Thanks so much for all help received on repeated measures. I am emboldened
to ask a further question. When one does a multivariate repeated measures analysis, there are two tables of multivariate results. The first is labelled Multivariate Tests and contains tests of the intercept and within subject factor. (For now lets assume just one repeated factor). The second table is labelled Tests of Within-Subjects Effects and contains just the tests of the repeated factor. I have not ever had the problem of these two tables giving substantially different answers, such as one significant and the other not. But the numbers and degrees of freedom etc. are different and I have never been sure why and particularly not sure which one should be reported. Thanks again, Susan QQ Consulting Toronto Canada [hidden email] |
The multivariate tests assume multiariate normality and homegenous
variance covariance matrices across groups, the univariate repeated measures assumes homegeneity of variance covariance matrices, sphericity of the variance covariance matrix and normality of the residuals. The univariate repeated measures is typically mor sensisitve since the violation of sphericity tends to result in more significant results. The tests are different and do have different degress of freedom. Paul R. Swank, Ph.D. Professor, Developmental Pediatrics Director of Research, University of Texas Health Science Center at Houston -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of S Elgie Sent: Wednesday, November 22, 2006 2:39 PM To: [hidden email] Subject: More on Repeated Measures Thanks so much for all help received on repeated measures. I am emboldened to ask a further question. When one does a multivariate repeated measures analysis, there are two tables of multivariate results. The first is labelled 'Multivariate Tests' and contains tests of the intercept and within subject factor. (For now let's assume just one repeated factor). The second table is labelled 'Tests of Within-Subjects Effects' and contains just the tests of the repeated factor. I have not ever had the problem of these two tables giving substantially different answers, such as one significant and the other not. But the numbers and degrees of freedom etc. are different and I have never been sure why and particularly not sure which one should be reported. Thanks again, Susan QQ Consulting Toronto Canada [hidden email] |
Hi Paul,
The design I am testing has eight measures at three time periods, and I am conducting a multivariate repeated meassures analysis. The design is completely repeated with no between-subjects factor(s). I do get univariate results, but what is puzzling me is the two tables containing multivariate tests (Pillais and so on), one labelled 'Multivariate Tests' and the other 'Tests of Within-Subjects Effects.' Thanks! Susan >From: "Swank, Paul R" <[hidden email]> >Reply-To: "Swank, Paul R" <[hidden email]> >To: [hidden email] >Subject: Re: More on Repeated Measures >Date: Wed, 22 Nov 2006 16:00:56 -0600 > >The multivariate tests assume multiariate normality and homegenous >variance covariance matrices across groups, the univariate repeated >measures assumes homegeneity of variance covariance matrices, sphericity >of the variance covariance matrix and normality of the residuals. The >univariate repeated measures is typically mor sensisitve since the >violation of sphericity tends to result in more significant results. The >tests are different and do have different degress of freedom. > > >Paul R. Swank, Ph.D. >Professor, Developmental Pediatrics >Director of Research, > > >University of Texas Health Science Center at Houston > >-----Original Message----- >From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of >S Elgie >Sent: Wednesday, November 22, 2006 2:39 PM >To: [hidden email] >Subject: More on Repeated Measures > >Thanks so much for all help received on repeated measures. I am >emboldened to ask a further question. When one does a multivariate >repeated measures analysis, there are two tables of multivariate >results. The first is labelled 'Multivariate Tests' and contains tests >of the intercept and within subject factor. (For now let's assume just >one repeated factor). The second table is labelled 'Tests of >Within-Subjects Effects' and contains just the tests of the repeated >factor. I have not ever had the problem of these two tables giving >substantially different answers, such as one significant and the other >not. But the numbers and degrees of freedom etc. >are different and I have never been sure why and particularly not sure >which one should be reported. > >Thanks again, > >Susan > >QQ Consulting >Toronto Canada >[hidden email] |
In reply to this post by S Elgie
Ahhhh.
