Between group mean differences

[am.i.wrig...@gmail.com]

Hi Evan, 

I am trying to calculate the mean difference between two groups. 

example: 

group 1 goes from 100 to 90

group 2 goes from 95 to 92

When I have them all in one dataset, using the inbuilt funciton I can subtract time 1 from time 0 to see the difference, then in the covariate I can see the difference by groups. This is good. 

The main result is giving a mean however and I would like a difference between groups rather than mean of both groups. So it is giving (100+95)/2 / (90+92)/2, whereas I'm wondering if you can do (100-90)-(95-92) with a few clicks? Is there a way to automate this with WIZARD?

Many thanks

[Evan Miller]

Hi!

I think you will be able to make use of the "Repeated Measures" function. 

https://www.wizardmac.com/help/summary/paired_t_test.html

If you have individual observations in two time periods, put each time period in a separate column. E.g.

ID  Group    M1  M2

===============

A   Group1  10   20

B  Group1  20   98

C  Group2  93   92

D  Group2  90   93

The paired t-test will let you test the increase of (M2 - M1) between the two groups.

Does that help?

Evan

--
You received this message because you are subscribed to the Google Groups "Wizard User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to wizard-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wizard-users/9f526db3-2dd6-4a28-9e5e-e0d73b3f2814%40googlegroups.com.

[am.i.wrig...@gmail.com]

Hi there Evan, 

Thank you for your response - this is a good idea and I can see even now that it will work. It does mean reformatting my tables as I had both groups in one line, but I will think of some way to speed this up (maybe insert new column between each group then copy the data from the second group across one column).

Many thanks for your help!

On Tuesday, April 28, 2020 at 2:20:27 AM UTC+12, Evan Miller wrote:

Hi!

I think you will be able to make use of the "Repeated Measures" function. 

https://www.wizardmac.com/help/summary/paired_t_test.html

If you have individual observations in two time periods, put each time period in a separate column. E.g.

ID  Group    M1  M2

===============

A   Group1  10   20

B  Group1  20   98

C  Group2  93   92

D  Group2  90   93

The paired t-test will let you test the increase of (M2 - M1) between the two groups.

Does that help?

Evan

On Apr 27, 2020, at 09:47, am.i.wr...@gmail.com wrote:

Hi Evan, 

I am trying to calculate the mean difference between two groups. 

example: 

group 1 goes from 100 to 90

group 2 goes from 95 to 92

When I have them all in one dataset, using the inbuilt funciton I can subtract time 1 from time 0 to see the difference, then in the covariate I can see the difference by groups. This is good. 

The main result is giving a mean however and I would like a difference between groups rather than mean of both groups. So it is giving (100+95)/2 / (90+92)/2, whereas I'm wondering if you can do (100-90)-(95-92) with a few clicks? Is there a way to automate this with WIZARD?

Many thanks

--
You received this message because you are subscribed to the Google Groups "Wizard User Group" group.

To unsubscribe from this group and stop receiving emails from it, send an email to wizard...@googlegroups.com.