected person is used for an uninfected person (16 years, Soweto Compound). 4 Heterosexual sex is the primary mode of HIV transmission in
Zambia, which may explain why participants spoke of women as primary transmitters of infection (UNAIDS 2002a).

7 Many participants shared folk explanations for transmission and attributed the likelihood of individuals becoming infected with HIV to the strength of their blood. They explained that individuals with weak blood are more likely to become infected and manifest symptoms quicker than individuals with strong blood. It depends on the blood; if its strong, it can take 2 years (to show signs of HIV infection). If on the other hand the blood isnt strong, itll take only months (16 years, Soweto Compound). Likewise, some participants believed that it takes longer for females to display symptoms of HIV infection than males because they are able to shed the virus during menstruation. One participant argued that females who have reached puberty cannot become infected with HIV/AIDS because menstruation removes the virus entirely from the body. A girl who hasnt reached puberty will be infected and one whos reached puberty wont be infected. Thats because for a girl whos reached puberty, the virus is removed through menstruation (18 years, Bauleni Compound). Several participants believed that being bitten by or kissing infected individuals could transmit HIV. Only some of these participants noted that such routes of transmission are only plausible if individuals have open sores in their mouths. I was told that if an infected person has a sore in the mouth and kisses a healthy person, the healthy person will be infected (15 years, Misisi Compound). Some participants used medical jargon when describing HIV transmission and explained that HIV is a virus or germ that can be passed through blood or other bodily fluids. Some participants believed that HIV is passed through saliva. A few believed that sharing food or toilets, or shaking hands with infected individuals presents risk for infection. In addition, several participants thought that mosquito bites pose a threat for transmission. A person, bitten by a mosquito, which had previously bitten an infected person, will be infected. The mosquito gets blood from the infected person and in biting the uninfected person, it leaves some blood in his body (15 years, Misisi Compound). A few participants said that HIV/AIDS is related to sorcery. They explained that people infected

6.2-75

PSI/AIDSMark Social Marketing Research Tool Kit 2004 with HIV act as if they have been bewitched. One participant said that some girls become infected with HIV by having sex with attractive ghosts who frequent bars. In drinking places girls see good-looking males who, in actual fact, are ghosts. At night (the ghost) wakes up to infect other people with HIV/AIDS. I dont know how the ghost became infected because its something I merely heard from other people (16 years, Kalingalinga Compound). 8 Participants identified condom use and abstinence as the most effective means of preventing STIs and HIV/AIDS. However, most participants identified one method to the exclusion of the other and failed to cite multiple forms of prevention. I think that people should just abstain from sex. This is the only way that we can prevent these diseases (19 years, Chainda Compound). Sources of information about sex, STIs, and HIV/AIDS Participants reported receiving information on sex, STIs, and HIV/AIDS from a variety of sources, notably interpersonal contacts; Information, Education and Communication Programme (IEC) agents; and mass media. Several reported knowing individuals who had had STIs and a few knew people living with HIV or AIDS. Most participants indicated that they rely on interpersonal contacts for information on infection. Such discussions with friends and relatives usually occur within the context of gossip or speculation about individuals thought to be infected. Participants were reluctant to ask parents for information for fear of being punished or accused of inappropriate behavior. Several identified elders, doctors, and clinics as trusted sources of information but were usually embarrassed to approach them with questions about sex and infection. Interpersonal contacts and IEC agents Almost all participants reported obtaining information about STIs and HIV/AIDS through interpersonal communication with friends or family members. Most explained that they avoid asking friends and family questions about STIs and HIV/AIDS out of fear, embarrassment, or shyness. As a result, most obtain information about infection from conversations that are overheard within the context of gossip. Much information appears to be based on hearsay garnered when friends and family speculate about individuals they know are infected. Everyone knows about AIDS. Ive heard about it from other people, especially adults. The first time I heard about it was from a certain adult who was talking about someone who was suspected to have contracted the disease (16 years, Chawama Compound). Several participants said that they were embarrassed to ask their friends about sexual matters or infection because they do not want them to presume that they are sexually active or have an STI or HIV/AIDS. Likewise, most participants said that they are reluctant to ask parents about sex, STIs, and HIV/AIDS because they would be reprimanded and suspected of having sex. I cant approach my parents because theyd simply tell me to shut up quickly and get out of their sight (15 years, Chazanga Compound). Despite this reluctance, some participants reported receiving information from family members about sex, STIs, and HIV/AIDS. Most information arrives in the form of warnings rather than open discussion with parents alerting youth to the consequences of sexual activity. 6.2-76

