Correlational Research Flashcards
Correlational Research
Research that identifies and describes the relationship (association) between two or more measured variables.
Association Claim
Describes the relationship found between two measured variables.
Bivariate Correlation
An association that involves exactly two variables.
What are the 3 results of association claims?
- Positive
-Negative
-Zero
Positive Association
As one variable increases, the other increases too.
Negative Association
As one variables increase, the other decreases.
Zero Association
No consistent relationship between the two variables.
What do you do if there are more than two associations in a study?
Look at different pairs separately to examine relationships.
How can an association between two variables be measured?
Using scatter plots and correlation r
What are the two qualities of r?
Direction and strength
Direction of r
Whether the association is positive, negative, or zero.
Strength of r
How closely related the two variables are
The stronger the relation, the closer the r value is to –1 or 1.
What are Categorial Variables best represented by?
Bar graphs
- Each person is not represented by one data point, instead the graph shows the mean (average)
What indicates an association in bar graphs?
The difference between the two variable means.
How are quantitative variables best represented?
Scatter plots
What does the study become when one variable is manipulated?
Experiment
What are the two most important validities in association claims?
Construct and statitical
Construct Validity
Analyze how well each of the two variables were measured.
What two questions examine construct validity?
- Does the measure have good reliability?
- Is the study measuring what it intended to measure?
Statistical Validity
Ask what factors might have affected the scatterplot, correlation coefficient r, bar graph, or difference score that led to the association claim.
What needs to be considered when analyzing statistical validity?
Strength/precision of point estimate
If the study was replicated
If there are any outliers
If range restriction occurred
Effect Size
Strength between two or more variables.
What value indicates strong correlation relationship?
Close to -1 or 1
When can you determine a study’s results is statistically significant?
When the p value is less than 0.05 and when the CI does not contain 0
