Research methods in psychology by B. Morling (third edition) – Chapter 8 summary

INTRODUCING BIVARIATE CORRELATIONS
A bivariate correlation or bivariate association, is an association that involves exactly two variables. The nature of the association can be described with scatterplots and the correlation coefficient. Associations between categorical variables are usually presented in a bar graph.

INTERROGATING ASSOCIATION CLAIMS
The two most important validities to interrogate are construct validity and statistical validity with an association claim. The construct validity checks how well each variable was measured. The statistical validity checks how well the data supports the conclusion.

There are five questions that can be asked in order to interrogate the statistical validity:

1. What is the effect size?
   The effect size is the strength of a relationship between two or more variables. Larger effect sizes allow more accurate predictions and large effect sizes are usually more important. Exceptions on this second rule depend on the context
2. Is the correlation statistically significant?
   Statistical significance refers to the conclusion a researcher reaches regarding the likelihood of getting a correlation of that size just by chance, assuming there is no correlation in the real world. Statistical significance calculations depend on effect size and sample size.
3. Could outliers be affecting the association?
   Outliers are extreme scores. Outliers matter the most when a sample is small.
4. Is there restriction of range?
   If there is not a full range of scores on one of the variables in the association, it can make the correlation appear smaller than it really is. One of the solutions for this is the statistical technique called correction for restriction of range.
5. Is the association curvilinear?
   A curvilinear association is an association in which the relationship between two variables is positive or negative up to some point and then changes.

INTERNAL VALIDITY: CAN WE MAKE A CAUSAL INFERENCE FROM AN ASSOCIATION?
The three requirements in order to establish causation are the following:

1. Covariance of cause and effect
2. Temporal precedence (directionality problem)
3. Internal validity (third-variable problem)

A third-variable problem can be exposed by checking the correlation with that third variable between the original two variables. If there is a third-variable present, the original association is then called a spurious association.

EXTERNAL VALIDITY: TO WHOM CAN THE ASSOCIATION BE GENERALISED
The external validity interrogation asks whether the association can generalize to other people, places and times. When the relationship between two variables changes depending on the level of another variable, that variable is called a moderator.