In-Depth Interviews

Chapter 6.2

9 Even my father told us about the disease. He usually tells all of us in the house that we should avoid having sex if were to avoid contracting AIDS (16 years, Chawama Compound). Reliance upon interpersonal contacts can result in the spread of inaccurate information and misconceptions about risk and STI/HIV transmission. Most inaccurate information appears to be obtained from peers and adults. When we meet with older people in the bars, they reveal things to us. They say that if a boy whos never had children has sex with a girl whos ever had a child, the boy will be hurt (18 years, Kabanana Compound). Only a few participants reported discussing HIV/AIDS with their girlfriends and sexual partners. I discuss HIV/AIDS with my girlfriend. I tell her that we should be careful about our sexual life because we wont know who between us will contract HIV/AIDS and infect the other partner (18 years, Kabanana Compound). Several participants reported receiving information from IEC agents during awareness-building events and presentations in compounds. Such events included football games, parades, and mobile video shows as well as peer education. Radio, television, and newspapers Most participants stated that they receive some information about STIs and HIV/AIDS from radio, television, or newspapers. Radio was the most frequently mentioned medium, and some participants recited prevention messages they had heard on radio programs. However, some participants admitted that they do not pay attention to programs about STIs and HIV/AIDS when they are on the air. I remember hearing about STIs from Radio One although I didnt pay much attention to what was being said (18 years, Chainda Compound). The inability to understand and read English may prevent some participants from obtaining accurate information from mass media sources. Some participants said that they are unable to understand complicated radio messages in English and would prefer to hear them in local languages. A few said that they rely on friends or family members to read newspaper articles to them. Ive heard about STIs so many times from the radio but also sometimes on televisioneven from newspapers my brother reads for me. I also know how to read so I read for myself sometimes (17 years, Jack Compound). 10 Personal experience or knowing someone with an STI or HIV/AIDS Some participants reported that their knowledge about sex and STIs came from firsthand experience with infection. I came to know about STIs when I suffered from the disease myself. I developed a swelling around my groin coupled with leaking (gonorrhea). It was at the hospital that I was told that what I had was an STI (19 years, Kalingalinga Compound). Several participants reported that they knew someone who had had an STI. Some of them said that their friends experiences allowed them to learn more about symptoms and transmission. My friend who suffered from an STI tells us that STIs arent good because theyre painful. He usually tells us this whenever he reads a brochure or other such material (15 years, Jack Compound). Fewer participants knew individuals who were HIV positive or who had died from AIDS. For the most part, infected individuals were family members or neighbors, and participants appeared to make judgments about risk behavior based on rumors that 6.2-77

PSI/AIDSMark Social Marketing Research Tool Kit 2004 they heard about them. We had a relative who suffered from AIDS. I was young then, but my parents have been telling us about him. He was very promiscuous and had many sexual partners (15 years, Misisi Compound). Trusted sources of information Participants identified adults, friends, IEC agents, doctors, and clinics as trusted sources for information about sex, STIs, and HIV/AIDS. Most argued that medical professionals provide the most accurate information and that friends and family might not always be reliable resources. I used to discuss HIV/AIDS with my friends about 2 years ago as a way of gaining and spreading knowledge. Weve stopped such discussions because we may just end up sharing (inaccurate information) because Im no longer in the Anti-AIDS Club (16 years, Chainda Compound). A few participants reported knowing no one who could serve as a trusted source of information. In their opinion, individuals outside of the medical profession have limited knowledge about STIs and HIV/AIDS, rendering them unable to provide accurate information. I dont know any person that I can approach for information about AIDS.Its difficult to approach just anyone for such information....When someones not a medical person, its difficult to determine whether hes knowledgeable about AIDS or not (19 years, Kabanana Compound). 11 The contribution of stigma and shame to denial Previous sections of this document identified misconceptions and folk explanations common among study participants for STIs and HIV/AIDS. Both sets of beliefs can impede personal risk perception and the adoption of safer sexual behavior. Another, more complicated process is one of denial. Denial of infection may occur under several circumstances, including when infection and the behaviors that lead to infection are stigmatized, and when a conflict exists between the values that local culture dictate and individuals behavior (Feldman et al. 1997; Prochaska et al. 1990). Participants comments revealed several rationalizations for risky sexual behavior, suggesting denial and an unwillingness to admit that current behavior increases their risk for infection with STIs and HIV/AIDS. Most participants spoke about the influence of religion in their communities and their parents unwillingness to discuss sexual matters with them. Their comments demonstrated that a great deal of stigma is attached to sexual behavior among Zambian youth, STIs, and HIV/AIDS. Most participants believed, however, that individuals need sex for emotional and physical development, and they acknowledged that a conflict exists between social taboos against youths sexual activity and their actual behavior. As a result, participants spoke about infection as punishment for moral shortcomings. Stigma and shame associated with sexual behavior, STIs, and HIV/AIDS In large part, participants denial of risk for STIs and HIV/AIDS appeared to be attributed to the stigma and shame associated with sexual behavior among youth. Because transmission is associated with sexual behavior, infection can be regarded as punishment for moral shortcomings and lack of self-control. Throughout interviews, participants spoke about the inappropriateness of youth being sexually active and how churches and adults discourage them from having sex. 6.2-78

In-Depth Interviews

Chapter 6.2

Some participants echoed these core values and said that youth should not be allowed to buy condoms because access to them could encourage sexual activity. Others said that youth who become infected with STIs and HIV/AIDS have only themselves to blame because they know that they should not be sexually active. Some participants appeared to provide interviewers with socially appropriate responses, reinforcing the idea that youth believe that their sexual desires and behavior are socially inappropriate. Adolescents are underage (and shouldnt be allowed to buy condoms). Even if at the moment young people are allowed to use condoms, its against the laws of God (18 years, Bauleni Compound). Because of the association between infection and reckless behavior, many participants said that they are reluctant to approach adults for information about STIs and HIV/AIDS. In their opinion, curiosity about these topics can be interpreted as an admission of participating in inappropriate behavior. Finally, the stigma associated with STIs and HIV/AIDS can cause not only denial, but also fear and reluctance to be diagnosed or treated. Those who dont get cured (for STIs) are the ones who hide the fact that they have an STI until its too late. For such people, taking medication is as good as taking poison (18 years, Chainda Compound). 12 Conflicts between core values and youths need for sex Participants denial of risk behavior also appeared to be a result of the conflict between local core values discouraging sex among youth and their own perceived need for sex. Even though most recognized sexual abstinence as one of the most effective methods of STI/HIV prevention, they felt it impractical to expect youth to abstain. They explained that sex is a fundamental human need, pleasurable, and essential for emotional and physical development. In their opinion, most youth find it impossible to abstain or to limit their number of sexual partners. Abstinence isnt good. Every human experiences a desire for sex. Those who feel like having sex mustnt abstain because everything needs to be pretested. Sex provides knowledge and experience thats helpful in the future (15 years, Chazanga Compound). Some participants recognized a conflict between church doctrine, adults expectations of youths sexual behavior, and youths desire to have sex. They acknowledged the difficulty of this situation and explained that youth hide sexual activity as a result. Young men can never brush aside the thought of sex. We think that sex is a natural thing and yet, the Bible tells us to abstain until we get married. What we see in reality is that once a boy sees a girl that he likes, hell stop at nothing until he sleeps with her (17 years, Jack Compound). Finally, some participants spoke of a need to have multiple sexual partners to establish their manhood. In their opinion, it is important for young men to demonstrate their prowess by having several sexual conquests. If hes really a man, he cant come away from a girls home without having sex (18 years, Bauleni Compound). How misconceptions, folk beliefs, and denial affect risk perception and strategies for avoiding infection Participant comments reveal that misconceptions and folk beliefs about STIs and HIV/AIDS render them ill prepared to evaluate their personal risk for infection. Perhaps more important, the stigma and shame associated with STIs and HIV/AIDS can result 6.2-79

