Unit 3 - Lena PDF

SOCIAL SCIENCE RESEARCH DESIGN
AND DATA ANALYSIS
PRESENTATION OUTLINE
1.
Introduction
2.
Social Science Research Methodologies
3.
Research Ethics
4.
Conclusion
SOCIAL SCIENCE RESEARCH -1
Social science research is concerned with people

and their life contexts in unpredicted fashion.
Examples:
A study of the socio-economic and environmental
effects of kerosene lanterns.
Food safety assurance in the Ghanaian food
manufacturing value chain.
Assessing the knowledge and practices of wayside

auto electricians in Ghana
3
SOCIAL SCIENCE RESEARCH- 2
The purpose of this unit is to introduce participants to the empirical

social science research.
Objectives of the unit:

Identify some of the research strategies available to researchers in
the social sciences;
Know the factors that affect the effectiveness of these research
designs; and
Operationalise these research approaches in ways that the

weaknesses do not limit the credibility of the research findings.
4
Social Science Research Strategies

1.
Survey Research
2.
Case Study
3.
Action Research
SURVEY RESEARCH DEFINED
Suited to research problems when the researcher aims to

describe the prevalence of a phenomenon or when there is the
need to be predictive about an outcome.
Allows for the drawing of a sample from a population of interest,

with the aim of generalising to the final population
It aims to collect information from a sample within a certain

degree of error.
Surveys tend to be quantitative in nature and usually use

structured and standardized questionnaires.
SURVEY RESEARCH PROCESS

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Clarify the purpose of the research

Resource assessment
Take a decision on the methods and procedures
Design the questionnaire
Pilot test the questionnaires
Prepare the sample
Train interviewers
Data collection
Process the data
Analyse the results
Interpret and disseminate results
Take action
7
DETAILED STEPS FOR CONDUCTING A SURVEY

RESEARCH
Step One: Clarify the Purpose of the Research
It spells out the reasons for conducting the survey.
It is necessary because clarity of the research purpose helps to know if

the survey is the most appropriate approach to use.

RESEARCH CONTD
Step Two: Resource Assessment
Evaluation of the adequacy of in-house resources to:

design a survey that is within the budget line.
know which resources one needs to contract out.
The resources include:

Staff availability and skills;
logistics (materials and equipment, etc).

9

RESEARCH CONTD
Step Three: Decision on the Methods
and Procedures
Should it be:
Administer face-to-face questionnaires?
Mailed and e-mailed questionnaires? or
Telephone and computerized telephone
interviews?
Administer Online?
10

RESEARCH CONTD
Step Four: Design the Questionnaire
Questionnaires are designed to address the research objectives or

questions.
Guidelines:
Language wording
Recall Bias
Order of the Questions
Length of the Questionnaire
11
Step four - contd
Type of Questions:
Structured Questions:Structured questionnaires consist of closed or
prompted questions with predefined answers. The researcher has to anticipate all
possible answers with pre-coded responses
Unstructured Questions:Unstructured questionnaires are made up

of questions that elicit free responses. These are guided conversations rather than
structured interviews and would often be referred to as a topic guide.
Contingency questions:Questions that are limited to a subset of

respondents for whom they are relevant are called "contingency questions
Matrix Questions:Identical response categories are assigned to

multiple questions
12
DETAILED STEPS FOR CONDUCTING A

SURVEY RESEARCH CONTD
Step Five: Pilot Test the Questionnaires
The pilot testing of survey instruments helps to:

Know if each question addresses the research questions;
Know if the questions are interpreted in a similar vein by
respondents;
Identify if options provided for close-ended questions are exhaustive;
Assess clarity and understandability of the questions;
Evaluate the time taken to administer a questionnaire; and
Know respondents reactions to some questions.
13

