2
$\begingroup$

This may be a relatively simpleton kind of question to ask, as this forum seems to be rather statistically sophisticated, but I'm rather mixed up right now.

I ran a study that involved individuals performing a group activity and then were measured on an individual difference variable (extroversion). These groups either allowed the individuals to communicate with one another or not(experimental manipulation). I only have the groups' scores on the task as the DV ( there is no individual score for the task).

I would like to see if there is a difference in the effect of group members extroversion on group performance, and whether that effect differs between group conditions. To sum up:

  • DV = Group performance (Level 2)
  • IV = Individual Group Members' Extroversion (Level 1)
  • Moderator = Group Communication Condition (Level 2)

I am not sure how to appropriately perform a statistical test relevant to the above question, or even if such a question can be assessed from this data (I learned a lot about the importance of clearly understanding the statistical tests prior to commencement of research from this mess). I have a hunch on a variety of possible approaches, but I am not quite sure on how to execute, and would greatly appreciate some advice. I have SPSS and HLM at my disposal, but I am not sure how best to proceed with either of these tools.

I would greatly appreciate any advice that anybody could offer me. Thank you in advance.

This is my first post, so go easy on me :)

Improve this question
$\endgroup$
3
  • $\begingroup$ I'm sure that there are many approaches to this problem. IMHO, hierarchical [aka multilevel] structural equation modeling (SEM) / latent variable modeling (LVM) is one of them, but likely not the simplest one (additionally, it requires AMOS - an optional and quite expensive SPSS module for SEM). Check this paper for some details on multilevel SEM. $\endgroup$
    – Aleksandr Blekh
    Commented Jan 19, 2015 at 4:48
  • $\begingroup$ I gather you have 2 conditions. How many groups, & individuals per group, do you have? $\endgroup$
    – gung - Reinstate Monica
    Commented Jan 19, 2015 at 4:54
  • $\begingroup$ In response to gung, 3 to 4 individuals per group and 75 groups. $\endgroup$
    – Lavish Logic
    Commented Jan 19, 2015 at 15:18

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.