Anova/ T-Test for different variables- thanks for your help!

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Elenieth
Posts: 3
Joined: Wed Mar 04, 2015 10:54 am

Anova/ T-Test for different variables- thanks for your help!

Postby Elenieth » Wed Mar 04, 2015 11:08 am

Good morning!

Thank you for reading this plea for help and apologies for my rudimentary descriptions.

I have only just started doing some SPSS for my PhD in education. I have the following data sets:

Batch number of questionnaire (of 2; beginning and end of the year),
Type of student (intervention; one year after intervention; control),
plus
gender and year group

I have measured how students feel about writing on five different scales, ranging from 1 (f.e.happy) to 5 (f.e. unhappy).

I am now trying to find out whether students of a particular gender and/or year group feel differently at the end of the year, depending on which group they belong to. For example:
- the changes between (batch 1 + intervention + boy) = x mean feeling VS (batch 2 + intervention + boy) = x mean feeling

I have got the means of these groups (see screenshot attached) but unfortunately I can't figure out how to run a t-test or an anova which can tell me about the significance of the changes- and which changes are the ones that I should take seriously.

I would be most grateful for some pointers; this has been doing my head in for a couple of weeks now and unfortunately both my supervisor and my uni are unable to help.

Thank you! :)
You do not have the required permissions to view the files attached to this post.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Anova/ T-Test for different variables- thanks for your h

Postby GerineL » Wed Mar 04, 2015 12:38 pm

Can you clarify your design?

i.e., I think you have a 1 between subject variable (intervention, 3 levels), 1 repeated measure (questionnaire, 2 timepoints), and 2 within subject variables (gender, year group). Is that correct?

In that case, you can used the MIXED procedure or Repeated Measures ANOVA. Look at some youtube videos.
iopsych
Posts: 25
Joined: Tue Feb 24, 2015 11:58 pm

Re: Anova/ T-Test for different variables- thanks for your h

Postby iopsych » Thu Mar 05, 2015 9:04 pm

Hello Elenieth,

I will help you to clarify:

Batch number of questionnaire (of 2; beginning and end of the year),
Did you give the questionnaire twice to the same students? At the beginning and at the end of the year? Or were these different students who tool it either at the beginning or at the end?

Type of student (intervention; one year after intervention; control),
I think these are different students. Intervention group got the questionnaire immediately after the intervention, one year after group took it one year after, and control just took the questionnaire without the intervention. Is this correct?

gender and year group
And these would be different students again. I cannot see a year group variable in you output; where is it?

Yalcin
Elenieth
Posts: 3
Joined: Wed Mar 04, 2015 10:54 am

Re: Anova/ T-Test for different variables- thanks for your h

Postby Elenieth » Sat Mar 21, 2015 9:49 am

Good morning!

Thank you for the answers, iopsych and GerineL! I am really sorry for the late response; have only just emerged from two weeks with a really bad stomach virus. Back to statistics and online now!

Here is more information:

Batch number of questionnaire (of 2; beginning and end of the year),
Yes, I handed out the same questionnaire twice: once at the beginning and once at the end of the year. The students I gave it to were a random selection out of 500ish; some took the same questionnaires twice but some only took the first or the second. I am interested in whether there is a change in mean overall.

Type of student (intervention; one year after intervention; control),
Intervention group got the questionnaire immediately after the intervention, one year after group took it one year after, and control just took the questionnaire without the intervention- Yes, this is correct.

Gender and year group
You are right, I have not included the year group variable yet, as I did not know how to integrate more than two on the y axis. Overall, I am more interested in (intervention + age) or (intervention + year) as variables, but if there is a way to do all three at the same time that would be great.

GerineL, your summary is correct. I have tried to find explanations for repeated measures ANOVA but have been unsuccessful so far. Is there maybe a video you can recommend?

Sorry again for the late reply- hopefully I am all fighting fit now and it won't happen again. This is very important to me and I appreciate your time and support! Thank you!
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Anova/ T-Test for different variables- thanks for your h

Postby GerineL » Fri Mar 27, 2015 3:39 pm

as I said earlier, I think you would benefit a lot from looking at some youtube videos for the tests I suggested, as this will help you understand how to run it on your variables.
Elenieth
Posts: 3
Joined: Wed Mar 04, 2015 10:54 am

Re: Anova/ T-Test for different variables- thanks for your h

Postby Elenieth » Tue Mar 31, 2015 5:44 pm

Hi GerineL, thanks for posting again.

I had a look, but I always get stuck when people only use two variables- or they don't use them in the same order as I do. Would you be able to recommend a video?

Thank you!
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Anova/ T-Test for different variables- thanks for your h

Postby GerineL » Wed Apr 01, 2015 11:09 am

Reading your post again, I think what you need is Repeated MEasures Anova, because you want to see changes.

The questionnaire would then be the repeated measure, type of student, gender and year group would be factors. If any of these are measures on a scale (e.g.: age is a number that varies, not a group), you add them as covariates.
I am just not clear on whether the gorups intervention, one year after intervention, and control are also students that filled out the questionnaire twice.

Who is online

Users browsing this forum: No registered users and 4 guests

cron