IBM SPSS for Intermediate Statistics by Nancy L. Leech Karen C. Barrett George A. Morgan & Karen C. Barrett & George A. Morgan

Author:Nancy L. Leech,Karen C. Barrett,George A. Morgan & Karen C. Barrett & George A. Morgan [Nancy L. Leech]
Language: eng
Format: epub
ISBN: 9781136334931
Publisher: Taylor & Francis Ltd

Interpretation of Output 7.1 continued

The first part of this includes redundant information. First, the Total effect of X on Y gives the b, the standard error, t, and p for math achievement predicting scholastic aptitude test – math, b = 11.16, p < .001. Second, the Direct effect of X on Y gives us the beta, the standard error, t, and p for the combination of math achievement and motivation predicting scholastic aptitude test – math. For these data, the direct effect is b = 10.95, p < .001.

New information in the output includes the Indirect effect of X on Y and is the model with the relationship between math achievement and scholastic aptitude test – math being mediated by motivation. The beta (found under Effect) for this effect is .20, with a bootstrapped standard error (BootSE) of .46, and a 95% confidence interval (BootLLCI and BootULCCI) ranging from -.4171 to 1.5853. Because this range includes zero, motivation did not statistically significantly mediate the relationship between math achievement and scholastic aptitude test- math, b = .20, BCa CI [-.417, 1.585].

The output includes multiple effect size measures. The effect size to report when the indirect effect is statistically significant is the Preacher and Kelley (2011) Kappa-squared. This is preferred to other effect sizes since it is more easily interpreted since it is a ratio (where as the other effect sizes are not but tend to be interpreted as if they are). For these data, the Kappa-squared is .027. Kappa-squared values range from 0 to 1, so they can be interpreted similarly to R. Therefore, the Kappa-squared of .027 is a small effect size. The 95% confidence interval for Kappa-squared can be found under BootLLCI and BootULCCI and ranges from .0003 to .1083. For these data, since the indirect effect was not statistically significant, an effect size should not be reported. If the indirect effect had been statistically significant, Kappa-squared would be reported as κ2 = .027, 95% BCa CI [.001, .108], which represents a small effect size according to Cohen (1988).

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.