Unit VI: Data collection and analysis

6.1 Data and its types

Sources of primary and secondary data
6.2 Questionnaire- principles, components and types, format and
types
6.3 Research interviews- principles and types;
6.4 Sources of qualitative data- observations, participant
observations, focus groups
6.5 E-research using internet and websites to collect data from
individuals
Web surveys, e-mail surveys
6.6 Getting data ready for analysis; data processing; presenting data
in graphs and tables;
6.7 Statistical analysis of data- descriptive and inferential statics,
hypothesis testing
6.8 Methods of analyzing qualitative data.
 What is data?
• Data can be defined as the values collected through
record-keeping or polling, observing, or measuring.
• In other words, data is facts, texts, or numbers that
can be collected.
• Hence, data can not only be in the form of numbers
or quantitative but it may be another form such as
interviews, maps, photographs, and videotapes of
social interactions.
• Data is a building block of any research.
 6.1 Data and its types

1. Primary and secondary data

2. Qualitative and quantitative data
 Primary Data

Primary data is information collected by

researcher directly through instruments such as
survey, interview or observation.

A primary data is collected from an original

source such as eyewitness to an event, a film or
audio recording of an event and field visit etc.

Primary sources tend to be recent, technical,

and very detailed.

Obtaining primary data can be expensive and

time consuming.

Saturday, May 3, 2025

• Sources of primary data
• Primary data collection is necessary when a
researcher cannot find the data needed in
secondary sources.
• Or when the data extracted from secondary
sources are not reliable or correct
Methods:
1. Questionnaire
2. Observation
3. Interview
4. Focus group discussion
 Secondary Sources
• If a researcher uses the data developed by other in the
past for their own purpose is known as secondary
data.
• Secondary data can be obtained from published and
unpublished sources.
• As a general rule, no research should be conducted
without a search for secondary data.

• Search for secondary data should be undertaken

during the exploratory investigation of the problem.

6
SOURCE OF SECONDADRYDATA
i. Published printed source
• Books
• Journals/periodicals
• Magazines/Newspapers
ii. Published Electronic Sources
• e-journals
• General websites
• Weblogs
iii. Unpublished Personal Records
• Diaries
• Letters
• Government Records
• Educational institutes
There are several reasons for the use
of secondary data:

• Secondary data may solve the

problem

• Secondary data is economical

• Secondary data help to define the problem and

formulate hypothesis.

8
• It can provide a better understanding of the
problem and its environment that may provide
solutions not considered previously.

• Secondary data also helps to plan the collection

of primary data.

• It is useful for selecting suitable research design

to solve the current problem.

9
MERIT OF SECONDARY DATA
• Use is very convenient.
• Saves time and finance.
• In some enquiries primary data cannot be collected.
• Longitudinal studies
• Comparative data
DEMERIT OF SECONDARYDATA
• Difficult to find secondary data which exactly fulfils the need of
present investigation.
• Extra caution is required to use secondary data.
• Not available for all types of enquiries.
• Collected for other purposes
• Unit of measurement may be different
• No control over data quality
• Access may be difficult
 Problems and precautions in the use of secondary
data

Availability and adequacy:

Organizations that have the data are found to be

reluctant to share the information with other
organizations.

The researcher may find them inadequate to solve the

current problem.

Working on the basis of incomplete information often

results in disastrous conclusions and this should
always be avoided.

Saturday, May 3, 2025

Relevancy:

The secondary data should be relevant to the problem.

It should fit the information needs of the current research.

The following precautions need to be undertaken while

evaluating the relevancy of the secondary data:

Unit of measurement

It should have the same unit of measurement adopted for the

current research.

Definition of classes

The secondary data should include the same demographic

characteristics as the current research.

Saturday, May 3, 2025

Time:

The secondary data should be recently collected and published.

Old data are irrelevant to solve the current problem.

Accuracy

The secondary data must be as accurate as possible.

It is important for the researcher to investigate the original

source of the data.

Precautions should be taken with regards to the quality of data

for which, the researcher should be confident about the
reputation of the organization and the data collection
procedure applied by the organization.

Saturday, May 3, 2025

 Questionnaire
• Questionnaire is a general term that includes
all techniques of data collection in which each
person is asked to respond to the same set of
questions in a pre-determined order.
• It therefore includes both structured interviews
and telephone questionnaires as well as those
in which the questions are answered without
an interviewer being present.

14
 Questionnaire
• As each person (respondent) is asked to respond to the
same set of questions, it provides an efficient way of
collecting responses from a large sample prior to
quantitative analyses.
• Questionnaire can be used as the only data collection
method, but also it is usually better to link them with
other methods, such as in-depth interviews.
• The design of a questionnaire differs according to how
it is administered and the amount of contact you have
with the respondents.