With this model, you get multivariate tests for the multiple DVs and multivariate tets for the repeated measures factors. This is known as a doubly multivariate analysis. You may either interpret the two mulitvariate tests, or the multivariate tests for the DVs and the univariate tests for the repeated measures portion of the design. The latter is sometimes called a multivariate mixed model because it combines both multivariate and univariate tests. The decsion to make is whether the repeated measures portions meets the sphericity assumptions. If not, the the doubly multivariate design may be better but is assumes multivariate normality across both sets of variates. Paul R. Swank, Ph.D. Professor, Developmental Pediatrics Director of Research, University of Texas Health Science Center at Houston -----Original Message----- From: S Elgie [mailto:[hidden email]] Sent: Wednesday, November 22, 2006 8:02 PM To: Swank, Paul R; [hidden email] Subject: Re: More on Repeated Measures Hi Paul, The design I am testing has eight measures at three time periods, and I am conducting a multivariate repeated meassures analysis. The design is completely repeated with no between-subjects factor(s). I do get univariate results, but what is puzzling me is the two tables containing multivariate tests (Pillais and so on), one labelled 'Multivariate Tests' and the other 'Tests of Within-Subjects Effects.' Thanks! Susan >From: "Swank, Paul R" <[hidden email]> >Reply-To: "Swank, Paul R" <[hidden email]> >To: [hidden email] >Subject: Re: More on Repeated Measures >Date: Wed, 22 Nov 2006 16:00:56 -0600 > >The multivariate tests assume multiariate normality and homegenous >variance covariance matrices across groups, the univariate repeated >measures assumes homegeneity of variance covariance matrices, >sphericity of the variance covariance matrix and normality of the >residuals. The univariate repeated measures is typically mor sensisitve >since the violation of sphericity tends to result in more significant >results. The tests are different and do have different degress of freedom. > > >Paul R. Swank, Ph.D. >Professor, Developmental Pediatrics >Director of Research, > > >University of Texas Health Science Center at Houston > >-----Original Message----- >From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf >Of S Elgie >Sent: Wednesday, November 22, 2006 2:39 PM >To: [hidden email] >Subject: More on Repeated Measures > >Thanks so much for all help received on repeated measures. I am >emboldened to ask a further question. When one does a multivariate >repeated measures analysis, there are two tables of multivariate >results. The first is labelled 'Multivariate Tests' and contains tests >of the intercept and within subject factor. (For now let's assume just >one repeated factor). The second table is labelled 'Tests of >Within-Subjects Effects' and contains just the tests of the repeated >factor. I have not ever had the problem of these two tables giving >substantially different answers, such as one significant and the other >not. But the numbers and degrees of freedom etc. >are different and I have never been sure why and particularly not sure >which one should be reported. > >Thanks again, > >Susan > >QQ Consulting >Toronto Canada >[hidden email] |
In reply to this post by S Elgie
Gang,
Since we're talking repeated measures...I just ran a simple 2-group by 2-time repeated measures analysis in SPSS v13 Mac, defining the syntax via the menus. I wanted to obtain a chart that showed the means on my DV by group and time--a typical type of chart with two lines (one line for each group) in two different colors or suchforth, with a legend that illustrated which line represents which group. However, the chart that I got has the 4 data *points* representing the 4 means, all in the same symbol (small open circles), and just one line that doesn't actually connect any of the points but appears to be connecting the average of the points at time 1 and the average of the points at time 2. My syntax came out as follows: GLM DV_time1 DV_time2 BY Group /WSFACTOR = time 2 Polynomial /METHOD = SSTYPE(3) /PLOT = PROFILE( time*Group ) /EMMEANS = TABLES(Group*time) /PRINT = DESCRIPTIVE ETASQ OPOWER HOMOGENEITY /CRITERIA = ALPHA(.05) /WSDESIGN = time /DESIGN = Group . Is there something wrong with the way that the plot is specified? thanks, Dennis -- ```````````'''''''''''' Dennis R. Wahlgren, M.A. Center for Behavioral Epidemiology and Community Health San Diego State University http://www.cbeach.org "Poets say science takes away from the beauty of the stars--mere globs of gas atoms. Nothing is 'mere.' I too can see the stars on a desert night, and feel them. But do I see less or more?" --Richard Feynman |
