08.01.2015 Views

A Step by Step Guide for SPSS and Exercise Studies

A Step by Step Guide for SPSS and Exercise Studies

A Step by Step Guide for SPSS and Exercise Studies

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Statistical tests 105<br />

Figure 32<br />

disengagement). For each main effect (anxiety intensity <strong>and</strong> direction) <strong>and</strong> <strong>for</strong><br />

the interaction effect (intensity direction) you should present the Wilk’s<br />

lambda, the associated F value, the degrees of freedom <strong>and</strong> the significance<br />

level of the F value (see Table 33). If any of the multivariate F values is<br />

significant, you should proceed <strong>and</strong> report the univariate effects (F value, df, <strong>and</strong><br />

significance level; see Table 34). If an independent variable has more than two<br />

groups you should mention the results of post-hoc tests or planned comparisons.<br />

Results of a MANOVA test can be reported in a table <strong>for</strong>mat similar to the one<br />

in Example 4.<br />

General Linear Model/Repeated Measures<br />

Pre-test/post-test designs have been described be<strong>for</strong>e in the Compare means/<br />

Paired Samples T Test option of the Analyze menu. However, such designs look<br />

at changes in the scores of one dependent variable over two periods of time.<br />

With the repeated measures test you can examine more complicated designs.<br />

Assumptions<br />

All the assumptions described <strong>for</strong> ANOVA <strong>and</strong> MANOVA tests are also<br />

applicable to repeated measures. To analyse repeated measures you can take<br />

either a multivariate approach or a univariate approach (Nevill, 2000). The

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!