PSI/AIDSMark Social Marketing Research Tool Kit 2004 in a denial of risk and shape youths strategies for avoiding infection. Overall, risk perception among study participants was low. Misconceptions and folk explanations appear to interfere with participants risk perception by competing with correct information, thereby encouraging unsafe behavior, and hindering the adoption of safer behavior. Most participants believed that avoiding individuals at highest risk for transmission is an effective prevention strategy. Participants, however, tended to base their assessment of others risk on outward appearances. In addition, some participants appeared to use misconceptions and folk explanations as justifications for continued risk behavior and were unwilling to admit that their sexual activity poses a risk for infection. This denial of risk was also manifested by participants reluctance to adopt effective prevention strategies, such as abstinence and condom use, or to be tested for STIs and HIV. 13 Perceived risk for STIs and HIV/AIDS Few study participants believed that they are at risk for STIs and HIV/AIDS. They considered their risk for infection to be negligible because they are not promiscuous or unhygienic, nor do they fit the profile of high-risk individuals. Participants appeared more concerned about pregnancy, which they perceived as a more likely and immediate threat than STIs or HIV/AIDS. In addition to minimizing their personal risk for infection, participants presumed that current partners present little risk for STIs and HIV/AIDS because they are sexually faithful and have forsaken past risk behavior. None of the participants mentioned the risk serial monogamy presents or recognized the link between past sexual encounters and potential risk in current relationships. Issues about HIV/AIDS dont bother me because I know that Im not promiscuous. I cant be infected. It never occurs to me (that I could be infected). Im not promiscuous and my girlfriends stopped moving around [having sex] with other boys. She cant have HIV/AIDS (18 years, Soweto Compound). A general misconception among participants was that so long as they remain sexually faithful to their partners, they can avoid STIs and HIV/AIDS. Consequently, participants viewed monogamous sexual relationships as risk-free even though the majority of their relationships lasted no longer than 6 months. I dont think Id be infected with any STIs or HIV/AIDS.I have only one girlfriend; I dont think she could be infected. Shes been my girlfriend for a very long time (19 years, Kabanana Compound). By externalizing the threat of infection to promiscuous, careless, or unhygienic individuals, participants placed the onus of behavior change on others and argued that policy makers should ensure that risk activities are controlled. There was little acknowledgement on the part of participants that their own behaviors merit change. At least nightclubs should be made to close early so that the girls who patronize them will be compelled to go home early (and avoid risk) (19 years, Chainda Compound). 14 Strategies for Avoiding Infection Avoiding those thought to be infected with STIs and HIV/AIDS A prevention strategy most participants appear to have adopted is avoiding individuals thought to be at risk for STIs and HIV/AIDS. They spoke of avoiding promiscuous and unhygienic females, only having sex with partners they trust, and ending relationships 6.2-80

In-Depth Interviews

Chapter 6.2

with partners who appear risky. I believe that (my girlfriend) has another boyfriend. Judging by her behavior, its not safe to continue my relationship with her because she could infect me with an STI that could cause my death (19 years, Misisi Compound). Abstinence and fidelity Participants also spoke about abstinence and sexual fidelity as methods of preventing STIs and HIV/AIDS. Even though participants recognize abstinence as the best form of protection against infection, it appears to be the hardest behavior to maintain. Some participants spoke about them ability to abstain from sex as a sign of being a good person who will avoid being punished with STIs or HIV/AIDS. Everyone wants to have sex but is restrained only by the fear of AIDS. Under these circumstances, the better person is the one who refrains from sex (19 years, Chainda Compound). Discussions of morality and controlling ones behavior were common during participant interviews. Some participants suggested that youth who find it difficult to abstain should go to church and heed adults warnings to avoid sexual encounters. One way of preventing (AIDS) is going to church. That way, youll fear sleeping with girls (19 years, Chainda Compound). Participants who mentioned sexual fidelity as a means of prevention spoke of it in the same vein as abstinence, implying a moral imperative for limiting ones sexual partners and saving oneself from STIs and HIV infection. Likewise, some participants mentioned getting married as a prevention method, using marriage as a proxy for sexual fidelity. I think that people should stop having sex if they think theyre to be saved from the disease. I think that I should also abstain from sex or get married quickly (19 years, Kalingalinga Compound). 15 Condom use Several participants spoke about condom use as an effective prevention method. However, few mentioned the effectiveness of adopting a dual strategy and spoke about condom use independent of sexual fidelity. Some participants also said that if they were to choose between remaining abstinent and using condoms, they would opt for condom use because they find it impossible to forgo sexual relations. The only way of preventing AIDS is through using condoms or abstaining. Of these two ways of prevention, the better way is using condoms. You know that, as a male, one cant abstain. He must have sex all the time (19 years, Misisi Compound). Proponents of condom use said that condoms should be used with partners who look promiscuous or at high risk for infection. Overall, participants failed to recognize the importance of consistent condom use with all sexual partners. I think its important for one to take care of his life and use condoms, especially if you suspect that the girl looks like someone whos promiscuous (17 years, Chainda Compound). The most common justification for not using condoms was trust in ones sexual partner. Most participants believed that their sexual partners present little to no risk for STIs or HIV because they are good, not promiscuous, known to participants, and sexually faithful. Some appeared to use trust as a bargaining tool during condom negotiation by convincing partners that using condoms signifies a lack of trust. In some cases, participants used trust as a justification for risk behavior that had occurred before sexual fidelity was established in a relationship. I said to myself that the girl that I had sex with was a trusted person, and it wasnt possible that she could have 6.2-81

