Embedded Files
(1) Get the Facts: Sample size (N = ), Pearson r (robt = )
(2) Calculate the degrees of freedom: df = (N-2)
Where N is the complete number of subjects for whom data were collected on both variables.
(3) Look up rcrit
(4) Draw Pearson r spread, place rcrit circumscribing 95% CI.
(5) Place robt on the spread --
(6) Write out the theory, and express it as an equation -- r0
(The equation conveys this concept: "The relationship is different from Zero."
(7) Formulate the Null, or express it as an equation: r = 0
(The equation conveys this concept: "The relationship is not different from Zero,")
(8) Test the null hypothesis: If (the absolute value of) robt is greater than rcrit, reject the null. In other words, if robt exceeds the 95% confidence interval circumscribed by rcrit on the spread, reject the null hypothesis.
Write the results of the test of the null hypothesis. The null will either be retained or rejected. Written in APA format like this: r(df) = , p = .
(6) Write the analysis. If you rejected the null hypothesis (the relationship is statistically significant). If, and only if, you rejected the null, calculate Magnitude of Effect (MOE) r2.
(7) Report the Magnitude of Effect (MOE): (This step is only done if you rejected the null.)
Compute r-squared (r2): r2 = robt * robt )
Google Sites
Report abuse