Skip to main content
Have a personal or library account? Click to login
An Unsupervised Learning Approach to Evaluate Questionnaire Data—What One Can Learn from Violations of Measurement Invariance Cover

An Unsupervised Learning Approach to Evaluate Questionnaire Data—What One Can Learn from Violations of Measurement Invariance

Open Access
|Mar 2024

Figures & Tables

Figure 1

Graphical representation of the gap statistic as well as the dendrogram corresponding to the goodness of the clustering of the questionnaires in data set D1. Moreover, the corresponding response types are shown as a spider plot.

Figure 2

The fingerprints of the different groups regarding the response types as spider plots. The radial y-axes are scaled to (0, 0.7). Also, the group similarity on data set D1 is given by a dendrogram.

Figure 3

Graphical representation of the gap statistic as well as the dendrogram corresponding to the goodness of the clustering of the questionnaires in data set D2. Moreover, the corresponding response types are shown as a spider plot.

Figure 4

The fingerprints of the different groups regarding the response types as spider plots. The radial y-axes are scaled to (0, 0.7). Moreover, the group similarity on data set D2 is shown as a dendrogram.

Figure 5

Graphical representation of the gap statistic as well as the dendrogram corresponding to the goodness of the clustering of the questionnaires in data set D3. Moreover, the corresponding response types are shown as a spider plot.

Figure 6

The fingerprints of the different groups regarding the response types as spider plots. The radial y-axes are scaled to (0, 0.7). Again, the group similarity on data set D3 is visualized by a dendrogram.

Figure 7

The similarity between the groups in data set D1 for a growing number of response types (left: 5 response types, right: 8 response types). The similarity based on the optimal number of response types (5) gives no significantly different result as the similarity based on more response types.

Figure 8

The eight response types on data set D1. While the similarity between group fingerprints does not vary if more response types are used, the response types themselves are much worse separated.

Language: English
Page range: 13 - 13
Submitted on: Dec 8, 2023
Accepted on: Feb 27, 2024
Published on: Mar 27, 2024
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2024 Max Hahn-Klimroth, Paul W. Dierkes, Matthias W. Kleespies, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.