15
16
 Questionnaire Design
• Objective of questionnaire is used to obtain
following information of respondents:
1. intention, 2. attitude, 3. demographic
characteristics, 4. behavior etc.
• Designing of questionnaire is technical job.
• Questionnaire should be straightforward so
that everyone can understand easily and
sufficient information can be obtained.

17
 Questionnaire Design…
Type of Questions:

Open Ended Questions– Allow the respondents to answer them in

any way they choose.
Closed Questions– Restrict respondents within the given
alternatives.
Structured Questions - Respondents have to answer within the given
alternatives.
Unstructured Questions - Respondents can give any answer as per
their desire. No restriction to answer.
Long vs. short questions – Long question increases the cost and time
as well as discourage respondent to complete the questionnaire.
18
 Questionnaire Design...
Wording – The word used in questionnaire should not be ambiguous
or complex. Easily understandable word to be used.
Questions should be logically ordered depending on the nature of the
research like:
• From general to specific
•From easy to difficult
•From Factual to abstract
Physical appearance:
Should be attractive, neat and clean, size of letter to be readable,
enough space for answer, appropriate introductions and instructions.
Also consider , quality of paper and size of paper, margin and spacing.

19
 Quality of Questionnaire
• Avoid vagueness and confusion :
Do you eat meat regularly?
Do you withdraw your deposit from this bank regularly?
• Avoid leading questions :
I am a leader of this college, I hope you will help me, Do you?
• Avoid double barrel questions : More than one questions
combined
Do you eat fruits and milk each day? Yes / No
Do you use ATM card and credit card of this bank? Yes / No
•Avoid technical words :
In your opinion which banks dividend payout ratio is higher?
20
 Quality of Questionnaire
• Avoid ‘recall dependent’ question:
In which subject did you obtain highest mark in class 8?
• Provide enough alternative:
Marital Status : Single and Married ( but if the research area is
on social issue other alternatives also should be used widow,
divorced and separated )
• Aware on sensitive issues social values & ethical issues
Religion, political belief, personal secrecy etc.
• End the questionnaire in a gentle and friendly manner
Express your gratitude to your respondents for their time and
effort

21
 Example : Likert Type Questions
Please check the number that corresponds to
your answer to the statement provided and
interpreted as follows. Please indicate the extent
to which you agree or disagree with the
following statement about the services you
received from your bank in your opinion.
[1=Strongly Agree, 2=Agree, 3=Neutral,
4=Disagree, 5=Strongly Disagree]

22
S.N. Reliability (timely and accurate) 1 2 3 4 5

a. The bank shows fair value

disclosure for loan.
b. The bank maintains error
free records.
c. The bank is able to deliver
all services in time.
d. The bank maintains every
transaction very
confidential.

23
 Contd…

S.N. Employee Motivation 1 2 3 4 5

a. The measurement system of performance

and incentives in the bank encouraged to
work well
b. The bank is keen to measure and reward
employee performance that contributes to
achieving its vision
c. The bank depends on the use of data and
information collected from staff to improve
their jobs
d. The bank informs the staff about the
importance of service roles that they do

e. Bank staffs who provide excellent services

get a good reward

24
 Contd…
S.N. Customer Satisfaction 1 2 3 4 5

a. The bank’s forms and slips are

easy to fill.
b. The bank provides timely service.

c. Bank charges are clearly

explained.
d. Internet banking and SMS
banking are easy to use
e. I am satisfied with the ATM
service of the bank.

25
 Features of Good Questionnaire
• Careful design of individual questions
• Clear layout of the questionnaire form
• Detailed explanation of the purpose of the
questionnaire
• Carefully planned and executed administration
• Linkages between previous and next questions
• In a sequential order

26
 Types of Interview
Interview means a purposeful discussion between
two or more people.
• Structured interviews
– use questionnaires and standardized questions
• Semi-structured interviews
– use non-standardized questions. Researcher uses
themes as a guide
• Unstructured interviews
– use non-standardized questions. Totally informal in
approach. Heavy use of open ended questions.
27
 Conducting Effective Interview

• opening the interview

• using appropriate language
• questioning and talking
• listening
• testing and summarising understanding
• noticing behavioural signals
• recording data

28
The interview method of gathering primary data
is not only popular but also very effective.

Interview method is adopted for investigating

factors such as opinions, attitudes, motivations
and perceptions.

There are three types of interview:

 Personal interview

 In-depth interview

 Focus group interview.

