Engagement Views Notes

Model

Our basic model is:

Motivation -> Engagement -> Achievement

We are trying to test this using time 1 self-reported variables, observed measures of engagement (through trace data), and final exam and grade scores.

Questions

Key questions concerns how to process the behavioral data, how to analyze it, and on what time scale to measure it, and then how to proceed with the overall data analysis.

How to process the behavioral data:

• Mean and standard deviation of time watched
• BLUP and standard error of time watched
• Longitudinal model using intercept and slope

How to analyze the behavioral data

• Using the variables (i.e., mean and standard deviation or intercept and slope) on their own
• Creating profiles of the variables

On what time scale to measure it:

• Before the first exam
• Over the entire semester

Overall data analysis:

• Using two separate regressions
• Using a path model as part of a latent variable model approach