2. The Application of Research
The service
provider
• How many
people are using
the service or
product?
• Why do different
people use
different
services?
The administrator
manager
• What are the
training needs of
the staff?
• How can the
effectiveness of
the workers be
evaluated?
The consumer
• Am I getting
value for my
money?
• How good are
the service
providers?
The Professional
• What is the most
effective
relationship
between x and
y?
• How valid is the
particular theory
in this contet?
3. What is Research?
Research is a structured inquiry that utilizes
acceptable scientific methodology to solve
problems and creates new knowledge that is
generally applicable (Grinnell, 1993)
4. What is Research?
Scientific methods consist of systematic
observation, classification, and interpretation of
data. Now, obviously this process is one in which
nearly all people engage in the course of their
daily lives. The main difference between our day
to day generalizations and the conclusions usually
recognized as scientific method lies in the degree
of formality, rigorousness, verifiability and general
validity of the latter (Lundberg, 1942)
5. Characteristics of Research
1. Testability/Verifiability/non-circularity
2. Empirical: based on real life experiences
3. Combination of Deductive-Inductive
4. Systematic : logical sequence, not haphazard
5. Rigorous: maximum possible application of
scientific methods
6. Control: minimizing the effects of other variables
affecting the relationship among main variables
7. Pure Research
Involves developing and testing theories and
hypotheses that are intellectually challenging
to the researcher but may or may not have
practical application at the present time or in
future. Thus such work often involves the
testing or hypotheses containing very abstract
and specialized concepts. (Kumar, 2005)
8. Pure Research
Developing a sampling technique
Developing a methodology to assess the
validity of procedure
Finding best way to measure people’s
attitudes
Usually helps in adding into existing body of
knowledge
9. Applied Research
Research techniques and procedures (…)
that are applied to the collection of information
about various aspects of a situation, issue or
problem or phenomenon so that information
gathered can be used in other ways- such as
for policy making, administration and
enhancement of understanding of
phenomenon.
10. Descriptive Research
Attempts to describe systematically a
situation, problem, phenomenon, service or
program
E.g. attempt to describe services provided by
an organization
Living conditions of village people in internal
Sindh
How a child feels living in a house with
domestic violence
11. Correlational Research
Discover or establish the existence of
relationship between 2 or more aspects of a
situation.
E.g. impact of advertisement on sale
Relationship between stress and heart attacks
Relationship of technology with
unemployment
12. Explanatory Research
Why and how there is a relationship between
two aspects
Cause-Effect relation
How home environment effects child’s
academic achievement.
13. Exploratory Research
To explore an area about which little is known
or to investigate the possibilities of
undertaking a particular research studies.
E.g. understanding the living conditions of sex
workers/ exploring beliefs of people going to
shrines etc…
14. Qualitative Research
Unstructured approach
Exploratory studies/descriptive studies
More flexible methodology
Usually information is gathered on nominal or
ordinal scales
Small samples/ getting info till saturation
Not highly objective
Description of observed situation, historical
analyses etc
Identifying themes/Content analysis
15. Quantitative Research
Structured approach
Usually to determine the extent of some well
established phenomenon
More hypothesis testing involved
More inferential statistics are used
Large samples
More objectivity
How many people hold a particular
attitude?/relationship between anxiety and exam
performance?
16. Research Process
Methods of
data
analyses…
Statistical
expertise
SPSS,
NVivo,
AMOS etc.
7. Data
Analysis
6. Collecting
Data
Contents of
the Research
Proposal
Usually APA
style
5. Writing a
Research
Proposal
Sampling
Theory and
Sampling
types
4. Selecting a
Sample
Methods and
tools of data
collection
Validity and
Reliability of
tools
3. Instrument
for data
collection
Study
Designs
2. Research
design
Literature
Review
Hypotheses
and
variables
1. Research
Problem
17. 1. Formulating a Research Problem
First and most important step
Should be precise, specific and clear
No ambiguity allowed
Everything in the study follows from the
research statement
Should thoroughly observe subject area
before formulating a problem
18. Literature Review
Provides a theoretical background to study
Helps in refining research methodology
Can create a link between what has already done
and what needs to be done
Enables to contextualize research findings
Can integrate new findings with old findings
Helps in saving time and energy for uselessly
repeating studies
19. How to do Literature Review?
Search for existing literature
1. Books
2. Journals
3. E-Journal
20.
