Interaction in group setting

pmeiners commented 3 years ago

Thank you for the great package. It is possible to compute non-linear interaction effects between treatments and a group-level moderator? Currently, both inter.raw and interflex only produce the error message: "Can't subset columns that don't exist. x Location 2 doesn't exist. ℹ There are only 1 column."

Using version 1.1.3

lzy318 commented 3 years ago

Hi, By now the package only allows group-level treatments (D) but not a group-level moderator (X). And in the setting of the paper, the moderator should be continuous. If both of your treatment and moderator are categorical, maybe a group-by-group comparison is enough.

pmeiners commented 3 years ago

I see, thanks for the info. I was looking at a situation with a group level continuous moderator.

Do you know if there is any specific reason, group level-moderators are not allowed?

lzy318 commented 3 years ago

font{
    line-height: 1.6;
}
ul,ol{
    padding-left: 20px;
    list-style-position: inside;
}

Hi. I don't catch up what your "group level continuous" moderator means here. Do you mean you have several ordered groups, such as "less than 10", "10 to 20" and "larger than 20", and you treat them as the moderator?

    font{
        line-height: 1.6;
    }

Ziyi Liu (刘子毅)Predoctoral Fellow at The University of Chicago Center in BeijingEPIC-ChinaTel: +86 18942900149

On 3/17/2021 ***@***.***> wrote：

I see, thanks for the info. I was looking at a situation with a group level continuous moderator. Do you know if there is any specific reason, group level-moderators are not allowed?

—You are receiving this because you commented.Reply to this email directly, view it on GitHub, or unsubscribe.

pmeiners commented 3 years ago

I'm sorry, let me be more clear. I am interested in the interaction between a group level moderator and an individual level covariate. You could also call it a cross-level interaction.

xuyiqing / interflex

Interaction in group setting #8