Analyzing and Interpreting Data

Data analysis and interpretation is the process of assigning meaning to

the collected information and determining the conclusions, significance, and imp
lications of the findings. The steps involved in data analysis are a function of
the type of information collected, however, returning to the purpose of the ass
essment and the assessment questions will provide a structure for the organizati
on of the data and a focus for the analysis.
The analysis of NUMERICAL (QUANTITATIVE) DATA is represented in mathematical ter
ms. The most common statistical terms include:
Mean The mean score represents a numerical average for a set of responses.
Standard deviation The standard deviation represents the distribution of the res
ponses around the mean. It indicates the degree of consistency among the respons
es. The standard deviation, in conjunction with the mean, provides a better unde
rstanding of the data. For example, if the mean is 3.3 with a standard deviation
(StD) of 0.4, then two-thirds of the responses lie between 2.9 (3.3 0.4) and 3.
7 (3.3 + 0.4).
Frequency distribution Frequency distribution indicates the frequency of each re
sponse. For example, if respondents answer a question using an agree/disagree sc
ale, the percentage of respondents who selected each response on the scale would
be indicated. The frequency distribution provides additional information beyond
the mean, since it allows for examining the level of consensus among the data.
Higher levels of statistical analysis (e.g., t-test, factor analysis, regression
, ANOVA) can be conducted on the data, but these are not frequently used in most
program/project assessments.
The analysis of NARRATIVE (QUALITATIVE) DATA is conducted by organizing the data
into common themes or categories. It is often more difficult to interpret narra
tive data since it lacks the built-in structure found in numerical data. Initial
ly, the narrative data appears to be a collection of random, unconnected stateme
nts. The assessment purpose and questions can help direct the focus of the data
organization. The following strategies may also be helpful when analyzing narrat
ive data.
Focus groups and Interviews:
Read and organize the data from each question separately. This approach permits
focusing on one question at a time (e.g., experiences with tutoring services, ch
aracteristics of tutor, student responsibility in the tutoring process).
Group the comments by themes, topics, or categories. This approach allows for fo
cusing on one area at a time (e.g., characteristics of tutor level of preparatio
n, knowledge of content area, availability).
Documents
Code content and characteristics of documents into various categories (e.g., tra
ining manual policies and procedures, communication, responsibilities).
Observations
Data analysis and interpretation is the process of assigning meaning to the coll
ected information and determining the conclusions, significance, and implication
s of the findings. The steps involved in data analysis are a function of the typ
e of information collected, however, returning to the purpose of the assessment
and the assessment questions will provide a structure for the organization of th
e data and a focus for the analysis.
Numerical vs. Narrative Quantitative vs. Qualititative
The analysis of numerical (quantitative) data is represented in mathematical ter
ms. The most common statistical terms include:
Mean The mean score represents a numerical average for a set of responses.For a
data set, the terms arithmetic mean, mathematical expectation, and sometimes ave
rage are used synonymously to refer to a central value of a discrete set of numb
ers: specifically, the sum of the values divided by the number of values. If the
data set were based on a series of observations obtained by sampling from a sta
tistical population, the arithmetic mean is termed the sample mean to distinguis
h it from the population mean.
Standard deviation The standard deviation represents the distribution of the res
ponses around the mean. It indicates the degree of consistency among the respons
es. The standard deviation, in conjunction with the mean, provides a better unde
rstanding of the data. For example, if the mean is 3.3 with a standard deviation
(StD) of 0.4, then two-thirds of the responses lie between 2.9 (3.3 0.4) and 3.
7 (3.3 + 0.4).
Frequency distribution Frequency distribution indicates the frequency of each re
sponse. For example, if respondents answer a question using an agree/disagree sc
ale, the percentage of respondents who selected each response on the scale would
be indicated. The frequency distribution provides additional information beyond
the mean, since it allows for examining the level of consensus among the data.
Higher levels of statistical analysis (e.g., t-test, factor analysis, regression
, ANOVA) can be conducted on the data, but these are not frequently used in most
program/project assessments.
Try this course for a comprehensive instruction on data analysis for excel.
The analysis of narrative (qualitative) data is conducted by organizing the data
into common themes or categories. It is often more difficult to interpret narra
tive data since it lacks the built-in structure found in numerical data. Initial
ly, the narrative data appears to be a collection of random, unconnected stateme
nts. The assessment purpose and questions can help direct the focus of the data
organization. The following strategies may also be helpful when analyzing narrat
ive data.
Focus groups and Interviews:
Read and organize the data from each question separately. This approach permits
focusing on one question at a time (e.g., experiences with tutoring services, ch
aracteristics of tutor, student responsibility in the tutoring process). Group t
he comments by themes, topics, or categories. This approach allows for focusing
on one area at a time (e.g., characteristics of tutor level of preparation, know
ledge of content area, availability).
Documents
Code content and characteristics of documents into various categories (e.g., tra
ining manual policies and procedures, communication, responsibilities). This app
roach keeps your information organized and easily accessible when you
Observations
Code patterns from the focus of the observation (e.g., behavioral patterns amoun
t of time engaged/not engaged in activity, type of engagement, communication, in
terpersonal skills).
Data Interpretation and Analysis Techniques
The analysis of the data via statistical measures and/or narrative themes should
provide answers to your assessment questions. Interpreting the analyzed data fr
om the appropriate perspective allows for determination of the significance and
implications of the assessment.
Analysis of data is a process of inspecting, cleaning, transforming, and modelin
g data with the goal of discovering useful information, suggesting conclusions,
and supporting decision making. Data analysis has multiple facets and approaches
, encompassing diverse techniques under a variety of names, in different busines
s, science, and social science domains.
Data mining is a particular data analysis technique that focuses on modeling and
knowledge discovery for predictive rather than purely descriptive purposes. Bus
iness intelligence covers data analysis that relies heavily on aggregation, focu
sing on business information. In statistical applications, some people divide da
ta analysis into descriptive statistics, exploratory data analysis (EDA), and co
nfirmatory data analysis (CDA). EDA focuses on discovering new features in the d
ata and CDA on confirming or falsifying existing hypotheses. Predictive analytic
s focuses on application of statistical or structural models for predictive fore
casting or classification, while text analytics applies statistical, linguistic,
and structural techniques to extract and classify information from textual sour
ces, a species of unstructured data. All are varieties of data analysis.
Some Data Interpretation and Analysis Tips
Consider the data from various perspectives. Whatever your project may be or wha
tever data you have collected from your business its always best to ask what that
data means for various actors or participants.
Think beyond the data but do not stray too far from the data. Be mindful that yo
u are not making too much of your data or too little. Make the link between the
data and your interpretations clear. Base your interpretations in your research.
Make visible the assumptions and beliefs, or mental models, that influence your
interpretation. We each carry images, assumptions, and stories in our minds abou
t ourselves, others, the organizations we work in, etc. As a composite, they rep
resent our view of our world. Because these models are generally unarticulated,
i.e., below our level of our awareness, if left unexamined, these assumptions an
d beliefs can lead to incorrect interpretations. Reflect on your own thinking an
d reasoning. Individually and/or collectively list your assumptions about the in
quiry focus.
Take care not to disregard outlying data or data that seems to be the exception.
Data that is surprising, contradictory or puzzling can lead to useful insights
Code patterns from the focus of the observation (e.g., behavioral patterns amoun
t of time engaged/not engaged in activity, type of engagement, communication, in
terpersonal skills).
The analysis of the data via statistical measures and/or narrative themes should
provide answers to the assessment questions. Interpreting the analyzed data fro
m the appropriate perspective allows for determination of the significance and i
mplications of the assessment.