Why Using Segmented Regression Analysis to Explore Change in Relations Between Variables Is Problematic

• In Event: Marriott Roundtable Session Five
  In Roundtable Session: Advanced Research Methodology (Table 1)

Abstract

Relations between variables can take different forms like linearity, piecewise linearity, or non-linearity. Segmented regression analyses (SRA) are specialized statistical methods that detect breaks in the relation between variables. They are commonly used for exploratory analyses. However, many relations may not be best described by a breakpoint, but rather by nonlinearity. In the present simulation study, we examined the application of SRA in the presence of various forms of nonlinearity. We found that moderate and strong degrees of nonlinearity led to a frequent identification of statistically significant breakpoints, clearly indicating that SRA cannot be used for exploratory analyses. We propose alternative statistical methods for exploratory analyses and outline the conditions for a legitimate use of SRA.

Authors

• Moritz Lion Breit, University of Trier
  Presenting Author

• Julian Preuß, University of Trier
  Non-Presenting Author

• Vsevolod Scherrer, University of Trier
  Non-Presenting Author

• Franzis Preckel, University of Trier
  Non-Presenting Author