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

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

78 Statistical tests<br />

have stopped here <strong>and</strong> reported that the three styles did not differ significantly in<br />

clearance height.<br />

The post-hoc Tukey test compares the clearance height of the three styles.<br />

Note that the use of the Tukey test is justified because the Levene Test was not<br />

significant (the actual value of the test is not shown here). If the Levene Test<br />

was significant, you should have used one of the post-hoc tests under the equal<br />

variances not assumed section of Dialog box 67. Table 24 shows that the<br />

clearance height differs significantly among all three styles. For example,<br />

when using the Fosbury style the clearance height is 13.76 cm <strong>and</strong> 5.36 cm<br />

smaller than the clearance height obtained with the Western Roll <strong>and</strong> the<br />

Straddle styles respectively. Table 24 shows the mean difference in clearance<br />

height, as well its st<strong>and</strong>ard error, significance level <strong>and</strong> 95% confidence<br />

intervals. These intervals show the values two st<strong>and</strong>ard errors below (Lower<br />

Bound) <strong>and</strong> above (Upper Bound) the mean difference respectively.<br />

Note that some statisticians (e.g., Pedhazur <strong>and</strong> Schmelkin, 1991) do not<br />

recommend the use of post-hoc tests, because these tests require a large<br />

number of mean comparisons which can increase the probability <strong>for</strong> Type I<br />

error (especially if the significance level in Dialog box 67 is not adjusted).<br />

Pedhazur <strong>and</strong> Schmelkin (1991) advocate the use of a priori planned<br />

comparisons to prevent Type I errors. A priori comparisons per<strong>for</strong>m only a<br />

limited number of comparisons between mean scores, because they are based<br />

on a theory that specifies which comparisons are important <strong>and</strong> which are not.<br />

For example, the post-hoc Tukey test above carried out three mean<br />

comparisons contrasting each style with the others. If there were five styles,<br />

you would have per<strong>for</strong>med 10 different mean comparisons. However, with a<br />

priori planned comparisons you could limit the comparisons to a certain<br />

number specified <strong>by</strong> a theory or previous research (e.g., compare the Straddle<br />

<strong>and</strong> Fosbury styles only).<br />

To carry out a priori planned comparisons, click Contrasts in Dialog box 66<br />

to open Dialog box 68. Select Polynomial <strong>and</strong> Linear under the Degree option.<br />

Each style should be given a comparison coefficient. The order of the<br />

coefficients is crucial because each coefficient corresponds to a different high<br />

jump style. Although there are many types of planned comparisons, orthogonal<br />

ones are most often used. Orthogonal planned comparisons require that the sum<br />

of the coefficients is zero in any given comparison. That is, if you want to<br />

compare the first <strong>and</strong> the third style, you should assign coefficient 1 to the first<br />

style (Western Roll), coefficient 0 to the second style (Straddle), <strong>and</strong> coefficient<br />

1 to the third style (Fosbury) (i.e., 1 ‡ 0 1 ˆ 0). If the first style was compared<br />

with the other two styles, the coefficients <strong>for</strong> the three styles should have been<br />

2, 1, <strong>and</strong> 1 respectively. Use the Add button to add each coefficient. For more<br />

complicated designs use the Next button to add another set of contrasts. Click<br />

Continue <strong>and</strong> when you get back to Dialog box 66 click OK. The output tables<br />

(Tables 25, 26) list the contrast coefficients <strong>and</strong> the results of the planned<br />

comparisons between the first <strong>and</strong> the third style. As you can see, the t test is<br />

significant which indicates that the difference in clearance height between the

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

Saved successfully!

Ooh no, something went wrong!