An

Assignment on

Errors in Research, Features of Good Research

and Research Design Process

Submitted to: Submitted by:

Dr. Mahesh Sharma Vinita
Assistant Professor Department of ECE
Department of Mechanical Engg.

GURU JAMBHESHWAR UNIVERSITY OF SCIENCE & TECHNOLOGY

HISAR-125001
Q1. Briefly explain errors in research.
 Population Specification Error—This error occurs when the researcher does
not understand who they should survey. For example, imagine a survey about
breakfast cereal consumption. Who to survey? It might be the entire family,
the mother, or the children. The mother might make the purchase decision,
but the children influence her choice.

 Sample Frame Error—A frame error occurs when the wrong sub-population
is used to select a sample. A classic frame error occurred in the 1936
presidential election between Roosevelt and Landon. The sample frame was
from car registrations and telephone directories. In 1936, many Americans did
not own cars or telephones, and those who did were largely Republicans. The
results wrongly predicted a Republican victory.

 Selection Error—This occurs when respondents self-select their participation

in the study – only those that are interested respond. Selection error can be
controlled by going extra lengths to get participation. A typical survey process
includes initiating pre-survey contact requesting cooperation, actual
surveying, and post-survey follow-up. If a response is not received, a second
survey request follows, and perhaps interviews using alternate modes such as
telephone or person-to-person.

 Non-Response—Non-response errors occur when respondents are different

than those who do not respond. This may occur because either the potential
respondent was not contacted or they refused to respond. The extent of this
non-response error can be checked through follow-up surveys using alternate
modes.

 Sampling Errors—These errors occur because of variation in the number or

representativeness of the sample that responds. Sampling errors can be
controlled by (1) careful sample designs, (2) large samples, and (3) multiple
contacts to assure representative response.
 Q2. What are the features of a good research?
Features of Research are:
1. Empirical - based on observations and experimentation on theories.
2. Systematic - follows orderly and sequential procedure.
3. Controlled - all variables except those that are tested/experimented upon are kept
constant.
4. Employs hypothesis - guides the investigation process
5. Analytical - There is critical analysis of all data used so that there is no error in their
interpretation
6. Objective, Unbiased, & Logical - all findings are logically based on empirical
7. Employs quantitative or statistical methods - data are transformed into numerical
measures and are treated statistical.

Criteria of a Good Research

Good Research: How to Identify the Right Topic?

Every research study, irrespective of its type, should meet some criterions so that it
can be classified as good research. Let us discuss some of these criterions.

