[Feature Request]: MixedModelsLMM with only ~1 in fixed part

[hugoquene]

Description

allow LMM with intercept-only fixed part

Purpose

1. To compare null model without and fuller model with predictor in their fixed part, while allowing random intercepts. 2. To assess ICC of empty model, as per Snijders & Bosker (1999, 1E).

Use-case

See e.g. Twisk (2010) Chapter 3?; Snijders & Bosker 1999 on ICC

Is your feature request related to a problem?

Able to compare model with and without predictor, while taking random intercepts into account.

Is your feature request related to a JASP module?

Mixed Models

Describe the solution you would like

In MixedModelsLMM, allow formula = y~1+(1|clusters), and report the LMM.

Describe alternatives that you have considered

add a column of constants (1), and use that as predictor

Additional context

Ideally I would also like to compare
y ~ 1 + (1|clusters),
y ~ 1 + (1+x|clusters),
and y ~ 1 + x + (1+x|clusters).
Currently only the latter LMM is possible. But comparing these three models is often relevant.

See also Faraway (2006) p.171 (ISBN 978-1-58488-424-8)

[R180]

I think my request is the same. JASP does not allow an intercept-only logistic linear mixed effects model. The use case is to ascertain whether a proportion is significantly different from 0.5, while accounting for a random effect. See the attached JASP file.
will not run intercept-only logistic generalized linear mixed effects model.zip