PSI/AIDSMark Social Marketing Research Tool Kit 2004 been promiscuous.I used to notice her behavior, and it seemed to me that she wasnt promiscuous (18 years, Chainda Compound). Several participants rationalized not using condoms by arguing that condoms are ineffective at preventing transmission of STIs and HIV. They stated that condoms contain holes and maintained that latex is an ineffective barrier to the virus. Several participants also said that condoms tear or break easily. Nevertheless, many admitted that they had not personally experienced any difficulties using condoms but based their observations on friends experiences. I dont trust condoms entirely because AIDS is a very destructive disease. Ive heard many people complain that condoms get torn (18 years, Chainda Compound). Some participants thought that although condoms can provide protection from some STIs, they are ineffective at preventing HIV transmission. (Condoms) are effective for diseases like kaswende (gonorrhea) and syphilis but not for AIDS (17 years, Misisi Compound). 16 Finally, some participants argued that condoms interfere with sexual pleasure. Several spoke about the importance of having skin-to-skin contact with their sexual partners. Only a few participants refuted this argument. In their opinion, young men use sexual pleasure as an excuse to forgo use, even when they know condoms are an effective means of protecting against infection. Many young people dont use condoms because they tell each other lies, such as sex isnt enjoyable with condoms. Yet theres no difference whether one uses condoms or doesnt use them (19 years, Kabanana Compound). HIV testing Some participants harbored misconceptions about HIV testing or had received incomplete information about the testing process. A few participants said that they had never heard of HIV testing and were unaware of its benefits. Some were poorly informed about the availability and cost of testing. I dont know whether or not young people like me are allowed to go for an HIV test (16 years, Kalingalinga Compound). Others had misconceptions about who should be tested and when testing should occur. A few believed that they would not require an HIV test because if they were infected, they would feel it inside of their bodies. A couple of participants thought that only those who are sick should obtain tests. I can only go for an HIV test if Im sick. Otherwise, theres no need to go for one (16 years, Jack Compound). Most study participants said that they assume they are HIV negative, so there is no need to get tested. In their opinion, testing is appropriate for adults and others who are at risk for infection. Some participants said that if they were to get tested, it would only be to confirm that they are HIV-negative. I would (get tested) in order to confirm my status. I know that I would test HIV negative (18 years, Soweto Compound). Nevertheless, some participants spoke frankly about their fear of testing and how a positive HIV diagnosis could be perceived as a death sentence, leaving them hopeless about the future. I wouldnt go for an HIV test because I wouldnt have the courage for it. If I were found to be HIV positive, Id be going around with the knowledge that Im about to die (18 years, Chainda Compound).

6.2-82

In-Depth Interviews 17

Chapter 6.2

Conclusions
This study examines young Zambian males beliefs about STIs and HIV/AIDS, their sources of information, and the reliability of those sources. Although the role of knowledge in increasing risk perception and changing behavior is unclear, it is considered a prerequisite for adopting protective behaviors, and programs should not overlook its importance. Our study identifies common misconceptions and folk explanations and distinguishes denial about infection from other sets of beliefs. Study findings offer insight into the influence these factors have on risk perception and youths strategies for avoiding infection. We also demonstrate how some youth use inaccurate knowledge and folk explanations as justifications for continued risk behavior. Although young men in Zambia appear to have received certain key messages about STIs and HIV/AIDS, they lack sufficient knowledge and opportunity to discuss risk behavior that would allow them to protect themselves adequately from infection. Despite high levels of awareness about STIs and HIV/AIDS, most participants lacked specific knowledge about modes of transmission, methods of prevention, symptoms, and risk factors for infection. Moreover, only a few acknowledged that everyone is at risk for these diseases; most identified promiscuous, careless, and unhygienic individuals, especially women, as likely transmitters of STIs/HIV. Misconceptions about HIV transmission through mosquito bites, kissing, and biting were common as were folk explanations that link infection to the strength of individuals blood, menstruation, and sorcery. Despite knowing that a healthy-looking person can be infected with STIs and HIV, most participants continue to rely upon outward appearances to identify individuals who are likely to be infected. The spread of misconceptions can be attributed to youths reliance upon interpersonal contacts as sources of information, individuals who are usually poorly informed themselves. Misconceptions and folk explanations about STIs and HIV/AIDS can impede personal risk perception for infection. Overall, risk perception among study participants was low and many considered pregnancy a more immediate threat than STIs or HIV/AIDS. In addition, several participant comments served as rationalizations for risky sexual behavior, suggesting that denial renders them unwilling to admit that current behavior increases their risk for infection. Denial appears to stem from the stigma attached to STIs and HIV/AIDS, behaviors associated with infection, and conflicts between local values and youths sexual behavior. Most participants spoke about the influence of religion in their communities and their parents unwillingness to discuss sexual matters with them. Most argued that individuals need sex for emotional and physical development but acknowledged that a conflict exists between social taboos against youths sexual activity and their actual behavior. As a result, participants spoke about infection as punishment for moral shortcomings. Misconceptions, folk beliefs, and denial can also interfere with the adoption of safer sexual behaviors. A prevention strategy common among most participants is to avoid individuals thought to be at risk for STIs and HIV/AIDS. They spoke of avoiding promiscuous and unhygienic females, only having sex with partners they trust, and ending relationships with partners who appear risky. Denial is also manifested in their reluctance to get tested for STIs or HIV and adopt prevention methods, such as abstinence and condom use.

