Modeling of confounding effects and so on

Welcome to the forum for MLwiN users. Feel free to post your question about MLwiN software here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Remember to check out our extensive software FAQs which may answer your question: ... port-faqs/
Post Reply
Posts: 1
Joined: Thu Jul 19, 2012 6:12 am

Modeling of confounding effects and so on

Post by MLMmouse »

Hi forum,

Ive inherited data from a, lets say, rather “tricky” quasi-experimental design. Its “basically” repeated measurement. 8 classes of 3 schools were measured 4 times.
Altogether, each student was measured 4 times. Each cluster has 20+ students. The outcome measures were school scores. That’s the easy part. Now comes the tricky part:
1. The four tests used were not standardized tests (different topics, difficulty may be also inconsistent). In fact, every time (t1, t2, t3, t4) a different test stimulation was applied.
2. Every second test situation was a different test motivation for the students. T1 and t3 were “easy” testing situations. T2 and t4 were “serious” testing situations.
Yes, before you wonder: No controls, no baseline, just 4times measured differently and confounding effects abound.
And, one of the favoruite resreach questions is: Are there differences between the uneven and even measurement times (t1,3 vs. t2,4)?
Any idea whether this could be modelled by HLM or mixed modelling?

I appreciate every bit of advice!

Post Reply