I loved that I saved myself the trouble of digging through a whole semester's worth of grad school notes from 8 years ago by watching a 2 1/2 minute video! Thanks so much!
Hey Loren, for my Bachelor-thesis I am doing resarch on 2 groups at 3 different points in time. First the entry level in 3 different motor tests are measured. Two of them include the stability of the knee-joint and one conductes the soccer-specific skill of dribbling. After the entry-level measurement group 1 undergoes an intervention-programm, while group 2 functions as a control group for one of the motor tests. The intervention runs for 4 weeks before testing both groups again. Then groups are switched so that people first assigned to the control group are now undergoing the 4-week-intervetion-programm and vice versa. I now want to find out which effects of the intervention can be analyzed for each group and the seperate time-levels (1-3). Do I have to do additional tests after the mixed ANOVA or do I extract these findings from pairwise comparisons? - Thank you for your work, really helps.
You would typically do pairwise tests between groups at each time point. Or, you could do pairwise tests between time points WITHIN each group. Or, you could do both. Assumptions of these approaches and type I error inflation implications differ in each case, but those are generally the options. You're right on track!
Thank you for the reply- I'm focussing on the pairwise comparisons within each group since the output of the ANOVA was non-significant for the between factor. Really appreciate the help. What do you mean by the different assumptions though?@@loren.toussaint
@@thatguyvince5703 so you would not want to use the between effect as a gateway to doing t-tests between groups. the overall between effect is an AVERAGED effect that will likely be nonsignificant because at baseline there shouldn't be a difference and this is averaged into the between effects across baseline, post and follow-up. do the between comparisons anyway. if you do the within comparisons. there are assumptions of any statistical test (often normality, homogeneity of variance, etc.). I would review them from a trusted text or website for both independent and dependent t-test as well as mixed anova.
Thanks again for the help. The output from the mixed Anova shows the pairwise comparisions between and within effects already or am I wrong? By the way I have an significant interaction effect of Time x Group while main-effect of time is also significant and main effect of group is not. Would it make sense to explore the main effect of time anyway or should I just focuse on the interaction effect? If I decide to analyze the within-effect of time further which t-test do I use since the same participants are measured but groups are switched? I also think I need to split the data-set in SPSS for the group-variable in order to conduct the t-test. For me the only reasonable answer would be a paired t-test I think to explore at which point in time differences can be found.@@loren.toussaint
@@thatguyvince5703 Oh, sorry. Yes, if you have requested the between and within effects, then they should already be there. I'd use those. If you have an interaction, you should focus on that in your interpretation, not time or betweeen groups effects. Yes, you are correct that you need to look at changes over time WITHIN each group OR differences BETWEEN groups at each time. You can request those tests in syntax I believe or do it by splitting by group.
Since time has 3 levels and Mauchly's is significant with Epsilon valued at >.750 wouldn't we need to look at Huynh-Feldt instead of sphericity assumed?
Hi, we are doing a research with 2 different groups (Group A and B). They undergo the SAME intervention (a rehab program) and we let them take a survey before, during and after the intervention. During and after the intervention, we expect group A scores better on the survey than group B. Will this 'mixed model ANOVA' work for our experiment? Our within subjects would be the time we take the survey and our between subjects consists of the two different groups.
Hi sorry to disturb you but....I have a problem with the anova for my experiment. In my experiment participant were exposed to a game played with a kinect and a joystick. Group 1 (Kinect and joystick) and Group 2 (joystick and Kinect). I want to see the influence of the order of the controls, the age, the skill and the gender on the score of two different test (presence and flow experience) So I have as IV the condition (K-J and J-K), gender (F-M), age (19-24, 25-30) and skill (skilled, non skilled) and as DV the score of the presence and flow experience test. I’m confused because my professor keeps saying that I need a mixed anova to do everything but I can’t find a way to put the variable inside the control panel. So I used a paired sample t-test for the condition and a between t-test for the age, gender, skill and the result look kinda possible. Why he keep saying mixed anova? Is it all because of the interaction of the factor?