21. How to do Literature Review?
Review the selected literature
Read critically and look for important themes and
theoretical framework
Read criticism as well to get more comprehensive
picture
Look for methodologies adopted in different studies,
research designs, instrumentation, sampling,
analyses etc.
Look for significant differences of opinion
Look for gaps in studies, areas where more work is
needed
22. How to do Literature Review?
Develop a theoretical framework
Use a Narrow Down Approach
A problem may be understood by various theoretical
frames
Highlight common and uncommon themes in order to
make a holistic picture
You may be interested in one approach more than others
and want to work in that area further
So focus on those aspects that look to you more
appropriate (make your own subjective framework)
Differentiate between Universal and Local trends in
theories and studies
23. How to do Literature Review?
Develop a Conceptual Framework
Develop your own conceptual framework that includes
theories and works that suit you most and upon which you
want to base your work
Use that framework in your study
24. Research Problem
Sources of Research Problem (4 Ps)
1. People: (individuals or groups that are studied)
2. Problems: (examining certain issues, rate of
suicide, marital discord)
3. Programs: (evaluation of interventions, like
therapies, advertisements)
4. Phenomena: (may be pure phenomena like
relationship of intelligence with speed of
reading)
25. Considerations in selecting a research
problem
Interest
Magnitude (manageable)
Measurement of concepts
Level of expertise
Relevance/ utility
Availability of data
Ethical Issues
26. Steps in formulation of Research
Problem
1. Identify a broad Field or area of interest (e.g.
mental illnesses, political instability, extremism,
terrorism etc.)
2. Narrow Down and Dissect broad area into
subareas
Mental illness:
types of illnesses,
prevalence of some specific illness e.g. Depression,
Profile of families in which depression occurs
Who are most affected by depression
Reasons for depression
Types of depression
Treatment of depression
Previous work done on depression
27. Steps in formulation of Research
Problem
3. Select area Most interesting for you
4. Make research questions (write down what you really
want to find out in this specific area and why?)
5. Formulate your objectives (Affirmative statements
about what you are going to do in your research)
6. Assess your objectives
28. Variables
An image, perception or concept that is
capable of measurement-hence capable of
taking on different values- is called a Variable
(Kumar, 2005)
A variable is a property that takes on different
values. Putting it redundantly, a variable is
something that varies… a variable is a symbol
to which numeral or values are attached
(Kerlinger, 1986)
29. Difference between a concept and
variable
Concepts are mental images or perceptions;
highly subjective, vary from individual to individual
(e.g. self-esteem, richness, achievement,
violence)
Concepts are not measureable per se
Variables can be subjected to measurement as
they are subjective/objective quantifiable units
(e.g. gender, income, weight, religion, age)
30. Difference between a concept and
variable
In order to measure the Concepts, we have to
convert them into variables first
This process is known as
“OPERATIONALIZATION”
To give an operational definition of concept,
first we need to identify the defining
INDICATORS of that concept
31. Difference between a concept and
variable
Some indicators are easy to establish, like
operationally defining Richness (we can take Annual
income and Assets as the operational definition)
More abstract concepts are more difficult to define (e.g.
Self-esteem, intelligence).
Intelligence consists of more underlying concepts like
speed of processing, comprehension, analytical ability
etc. that are claimed to be measured through some IQ
tests… So the IQ test becomes the Operational
definition of the concept Intelligence.
32. Types of Variables
Independent Variables
Supposed to bring out change/manpulateable
Dependent variables
Outcome of change due to IVs
Extraneous Variables
Several other factors in real life situations like noise, temperature etc.
that are not measured but might affect the results
Confounding variables
Attached with IVs, cannot be controlled and IV has a cumulative affect
(Mortality-Fertility relationship)
Attribute Variables
Cannot be changed or manipulated like age, gender, education etc.
33. Types of measurement for variables
S S Stevens’s classification of levels of
measurement
1. Nominal/categorical scales
Two or more subcategories
For identification and classification
E.g. gender, religions, political orientation etc.
