You are on page 1of 21


Ayaz Muhammad Khan Assistant Professor University of Education, Lahore

Causal Comparative Research

Determine the cause of existing differences among groups.
The Aim

It is a quantitative research
To determine the cause or consequences of the

differences that already exist between or among group of individuals- ex post facto

At least two different groups are compared on a

dependent variable or measure of performance (called the effect) because the independent variable (called the cause) has already occurred or cannot be manipulated. Dependent variable-the change or difference occurring as a result of the independent variable. Independent variable- an activity of characteristic believed to make a difference with respect to some behavior.

The researcher attempts to determine the cause, or

reason, for pre existing differences in groups of individuals. Attempts to identify cause and effect relationships. Involve two or more group variables. Involve making comparison. Individuals are not randomly selected and assigned to two or more groups. Cannot manipulate the independent variable Less costly and time consuming

Steps Taken

1. Problem 2. Sample 3. Design and procedure 4.Data Analysis

Problem formulation Identify and define particular phenomena of interest and then to consider possible causes for, or consequences of these phenomena Sample Define carefully the characteristic to be studied and select group that differ in this characteristic Instrumentation No limitation Design Involve selecting groups that differ on particular variable of interest, compare and remember no manipulation!

DESIGN Select two groups that differ on some independent

variable I. One group possesses some characteristics that the other does not II. Each group possesses the characteristic but in differing amounts

The Basic Causal-Comparative Designs

Group (a) I Independent variable C (Group possesses characteristic) C (Group does not possess characteristic) C1 (Group possesses characteristic 1) C2 (Group possesses characteristic 2) Dependent variable O (Measurement) O (Measurement)



O (Measurement) O (Measurement)


Examples of the Basic Causal-Comparative Design

(Figure 16.1)

Data Analysis Construct frequency polygons Calculate means and standard deviations T-test to show differences between means The result do not prove cause and effect, but only identifying the relationship

Example: The Relationship between Years of

Experience and Job Satisfaction Causal Comparative Design Hypotheses Alternative- Teachers with a high level of experience will be more satisfied with their jobs than teachers with low levels of experience. Null- Teachers with a high level of experience will be equally satisfied with their jobs when compared to teachers with low levels of experience. Variables Dependent- Job satisfaction Independent- Years of experience

Two levels (high & low) Exists naturally in the population of teachers at the

start of study. Sample Two groups sampled, one for each level of the independent variable High Experience Low Experience

Design and Procedure Select two groups that differ on some independent variable One group possesses some characteristic that the other does not Each group possesses the characteristic but in differing amount The independent variable must be clearly operationally defined * Randomly sample subjects from each of the two groups Collect background information on subjects to determine the equality of the groups Compare groups on the dependent variable

Design and Procedure Control of Extraneous variable What other variable besides years of experience could explain job satisfaction among teachers? Matching: Each subject in the high experience group is matched with a subject with a low experience group along the variable of class size. Each high experience teacher who teachers a large class is matched with a low experience teacher who teaches a large class. Each high experience teacher who teaches a small class is matched with a low experience teacher who teaches a small class.

Data Analysis Mean- job satisfaction ratings for High Experience and Low Experience subjects are compared using t-test, ANOVA or other appropriate statistical test. Rejection of the null hypothesis supports the alternative hypothesis that years of experience result in increased job satisfaction.


WEAKNESSES: Lack of randomization Inability to manipulate an independent variable Loss of subjects Location Instrumentation History Maturation

Data collector bias
Instrument decay Attitude

Pre-test/treatment interaction effect


Matching of Subjects
Finding or Creating Homogeneous Subgroups Statistical Matching


Step 1: ask: What specific factors either are known to

affect or may logically be expected to affect the variable on which groups are being compared? Step 2: ask: What is the likelihood of the comparison groups differing on each factor? Step 3: Evaluate the threats on the basis of how likely they are to have an effect.