RESEARCH CONTD
Step Six: Preparation of the Sample
Representativeness of the sample is most relevant.
Guided questions:
How many will be included (the sample size)?
How will the survey respondents be selected?
Determining the Sample Size
Some relevant questions to consider:
What is the size of your target population?
What can the budget allow?
How confident do you need to be with the results?
Do you need to look at any subgroups?
14
Step Six: contd
The sample size is often determined statistically if the sample

frame in known.
In a frame of 10 000 students, @ 95% confidence level, and

a confidence interval of 0.05 the sample size can be
determined as follows:
Where n is the sample size; N is the sample frame and is the
confidence interval.
= 10 000 / 1 + 10 000 (0.05)2

n = 384.6
the sample size is 385 students.
15
Step Six: contd

Sampling Methods
Two methods:
1) Probability sampling every units has a known and non-zero
chance of being selected.
2) Non-probability Some units are selected while others are not.
There are several types of probability and non-probability sampling:
16
Sampling methods contd

Probability sampling Non probability
sampling
Random sampling
Snowball sampling
Stratified Sampling
Quota sampling
Cluster Sampling
Purposive Sampling
17
Merits and Demerits of the Sampling

Techniques
Technique Brief Descriptions
Advantages
Simple
Random sample from Highly representative if all
random
whole population
subjects participate; the ideal
Stratified
random
Cluster
Random sample from

identifiable groups
(strata), subgroups,
etc.
Can ensure that specific groups

are represented, even
proportionally, in the sample(s)
(e.g., by gender), by selecting
individuals from strata list
Random samples of Possible to select randomly when
successive clusters of no single list of population
subjects (e.g., by
members exists, but local lists do;
institution) until small data collected on groups may
groups are chosen as avoid introduction of confounding
units
by isolating members
Disadvantages
Not possible without complete list
of population members;
potentially uneconomical to
achieve; can be disruptive to
isolate members from a group;
time-scale may be too long.
More complex, requires greater
effort than simple random.
Clusters in a level must be

equivalent and some natural
ones are not for essential
characteristics (e.g., geographic:
numbers equal, but
unemployment rates differ)
18
Merits and Demerits of the Sampling

Techniques contd
Purposive
Hand-pick subjects on the basis Ensures balance of group

of specific characteristics
sizes when multiple groups
are to be selected
Quota
Select individuals as they come Ensures selection of adequate Not possible to prove
to fill a quota by characteristics numbers of subjects with
that the sample is
proportional to populations
appropriate characteristics representative of
designated population
Snowball
Subjects with desired traits or

characteristics give names of
further appropriate subjects
Possible to include members

of groups where no lists or
identifiable clusters even
exist (e.g., drug abusers,
criminals)
Samples are not

easily defensible as
being representative
of populations due to
potential subjectivity
of researcher
No way of knowing
whether the sample is
representative of the
population
19

Step Seven: Train Interviewers
This involves providing them with the skills needed to

undertake successful interviewing.
Interviewers have a tremendous amount of influence on the

quality of the research.
A good interviewer can make all the difference in the world to

the usefulness of the data collected.
20

Step Eight: Data Collection
Describes how the information is gathered from respondents.
This is an important step, that must be done right in order to ensure

the integrity of the information collected.
The following procedures are to be observed in data collection:

Face-to-face Interviews
Using Telephone Surveys
Using Mail Surveys
Online
21

RESEARCH CONTD
Step Nine: Processing the Data
This involves preparing and translating the data for analysis.
Involves putting the completed questionnaires into a format

that can be summarized and interpreted.
The procedure includes:

Coding
Data Entry
22

Step Ten: Analysis of Results
A crucial step in ensuring that the findings reflect

the opinions and views of respondents.
Both quantitative and qualitative methods are

used.
The qualitative inquiries capture areas where indepth information is required for better
understanding of issues.
The quantitative analysis is can be done using

SPSS, STATA or any other statistical software.
23

Step Eleven: Interpret and Disseminate
Results
It is important to feed back the results of the

survey to management, staff, interested
participants and other stakeholders.
This is to keep them informed and establish buy-in

for implementing any changes resulting from the
survey.
24

Step Twelve: Take Action
This refers to implementing the changes suggested

by the results of your survey.
It is important to take action and implement

changes in order to make improvements to
subjects understudied.
25
CASE STUDY RESEARCH
A case study is an in-depth

investigation of an individual, group,
institution or phenomenon.
Case studies are often based on the

premise that locating one case is
enough to make a conclusion for
other cases since a case can be
typified for similar other cases.
26
Purpose of Case Studies
The primary purpose of case study is to

determine factors that have resulted in
the behaviour understudy.
The investigation involves a detailed

examination of a single subject, group or
phenomenon.
27
TECHNIQUES FOR SELECTION OF

