Correlational Research

Correlational Research

Example #1

Harlow, R. E., & Cantor, N. (1996). Still participating after all these years: A study of life task participation in later life. Journal of Personality & Social Psychology, 71 1235-1249.

Summary

This study investigated the relationship between variables such as social life and organizational affiliation to life satisfaction in 618 older Americans. They found that social life was critically important in predicting life satisfaction, especially in those older Americans who were not working. These finding held up even when factors such as health, self-reported vitality, social support, and congeniality were controlled.

It helps researchers to see the relationship between two or more variables.

Instruments must be reliable and must provide a whole range of responses.

It helps narrow down possible causes.

Researchers must select the subjects to provide a range of responses on the variables.

Practical Considerations/ Tips/ Insights/ Suggestions

Practical Considerations/ Tips/ Insights

Conduct a Pilot Test.

"Spurious Correlations".

Short term practitioners are more accurate than long term predictions.

Correlational Research

relationships among two or more variables are studied without any attempt to influence them

TYPES OF CORRELATIONAL RESEARCH DESIGN

describes and measures the degree of association between two or more variables or sets of scores

plays an important role in the development and testing of theoretical models

Unrelated variables can be eliminated from further consideration.

Explanatory Design

Prediction Design

looks for simple associations between variables and investigates the extent to which the variables are related

data collected at one time

single group

researcher draw from statistics alone

used to predict outcomes in one variable from another variable that serves as the predictor

Purpose of Correlational Research

to help explain important human behaviors

to determine the relationship between two or more variables

to predict likely outcomes

entrance test scores and academic achievement

alcohol and reaction time

smoking and lung cancer

palm’s lifeline and longevity

brain size and intelligence

height and weight

Direction and Strenght of Relationship

Negative Correlation (Indirect)

Positive Correlation (Direct)

Correlation

Strenght of

Relationship

Coefficient

high scores on one variable is associated with low scores on the other variable

scores change in opposite direction

high scores on one variable tend to be associated with high scores on the other variable

scores change in one direction

0.7-1.00 Strong

.30-.69 Moderate

.29-.00 Weak

Advantages

Can collect as much information from many subjects.

Can study a wide range of variables.

Can make predictions based on correlated variables.

Allows testing of expected relationships between and among variables and the making of predictions.

Disadvantages

Correlation does not imply causation.

Cannot be used to draw conclusions about the causal relationships between and among the variables.

Other variables might explain any results that are obtained.

Define a Problem

Data Collection

Sample

Instruments

Data Analysis and Interpretation

Design and Procedures

Observations

Subj. Self Math

-esteem Achievement

A 25 95

B 23 88

C 25 96

D 18 81

E 12 65

*Example of data obtained

Identify an appropriate population.

Minimum sample size should be no less than thirty.

Must yield quantitative data.

Must yield reliable scores

Instruments must show evidence of validity.

Data on both variables are acquired with ease.

Instruments used are administered in a single

session or two, immediately after the other.

A correlational study must be chosen based on a sound rationale, growing out of experience or theory.

The researcher should also have some reason for thinking that certain variables may be related.

Define certain variables.

Two or more scores are obtained from each subject, one score for each variable.

Scores are then correlated.

The results indicates the degree of relationship between variables.

Scores must be plotted in a graph(scatterplot).

A correlation coefficient (r) is produced.

The coefficient will be in decimal form, somewhere between 0.00 and -1.00 or +1.00