Abstract
Background: Mediation modeling can explain the nature of the relation among three or more variables. In addition, it can be used to show how a variable mediates the relation between levels of intervention and outcome. The Sobel test, developed in 1990, provides a statistical method for determining the influence of a mediator on an intervention or outcome. Although interactive Web-based and stand-alone methods exist for computing the Sobel test, SPSS and SAS programs that automatically run the required regression analyses and computations increase the accessibility of mediation modeling to nursing researchers.
Objectives: To illustrate the utility of the Sobel test and to make this programming available to the Nursing Research audience in both SAS and SPSS.
Methods: The history, logic, and technical aspects of mediation testing are introduced. The syntax files sobel.sps and sobel.sas, created to automate the computation of the regression analysis and test statistic, are available from the corresponding author.
Results: The reported programming allows the user to complete mediation testing with the user's own data in a single-step fashion. A technical manual included with the programming provides instruction on program use and interpretation of the output.
Conclusion: Mediation modeling is a useful tool for describing the relation between three or more variables. Programming and manuals for using this model are made available.
The early work on mediation and moderation by Baron and Kenny (1986) provided behavioral researchers with a powerful analytical tool: mediation modeling. This type of statistical modeling has recently gained attention in nursing research literature. The research of Bennett (2000) and Bennett, Stewart, Kayser-Jones, and Glaser (2002) provided an application for mediation modeling. The case for using mediation modeling in cancer nursing to test models that explain why cancer patients often experience clusters of symptoms was made by Dudley, Beck, and Barsevick (2003).
A variable is defined as a mediating variable if it accounts for the relationship between two other variables (Baron & Kenny, 1986;MacKinnon, 1994).
Mediation is often useful for conceptualizing the relation among three or more variables as a way to explain the nature of the relation. For instance, Williamson and Schulz (1995) showed that the relation between pain and depression in patients with cancer was attributable to the mediating variable, function. In their study, patients in pain experienced a decline in function as measured by their ability to accomplish activities of daily living, and this decline led to increased depression. Thus function, or in this case a decrease in function, was a mediating variable in that it accounted for the relation between pain and depression.
This finding has important implications in that it opens the possibility of nonpharmacologic approaches to easing pain-related depression (Williamson & Schulz, 1995). For example, interventions to improve a patient's ability to function within the limitations imposed by pain (e.g., patient education or restructuring of the environment) may alleviate depression.