CASES IN CASE STUDY RESEARCH
Case selection in case study research has similar

objectives as random sampling.
In case selection, a researcher desires a

representative sample which has useful variation on
the dimensions of theoretical interest.
Ones choice of cases is therefore driven by the way

a case is situated along these dimensions within
the population of interest. That is, how the case fits
into the theoretically specified population.
28
Techniques for selection of cases

in Case Study Research contd
Cases should be selected in the same way as

the topic of an experiment is selected;
Developed preliminary theory is used as a

template with which to compare the
characteristics and empirical findings from
the case(s); and
Selected cases should reflect characteristics

and problems identified in the underlying
theoretical propositions/conceptual
framework.
29
ETHICS IN RESEARCH
Informed Consent
Competence
Privacy
Sensitivity of information
Settings being observed
Dissemination of the information
Anonymity and Confidentiality
30
SUMMARY
Research in the social sciences deal

directly with the human being.
Several approaches can be used to

investigate the subject.
The onus is on the researcher to use the

most appropriate method by clarifying
the purpose of the research.
It is important to observe the code of

ethics in research.
31
Thank you
32

Unit 3 - Lena PDF

Uploaded by

Copyright:

Available Formats

Unit 3 - Lena PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Unit 3 - Lena PDF

Uploaded by

Copyright:

Available Formats

SOCIAL SCIENCE RESEARCH DESIGN

AND DATA ANALYSIS

Social Science Research Methodologies

SOCIAL SCIENCE RESEARCH -1

Social science research is concerned with people

Assessing the knowledge and practices of wayside

SOCIAL SCIENCE RESEARCH- 2

The purpose of this unit is to introduce participants to the empirical

Objectives of the unit:

Operationalise these research approaches in ways that the

Social Science Research Strategies

SURVEY RESEARCH DEFINED

Suited to research problems when the researcher aims to

Allows for the drawing of a sample from a population of interest,

It aims to collect information from a sample within a certain

Surveys tend to be quantitative in nature and usually use

SURVEY RESEARCH PROCESS

Clarify the purpose of the research

DETAILED STEPS FOR CONDUCTING A SURVEY

It spells out the reasons for conducting the survey.

It is necessary because clarity of the research purpose helps to know if

DETAILED STEPS FOR CONDUCTING A SURVEY

Evaluation of the adequacy of in-house resources to:

know which resources one needs to contract out.

The resources include:

logistics (materials and equipment, etc).

DETAILED STEPS FOR CONDUCTING A SURVEY

DETAILED STEPS FOR CONDUCTING A SURVEY

Questionnaires are designed to address the research objectives or

Step four - contd

Unstructured Questions:Unstructured questionnaires are made up

Contingency questions:Questions that are limited to a subset of

Matrix Questions:Identical response categories are assigned to

DETAILED STEPS FOR CONDUCTING A

Step Five: Pilot Test the Questionnaires

The pilot testing of survey instruments helps to:

DETAILED STEPS FOR CONDUCTING A SURVEY

Step Six: contd

The sample size is often determined statistically if the sample

In a frame of 10 000 students, @ 95% confidence level, and

= 10 000 / 1 + 10 000 (0.05)2

Step Six: contd

There are several types of probability and non-probability sampling:

Sampling methods contd

Merits and Demerits of the Sampling

Random sample from

Can ensure that specific groups

Clusters in a level must be

Merits and Demerits of the Sampling

Hand-pick subjects on the basis Ensures balance of group

Subjects with desired traits or

Possible to include members

Samples are not

DETAILED STEPS FOR CONDUCTING A

This involves providing them with the skills needed to

Interviewers have a tremendous amount of influence on the

A good interviewer can make all the difference in the world to

DETAILED STEPS FOR CONDUCTING A

This is an important step, that must be done right in order to ensure