I'm assuming your data are set up in 3 columns: group, dv_time1, and
dv_time2? I dummied up some data and did an analysis using the GUI (my syntax matched your exactly) and got what I believe to be precisely the plot you were looking for. There is a red line connecting time 1 and time 2 means for group1 and a green line connecting time 1 and time 2 means for group 2. I'm using v11.0.3 for Mac, if that has anything to do with it, but your syntax appears to be fine. Sorry I can't be of much help. Fred -- Fredric E. Rose, Ph.D. Assistant Professor of Psychology Palomar College 760-744-1150 x2344 [hidden email] On 11/29/06 10:14 AM, "D.R. Wahlgren" <[hidden email]> wrote: > Gang, > Since we're talking repeated measures...I just ran a simple 2-group > by 2-time repeated measures analysis in SPSS v13 Mac, defining the > syntax via the menus. > > I wanted to obtain a chart that showed the means on my DV by group > and time--a typical type of chart with two lines (one line for each > group) in two different colors or suchforth, with a legend that > illustrated which line represents which group. > > However, the chart that I got has the 4 data *points* representing > the 4 means, all in the same symbol (small open circles), and just > one line that doesn't actually connect any of the points but appears > to be connecting the average of the points at time 1 and the average > of the points at time 2. > > My syntax came out as follows: > > > GLM > DV_time1 DV_time2 BY Group > /WSFACTOR = time 2 Polynomial > /METHOD = SSTYPE(3) > /PLOT = PROFILE( time*Group ) > /EMMEANS = TABLES(Group*time) > /PRINT = DESCRIPTIVE ETASQ OPOWER HOMOGENEITY > /CRITERIA = ALPHA(.05) > /WSDESIGN = time > /DESIGN = Group . > > Is there something wrong with the way that the plot is specified? > > thanks, > Dennis > -- > > ```````````'''''''''''' > Dennis R. Wahlgren, M.A. > > Center for Behavioral Epidemiology and Community Health > San Diego State University > http://www.cbeach.org > > > "Poets say science takes away from the beauty of the stars--mere > globs of gas atoms. Nothing is 'mere.' I too can see the stars on a > desert night, and feel them. But do I see less or more?" > --Richard Feynman |
In reply to this post by S Elgie
At 12:02 AM -0500 11/30/06, Automatic digest processor wrote:
> >Date: Wed, 29 Nov 2006 11:02:32 -0800 >From: "Fredric E. Rose, Ph.D." <[hidden email]> >Subject: Re: More on Repeated Measures > >I'm assuming your data are set up in 3 columns: group, dv_time1, and >dv_time2? I dummied up some data and did an analysis using the GUI (my >syntax matched your exactly) and got what I believe to be precisely the plot >you were looking for. There is a red line connecting time 1 and time 2 >means for group1 and a green line connecting time 1 and time 2 means for >group 2. I'm using v11.0.3 for Mac, if that has anything to do with it, >but your syntax appears to be fine. Yes, I have my data structured in 3 columns, one participant per row. Apparently something changed from v 11 to v 13? I've gone back through the menus to set up the syntax again and clicked the help button in various places, but didn't see anything to suggest why my charts came out this way. Maybe there's a general SPSS pref that needs to be changed? I can't imagine why they would set the default to the type of chart I'm getting, though. _dennis -- ```````````'''''''''''' Dennis R. Wahlgren, M.A. Center for Behavioral Epidemiology and Community Health San Diego State University http://www.cbeach.org "Poets say science takes away from the beauty of the stars--mere globs of gas atoms. Nothing is 'mere.' I too can see the stars on a desert night, and feel them. But do I see less or more?" --Richard Feynman |
Free forum by Nabble | Edit this page |