Sunteți pe pagina 1din 19

CAUSAL-COMPARATIVE

RESEARCH
a non-experimental Quantitative Research
-compares one or more measurable
characteristics of two or more groups to find
the similarities and the differences between
them.
-”ex post facto”
GOAL
-determine whether the independent variable
affected the outcome (dependent variable)
-attempt to determine the cause or
consequences of differences that already exist
between or among groups of individuals.
COMPARISONS WITH CORRELATIONAL RESEARCH

-both are useful when experimental research is


deemed impossible
-both designs attempt to determine relationships
among variables, but neither allows for actual
manipulation
-neither randomly places subjects intro control and
experimental groups
CONTRASTING WITH CORRELATIONAL RESEARCH
Causal-comparative

-researcher investigates effect of an IV on a DV


ex: Whether a computer-based ACT program has a
positive effect on ACT scores
*in correlational, it only words with one group of
individuals
COMPARISON WITH EXPERIMENTAL RESEARCH
-both typically compare two or more groups of subjects
-both determine what effect the IV may or may not have on
the DV

CONTRASTING WITH EXPERIMENTAL RESEARCH


Causal-comparative
-occurs after event or action has been completed
-research subjects are already in groups
WHEN TO USE
-particular IV that are not capable of being
manipulated (e.g. gender, ethnicity, socioeconomic
level, education level, and religious preferences)
- Some IV should not be manipulated

*usually a precursor to Experimental research


STEPS:
1. Determine the focus of research
-research question; generate hypothesis
-identifies DV and IV
2. Participant Sampling
-strengthen research design and counter threats to
internal validity
-impose selection techniques
Matching
-identifies one or more characteristics; selects
participants who fit in the characteristics of
control and experimental group

Comparing Homogenous subgroups


-compare subgroups that are clustered
according to a particular variable
Analysis of Covariance (ANCOVA)
-able to discuss previously disproportionate
scores on a pretest to equalize the groups on
the control variable.
3. Instrumentation and Data collection
- QUESTIONNARIES, PRETESTS AND
POSTTESTS, CARIOUS ASSESSMENTS,
BEHAVIORAL OBSERVATIONS

4. Data Analysis and Interpretation


3 MOST USED METHODS
1. chi-square test
- Determine whether there is a statistically
significant relationship between
experimental and control groups based on
frequency counts.
- Best with nominal data (categories of
treatment or participant characteristics)
3 MOST USED METHODS
2. Independent or dependent t tests
- Significant difference between the control
and experimental groups
*Independent t test: NO CONTROLS ARE
APPLIED TO SAMPLES
*Dependent t test: MATCHING HAS BEEN
APPLIED TO SAMPLES
3 MOST USED METHODS
3. ANCOVA
-find significant difference between groups
based on mean scores
LIMITATIONS
-no control over variables
- reversal causation
-inability to construct random samples
TWO TYPES:

1. Retrospective Causal-Comparative Research


2. Prospective Causal-Comparative Research
Retrospective Causal-Comparative
Research
-researcher begins investigating a particular
question when the effects have already
occurred and the researcher attempts to
determine whether one variable may have
influenced another variable
Prospective Causal-Comparative
Research
-researcher initiates a study beginning with the
causes and is determined to investigate the
effects of a condition.
-more common

S-ar putea să vă placă și