2007
DOI: 10.1177/1094428107300344
|View full text |Cite
|
Sign up to set email alerts
|

Tests of the Three-Path Mediated Effect

Abstract: In a three-path mediational model, two mediators intervene in a series between an independent and a dependent variable. Methods of testing for mediation in such a model are generalized from the more often used single-mediator model. Six such methods are introduced and compared in a Monte Carlo study in terms of their Type I error, power, and coverage. Based on its results, the joint significance test is preferred when only a hypothesis test is of interest. The percentile bootstrap and bias-corrected bootstrap … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

11
853
2
9

Year Published

2011
2011
2023
2023

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 1,006 publications
(875 citation statements)
references
References 28 publications
11
853
2
9
Order By: Relevance
“…B. Taylor, MacKinnon, & Tein, 2008). Both supported three-path mediation of the layoff effect through wage underemployment and job satisfaction, although we note that a relatively small proportion of the total effect operates through this causal chain (i.e., less than 1%).…”
Section: Mediationmentioning
confidence: 63%
“…B. Taylor, MacKinnon, & Tein, 2008). Both supported three-path mediation of the layoff effect through wage underemployment and job satisfaction, although we note that a relatively small proportion of the total effect operates through this causal chain (i.e., less than 1%).…”
Section: Mediationmentioning
confidence: 63%
“…Replicating our mediation analysis in study 3, a threepath mediation test (Taylor et al 2008) supports hypothesis 3, showing that perceived fit significantly mediates the effect of the frequency cue on relative usage frequency, which then mediates the effect of perceived fit on willingness to pay. Table 5 shows the three regressions required for the joint significance test.…”
Section: Resultsmentioning
confidence: 79%
“….001 in the high-frequency condition). As shown in table 3, supporting hypothesis 3, a three-path mediation test (Taylor, MacKinnon, and Tein 2008) shows that perceived fit significantly mediates the effect of the frequency cue on perceived relative usage frequency, which then mediates the effect of perceived fit on product interest (to demonstrate three-path mediation, the joint significance test requires that b 1 , b 2 , and b 3 are all significant). In contrast, the frequency scale manipulation did not affect cost per usage inferences significantly (p 1 .25), suggesting that this is not the mechanism being invoked by the manipulation of the frequency scale.…”
Section: Perceived Fit and Cost Per Usage Inferencesmentioning
confidence: 99%
“…Techniques are also available for analyzing complex multilevel mediation in which indirect effects occur in clustered data that include both between-groups and nested within-subjects measurements in the same analysis (Preacher, Zyphur, & Zhang, 2010). Researchers have now developed statistical procedures for testing the statistical significance of three-step path mediation, in which A  B  C  D (Taylor, MacKinnon, & Tein, 2008). In addition, researchers have developed statistical procedures for testing interactional hypotheses about ‚moderated mediation‛ (Preacher, Rucker, & Hayes, 2007), in which another independent variable influences the strength of a particular mediating mechanism, such that mediation is stronger under certain circumstances, for certain types of events, or for certain types of people, compared to others.…”
Section: Mediational Effectsmentioning
confidence: 99%