2003
DOI: 10.2307/30036541
|View full text |Cite
|
Sign up to set email alerts
|

In Pursuit of Moderation: Nine Common Errors and Their Solutions

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

10
263
0
3

Year Published

2006
2006
2019
2019

Publication Types

Select...
8
1
1

Relationship

0
10

Authors

Journals

citations
Cited by 422 publications
(276 citation statements)
references
References 74 publications
10
263
0
3
Order By: Relevance
“…As detailed in Appendix EC-3 (electronic companion), we (1) examined partial models and included additional twoway and three-way interactions in the models (Carte and Russell 2003), (2) used group analysis by splitting CSEs with high/low SEP or high/low WSC (Venkatesh 2000), (3) used an alternative ServQual measure to double check the effects of the feedback mechanisms (Zeithaml et al 1996), and (4) winsorized the two interaction terms (SEP × TechQual, WSC × ServQual) to reduce their variations (Carte and Russell 2003). The results of these additional tests suggest that the results in Figure 2 are robust to alternative model specifications.…”
Section: Robustness Of the Feedback Mechanismsmentioning
confidence: 99%
“…As detailed in Appendix EC-3 (electronic companion), we (1) examined partial models and included additional twoway and three-way interactions in the models (Carte and Russell 2003), (2) used group analysis by splitting CSEs with high/low SEP or high/low WSC (Venkatesh 2000), (3) used an alternative ServQual measure to double check the effects of the feedback mechanisms (Zeithaml et al 1996), and (4) winsorized the two interaction terms (SEP × TechQual, WSC × ServQual) to reduce their variations (Carte and Russell 2003). The results of these additional tests suggest that the results in Figure 2 are robust to alternative model specifications.…”
Section: Robustness Of the Feedback Mechanismsmentioning
confidence: 99%
“…To test H 2 and H 3 relative to the moderating effect of IT integration, a hierarchical procedure was followed (Carte and Russell, 2003), wherein a model that excluded the interaction construct (Innovation capability x IT integration) was first estimated. These first results were then compared to the results of estimating the "full" model in which both the main effects and the interaction effects were included, as presented in Table 4.…”
Section: Test Of the Research Hypothesesmentioning
confidence: 99%
“…In order to test the moderation effects, we estimated a "main effect" model (Model 3) with a moderating effect model (Model 4) (Carte and Russell 2003) and meeting nine conditions that no errors have been made. Our analysis concluded that no errors of commission were made.…”
mentioning
confidence: 99%