Hi, Your K-J J-K variable is a within subjects factor but gender, age, and skill are between subjects factors. That is why it is mixed. I'm guessing your instructor may want you to arrange the data as I have in the video for anxiety1(K-J) and anxiety2 (J-K). That is, put the data in two columns, then you'll have gender, age, and skill columns for all the participants too. Then you'll be able to examine for instance gender by order effects. Plot the order effect on the X axis and then put gender on the separate lines. These are my thoughts, but its always wise to talk with your instructor about these types of things :-)
I think you're probably referring to a "doubly multivariate" design and analysis in which case you'd want to report the lambda, F, dfs, and p value (eta squared for effect size). tables of means and sds, or a figure would usually accompany this.
hello, can you please show me how to perform 2x2x8 Mixed ANOVA in SPSS between subject factor 2 levels 1)CAI 2) Control, within subject factor 2 levels 1) Injured 2) Uninjured , with 8 levels of directions (AL,A,AM,M,PM,P,PL,L) I am Sports Physio. & i am doing study on tennis players with ankle instability for my Thesis ( masters). I have done the experiment, I have collected the data, but i am facing huge difficulty in feeding it into SPSS and analyzing it. Can you please help me with it. So that i can submit
Zubin Bhavsar Hi, I think you can follow the same general procedure, you just need to add another between subjects factor for directions. I hope that makes some sense.
but between subject factor is @ 2 levels (1) CAI - Those with Chronic ankle instability & (2) CONTROL - Those without CAI i:e Normal within subject factor is @ 2 levels (1) injured - those with CAI but having one leg involved either Right or Left (2) Uninjured - the other leg other than injured. with 8 directions of measurement for each leg. so i have to take measurement of one subject 2 times one each for leg . so there will be 16 measurement for each subjects. in each group. can u hlep.
First, we need to be clear that it is the normality of the residuals that is of interest. There are also several other assumptions of mixed model ANOVA. I found this link helpful: psych.wisc.edu/Brauer/BrauerLab/wp-content/uploads/2014/04/Murrar-Brauer-2018-MM-ANOVA.pdf. If assumptions are not met there are varying degrees of concern, depending on which assumptions are not met and how badly they are violated. I think you're going to find that the search for an exact nonparametric alternative is going to be a relatively difficult one. I might consider looking into generalized estimating equations, if needed. Hope that helps.
I loved that I saved myself the trouble of digging through a whole semester's worth of grad school notes from 8 years ago by watching a 2 1/2 minute video! Thanks so much!
Warms my heart to hear this! Thanks for your kind words. I hope it was helpful.
Hey Loren, for my Bachelor-thesis I am doing resarch on 2 groups at 3 different points in time. First the entry level in 3 different motor tests are measured. Two of them include the stability of the knee-joint and one conductes the soccer-specific skill of dribbling. After the entry-level measurement group 1 undergoes an intervention-programm, while group 2 functions as a control group for one of the motor tests. The intervention runs for 4 weeks before testing both groups again. Then groups are switched so that people first assigned to the control group are now undergoing the 4-week-intervetion-programm and vice versa. I now want to find out which effects of the intervention can be analyzed for each group and the seperate time-levels (1-3). Do I have to do additional tests after the mixed ANOVA or do I extract these findings from pairwise comparisons?
- Thank you for your work, really helps.
You would typically do pairwise tests between groups at each time point. Or, you could do pairwise tests between time points WITHIN each group. Or, you could do both. Assumptions of these approaches and type I error inflation implications differ in each case, but those are generally the options. You're right on track!
Thank you for the reply- I'm focussing on the pairwise comparisons within each group since the output of the ANOVA was non-significant for the between factor. Really appreciate the help. What do you mean by the different assumptions though?@@loren.toussaint
@@thatguyvince5703 so you would not want to use the between effect as a gateway to doing t-tests between groups. the overall between effect is an AVERAGED effect that will likely be nonsignificant because at baseline there shouldn't be a difference and this is averaged into the between effects across baseline, post and follow-up. do the between comparisons anyway. if you do the within comparisons. there are assumptions of any statistical test (often normality, homogeneity of variance, etc.). I would review them from a trusted text or website for both independent and dependent t-test as well as mixed anova.
Thanks again for the help. The output from the mixed Anova shows the pairwise comparisions between and within effects already or am I wrong? By the way I have an significant interaction effect of Time x Group while main-effect of time is also significant and main effect of group is not. Would it make sense to explore the main effect of time anyway or should I just focuse on the interaction effect? If I decide to analyze the within-effect of time further which t-test do I use since the same participants are measured but groups are switched? I also think I need to split the data-set in SPSS for the group-variable in order to conduct the t-test. For me the only reasonable answer would be a paired t-test I think to explore at which point in time differences can be found.@@loren.toussaint
@@thatguyvince5703 Oh, sorry. Yes, if you have requested the between and within effects, then they should already be there. I'd use those. If you have an interaction, you should focus on that in your interpretation, not time or betweeen groups effects. Yes, you are correct that you need to look at changes over time WITHIN each group OR differences BETWEEN groups at each time. You can request those tests in syntax I believe or do it by splitting by group.