2. Ordinal/ranking Scales
Categorical in nature
Categories are arranged in ranks
Ranks are discrete not continuous
E.g. class positions, high, average, low income groups
34. Types of measurement for variables
3. Interval Scale
All characteristics of nominal and ordinal scale + unit of
measurement with arbitrary starting and ending point
Relative scale that plots position of individuals in relation to
one another with respect to magnitude
No equal intervals
E.g. Celsius scale with arbitrary zero point
4. Ratio Scale
All the properties of previous scales + plus an absolute fixed
zero point where some property is non existent
Age, salary etc
35. Constructing Hypotheses
A proposition that is stated in a testable form
and that predicts a particular relationship
between two (or more) variables. In other
words, if we think that a relationship exists,
we first state it as a hypothesis and then test
the hypothesis in the field. (Bailey, 1978)
36. Characteristics of Hypotheses
Simple, specific, clear
Testable, verifiable
Parsimonious (emerge from existing body of
knowledge and relate to it)
Operationalizable/measureable
37. Types of Hypotheses
Null Hypothesis (Ho: no
difference/relationship exists among
variables)
Research Hypothesis (main hypothesis)
Hypothesis of difference
Hypothesis of Point-prevalence
Hypothesis of association
38. Research Design
A traditional research design is a blueprint or
detailed plan for how a research study is to be
completed-operationalizing variables so the
can be measured, selecting a sample of
interest to study, collecting data to be used as
a basis for testing hypotheses, and analyzing
the results (Thyer, 1993)
39. Types of Research design
Cross Sectional Study Design
One shot studies, only one contact with study
population
Studies conducted to find out prevalence of
some phenomenon, problem, issue etc. by
taking cross-section of a population
Cheaper study
Cannot measure change as only one contact
with population
40. Types of Research design
Before-and-after study design
Pretest-posttest studies
Two times contact with population
Good for assessing impact of some program, intervention
Differences are compared
Disadvantages can be
Only total change can be measured (so difficult to assess
contribution of extraneous variables)
Maturation effect
Reactive effect (research instrument educates sometime)
Regression effect (from initial extreme position to tending toward
mean)
41. Types of Research design
Longitudinal Research Design
Contact with population a number of times
To assess reduction of disease like polio with
passage of time
To assess the effectiveness of some therapy for drug
addicts
Assessing performance of students during particular
session
Disadvantages can be
Same as in before-and-after study design
Conditioning effect (respondent might become aware of the
process and give casual responses)
mortality
42. Types of Research design
Replicated Cross-Sectional Design
Cross sectional study substituting Longitudinal
study
Choosing clients who are at different phases of a
program
Choosing students from the same institute who
are studying with different durations, 1st year, 2nd
year, 3rd year etc.
43. Types of Research design
Experimental designs
1. After-only design
2. Before-and-after design
3. Control-group design
4. Double-control design
5. Comparative design
6. Matched control design
7. Placebo design
44. Types of Research design
Experimental designs
1. After-only design (post-test only design)
1. Intervention is introduced and its effect measured
2. Baseline is usually created by past records or
subjects’ recall of the situation before the intervention
Disadvantage: No actual baseline to compare
45. Types of Research design
Control-group design
2 population groups are selected (experimental
and control groups)
Control group is introduced for comparison
The groups are comparable in every respect
except the intervention which is introduced to
Experimental group only
Before and after measurements are conducted for
both groups
Effect of extraneous variables is quantified
46. Types of Research design
Double-Control design
2 control groups are used to check the effects of
maturation, regression or reactive effect
Researcher excludes one of the control groups
from the ‘before’ observation.
Exp1-exp2 (mixed difference)
Con1-con2 (difference without treatment effect)
Only one Con (no reactive effect)
47. Types of Research design
Comparative Design
To compare effectiveness of different treatment
modalities
Study population is divided into same number of
subgroups as the number of treatments need to be
tested
Before and after measurements are done while
introducing treatments to subgroups
Degree of change in the DV in different populations is
compared to establish the relative effectiveness of
various interventions
48. Types of Research design
Matched-Control Experimental design
Comparative groups are formed on matched
characteristics e.g. socioeconomic status, age,
disease (diabetic) etc.
To reduce individual differences
Disadvantages can be
On more than one variables matching becomes
difficult
On abstract concepts (self-esteem, EQ, etc matching
is very difficult)
49. Types of Research design
Placebo Design
Patients’ beliefs sometimes effect the treatment
they are receiving
This design contains experimental, control and
placebo groups (who receive a fake treatment)
The groups are compared
50. Types of Research design
Cohort studies
These are based on the existence of a common
characteristic such as year of birth, graduation or
marriage, within a subgroup of a population.