6.2-83

PSI/AIDSMark Social Marketing Research Tool Kit 2004 18 As with all research, this study faced certain limitations that must be taken into account when interpreting results. Some participant comments appeared to reflect courtesy bias, especially those in which participants explained how their behavior was socially appropriate and concurrent with religious teachings. Better pretesting of the discussion guide might have identified questions that were vague or poorly interpreted by participants. A random sample of study participants may have yielded different results than one in which participants were purposively selected. It is difficult to know if randomly selected participants were likely to have opinions typical of their counterparts or if selecting participants according to their experience and willingness to discuss sensitive topics would have strengthened study results. Finally, like all studies reliant upon self-reported data for sexual activity, participants may have under- or overrepo rted certain behaviors. The findings from this study have several research and programmatic implications. Using qualitative research to inform survey instruments can strengthen data collection and improve the quality of data available for program development. Surveys should assess more than basic awareness and knowledge about STIs and HIV/AIDS: They should uncover conflicting beliefs, local terminology, and folk beliefs surrounding infection. Questions should also address stigma and denial by asking youth about community perceptions of STIs and HIV/AIDS as well as differences between social expectations of youth and their actual behavior. Using scales rather than traditional question design can measure dimensions of knowledge and expose a range in attitudes and beliefs about risk for infection. Several opportunities are available for targeting youth and relaying accurate information to prevent STIs and HIV/AIDS. Given youths reliance upon interpersonal contacts for information, peer-based interventions should be enhanced to ensure that youth have access to trusted and well-informed sources of advice. Communication campaigns should also target adults, equipping them with accurate information and encouraging them to talk openly with youth about STIs and HIV/AIDS. Mass media messages should be more accessible to populations with limited education and distributed in local languages. Likewise, media messages should complement face-to-face interventions in which participants have an opportunity to ask trained agents questions about STI/HIV transmission and prevention. Communication campaigns should correct youths incomplete knowledge about STIs and HIV/AIDS and challenge their misconceptions. Messages should communicate the availability of HIV and STI testing and the importance of knowing ones serostatus. Programs should also present science-based explanations for infection without dismissing local beliefs. For example, campaigns could present information that works within local understandings of health and illness concerning the difference between HIV and AIDS, physical manifestations of illness, and viral loads in blood. Such a process could ensure that messages are culturally appropriate and accessible to audiences. Programs should also work with communities and leaders to decrease the stigma associated with STIs and HIV infection. Finally, programs should collaborate with communities to create long-term strategies for increasing youths access to educational and economic opportunities, rendering them unwilling to sacrifice personal achievement for STIs or HIV infection.

6.2-84

In-Depth Interviews 19

Chapter 6.2

Reference List
Agha, Sohail. 2001. An Evaluation of the Effectiveness of a Peer Sexual Health Intervention among Secondary School Students in Zambia. AIDS Education and Prevention 14(4): 269-281. Agha, Sohail. 2000. Changes in Casual Sex and Condom Use in Lusaka, Zambia: 19961999. Working Paper N. 33. Washington, DC: PSI. Alan Guttmacher Institute (AGI). 1998. Sexually Transmitted Diseases Hamper Development Efforts. Issues in Brief, 1998 Series, No. 2. New York: AGI. Astatke, Hibist and Robert Serpell. 2000. Testing the application of a western scientific theory of AIDS risk behavior among adolescents in Ethiopia. Journal of Pediatric Psychology 25(6): 367- 379. Bandura, Albert. 1994. Social cognitive theory and exercise of control over HIV infection. In Prevention AIDS: Theories and Methods of Behavioral Interventions, (eds. DiClemente, Ralph J., and John L. Peterson) New York: Plenum Press, pp. 25-59. Bernardi, Laura. 2002. Determinants of Individual AIDS Risk Perception: Knowledge, Behavioural Control, and Social Influence. Max Planck Institute for Demographic Research Working Paper 2002-029. Rostock, Germany: MPIDR. Central Statistical Office (CSO) Zambia, Central Board of Health Zambia, and MEASURE DHS+ (ORC Macro). 2002. Zambia Demographic and Health Survey, 2001-2002: Preliminary Report. Lusaka Zambia: CSO and ORC Macro. Central Statistical Office (CSO), Zambia Ministry of Health, and MEASURE/Evaluation. 2002. Zambia Sexual Behaviour Survey 2000. Lusaka, Zambia: CSO. Dowsett, G. and P. Aggleton. 2000. Young People and Risk Taking in Sexual Relations. Retrieved November 12, 2002 http:\\www.unaids.org/publications/documents/ children/index.html Feldman, Douglas A, Peggy OHara, K.S. Baboo, Ndashi W. Citalu, Ying Lu. 1997. HIV prevention among Zambian adolescents: developing a value utilization/norm change model. Social Science and Medicine 44(4): 455-468. Fetters, T., E. Mupela, N. Rutenberg. 1998. Youth Talk about Sexuality: A Participatory Assessment of Adolescent Sexual and Reproductive Health in Lusaka, Zambia. Lusaka, Zambia: Care International. Fishbein, Martin, Susan E. Middlestadt, and Penelope J. Hitchcock. 1994. Using information to change sexually transmitted disease-related behaviors. In Prevention

