|
Size: 3844
Comment:
|
Size: 2221
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 30: | Line 30: |
| * [:FAQ/subtractmn: Reconstructing sums of squares bu subtracting means] It is possible to reconstruct the sums of squares in repeated measures ANOVA using multiple regression using the GLM parameterisation (see the GLM Graduate Statistics talk) or univariate (between subjects) ANOVA but thse are again more cumbersome approaches than using the repeated measures ANOVA. For example suppose we wish to assess the correlation between a measure, ECAP, and outcome, PERF which are each recorded on eight subjects at the same four different rates. This can be done by correlating the 8 x 4 differences obtained by subtracting each subject's respective ECAP and PERF means from each of that subject's four ECAP and PERF observations. Subtracting the different subject rate means removes the within subject factor of rate (see Boniface (1995)) and allows us to then perform a 'zero-order' Pearson correlation on the ECAP and PERF difference scores. This resultant correlation is equivalent to the signed square root of the partial eta-squared for ECAP entered as acovariate obtained from a one-way repeated measures ANOVA with rate as a four category within subjects factor. The above assumes the rate means vary within subject. If there is no difference between the within subjects factor rate then the within subject variance may be regarded as random white noise and it is, therefore, not necessary, as described in the preceding paragraph, to adjust the ECAP and PERF scores by subtracting respective ECAP and PERF subject means. |
* [:FAQ/subtractmn: Reconstructing sums of squares bu subtracting means] |
| Line 37: | Line 33: |
__Reference__ Boniface DR (1995) Experiment design and statistical methods for behvioural and social research. Chapman and Hall:London. (In CBSU library). |
How do I perform a repeated measures analysis of variance in SPSS?
The key thing here is that each combination of the repeated measures factors is represented by a separate column. By contrast, all the levels of a between subject factor are entered in the same column.
For example suppose we wish to compare activity in four areas of the brain - the hippocampus, area 23, areas 29 and 30 and acingulate congulate. Activity in each of these areas is entered in four columns called hippo, a23, a29 and acin respectively.
The syntax below does this by specifying a within subjects factor, region, with four levels made up of each of the four columns. You can also use the repeated measures option in the General Linear Model menu. If you use the menu interface note that you can select more than one variable at a time to define the repeated measures factor by holding the Alt and Ctrl keys down and left mouse clicking on the variables. This can be quicker than defining one variable at a time especially if you have a lot of variables.
If in addition you think that the acingulate congulate will show less activity than the other three areas you can test this using /MMATRIX subcommand. It specifies user defined contrast coefficients for each of the levels of the repeated measures factor and is only available using syntax.
filter by group. GLM hippo a23 a29 acin /WSFACTOR = region 4 Polynomial /METHOD = SSTYPE(3) /CRITERIA = ALPHA(.05) /MMATRIX "Hip_a23_a29v.acin" all -0.333 -0.333 -0.333 1 /WSDESIGN = region.
An equivalent form of the /MMATRIX subcommand above is
/MMATRIX "Hip_a23_a29v.acin" hippo -0.333 a23 -0.333 a29 -0.333 acin 1
Note the contrast coefficients sum to zero and specify a paired t-test comparing the activity means of the hippocompus, areas 23, 29/30 with that of the acingulate congulate. A slightly more cumbersome approach uses the [:FAQ/rpcon: SPSS MANOVA procedure.]
- [:FAQ/subtractmn: Reconstructing sums of squares bu subtracting means]
[:FAQ/glm99rm: How to fit a mixed repeated measures ANOVA in SPSS with two or more repeated measures factors and/or repeated measures factors with > 99 levels]