Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Unable to define between-subject basic unit in a mixed design !!!

From   Bastien Christian <>
Subject   st: Unable to define between-subject basic unit in a mixed design !!!
Date   Mon, 4 Feb 2013 17:38:16 +0100

Dear Colleagues,

In a research dedicated to evaluating the errors subjects make in evaluating distances in a real and augmented reality, we used a mixed experimental design with 1 between-subject factor and 3 within-subject factors.

The between-subject factor, "Groupe", has two levels (Reel-RA, RA-Reel). It consists in counterbalancing the order of the context the subjects are confronted to (Real-augmented reality, Augmented reality-real).
There are three repeated variables, "Phase" (1 and 2), Distance (5 levels) and Repetition (8 levels).
Each subject has 80 values recorded (Phase (2) * Distance (5) * Repetition (8) = 80).

In other words subjects are submitted to two phases. In each phase they are submitted to 5 distances and for each distance a stimulus is presented 8 times.
As I said in one group subjects go through a real than an augmented reality, and the participants of group 2 go through an augmented reality than a real context.
The dependent variable is "Errreur_Absolue".

I tried this model but got the following error message "could not determine between-subject basic unit; use bseunit() option" !

anova Erreur_Absolue Groupe / Participant|Groupe Phase Groupe#Phase / Phase#Participant|Groupe Distance Groupe#Distance / Distance#Participant|Groupe Repetition Groupe#Repetition / Repetition#Participant|Groupe Phase#Distance Phase#Repetition Distance#Repetition Phase#Distance#Repetition, repeated(Phase Distance Repetition) bse(Participant|Groupe) baseunit(Participant)

I just can't figure out what is wrong !!!

By the way, how do you do "contrasts" analyses so as to answer the following question : Are the participants' Erreur_Absolue different between real and augmented situations which consists in comparing  [(Group1, Phase1) and (Group2, Phase2)] to [(Group2, Phase 1) and (Group1, Phase2)]

Thanks a logy for your help,


*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index