6.2-85

PSI/AIDSMark Social Marketing Research Tool Kit 2004 AIDS: Theories and methods of behavioral interventions, (eds. DiClemente, Ralph J., and John L. Peterson). New York: Plenum Press, pp. 61-78. 20 Fylkesnes, Knut, Rosemary M. Musonda, Moses Sichone, Zacchaeus Ndhlovu, Francis Tembo and Mwaka Monze. 2001. Declining HIV prevalence and risk behaviours in Zambia: evidence from surveillance and population-based surveys. AIDS 15: 907-916. Johns Hopkins University Center for Communication Programs (JHUCCP). 2001. Youth Crucial to Stopping HIV/AIDS. Population Reports, Series L, No. 12. Baltimore, MD: JHUCCP. Retrieved November 1, 2002 http://www.jhuccp.org. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2002a. Report on the Global HIV/AIDS Epidemic (July 2002). Geneva, Switzerland: UNAIDS. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2002b. Zambia: Epidemiological Fact Sheet on HIV/AIDS and Sexually Transmitted Infections. (2002 update). Geneva, Switzerland: UNAIDS. Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO). 1997. Sexually Transmitted Diseases: Policies and Principles for Prevention and Care. Geneva, Switzerland: UNAIDS. Kaiser Family Foundation. 2002. HIV/AIDS Policy Fact Sheet: The Global Impact of HIV/AIDS on Youth (May 2002). Menlo Park, CA: Henry J. Kaiser Family Foundation. Retrieved August 16, 2002 http://www.kff.org/content/2002/6039/Global_HIVAIDS_Youth_ Fact_Sheet.pdf. Kelly, MJ. The Encounter between HIV/AIDS and Education. 2000. Lusaka, Zambia: University of Zambia. Kirby, Douglas. 1999. Antecedents of Adolescent Sexual Risk-taking, Pregnancy and Childbearing: Implications for Research and Programs. ETR Associates. Kirby, Douglas. 1997. No Easy Answers: Research Findings on Programs to Reduce Teen Pregnancy. Washington D.C.: National Campaign to Prevent Teen Pregnancy. MacIntyre, Kate, Lisanne Brown, and Stephen Sosler. 2001. Its not what you know, but who you knew: Examining the relationship between behavior change and AIDS mortality in Africa.AIDS Education and Prevention 13(2): 160-174. Magnani, Robert J., Ali M. Karim, Lisa A. Weiss, Katherine C. Bond, Musonda Lemba, and Gwendolyn T. Morgan. 2002. Reproductive health risk and protective factors among youth in Lusaka, Zambia. Journal of Adolescent Health 30: 76-86. Monitoring the AIDS Pandemic (MAP) Network. 2002. The Status and Trends of the HIV/AIDSEpidemics in the World (Provisional Report July 7, 2002). Barcelona, 6.2-86

In-Depth Interviews

Chapter 6.2

Spain:The Joint United Nations Programme on HIV/AIDS (UNAIDS) and United States Agency for International Development (USAID). Mshana, G., J. Wamoyi, M. Plummer, D. Wight, D. Ross, and Z Salam. 2002. Perceptions of AIDS among Young People in Rural Mwanza, Tanzania. Presentation at 14th International AIDS Conference (July 712, 2002). Barcelona, Spain. Ndubani, Phillimon and Bengt Hojer. 2001. Sexual behaviour and sexually transmitted disease among young men in Zambia. Health Policy and Planning 16(1): 107-112. Nzioka, C. 1996. Lay perceptions of risk of HIV infection and the social construction of safer sex: some experiences from Kenya. AIDS Care 8(5): 565-579. Peruga, Armando and David D. Celentano. 1993. Correlates of AIDS knowledge in samples of the general population. Social Science and Medicine 36(4): 509-524. Population Council. 2001. Reducing HIV infection among youth: what can schools do? Horizons Report, Fall 2001. Washington, DC: Population Council. Population Reference Bureau (PRB). 2001. Youth in Sub-Saharan Africa: A Chartbook on Sexual Experience and Reproductive Health. Washington, DC: PRB. Prochaska, Thomas.R., Gary Albrecht, Judith A. Levy, Noreen Sugrue, Joung-Hwa Kim. 1990. Determinants of self-perceived risk for AIDS. Journal of Health and Social Behavior 31(4): 384- 94. Rosenstock, Irwin M., Victor J. Strecher, and Marshall H. Becker. 1994. Changing HIV/AIDS risk behaviors: the role of behavioral interventions. In Prevention AIDS: Theories and Methods of Behavioral Interventions, (eds. DiClemente, Ralph J., and John L. Peterson). New York: Plenum Press, pp. 524. Summers, Todd, Jennifer Kates, and Gillian Murphy. 2002. The Tip of the Iceberg: The Global Impact of HIV/AIDS on Youth. Menlo Park, CA: Henry J. Kaiser Family Foundation. Retrieved August 16, 2002 http://www.kff.org/content/2002/6043/. Sweat, Michael D. and Julie A Denison. 1995. Reducing HIV incidence in developing counties with structural and environmental interventions. AIDS 9(suppl A): S251-S257. Trigg, Caroline. 2001. Investigation of Generic Issues Regarding Perceptions of Condoms and Risk from HIV amongst Urban and Semi-Rural Youth in Zimbabwe. Target Research (Pvt) Limited on behalf of PSI/Zimbabwe. TVT Associates. 2002. HIV/AIDS in Zambia: A USAID Brief. Washington, DC: USAID and Synergy Project. United Nations Childrens Fund (UNICEF), Joint United Nations Programme on HIV/AIDS (UNAIDS), and World Health Organization (WHO). 2002. Young People and HIV/AIDS: Opportunity in Crisis. New York: UNICEF. 22 6.2-87

