Option of assuming level-1 error variance differs by treatment condition.

[jepusto]

In the web-app, provide option of assuming level-1 error variance differs by treatment condition

• Default to assuming homoskedasticity.
• Allow user to select a model with heteroskedastic level-1 error.

[manchen07]

It seems that this option can be added without relying on the lmeInfo updates?

[jepusto]

Put this option and the dependence structure options in a box called "Within-case error structure assumptions" that appears below the boxes with random effects options.

[jepusto]

Ha, yeah that seems right. Because we only need the baseline within-phase variance.

…

On Jun 5, 2022, at 6:29 PM, Man Chen ***@***.***> wrote:  It seems that this option can be added without relying on the lmeInfo updates? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.