Saturday, May 3, 2025
 Personal Interview

Personal interview is a structured, face to face

interview where the interviewer conducts the
interview with a single individual at a time.

The interviewer uses the interview schedule as a

guide in questioning.

The interviewer uses notebook or audio or video for

recording the respondent's responses.

Sometimes, the interview may be conducted with a

structured questionnaire.

Personal interview can be conducted in three

modalities: in-home, telephone and intercept
interviews.
Saturday, May 3, 2025
 In-home Interview

The interviewer visits the respondent's place of

residence to conduct the interview.

This method is losing popularity in recent years on

account of several factors:

Difficulty in contacting the respondents at home

during the daytime as both the couple work
outside the home.

Interviewers are reluctant to visit respondents'

homes at night.

In-home interviews are also expensive and

difficult to manage.
Saturday, May 3, 2025
 Telephone Interview

Using telephone to conduct the interview is gaining

popularity in recent years.

This method has the advantages of low cost in data

collection.

Hundreds of telephone interviews can be conducted on

a single day.

Telephone surveys allow sampling through the use of

telephone directories.

Telephone interview can be recorded on an audio

recorder or directly on the computer.
Saturday, May 3, 2025
 Intercept Interview

The interview is conducted on the sidewalk in the

market place by intercepting the respondent.

This method is suitable for very short interviews,

possibly to be completed within five minutes.

In-depth Interview

In-depth interview is one-to-one unstructured

interview.

It is conducted for a prolonged period of time, usually

lasting two to four hours.
Saturday, May 3, 2025
Only a well-trained and experienced researcher can
conduct such interviews.

Out of the total interview time, the researcher

spends half of the time talking in the subject of the
interest of the respondent.

The interview is conducted without a discussion

guide although the researcher has the topic guide in
his memory.

The interview needs to be conducted in a quiet but

informal place. The interviewer has to put the topic
very covertly in between the interesting discussion.

Saturday, May 3, 2025

No recorders are used and all the responses
have to be recorded by the interviewer in his
memory system later to be retrieved and
noted down for analysis.

c) Focus Group Discussion

FGD can be conducted only among the

demographically homogeneous group, at least
similar in terms of gender, age and income or
social class.

Preferably, the gender of the moderator

should match the gender of the group.
Saturday, May 3, 2025
The interview is conducted with the help of a
discussion guide.

The moderator throws the issue to the group; the

group discusses the issue and come to a definite
conclusion.

The role of the moderator is to keep the discussion in

track and allow all group members to have their say in
the group.

Most FGD are tape or video recorded.

The discussions are later transcribed (written down)

for analysis.

Saturday, May 3, 2025

Principles of interviewing
1.Unbiaseness
2. Relevance (should not ask personal questions)
3. Good behavior
4.Flexibility
5. Do not give directions
6. Co-operative
7. Sincerity
8. Promote (positively add some words like yes, umm..
etc)
9.Control: limiting interview within the subject matter.
10. listening
 Participant Observation
In this method, researcher records the
respondent's overt behavior, taking note of
the physical conditions and events.
• complete participation
– researcher attempts to become a member of the group
being studied, and does not reveal true purpose
• complete observation
– watching what is going on, but does not become a
participant
• observer as participant
– the researcher has the role of a spectator/ he or she may
act as a member of the group; although he or she is really
not one. 38
 Advantages of participant observation

• Good for explaining what is going on in social

situations
• Heightened the researcher’s awareness of
significant social processes
• Provide the opportunity to experience the
emotions of those being studied
• All data collected are likely to be useful

39
 Disadvantages of Participant Observation

• Time consuming
• Poses difficult ethical dilemmas
• Can be high levels of conflict for the researcher
(are you a colleague or a researcher first?)
• Closeness to situation can lead to researcher
bias
• Access can be difficult
• Data recording can be difficult
40
 Observational tools
Non-mechanical tools