PSI/AIDSMark Social Marketing Research Tool Kit 2004 United States Agency for International Development (USAID) and Implementing AIDS Prevention and Care (IMPACT) Project. 1999. Zambia and HIV/AIDS: Key Talking Points. Washington, DC: USAID, Population, Health and Nutrition Programs, HIV/AIDS Division. Warwick, Ian and Peter Aggleton. 2001. Learning From What Young People Say about Sex, Relationships and Health. Safe Passages to Adulthood. London: Thomas Coram Research Unit, Institute of Education, University of London. Weiss, H.A., A. Buv, N.J. Robinson, E. Van Dyck, M. Kahindo, S. Anagonou, R. Musonda, L. Zekeng, L. Morison, M. Carael, M. Laga and R.J. Hayes for the Study Group on Heterogeneity of HIV Epidemics in African Cities. 2001. AIDS 15(suppl 4): S97-S108. PSI RESEARCH DIVISION WORKING PAPERS _____________________________________________________________________ ___ 1997 1 Meekers, Dominique and Ghyasuddin Ahmed. 1997. Pregnancy-Related School Dropouts in Botswana. 2 Meekers, Dominique and Ghyasuddin Ahmed. 1997. Adolescent Sexuality in Southern Africa: Cultural Norms and Contemporary Behavior. 3 Meekers, Dominique, Guy Stallworthy, and John Harris. 1997. Changing Adolescents Beliefs about Protective Sexual Behavior: The Botswana Tsa Banana Program. 4 Meekers, Dominique and Anne-Emmanule Calvs. 1997. Gender Differentials in Adolescent Sexual Activity and Reproductive Health Risks in Cameroon. 5 Meekers, Dominique and Anne-Emmanule Calvs. 1997. Main Girlfriends, Girlfriends, Marriage, and Money: The Social Context of HIV Risk Behavior in SubSaharan Africa. 6 Agha, Sohail. 1997. Sexual Activity and Condom Use in Lusaka, Zambia. 7 Davies, John and Sohail Agha. 1997. Ten Years of Contraceptive Social Marketing in Pakistan: An Assessment of Management, Outputs, Effects, Costs and CostEfficiency, 1987-1996. 8 Parker, Warren. 1997. Action Media: Consultation, Collaboration, and Empowerment in Health Promotion. 9 Meekers, Dominique. 1997. The Implications of Free and Commercial Distribution for Condom Use: Evidence From Cameroon. 6.2-88

In-Depth Interviews

Chapter 6.2

10 Calvs, Anne-Emmanule and Dominique Meekers. 1997. Gender Differentials in Premarital Sex, Condom Use, and Abortion: A Case Study of Yaound, Cameroon. 11 Meekers, Dominique, Michael Holscher, and Anemona Munteanu. 1997. Sexual and Reproductive Health Behavior among Romanian Adolescents: An Exploratory Narrative Research Analysis. 12 Meekers, Dominique, Ghyasuddin Ahmed, and Tinah Molathegi. 1997. Understanding Constraints to Adolescent Condom Procurement: The Case of Urban Botswana. 13 Meekers, Dominique. 1997. Going Underground and Going After Women: Combating Sexual Risk Behavior among Gold Miners in South Africa. 1998 14 Agha, Sohail and John Davies. 1998. Contraceptive Social Marketing in Pakistan: Assessing the Impact of the 1991 Condom Price Increase on Sales and Consumption. 15 Agha, Sohail. 1998. Is Low Income a Constraint to Contraceptive Use among the Pakistani Poor? 16 Meekers, Dominique. 1998. The Effectiveness of Targeted Social Marketing to Promote Adolescent Reproductive Health: The Case of Soweto, South Africa. 17 Eloundou-Enyegue, Parfait, Dominique Meekers, and Anne-Emmanule Calvs. 1998. From Awareness to Adoption: The Effect of AIDS Education and Condom Social Marketing on Condom Use in Tanzania (1993-1996). 18 Meekers, Dominique. 1998. Improving Condom Social Marketing in Malawi: Evidence from a Consumer Profile Survey. 1999 19 Van Rossem, Ronan and Dominique Meekers. 1999. An Evaluation of the Effectiveness of Targeted Social Marketing to Promote Adolescent and Young Adult Reproductive Health in Cameroon. 20 Shapiro, David and Dominique Meekers. 1999. The Reach of the SIDA dans la Cit AIDS Prevention Television Series in Cte dIvoire. 21 Agha, Sohail, Andrew Karlyn, and Dominique Meekers. 1999. The Promotion of Safer Sex Among High Risk Individuals in Mozambique. 22 Calvs, Anne E. 1999. Condom Use and Risk Perceptions among Male and Female Adolescents in Cameroon: Qualitative Evidence from Eda.