1. One of the important characteristics of a good research is that the purpose of the
research is clearly defined. A research study with clearly defined purpose finds
a wider acceptance and acknowledgement within the research community.
2. Second important characteristic of a good research is that the research method
should be defined in a clear manner with sufficient detail. This will allow the
repetition of the study in future for further advancement, while maintaining the
continuity of what has been done in the past.
3. The third thing to remember is that any limitations and assumptions made by the
researcher during the course of the study should be clearly highlighted in the
 research. This will support the findings of the research study, in case someone
tries to validate the study findings.
4. The fourth thing to remember is that, as far as possible, the research design
should be planned in a way that the results generated are as objective as possible.
This will provide an easier understanding about the findings of the research.
5. Another thing to be considered by the researcher is that there should be sufficient
data to investigate the research topic. And the researcher should carefully check
the reliability and validity of the data.
6. Further, in order to deliver a good research, a researcher should confine the
conclusions to those justified by the data.
7. Lastly, a good research depends a great deal on the integrity and commitment of
the researcher.
Q3. Explain Research Design Process.
Once the objective of research is clear, the research process enters the Research
Design phase. In this phase, the researcher will have to detail a plan in which
alternatives are going to be chosen at each of the following stages of research.
1. Selection of the type of research
2. Selection of the measures and the measurement techniques.
3. The kind and the number of subjects sampled, that is, sample design.
4. Selection of the data collection procedures.
5. The selection of methods of Analysis of data.
Selection of the Type of Research
The type of research determines the strategy of collecting research data and
information directly from the respondents by one of the following methods.
A. Field survey research in this method data/ information are obtained from a
general study of a large number of respondents (large sample), eliciting direct
responses to specific questions. The responses may be collected orally, in face-to-
face meetings, or remotely, through mail in written form or telephone in oral form.
The data obtained can be generalized and is representative of a large population.
These services are generally carried out at a particular point of time and are cross-
sectional. They produce ‘thin’ data. They are generally used for descriptive analysis,
correlational type hypothesis testing, and often for exploration.
B. Field study this is an in-depth study of a single respondent or a very small number
of respondents to obtain rich data of specific instances by using methods of probing.
It uses combinations of several Data Collection procedures. It provides greater
understanding of specific instances but its results cannot be easily generalized like
research survey results.
C Experiment In an experiment there is controlled manipulation of one or more
independent variables so that its effects on one or more dependent variables can be
measured. There are two major type of experimental designs.
1. Basic design this considers the effects of manipulating only one independent
variable
2. Statistical design this considers the effect of manipulating concomitantly
more than one independent variable.
There are three kinds of experiments
i) Laboratory experiments: experiments in which manipulation of independent
variable is carried out in an artificial environment away from the location of the
phenomena.
ii) Simulation experiments: selective manipulation of independent variable of a
model of the phenomena or system is carried out.
iii) Experiments in field setting: the manipulation of independent variable is
carried out in the natural setting but control is not as rigorous as in laboratory
experiments.
Experiment are used for finding cause-effect relationship, which is one of the major
objectives of science research.
Mathematical models of a phenomena
Another type of research assuming great importance in management problems is the
mathematical modelling approach in which the decision problem is represented by
a mathematical model, solution are obtained, and conclusions are drawn from that.
A high degree of understanding of the systems is necessary to model them. In this
sense, they are closer to experiments in traditional research they are only
representative and are Approximate and they have to be verified, tested, and
validated before drawing any interferences from them. The data collection processes
are similar to the one in other types. Secondary and primary data are used for
modelling purposes.
2. Measurement and Measurement techniques
Measurement is defined as the assignment of numbers to characteristics of objects,
persons, states or events according to rules. The most critical aspect of measurement
is the development of rules for assigning numbers to the characteristics. This
problem is particularly tricky and difficult in social science and organisational
research in which the definition of concepts and variables are often neither easy nor
direct. To overcome this difficulty, many techniques of measurement have been
developed by Social Science researchers. These are very relevant in Management
Research too.
Technique of measurement Based on the Research Design characteristics and the
kind of variables defined, a suitable measurement technique is selected. In general,
there are four techniques available to a researcher- questionnaires, attitude scales,
observation sheets, and depth interview schedules.
Questionnaire this is a set of questions, used as an instrument for seeking
information directly from respondents. The questions pertain to one or more
characteristics of the respondent like behavior, demographic characteristics,
knowledge, opinions, attitudes, beliefs, and feelings. Generally a question or a set of
questions represent a variable used in research. These are usually specially designed
for a particular Research and then suitably validated for use. However in many
studies standard inventory/ test designed and tested by others may also be used
Attitude scales these scale elicit self-reports of belief and feelings towards an
object. There are different type of attitude scale
1 rating scales that requires the respondent to place the object at some point on a
Continuum that is numerically ordered
2 composite scales require a respondent to express a degree of belief with regard to
several attributes of an object
3 multidimensional scale and skills developed using conjoint analysis are
mathematically developed scales to be used for interfering specific aspects of an
individual's attitude towards an object as against direct evaluation of the respondent
Observation: this is the direct examination of behavior or results of behavior.
Depth interviews: these are interviews in which individuals are made to express
their feelings freely and without fear of dispute or disapproval. The details are
recorded in specifically designed sheets.
Errors of measurement
A number of errors tend to vitiate measurement. The researcher has to ensure that
the desired accuracy levels are achieved by conducting suitable tests. The errors in
measurement can be systematic or variable. Systematic errors, which are consistent,
constitute the bias in measurement. Validity refers to bias and is the degree to which
measurement is free from systematic error. The variable error is associated with each
replication of measurement and the term reliability refers to variable error. It is
defined as the extent to which the measurement is free of variable error.
Therefore, unless a pre-evaluated and reliable instrument is used for data collection
the validity and reliability of a measurement technique or instrument designed by
the researcher must be established.
Selection of sample
The next step in Research Design in the selection of a sample of subjects for study.
In most cases of research, sampling is needed. Sampling is a necessary and an
inescapable part of any human activity life purchasing commodities, selection of a
television program to watch or even a book to read. If the population is small enough
instead of sampling a Census can be carried out. But usually population a large and
there is limited time and resources available with the researcher for data collection.
Therefore selecting a sample become necessary. Further, the sampling by reducing
the data collection effort makes data collection more efficient and accurate.
In spite of statistical methods being used in the selection of a sample, judgement is
Central to all stages of sampling. Sampling designs are aimed at two major
objectives (i) The sample is representative of the population (ii) the size of the
sample is adequate to get the desired accuracy. In general the sampling process
consists of
 Definition of the target population in terms of elements, sampling units,
domains, and period.
 Specification of is frame of sampling if probability sampling is used for
example telephone directories, map or listings.
 Specified sampling unit (e.g. a film, department, group, or an individual that
is addressed in sample).
 The sampling method(e.g. probability versus non probability)
 Determination of sample size, which is the number of elements in the sample,
using statistical method but open motivated by judgement based on other
consideration like availability, cost and accessibility
 Implementation of the sampling plan by ensuring the various controls
required in the field to attend the sampling objective and by contacting the
sample members.
Selection of Data Collection procedure
The data collection phase itself is considerably developed in contemporary
research. Data Collection will involve the development of the instruments for Data
Collection, identification of sources of data, and the context in which the sampling
has to be done. The sources of data are usually people and existing records. To get
information from people, it is either necessary to use interviews, where the
information may be given readily, or questionnaires, where the information may
have to be given after careful reflection on the part of the respondent. There are
several processes of Data Collection available to the researcher. Depending on the
problem, he may choose one or a combination of more than one procedure.
There are two sources of data- secondary data and primary data.
Secondary data: This kind of data is generated for purposes other than for solving
the problem under study. There are three methods of obtaining secondary data:
1. The data is available in published research journals, reports and books open
to the public in libraries.
2. Search of data generated within the organization through reports, log books,
records of union, minutes of meetings, proceedings, accounting documents,
home journals and so on.
3. Computer search of databases and the world wide web
Primary data: primary data the processes used for collecting primary data in a
research study are those of the research type already discussed. They are, (a)
questionnaire mail survey, (b) interviews of several kinds, (c) observation of
phenomena or subject,(d) special techniques like video/audio recording or
projective methods.
In general a researcher main use a pure strategy (one single type research) or a
combination of few types as in mixed design. For example cross sectional research
may be repeated at many points of time in a longitudinal study. As exploration may
be used before a descriptive study or a field study or an experiment.
Selection of methods of analysis
Data is useful only after analysis. Data analysis deals with the conversion of a series
of data gathered into information statements
1. Which descriptively state the information in terms of the means, percentage
,classification or distribution, or
 2. which make assertion about relationships conjectured prior to data collection,
or
3. which provide estimates of the purpose of prediction.
The selection of methods or techniques of analysis must generally precede the
collection of data in any good research. Dummy data may be used with the designed
instrument and subjected to analysis as per the selected methods to test whether the
results provide the desired information for the solution of the problem at hand. There
are a large number of statistical methods available analyzing the research data
collected.

