Conceptual question on Machine Learning - Clustering

EmilieC

Dear all,

I have a data set with 3 distinct groups of participants, and several dependent variables. I want to do a cluster analysis with the neighborhood-based clustering method to see if based on the dependent variable, it is possible for it to discriminate between my three groups accurately or not (in other words, if their data for several neuro-cognitive processes allow to distinguish the group they belong too).

My existing data file is made for linear mixed models, with several lines per participants. As a results, when I run the analysis, I have my requested 3 clusters, but all lines of a single participant are assigned to different clusters (see picture), which actually make sense.

I was wondering if I had to re-do another file with for instance only the means, and so have 1 line/participant, or if there was a way to deal with that in the existing functions.

Thank you

Emilie

Capture d’écran 2024-11-06 à 12.09.03.png

koenderks

Hi Emilie,

No, there is no way. You need a datafile in which all data for a participant are in a single row.

EmilieC

Thank you!