6.2-89

PSI/AIDSMark Social Marketing Research Tool Kit 2004 23 Van Rossem, Ron and Dominique Meekers. 1999. An Evaluation of the Effectiveness of Targeted Social Marketing to Promote Adolescent Reproductive Health in Guinea. 24 Agha, Sohail. 1999. Sexual Behavior of Truck-Drivers in Pakistan: Implications for AIDS Prevention Programs. 25 Agha, Sohail. 1999. Patterns of Use of the Female Condom in Lusaka, Zambia. 26 Agha, Sohail. 1999. Consumer Intentions to Use the Female Condom after One Year of Mass-Marketing 27 Meekers, Dominique and Edna Ogada. 1999. Explaining Discrepancies in Reproductive Health Indicators from Population-Based Surveys and Exit Surveys. 28 Meekers, Dominique. 1999. Patterns of Use of the Female Condom in Urban Zimbabwe. 2000 29 Agha, Sohail. 2000. An Evaluation of Adolescent Sexual Health Programs in Cameroon, Botswana, South Africa, and Guinea. 30 Adetunji, Jacob and Dominique Meekers. 2000. Consistency in Condom Use in the Context of HIV/AIDS in Zimbabwe. 31 Van Rossem, Ronan, Dominique Meekers, and Zacch Akinyemi. 2000. Condom Use in Nigeria: Evidence from Two Waves of a Sexual Behavior and Condom Use Survey. 32 Agha, Sohail, Thankian Kusanthan. 2000. Equity in Access to Condoms in Urban Zambia. 33 Agha, Sohail. 2000. Changes in Casual Sex and Condom Use in Lusaka, Zambia: 1996- 1999. 34 Meekers, Dominique. 2000. The Role of Social Marketing in STD/HIV Protection in 4,600 Sexual Contacts in Urban Zimbabwe. 35 Shapiro, David, Dominique Meekers, and Basile Tambashe. 2000. Exposure to the SIDA dans la Cit AIDS Prevention Television Series in Cote dIvoire, Sexual Behavior, and Condom Use. 36 Agha, Sohail. 2000. The Availability of Social Marketed Condoms in Urban Tanzania, from 1997 to 1999. 2001 37 Meekers, Dominique. 2001. Patterns of Condom use in Urban Males in Zimbabwe: 6.2-90

In-Depth Interviews Evidence from 4,600 Sexual Contacts.

Chapter 6.2

38 Agha, Sohail and Mwaba Chulu Nchima. 2001. HIV Risk Among Street and Nightclubbased Sex Workers in Lusaka, Zambia: Implications for HIV Prevention and Interventions. 39 Oladosu, Muyiwa and Olaronke Ladipo. 2001. Consistent Condom Use among Sex Workers in Nigeria. 40 Karlyn, Andrew S. 2001 The Impact of a Targeted Radio Campaign to Prevent STIs and HIV/AIDS in Mozambique 41 Agha, Sohail. 2001. An Evaluation of the Effectiveness of a Peer Sexual Health Intervention Among Secondary School Students in Zambia. 42 Meekers, Dominique and Megan Klein. 2001. Understanding Gender Differences in Condom Use Self-Efficacy among Youth in Urban Cameroon. 43 Hovig, Dana. 2001. The Conflict between Profits and Public Health: A Comparison of Contraceptive Social Marketing Models. 44 Agha, Sohail and Ronan Van Rossem. 2001. The Impact of Mass Media Campaigns on Intentions to Use the Female Condom in Tanzania. 45 Agha, Sohail. 2001. The Impact of the Kenya Social Marketing Program on Personal Risk Perception, Perceived Self-efficacy and on other Behavioral Predictors. 46 Meekers, Dominique, Megan Klein, Leger Foyet 2001. Patterns of HIV Risk Behavior and Condom Use among Youth in Yaound and Douala, Cameroon. 47 Meekers, Dominique and Megan Klein. 2001. Determinants of Condom Use among Unmarried Youth in Yaound and Douala, Cameroon. 48 Meekers, Dominique and Megan Klein. 2001. Patterns of Sexual Behavior and Condom Use among High School and University Students in Butare and Gitarama Province, Rwanda. 2002 49 Agha, Sohail, Thankian Kusanthan, Kim Longfield, Megan Klein and John Berman. 2002. Reasons for Non-Use of Condoms in Eight Countries in Sub-Saharan Africa. 50 Agha, Sohail and Ronan Van Rossem. 2002. The Impact of a School-Based Peer Sexual Health Intervention on Normative Beliefs, Risk Perceptions and Sexual Behavior of Zambian Adolescents. 51 Longfield, Kim, Megan Klein and John Berman. 2002. Criteria for Trust and how Trust affects Sexual Decision-Making among Youth. 6.2-91

PSI/AIDSMark Social Marketing Research Tool Kit 2004 52 Longfield, Kim, Anne Glick, Margaret Waithaka and John Berman. 2002. CrossGenerational Relationships in Kenya: Couples Motivations, Risk Perception for STIs/HIV and Condom Use.

Other Interview Resources Available from PSI Research Division Interview Guides In-depth interview guide from CMS Nepal family planning program (English) In-depth interview guide from Togo on knowledge about STIs and HIV/AIDS (French) In-depth interview guide from Togo on sexual behavior (French) In-depth interview guide from Togo on knowledge & attitudes toward condom use (French) In-depth interview guide from Togo on condom use & reasons for non-use (French) Interview Reports Longfield, Kim, Anne Glick, Margaret Waithaka, and John Berman. CrossGenerational Relationships in Kenya: Couples Motivations, Risk Perception for STIs/HIV and Condom Use. Working Paper N. 52. PSI: Washington, DC. Richter, Kerry and Natacha Bobin. 2002. Sweetheart Relationships in Cambodia: Love, Sex, and Condoms in the Time of HIV. Report prepared for PSI/Cambodia, Phnom Penh, Cambodia.

6.2-92