Data analysis method in general

 Data analysis aims at the levels of variables and their variability when is single
variable is used in the analysis. Univariate hypothesis testing is a typical
analysis. An important aspect is when small samples of data are used, non-
parametric tests are used, and parametric tests are used on large samples.
 It aims at associations in the case of two variables. Correlation and regression
analysis are performed and the significance of regression or correlation
Coefficient are tested to confirm the results.
 It aims at dependent relationship (in general a set of independent variables)
or interdependence relationship (a set of independent variables are present but
there is no dependent variable). The former are more useful to establishing
relationships and the letter in developing concepts and constructs.
Four basic processes are involved they are as follows:
A. Data reduction this includes the following steps-
1. Field controls to minimize error in data collection
2. Editing to ensure readable and accurate data
3. Coding to categories the edited data
 4. Transferring data to usable media like tapes
5. Generation of new variables by aggregation, scale changing, and data
transformation and
6. Calculating summary statistics like the mean, standard deviation, proportion
and so on.
B. Hypothesis testing this includes-
1. Hypothesis testing requires interval data; test for single sample- sample mean
and sample proportion; tests for multiple samples involving differences in
means, differences in proportions of both Independent and related sample
2. Hypothesis testing using ordinal data
3. Hypothesis testing using nominal data
4. Multivariate hypothesis testing, including hypothesis tests of difference
between groups to test interaction effects, for example, ANOVA in
experimental situations.
C. Bivariate measures of association this includes -
1. Simple correlation and regression analysis using ratio/ interval data
2. Rank correlation analysis using ordinal data
3. Contingency Coefficient determination for nominal data
D. multivariate measures of association the flow diagram relates to the choices of
various multivariate measures of association. The diagram is self-explanatory. The
key equations to be answered in order to make the choice of a particular technique
are (1) the number of independent variables (2) whether there is a dependent
variable, (3) the level of measurement; nominal (category) ordinal (ranked) or
internal or ratio, with respect to independent variables in the analysis of dependence
relationship, and (4) The level of measurement, nominal, ordinal or interval of the
independent variable whose interdependence is to be analyzed. A brief outline of
each technique is given below

 Multiple regression analysis examines relationships between two or more

interval is scaled predictor variables and one interval is scaled Criterion
variable (ordinal data that are near interval can also be used). This is an
extension of the bivariate regression analysis
 Discriminant analysis is used in place of regression analysis when the
Criterion variable is normally scaled and the predictor variables are interval is
called the objective is to group the Criterion variable into two or more
category is like good or bad, high risk or low risk and so on.
 Path analysis is a technique for refining casual relationships in theory
building or understanding influencing factors. It uses a series of regression
analysis conducted simultaneously to determine if a set of proposed causal
relationship exist in a sample data.
 Factor analysis is helpful in summarizing a large number of original
variables into a smaller number of factors in order to achieve parsimony in
presenting phenomena. It can help (i) in determining underline dimensions of
data (ii) in condensing and simplifying data and (iii) in hypothesis testing and
 the structuring of 02data that is a set of variables come from a specified factor.
It does not use Criterion and predictor variables on their relationship. it the
relationship among is set of variable
 Cluster analysis is used in segregating object into groups such that the group
are relatively homogeneous. Examples are grouping of product and market
segmentation.