Transcript Causal - Comparative Research Overview

EDU702 Research
Methodology
Quantitative Research
Prepared by
Azurawati Binti Wok Zaki
2011313879
ED7701A2
Chapter 16:
Causal-Comparative
Research
(Ex post facto Research
Non-experimental research)
to determine the cause
of existing differences
among groups.
Causal-Comparative Research is
Differentiated from Experimental
Research
• In an experiment, the independent
variable is manipulated by the
researcher.
• In causal comparative research the
independent has already occurred.
– Examples of independent variables include
socioeconomic status, pre-school history,
number of siblings, and so on.
Three types of causal-comparative research
Type 1
Type 2
Type 3
• Exploration of
effects (dependent
variable) cause by
membership in a given
group
• Question: What
differences in
abilities are caused by
gender?
• Research hypothesis:
Females have a
greater amount of
linguistic ability than
males.
• Exploration of causes
(independent variable)
of group membership
• Question: What
causes individuals to
join a gang?
• Research hypothesis:
Individuals who are
members of gangs
have more aggressive
personalities than
individuals who are
not members of
gangs.
• Exploration of the
consequences
(dependent variable)
of an intervention
• Question: How do
students taught by
the inquiry method
react to propaganda?
• Research hypothesis:
Students who were
taught by the inquiry
method are more
critical of propaganda
than are those who
were taught by the
lecture method.
Causal- comparative and
Correlational Research
• Similarities:
researchers explore relationship
among variables
seek identify variables that are
worthy of later exploration through
experimental research.
provide guidance for subsequent
experimental studies.
attempt to explore causation
Differences:
Causal-Comparative vs Correlational
- Compare 2 or more
groups of subjects
- Require a score on each
variable for each subject.
-Involve at least one
categorical variables
(group membership)
- Investigate 2 (or more)
quantitative variables
- Often compare averages - Analyse data using
or use crossbreak tables.
scatterplots
Causal-comparative and
Experimental Research
• Similarities:
• - Require at least one
categorical variable (group
membership).
• - Compare group
performances (average score)
to determine relationship.
• - Both typically compare
separate group of subjects.
Differences: Causal-Comparative vs
Experimental
No manipulation takes
place
-
Independent variable is
manipulated
-
- Provide much weaker evidence for
causation than do experimental studies.
- The group are already formed (the
researcher must locate them)
- The researcher can assign subjects to
treatment groups
- The researcher has much greater
flexibility in formulating the structure
of the design.
Value of Causal
Comparative Research
• Uncovers relationships to be
investigated experimentally.
• Used to establish cause-effect
when experimental design not
possible.
• Less expensive and time consuming
than experimental research.
• Note: if you conduct a quantitative
research study it most likely will be
a causal-comparative study.
• More Examples:
• How does preschool attendance affect
social maturity at the end of the first
grade?
• How does having a working mother
affect a child’s school absenteeism?
STEPS
TAKEN
1) Problem
Formulation
2) Sample
3) Instrumentation
4) Design
• Subject Characteristics
• - Matching the subjects
• Finding
or
Creating
Homogeneous Subgroups
• - Statistical Matching
Data Analysis in CausalComparative Studies
• To construct frequency polygon
• Means and standard deviations are
usually calculated in the variables
involved are quantitative.
• The most used test is a t-test for
differences between means.
• Analysis of covariance is useful.
• The
results
should
always
be
interpreted with caution.
Chapter 17:
Survey Research
The most popular technique for
gathering primary data in which a
researcher interacts with people to
obtain facts, opinions,
and attitudes.
The Purpose of Survey Research
To describe the characteristics of a
population
To find out how the members of a
population distribute themselves on one
or more variables
Rarely is the population as whole studies,
however. Instead, a sample is surveyed
and a description of the population is
inferred from what the sample reveals.
3 Characteristics
Information is
collected from a
group of people in
order to describe
some aspects of
characteristics of
the population of
which that group
is a part.