Observation diary

Checklist

Rating scales

Score cards

Mechanical tools

Camera

Audio and video recorders

Electric scanners

Counting machines

Saturday, May 3, 2025

 6.5 E-research using internet and websites to collect data

INTERNET
• As an umbrella term that includes the associated
terms cyberspace and the Web (World Wide
Web).
• Refer to the actual network and the exchange of
data between computers.
• Email has been the most frequently used
Internet tools in the survey research for its
convenience.
TYPES OF INTERNET SOURCES
• Youtube
• Skype (Online Conferencing)
• E-mail
• Web Survey Application
• Facebook
STRENGTHS
1. Communication facilitation
2. Easy access and convenience
3. Practical and economic benefits
4. Access to diverse participants from home
computers
LIMITATIONS
5. Technological limitations
6. Receiving of messages in a non chronological
sequence can disrupt the flow of topic
discussion
6.6 Getting data ready for analysis;
data processing; presenting data in
graphs and tables;
Data processing procedures are:
1. Editing: detects errors in the raw data and rectifies
the errors and simplifies to the act of coding is
known as editing.
2. Coding: act of assigning numbers or other symbols
to the responses of respondents so that the
responses can be grouped into a limited number of
categories.
3. Classification: data may be classified in different
forms such as age, geographical classification etc.
4. Tabulation: process of arranging data in a systematic
manner into rows and columns.
5. Summarizing of data
6.8 Methods of analyzing qualitative data.
In Qualitative Research:
 We do not test hypothesis or previous theories.
 We may try to develop new theories based on what
happens in specific situations.
 We do not try to generalize our findings.
 We often rely on data collected from interviews,
observations, and content analysis of newspapers,
books, videos, case records, and other already
developed documents.
 We usually do not know or try to develop response
categories prior to conducting the study. 47
Qualitative Research Tends To

 Answer research questions rather than

test a hypothesis.
 Examine the perceptions, actions, and
feelings of participants.
 Obtained detailed information from
interviews, content analysis, or
observations.

48
Steps in Analyzing Qualitative Data:
 Data reduction: inter-related data should be
integrated and non-related data is to be
removed to reduce the data.
 Data display: data are to be presented in a
certain format for the integration of data.
 Drawing conclusions
 Methods of analyzing qualitative data
1. Content analysis: Broad, general set of methods for
analyzing the content of some qualitative material to
build or support an argument.
Features:
i. Systematic
ii. Objectivity
iii. generalizability
Goal of content analysis are:
• Identify important aspects of the content
• Present them clearly and effectively
• In support of some argument
• That will persuade the reader and contribute to the field.
Guidelines for content analysis
1. Clear operational definition of the units of analysis.
2. Clear definition for the response category
3. Analysis of material
4. Decision on developing categories
5. Maintaining impartiality
6. The validity of the content analysis should be assessed
Limitations
7. non-reliable result
8. Difficult to categorize data
9. Difficult to get clear and appropriate information
10. Costly
11. Difficult to generalize
Steps:
12. Identify the essential data
13. Develop bases for tabulation
14. Develop bases for content analysis
15. Develop the layout for the construction of design
16. Classify various variables into various groups
17. Establish procedures for the use of materials
18. Prepare outline of analysis and utilizing them.
2. Narrative analysis
A technique of recording and analyzing the information and subject based on
the story of the respondents or people related to an event or subject matter.
Elements of narrative analysis
1. Understanding level: in this method respondents tell the stories on the
basis of their experience.
2. Data collection: through discussion, interview, FGD.
3. Analysis: can be analyzed by summaries of the stories, case wise
comparison etc.
4. People’s understanding over events
5. Key actors and events: (of their life) such as career, remoteness, relations
etc.
Steps:
6. Obtaining data
7. Focusing on analysis of data obtained from autobiography, interviews, FGD
etc.
8. Codify data using sign or symbols to the classes of data
9. Identify the relationship among the various classes.
3. Thematic analysis
It emphasizes pinpointing, examining, and recording patterns
(themes) within data.
It is a work of searching theme of the data, event or subjects
that is important for the description of the phenomena.
Thematic analysis is performed through the process of coding
in various phases to create established, meaningful patterns.
These phases are:
• familiarization with data,
• generating initial codes,
• searching for themes among codes,
• reviewing themes,
• defining and naming themes, and
• producing the final report.
• Reading and re-reading material until researcher is comfortable is crucial to the initial
phase of analysis. While becoming familiar with material, note-taking is a crucial part
of this step in order begin developing potential codes.
• The second step in thematic analysis is generating an initial list of items from data
set that have a reoccurring pattern. It is called coding. The coding process is rarely
completed the first time. Each time, researchers should strive to refine codes by
adding, subtracting, combining or splitting potential codes.
• In this phase, it is important to begin by examining how codes combine to form
themes in the data. Researchers begin considering how relationships are formed
between codes and themes and between different levels of existing themes. Themes
consist of ideas and descriptions within a culture that can be used to explain causal
events, statements, and morals derived from the participants' stories.
• This phase requires the researchers to search for data that supports or refutes the
proposed theory. This allows for further expansion on and revision of themes as they
develop. Some existing themes may collapse into each other, other themes may need
to be condensed into smaller units.
• After finalizing the themes to be discussed, researcher should name and define
themes.
• Finally, researchers begin the process of writing the final report. While writing final
report, researchers should decide on themes that make meaningful contributions to
answering research questions which should be refined later as final themes.