Applying readily accessible graphical techniques to assess curvilinear relationships and detect outliers. The case of protective family processes

Eval Health Prof. 1997 Sep;20(3):353-64. doi: 10.1177/016327879702000307.


The purpose of this article is to use data from a study of protective family processes to illustrate how readily accessible graphical techniques can supplement correlation and regression analyses. Graphical techniques can help to (a) better understand family process data from a descriptive standpoint, (b) minimize erroneous statistical conclusions, and (c) clarify theoretically important relationships among variables. Following a brief description of the substantive area selected for illustration of the graphical techniques, data from a study employing multimethod measurement procedures (N = 146) are employed to illustrate two graphical technique applications in the analysis of a predictor (marital quality) of a study outcome variable (parent-child affective quality). Instructions for the application of relevant procedures in SAS and SPSS statistical packages are provided.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Child
  • Humans
  • Marriage / psychology*
  • Models, Theoretical
  • Parent-Child Relations*
  • Psychometrics / methods*
  • Regression Analysis
  • Surveys and Questionnaires