research method by lena
TRANSCRIPT
-
7/26/2019 research method by lena
1/36
SOCIAL SCIENCE RESEARCH DESIGN
AND DATA ANALYSIS
1
-
7/26/2019 research method by lena
2/36
PRESENTATION OUTLINE
1. Introduction
2. Social Science Research Strategies
3. Research Ethics
4. Conclusion
2
-
7/26/2019 research method by lena
3/36
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 environmentaleffects of kerosene lanterns.
Food safety assurance in the Ghanaian food
manufacturing value chain.
Assessing the knowledge and practices of wayside
auto electricians in Ghana3
-
7/26/2019 research method by lena
4/36
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
-
7/26/2019 research method by lena
5/36
Social Science Research Strategies
1. Survey Research
2. Case Study3. Archival Analysis
4. Experiment
5. Action Research
5
-
7/26/2019 research method by lena
6/36
SURVEY RESEARCH DEFINED Suited to research problems in which the researcher aims to
investigate the incidence and prevalence of a phenomenon orwhen 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 certaindegree of error.
Surveys tend to be quantitative in nature and usually usesstructured and standardized questionnaires.
6
-
7/26/2019 research method by lena
7/36
SURVEY RESEARCH PROCESS1. Clarify the purpose of the research
2. Resource assessment
3. Take a decision on the methods and procedures
4. Design the questionnaire
5. Pilot test the questionnaires
6. Prepare the sample
7. Train interviewers
8. Data collection
9. Process the data
10. Analyse the results11. Interpret and disseminate results
12. Take action
7
-
7/26/2019 research method by lena
8/36
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.
8
-
7/26/2019 research method by lena
9/36
DETAILED STEPS FOR CONDUCTING A SURVEY
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
-
7/26/2019 research method by lena
10/36
DETAILED STEPS FOR CONDUCTING A SURVEY
RESEARCH CONTD
Step Three: Decision on the Methodsand Procedures
Should it be:Administer face-to-face questionnaires?
Mailed and e-mailed questionnaires? or
Telephone and computerized telephone
interviews?Administer Online?
10
-
7/26/2019 research method by lena
11/36
DETAILED STEPS FOR CONDUCTING A SURVEY
RESEARCH CONTD
Step Four: Design the Questionnaire Questionnaires are designed to address the research objectives or
questions.
Guidelines:
Language wording
Bias
Order of the Questions
Length of the Questionnaire
11
-
7/26/2019 research method by lena
12/36
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
-
7/26/2019 research method by lena
13/36
DETAILED STEPS FOR CONDUCTING ASURVEY 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 byrespondents;
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
-
7/26/2019 research method by lena
14/36
DETAILED STEPS FOR CONDUCTING A SURVEY
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
-
7/26/2019 research method by lena
15/36
-
7/26/2019 research method by lena
16/36
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
-
7/26/2019 research method by lena
17/36
Sampling methods contd
Probability sampling Non probabilitysampling
Random sampling Snowball sampling
Stratified Sampling Quota sampling
Cluster Sampling Purposive Sampling
17
-
7/26/2019 research method by lena
18/36
Merits and Demerits of the SamplingTechniques
Technique Brief Descriptions Advantages Disadvantages
Simplerandom
Random sample fromwhole population
Highly representative if allsubjects participate; the ideal
Not possible without complete listof population members;
potentially uneconomical to
achieve; can be disruptive to
isolate members from a group;
time-scale may be too long.
Stratifiedrandom
Random sample fromidentifiable groups
(strata), subgroups,
etc.
Can ensure that specific groupsare represented, even
proportionally, in the sample(s)
(e.g., by gender), by selecting
individuals from strata list
More complex, requires greatereffort than simple random.
Cluster Random samples of
successive clusters of
subjects (e.g., by
institution) until small
groups are chosen as
units
Possible to select randomly when
no single list of population
members exists, but local lists do;
data collected on groups may
avoid introduction of confounding
by isolating members
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
-
7/26/2019 research method by lena
19/36
Merits and Demerits of the SamplingTechniques contd
Purposive Hand-pick subjects on the basisof specific characteristics
Ensures balance of groupsizes when multiple groups
are to be selected
Samples are noteasily defensible as
being representative
of populations due to
potential subjectivity
of researcher
Quota Select individuals as they come
to fill a quota by characteristics
proportional to populations
Ensures selection of adequate
numbers of subjects with
appropriate characteristics
Not possible to prove
that the sample is
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)
No way of knowing
whether the sample is
representative of the
population
19
-
7/26/2019 research method by lena
20/36
DETAILED STEPS FOR CONDUCTING ASURVEY RESEARCH CONTD
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
-
7/26/2019 research method by lena
21/36
DETAILED STEPS FOR CONDUCTING ASURVEY RESEARCH CONTD
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
-
7/26/2019 research method by lena
22/36
DETAILED STEPS FOR CONDUCTING A SURVEY
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
-
7/26/2019 research method by lena
23/36
DETAILED STEPS FOR CONDUCTING ASURVEY RESEARCH CONTD
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 areused.