The main wayasking questions
Information is
collected from a
sample rather
than from every
member of the
population.
Types of Surveys
i) Cross-sectional study
• A study in which various segments of
a population are sampled. E.g.
managers and non-managers.
• Data are collected at a single moment
in time.
• When an entire population is
surveyed: Census
example
• A professor of Mathematics might
collect data from a sample of all the
high school mathematics teachers in a
particular state about their interests in
earning a master’s degree in
Mathematics from his university.
ii) Longitudinal study
• Longitudinal survey
• - information is collected at different
points in time in order to study
changes over time.
3 Longitudinal designs
Trend study
Cohort
Study
Panel Study
different
samples from a
population
whose member
may change are
surveyed at
different
points in time.
a particular
population
whose members
do not change
over the course
of the survey.
the researcher
surveys the
same sample of
individuals at
different times
during the
course of the
survey.
Examples:
• Trend study:
- A researcher might be interested in the
attitudes of high school principals
towards the use of flexible scheduling.
He would select a sample each year
from a current listing of high school
principals throughout the year. The
same individual would not be the
sampled each year, he would compare
responses from year to year to see
whether any trends were apparent.
Cohort study
• A researcher would like to study growth
in teaching effectiveness of all the 1st
year teachers who had graduated in the
past 5 years from a university. The
names of all would be listed, and
different sample would be selected
from this listing at different times.
Panel study
• A researcher select a sample of last
year’s graduates from a university who
are 1st year teachers and survey on the
same individual several times during the
teaching years.
Steps taken
Defining the problem
- It
should be
interesting and
important to motivate
individuals to respond.
Identifying the
target population
Forms of Survey Research
Internet surveys
Personal Interviews
Mail Surveys
Telephone surveys
Comparison of three
kinds of surveys
COMPARISON
OF SURVEY TECHNIQUES
Advantages and Disadvantages
Direct
Administration
Telephone
Mail
Interview
Comparative Cost
Lowest
Intermediate
Intermediate
High
Facilities needed?
Yes
No
No
Yes
Require training of questioner?
Yes
Yes
No
Yes
Data collection time
Shortest
Short
Longer
Longest
Response rate
Very high
Good
Poorest
Very
High
Group administration possible
Yes
No
No
Yes
Allow for random sampling?
Possibly
Yes
Yes
Yes
Require literate sample?
Yes
No
No
No
Permit follow-up-questions?
No
Yes
Yes
Yes
Encourage response to
sensitive topics?
Somewhat
Somewhat Somewhat Weak
Questionnaire Design
Open-Ended
Question
An interview question that encourages
an answer phrased in respondent’s
own words.
Closed-Ended
Question
An interview question that asks
the respondent to make a selection
from a limited list of responses.
Advantages
Closed-ended
Open-ended
-Enhance consistency of response
across respondents
- Allow more freedom of response
-Easier and faster to tabulate
- Easier to construct
- More popular with respondents
- Permit follow-up by interviewer
Disadvantages
- May limit breadth of response
- Tend to produce responses that are
consistent in length and content
across respondents
- Take more time to construct
- Both questions and responses
subjects to misinterpretation
- Require more questions to cover the
research topic
- Harder to tabulate and synthesize
Typical problems in wording questions
TYPICAL PROBLEMS IN WORDING QUESTIONS
Respondent Error
• A classification of
sample bias resulting
from some respondent
action
Non-response
-Total Nonresponse
Non-response
-total nonresponse
• Not enough people respond or
refuse to respond
• NONRESPONDENTS - People
who refuse to cooperate
• NOT-AT-HOMES
• SELF-SELECTION BIAS
especially in case of selfadministered questionnaire.
People respond to only those
they like.
Item Nonresponse
- The respondent may not know
the answer to a particular
question, the respondent may
find the questions
embarrassing or irrelevant.
The END
Thank You