Measurement Error in Panel Analyses PreviousNext
Mplus Discussion > Structural Equation Modeling >
 seth goldman posted on Friday, February 26, 2010 - 6:56 am
Hi. I read other posts that gave some direction on how to incorporate estimates of reliability into analyses, but I'm not clear on how to do so in the case of panel data involving multiple measurements of the same individuals over time. In my particular case, I have three waves of data. Thanks, Seth
 Linda K. Muthen posted on Friday, February 26, 2010 - 9:39 am
You would do it the same way as for a variable in a cross-sectional study. See Slide 44 of the Topic 1 course handout.
 seth goldman posted on Friday, February 26, 2010 - 10:13 am
Would that mean taking the average of the sample variances of the three items--that is, the variable from each of the three waves of the survey? For instance, racial prejudice was assessed at each wave, so there are three items: prejudice_1 prejudice_2 prejudice_3.
 Bengt O. Muthen posted on Saturday, February 27, 2010 - 6:18 am
You could do it that way, but you could also do the reliability setting for each of the 3 waves, so using the 3 different sample variances. That accomodates allowing for different error variances at the 3 waves.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message