Week 7 - Research Methods - Correlational Analysis

Association

= a relationship between two variables

• direction - is it positive or negative?

• strength - how closely are the variables related?

• shape - is the relationship linear, U-shaped, inverted U-shaped?

What graph is used to visualise correlations?

= scatterplot

Correlation Coefficient

= a number that describes the strength and direction of a relationship

• ranges from -1 to +1

• the closer to the ones, the stronger the correlation

• eg, -0.85 is stronger than +0.40, because +0.40 closer to 0

Pearson’s Correlation

= measures the strength and direction of a linear relationship

• used when both variables are interval or ratio level

• used when the data are approx normally distributed

• used when the relationship is linear

Spearman’s Correlation

= used when the data are ordinal or when interval/ratio data are not normally distributed

• uses RANKS instead of RAW scores

Which correlation is appropriate for skewed data?

= Spearman’s rank correlation

What does p < .05 mean in correlation test?

= the result is statistically significant at the 5% level

What are the small, medium and large correlation benchmarks?

= .10, .30, .50 respectively

What must be included when reporting a correlation in APA style?

= test type, variables, direction, significance, r, df, and p-value

How are degrees of freedom calculated for a correlation test?

df = n - 2

(n is no of ppts)

Two-tailed test (non-directional)

= test either direction

• these are the default

• must be justified in advance, not chosen afterwards to make a result significant

One-tailed test (directional)

= predict a direction

Why is half a correlation table usually empty?

= correlations are symmetrical, so duplicate values are unnecessary

Why does correlation not imply causation?

= a third variable or reverse causation could explain the association