Impact of Time Intervals on Lagged Moderated Regression Effects
Author(s) / Creator(s)
Dormann, Christian
Cortina, Jose
Abstract / Description
(a) Background: Moderated regression analysis is the most frequently applied statistical method to analyze interaction/moderation effects in the applied psychology literature, and one of the most common statistical techniques of any kind (e.g., Aguinis, & Stone-Romero, 1997; Cohen, Cohen, West, & Aiken, 2003). Further, appreciation of the need for longitudinal studies has led to an increase in the number of studies that used lagged moderated regression analysis or related methods such as multi-sample structural equation models. It is, however, not well-known that results of lagged moderation analysis could be misleading if time intervals are not appropriately modelled. This was shown, e.g., in a recent meta-analysis of longitudinal studies (Guthier, Dormann, & Voelkle, 2020), but it also applies to primary studies.
(b) Objectives/Research question(s): The objective is to identify conditions that lead to misleading results (e.g., wrong signs) from lagged moderation analysis and provide a solution.
(d) Method/Approach: Monte Carlo Simulation of longitudinal data and analysis of generated data with different multiple regression models and moderated ctsem
(e) Results/Findings: If more than a single effect in a causal system is moderated, length of interval is particularly consequential. Truly positive moderating effects can manifest as negative moderating effects and vice versa (sign-flipping) if moderated regression models are used. In particular, this happens if more than the focal lagged effect is moderated (e.g., a lagged effect in the 'reversed' causal direction) but with a different sign. Contrary, moderated ctsem yields less biased and in some cases unbiased estimates.
(f) Conclusions and implications: Interpretations of previously published longitudinal moderation analyses should be treated with caution, and moderated ctsem instead of moderated regression analysis should be to analyze moderation with longitudinal data.
Keyword(s)
continuous time structural equation modeling moderationPersistent Identifier
Date of first publication
2021-05-19
Is part of
Research Synthesis & Big Data, 2021, online
Publisher
ZPID (Leibniz Institute for Psychology)
Citation
Dormann, C., & Cortina, J. (2021). Impact of Time Intervals on Lagged Moderated Regression Effects. ZPID (Leibniz Institute for Psychology). https://doi.org/10.23668/PSYCHARCHIVES.4836
-
Dormann Cortina 2021 Research Synthesis Big Data Conference V1.2.pptxMicrosoft Powerpoint XML - 997KBMD5: 6c664a418874d0eae8c48820fc99bae8Description: Presentation
-
Dormann Cortina 2021 Research Synthesis Big Data Conference V1.2.pdfAdobe PDF - 4.02MBMD5: d5e4dd94628715644378ac1bfe445afcDescription: Presentation
-
There are no other versions of this object.
-
Author(s) / Creator(s)Dormann, Christian
-
Author(s) / Creator(s)Cortina, Jose
-
PsychArchives acquisition timestamp2021-05-14T16:13:20Z
-
Made available on2021-05-14T16:13:20Z
-
Date of first publication2021-05-19
-
Abstract / Description(a) Background: Moderated regression analysis is the most frequently applied statistical method to analyze interaction/moderation effects in the applied psychology literature, and one of the most common statistical techniques of any kind (e.g., Aguinis, & Stone-Romero, 1997; Cohen, Cohen, West, & Aiken, 2003). Further, appreciation of the need for longitudinal studies has led to an increase in the number of studies that used lagged moderated regression analysis or related methods such as multi-sample structural equation models. It is, however, not well-known that results of lagged moderation analysis could be misleading if time intervals are not appropriately modelled. This was shown, e.g., in a recent meta-analysis of longitudinal studies (Guthier, Dormann, & Voelkle, 2020), but it also applies to primary studies. (b) Objectives/Research question(s): The objective is to identify conditions that lead to misleading results (e.g., wrong signs) from lagged moderation analysis and provide a solution. (d) Method/Approach: Monte Carlo Simulation of longitudinal data and analysis of generated data with different multiple regression models and moderated ctsem (e) Results/Findings: If more than a single effect in a causal system is moderated, length of interval is particularly consequential. Truly positive moderating effects can manifest as negative moderating effects and vice versa (sign-flipping) if moderated regression models are used. In particular, this happens if more than the focal lagged effect is moderated (e.g., a lagged effect in the 'reversed' causal direction) but with a different sign. Contrary, moderated ctsem yields less biased and in some cases unbiased estimates. (f) Conclusions and implications: Interpretations of previously published longitudinal moderation analyses should be treated with caution, and moderated ctsem instead of moderated regression analysis should be to analyze moderation with longitudinal data.
-
Publication statusunknown
-
Review statusunknown
-
CitationDormann, C., & Cortina, J. (2021). Impact of Time Intervals on Lagged Moderated Regression Effects. ZPID (Leibniz Institute for Psychology). https://doi.org/10.23668/PSYCHARCHIVES.4836en
-
Persistent Identifierhttps://hdl.handle.net/20.500.12034/4273
-
Persistent Identifierhttps://doi.org/10.23668/psycharchives.4836
-
Language of contenteng
-
PublisherZPID (Leibniz Institute for Psychology)
-
Is part ofResearch Synthesis & Big Data, 2021, online
-
Keyword(s)continuous time structural equation modeling
-
Keyword(s)moderation
-
Dewey Decimal Classification number(s)150
-
TitleImpact of Time Intervals on Lagged Moderated Regression Effects
-
DRO typeconferenceObject
-
Visible tag(s)ZPID Conferences and Workshops