Since time has 3 levels and Mauchly's is significant with Epsilon valued at >.750 wouldn't we need to look at Huynh-Feldt instead of sphericity assumed?
Hi, so would this be described as a 2 x 3 (programme x time) mixed method or a 3 x 2? How do you know which way round they go?
correct. it can be described in either order. doesn't matter.
Hi, we are doing a research with 2 different groups (Group A and B). They undergo the SAME intervention (a rehab program) and we let them take a survey before, during and after the intervention. During and after the intervention, we expect group A scores better on the survey than group B.
Will this 'mixed model ANOVA' work for our experiment?
Our within subjects would be the time we take the survey and our between subjects consists of the two different groups.
this sounds exactly like what you need! Good fit of this statistic to your design.
Hi sorry to disturb you but....I have a problem with the anova for my experiment.
In my experiment participant were exposed to a game played with a kinect and a joystick. Group 1 (Kinect and joystick) and Group 2 (joystick and Kinect). I want to see the influence of the order of the controls, the age, the skill and the gender on the score of two different test (presence and flow experience)
So I have as IV the condition (K-J and J-K), gender (F-M), age (19-24, 25-30) and skill (skilled, non skilled) and as DV the score of the presence and flow experience test.
I’m confused because my professor keeps saying that I need a mixed anova to do everything but I can’t find a way to put the variable inside the control panel.
So I used a paired sample t-test for the condition and a between t-test for the age, gender, skill and the result look kinda possible.
Why he keep saying mixed anova? Is it all because of the interaction of the factor?
Hi,
Your K-J J-K variable is a within subjects factor but gender, age, and skill are between subjects factors. That is why it is mixed. I'm guessing your instructor may want you to arrange the data as I have in the video for anxiety1(K-J) and anxiety2 (J-K). That is, put the data in two columns, then you'll have gender, age, and skill columns for all the participants too. Then you'll be able to examine for instance gender by order effects. Plot the order effect on the X axis and then put gender on the separate lines. These are my thoughts, but its always wise to talk with your instructor about these types of things :-)
can go for GEE, which helps you to adjust for the covariates
How do we report the results of a mixed MANOVA?
I think you're probably referring to a "doubly multivariate" design and analysis in which case you'd want to report the lambda, F, dfs, and p value (eta squared for effect size). tables of means and sds, or a figure would usually accompany this.
hello, can you please show me how to perform 2x2x8 Mixed ANOVA in SPSS
between subject factor 2 levels 1)CAI 2) Control, within subject factor 2 levels 1) Injured 2) Uninjured , with 8 levels of directions (AL,A,AM,M,PM,P,PL,L)
I am Sports Physio. & i am doing study on tennis players with ankle instability for my Thesis ( masters). I have done the experiment, I have collected the data, but i am facing huge difficulty in feeding it into SPSS and analyzing it. Can you please help me with it. So that i can submit
Zubin Bhavsar
Hi, I think you can follow the same general procedure, you just need to add another between subjects factor for directions. I hope that makes some sense.
but between subject factor is @ 2 levels (1) CAI - Those with Chronic ankle instability & (2) CONTROL - Those without CAI i:e Normal
within subject factor is @ 2 levels (1) injured - those with CAI but having one leg involved either Right or Left (2) Uninjured - the other leg other than injured.
with 8 directions of measurement for each leg. so i have to take measurement of one subject 2 times one each for leg . so there will be 16 measurement for each subjects. in each group. can u hlep.
its 2x2x8 anova
What if normality is violated and transforming doesn’t fix this? What is the non-parametric model?
First, we need to be clear that it is the normality of the residuals that is of interest. There are also several other assumptions of mixed model ANOVA. I found this link helpful: psych.wisc.edu/Brauer/BrauerLab/wp-content/uploads/2014/04/Murrar-Brauer-2018-MM-ANOVA.pdf. If assumptions are not met there are varying degrees of concern, depending on which assumptions are not met and how badly they are violated. I think you're going to find that the search for an exact nonparametric alternative is going to be a relatively difficult one. I might consider looking into generalized estimating equations, if needed. Hope that helps.
@@loren.toussaint thanks a lot, will have a look into this