The qualitative inquiries capture areas where in-depth information is required for better
understanding of issues.
The quantitative analysis is can be done usingSPSS, STATA or any other statistical software.
23
-
7/26/2019 research method by lena
24/36
DETAILED STEPS FOR CONDUCTING ASURVEY RESEARCH CONTD
Step Eleven: Interpret and DisseminateResults
It is important to feed back the results of thesurvey to management, staff, interested
participants and other stakeholders.
This is to keep them informed and establish buy-infor implementing any changes resulting from the
survey.
24
-
7/26/2019 research method by lena
25/36
DETAILED STEPS FOR CONDUCTING ASURVEY RESEARCH CONTD
Step Twelve: Take Action This refers to implementing the changes suggested
by the results of your survey.
It is important to take action and implementchanges in order to make improvements tosubjects understudied.
25
-
7/26/2019 research method by lena
26/36
CASE STUDY RESEARCH
A case study is an in-depth investigation of an individual,group, institution or phenomenon.
It is better suited to enquiries into a phenomenon in its
natural context.
Case studies are often based on the premise that locating
one case is enough to make a conclusion for other casessince a case can be typified for similar other cases.
26
-
7/26/2019 research method by lena
27/36
Purpose of Case Studies
The primary purpose of case study is todetermine factors that have resulted inthe behaviour understudy.
The investigation involves a detailedexamination of a single subject, group orphenomenon.
27
-
7/26/2019 research method by lena
28/36
TECHNIQUES FOR SELECTION OFCASES IN CASE STUDY RESEARCH
28
Context
Context
Context Context
Context Context
Context Context
Case
Case
Case Case
Case Case
Embedded Unit ofAnalysis 1
Embedded Unit ofAnalysis 2
Embedded Unit ofAnalysis 1
Embedded Unit ofAnalysis 2
Embedded Unit ofAnalysis 1
Embedded Unit ofAnalysis 2
Embedded Unit ofAnalysis 1
Embedded Unit ofAnalysis 2
Embedded Unit ofAnalysis 1
Embedded Unit ofAnalysis 2
Type 1 Type 2
Type 3 Type 4
Single-case designs Multiple-case designs
Holistic(Single-unitof analysis)
Embedded(Multiple
unitsof analysis)
Case Case
Context Context
-
7/26/2019 research method by lena
29/36
Data Collection
29
Interviews
Surveys
Observation
Archived data or institutional reports
-
7/26/2019 research method by lena
30/36
Data Analysis
30
Coding data & and aggregating it
Pattern identification
Trend Identification
There are software packages for qualitative data analysis, e.g. Atlas
-
7/26/2019 research method by lena
31/36
ETHICS IN RESEARCH
Informed Consent
Competence
Privacy Sensitivity of information Settings being observed
Dissemination of the information
Anonymity and Confidentiality
31
-
7/26/2019 research method by lena
32/36
Tell participants who is conducting study
Why was subject singled out for participation?e.g. random sample
e.g. recently given birth to 1st child etc.
What is the time commitment?e.g. 45 minutes to complete the survey
Any benefits for the participant to be expected?Realistically, there are often few direct benefits.
Any potential risks, and how have these been managed?
ETHICS IN RESEARCH: INFORMED
CONCENT
-
7/26/2019 research method by lena
33/36
Informed Consent
Explain the study and offer to answer questions
Participation is always voluntary.
Provide participants with copy of informed consent form (ifrelevant).
Confidentiality(Anonymity, on the other hand, means that no one,
including the researcher, will know the subjects identity).
-
7/26/2019 research method by lena
34/36
Sometimes Consent is not Possible
Fake an assault in the street to gauge reaction of bystanders e.g. who willintervene and who will do nothing.
Element of surprise is essential.
Problem with the preceding:
witnessing such an attack may be very disturbing to some.
Those who do not intervene may be upset and suffer feelings of lowworth.
Those who do intervene may be injured.
Debriefing is mandatory.
-
7/26/2019 research method by lena
35/36
SUMMARY
Research in the social sciences dealdirectly with the human being.
Several approaches can be used toinvestigate the subject.
The onus is on the researcher to use themost appropriate method by clarifyingthe purpose of the research.
It is important to observe the code ofethics in research.
35
-
7/26/2019 research method by lena
36/36
Thank you
36