E.g. finding marital satisfaction among couples
who married in 1990s
51. Types of Research design
Case Studies
An approach to studying a social phenomenon
through a thorough analysis of an individual
case.
The case may be an individual, group, episode,
process, organization, community, society or
any other unit of social life
52. Types of Research design
Retrospective Study Design
These studies investigate a phenomenon,
situation, problem or issue that has happened in
the past
E.g. relationship between levels of unemployment
(past event) and street crime
Effect of early childhood experiences on certain
mental disorders
53. Methods of Data Collection
Methods of data
collection
Secondary Sources
Documents
Govt. documents
Earlier research
Personal records
Client histories
Primary Sources
Observation
Participant/non
participant
Interview
Structured/Unstructur
ed
Questionnaires
Mailed/Collective
54. Methods of Data Collection
Observation
A purposeful, systematic and selective way of
watching and listening to an interaction or
phenomenon as it takes place
E.g. interaction of a group, dietary patterns of a
population, studying behavior or personality traits
of individuals
More appropriate when information can’t be
accessed by questioning
55. Methods of Data Collection
Types of Observation
Participant: when researcher participates in the
activities of the group being observed, with or
without the knowledge of the group the group that he
is being observed (e.g. attitude of people toward
people on wheelchairs)
Non-Participant: Researcher does not get involved
in the activities of the group and remains a passive
observer, e.g. observation of students in class rooms
with cameras
56. Methods of Data Collection
Disadvantages
Hawthorne Effect: If people get to know that they
are being observed they may change their behavior
Observer Bias
Difference in interpretation among observers
Possibility of incomplete observation
Error of Central Tendency: (Less experienced
observers may record findings toward the mean as a
precaution)
Halo effect: Judgment in one area may generalize
into other areas as well
57. Methods of Data Collection
Interview
Semi-Structured:
Flexible structure
Flexible contents
Flexibility in
questions (mostly
open ended)
Semi-Structured:
Relatively flexible
but structured
Structured:
Rigid Interview
structure
Rigid Contents
Rigid format of
questions (close
ended)
58. Methods of Data Collection
Unstructured Interviews
In-depth Interviews: repeated face to face encounters
between the researcher and informants directed
towards understanding informants’ perspectives on
their lives, experiences, or situations as expressed in
their own words (Taylor & Bogdan, 1984)
E.g. client-therapist relationship
Focus Group Interviews: same as in-depth
interviews but taken with a group instead of individuals.
To explore the perceptions, feelings of those people who
have some common experiences, e.g domestic violence,
refugees etc.
59. Methods of Data Collection
Unstructured Interviews
Narratives: No predetermined content, researcher
seeks to hear personal experiences of a person with an
incident or happening in his/her life, researcher is
passive and uses techniques of ‘active listening’ saying
words like ‘hmm, huh, ok’
Good for exploring sensitive personal issues like child
sexual abuse, domestic violence
Oral Histories: for learning about past events, to gain
cultural or historical understanding
E.g. to understand political conditions of Pakistan during
1970s, people who were young at 1970 could report their
perceptions.
60. Methods of Data Collection
Advantages of Interview
Appropriate for complex situations
Useful for in-depth information
Information is supplemented by non-verbal cues
Questions can be explained
Disadvantages
Time consuming, expensive
Depends on nature of interaction
Expertise and experienced required
Subjectivity, different interpretations by different
interviewers
Researcher/interviewer bias
61. Methods of Data Collection
Questionnaire
A questionnaire is a written list of questions, the
answers to which are recorded by respondents.
Ways of Administering
Mailed questionnaires
Collective Administration
Administering in a public place
Questionnaires can be open ended or closed ended
62. Methods of Data Collection
Advantages in questionnaires:
Less expensive
Offers greater anonymity
Can be administered on large samples
Disadvantages
Application is limited (only literate)
Response rate is low (esp. mailed quest..)
