Effektstärkevarianten beim Eingruppen-Prä-Post-Design: Eine kritische Betrachtung
The DOI matches its Crossref record.
Abstract
Rehabilitationswissenschaftliche Untersuchungen - insbesondere solche zum Zweck der Evaluation und zur Überprüfung der Ergebnisqualität - folgen bisher überwiegend dem „naturalistischen” Eingruppen-Prä-Post-Design. Die Beurteilung der Ergebnisse orientiert sich gewöhnlich an den Mittelwertdifferenzen der Outcome-Variablen, wobei die Prä-Post-Unterschiede zufallskritisch geprüft und - falls signifikant - zunehmend als Effektstärken präsentiert werden. Aus diesem Anlass werden in dem Beitrag - trotz der grundsätzlichen methodischen Bedenken, die in Bezug auf Eingruppen-Prä-Post-Untersuchungen aufgrund ihrer geringen internen Validität vorliegen - verschiedene Effektstärke-Indizes vorgestellt, die für dieses Design Verwendung finden. Während der Zähler bei allen Effektstärkevarianten des Eingruppen-Prä-Post-Designs einheitlich als Differenz zwischen den Mittelwerten der untersuchten Stichprobe zu den beiden Messzeitpunkten ausgewiesen wird, existieren bezüglich des Nenners und damit der Standardabweichung, anhand derer die Mittelwertdifferenz standardisiert werden soll, unterschiedliche Möglichkeiten und Empfehlungen. Verwendung finden vor allem die Standardisierung an der Standardabweichung der Prä-Werte, die Standardisierung an der gepoolten Standardabweichung der Prä- und Post-Werte und die Standardisierung an der Standardabweichung der Prä-Post-Differenzen. Anhand von zwei Datenbeispielen wird gezeigt, dass die unterschiedlichen Berechnungsvarianten für Effektstärke-Indizes im Eingruppen-Prä-Post-Design zu höchst unterschiedlichen Ergebnismustern führen können. Darüber hinaus werden Hinweise dafür vorgestellt, dass Effektstärken für unkontrollierte Prä-Post-Designs im Allgemeinen größer ausfallen als Effektstärken aus Kontrollgruppenuntersuchungen. Abschließend werden die jeweiligen Vor- und Nachteile der verschiedenen Prä-Post-Effektstärkevarianten diskutiert und Empfehlungen abgeleitet. Effect Size Variations in the Single Group Pre-Post Study Design: A Critical View: In Germany, studies in rehabilitation research - in particular evaluation studies and examinations of quality of outcome - have so far mostly been executed according to the uncontrolled one-group pre-post design. Assessment of outcome is usually made by comparing the pre- and post-treatment means of the outcome variables. The pre-post differences are checked, and in case of significance, the results are increasingly presented in form of effect sizes. For this reason, this contribution presents different effect size indices used for the one-group pre-post design - in spite of fundamental doubts which exist in relation to that design due to its limited internal validity. The numerator concerning all effect size indices of the one-group pre-post design is defined as difference between the pre- and post-treatment means, whereas there are different possibilities and recommendations with regard to the denominator and hence the standard deviation that serves as the basis for standardizing the difference of the means. Used above all are standardization oriented towards the standard deviation of the pre-treatment scores, standardization oriented towards the pooled standard deviation of the pre- and post-treatment scores, and standardization oriented towards the standard deviation of the pre-post differences. Two examples are given to demonstrate that the different modes of calculating effect size indices in the one-group pre-post design may lead to very different outcome patterns. Additionally, it is pointed out that effect sizes from the uncontrolled one-group pre-post design generally tend to be higher than effect sizes from studies conducted with control groups. Finally, the pros and cons of the different effect size indices are discussed and recommendations are given.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cited by
Hans, E., & Hiller, W. (2013) · Clinical Psychology Review · doi:10.1016/j.cpr.2013.07.003 · Cited by 158
Plueck, J., Eichelberger, I., Hautmann, C., Hanisch, C., Jaenen, N., & Doepfner, M. (2015) · Prevention Science · doi:10.1007/s11121-014-0487-x · Cited by 30
Washburn, M., Yu, M., Rubin, A., & Zhou, S. (2021) · Journal of Telemedicine and Telecare · doi:10.1177/1357633x211009647 · Cited by 5
(2011) · Verhaltenstherapie · Cited by 3
Metadata
- Authors
- Maier-Riehle, B., & Zwingmann, C.
- Venue
- Die Rehabilitation
- Publisher
- Georg Thieme Verlag KG
- Published
- 2000-08
- Volume
- 39
- Issue
- 4
- Pages
- 189-199
- ISSN
- 0034-3536 (ISSN-L)
- 1439-1309
- OpenAlex
- W2083242658
- Cited by
- 66
- References
- 0
- FWCI
- 0.62
- Type
- journal article
- Language
- de
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.