Self-selecting bias (more motivated individuals return questionnaires)
Opportunity to clarify issues is lacking
Spontaneous responses are not allowed usually
Consultation with others is possible
Response to one question may be influenced by another question
63. Establishing Reliability and Validity of
Research Instrument
Validity of Research instrument
The degree to which the researcher has
measured what he has set out to measure (Smith,
1991)
Are we measuring what we think we are
measuring? (Kerlinger, 1973)
64. Establishing Reliability and Validity of
Research Instrument
There are generally 2 ways of establishing validity
of a research instrument, Rational and empirical
(theory based tests and empirical tests)
Types of Validity
1. Face and content validity
2. Criterion validity (Concurrent and predictive)
3. Construct validity (Convergent and Divergent)
65. Establishing Reliability and Validity of
Research Instrument
Face and content validity
Face Validity: The judgment that there is a logical
link between contents of the test and what it is
supposed to measure
Content Validity: Items and questions (i.e. test
content) should cover the full range of issue or
attitude being measured
Content is usually assessed by panel of experts
66. Establishing Reliability and Validity of
Research Instrument
Criterion validity
Predictive Validity: the degree to which an
instrument can forecast an outcome (SAT test,
Mechanical Aptitude test etc)
Concurrent Validity: How well an instrument
compares with a second assessment concurrently
done (two assessments matched, mechanical
aptitude test comparison with mechanical tasks)
67. Establishing Reliability and Validity of
Research Instrument
Construct validity
Determined by ascertaining the contribution of
each construct to the total variance observed in
a phenomenon (e.g. Factor Analysis)
Convergent Validity: Similar constructs should
converge or correlate positively
Divergent Validity: Dissimilar constructs
should diverge or correlate negatively
68. Establishing Reliability and Validity of
Research Instrument
Reliability of research instrument
Degree of accuracy and precision in the
measurements made by a research instrument
Types of reliability
Test/retest
Parallel forms
Internal consistency (Split-half, Chronbach alpha,
inter-item and item-total correlation)
69. Concept of Sampling
Sampling is a process of selecting a few (a
sample) from a bigger group (the sampling
population) to become the basis for estimating
or predicting the prevalence of an unknown
piece of information, situation or outcome
regarding the bigger group
70. Concept of Sampling
Population: The group of individuals from whom you select a
small group for your study (N)
Sample: the small group of individuals from whom you collected
required information to estimate trends in population (n)
Sampling design/strategy: the way you select individuals
Sampling element/unit: each individual in the sample
Sampling Frame: a list identifying each element in the
population
Sample statistic: findings based on the information obtained
from the sample
Population Parameters: the estimates arrived at from the
sample statistics
71. Types of Sampling
Sampling in Qualitative research
No predefined rigid rules
We are more interested in exploring phenomena
and don’t go for inferences
Even one individual can be a sample
Saturation Point: in order to explore diversity,
you go on interviewing as long as you keep
getting new information(until you reach a
saturation point)
72. Types of Sampling
Probability Sampling: (each individual has
an equal chance to be included or excluded
from the sample)
Non-Probability Sampling: (Individuals don’t
have an equal opportunity to be included or
excluded in the sample)
Mixed Sampling
73. Probability Sampling
Simple Random Sampling
Each individual has an equal chance…
Lottery method, computer programs, random number
tables
Stratified Random Sampling
For heterogeneous populations
Different homogenous strata are made
Simple random sampling is used within each strata
E.g. Study on Physicians (defining strata on the basis of
specialization such as psychiatrists, cardiologists,
nephrologists etc.)
74. Probability Sampling
Cluster Sampling
Less detailed than stratified sampling
For large populations
Clusters are formed on some common characteristics
such as geographical proximity
E.g. three clusters of University, College and Madrassah
One-Stage/two-stage/Multi-Stage
75. Non-Probability Sampling
Convenience Sampling
Ease of access to sample population
Selected from the location convenient to the researcher
The process continues until the researcher has been
able to contact the required number of
respondents(quota)
Accidental sampling
Accidental sampling makes no attempt to include people
possessing an obvious characteristic
Such as in marketing research
76. Non-Probability Sampling
Judgmental or Purposive Sampling
Judgment of the researcher as to who can provide
the best information to achieve the objectives of the
study
To create a historical reality, to understand a rare
phenomenon
Snowball Sampling
Selecting a sample using networks
Used when little is known about the target population
Like prostitutes, gamblers etc
77. Mixed Sampling
Systematic Random Sampling
Sampling frame is first divided into a number of
segments called intervals. Then from first interval
using the SRS technique, one element is selected.
Selection of other elements is totally dependent
on the selection of first element