Suppressor Variables in Social Work Research: Ways to Identify in Multiple Regression Models

Pandey, Shanta; Elliott, William

doi:10.5243/jsswr.2010.2

Cited by 184 publications

(156 citation statements)

References 35 publications

Supporting

Mentioning

149

Contrasting

Order By: Relevance

“…Ironically, suppressor variables may more accurately be named 'enhancers' because they suppress the 'noise' they share with one (or multiple) predictors and therefore clarify relationships between predictors and outcome variables, increasing the predictive power of the regression model (Pandey & Elliott, 2010). We contend that the positive variance shared between affective rumination and problemsolving pondering reflects the fact that sometimes people who ruminate with a problemsolving focus will also have an emotional response (which is fundamental to affective rumination).…”

Section: Discussionmentioning

confidence: 99%

“…If either of these variables was removed from the model, this would present an incomplete picture of the relationship of the two types of rumination to fatigue. This highlights the importance of ensuring suppressor variables are included in multiple regression models because they improve the explanatory power of the model and remove irrelevant shared 'noise' between predictor variables (Pandey & Elliott, 2010). Our considered position is that while both forms of rumination can interfere with recovery, it is when problem-solving ruminators also have an emotional response that recovery is most compromised.…”

Section: Discussionmentioning

confidence: 99%

“…However, when they are entered into a regression equation together, problem-solving pondering takes on a negative sign. According to Pandey & Elliott (2010), problem-solving pondering meets the criteria of a negative suppressor variable. It's inclusion in the regression equation clarifies the relationship between affective rumination and the two forms of fatigue.…”

Section: [Insert Table 4 About Here]mentioning

confidence: 99%

See 2 more Smart Citations

Exploring the relationship between work-related rumination, sleep quality, and work-related fatigue.

Querstret

Cropley

2012

Journal of Occupational Health Psychology

283

353

View full text Add to dashboard Cite

Objective: This study examined the association between three conceptualisations of work-related rumination (affective rumination, problem-solving pondering and detachment) with sleep quality and work-related fatigue. It was hypothesised that affective rumination and poor sleep quality would be associated with increased fatigue; and problem-solving pondering, and detachment would be associated with decreased fatigue. The mediating effect of sleep quality on the relationship between work-related rumination and fatigue was also tested. Method: An on-line questionnaire was completed by a heterogeneous sample of 719 adult workers in diverse occupations. Results: The following variables were entered as predictors in a regression model: affective rumination, problem-solving pondering, detachment, and sleep quality. The dependent variables were chronic work-related fatigue (CF) and acute work-related fatigue (AF). Affective rumination was the strongest predictor of increased CF and AF. Problem-solving pondering was a significant predictor of decreased CF and AF. Poor sleep quality was predictive of increased CF and AF. Detachment was significantly negatively predictive for AF. Sleep quality partially mediated the relationship between affective rumination and fatigue; and between problem-solving pondering and fatigue. Conclusions: Work-related affective rumination appears more detrimental to an individual's ability to recover from work than problem-solving pondering. In the context of identifying mechanisms by which demands at work are translated into ill-health, this appears to be a key finding; and suggests that it is the type of work-related rumination, not rumination per se, that is important.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: [Insert Table 4 About Here]mentioning

confidence: 99%

See 1 more Smart Citation

Exploring the relationship between work-related rumination, sleep quality, and work-related fatigue.

Querstret

Cropley

2012

Journal of Occupational Health Psychology

283

353

View full text Add to dashboard Cite

show abstract

“…It does not relate to the DV in the way theorized, but operates "as a measure of the sources of error" in the other predictor (Darlington, 1990, p. 155), whose effect is stronger. Put another way, the predictor whose sign has changed accounts for (or suppresses) a portion of the variance in the other predictor that is unrelated to the DV (Pandey & Elliott, 2010).…”

Section: Resultsmentioning

confidence: 99%

What residualizing predictors in regression analyses does (and what it does not do)

Wurm

Fisicaro

2014

Journal of Memory and Language

141

125

View full text Add to dashboard Cite

“…Some other candidate independent variables proven not to be significantly correlated with the dependent variable but were still kept for subsequent analysis, to account for possible suppressor effects [Pandey et al 2010]. …”

Section: Conceptmentioning

confidence: 99%

Information Systems Design and User Satisfaction: A Banking Case Study

Padinha

O’Neill

2015

Atas Da 15ª Conferência Da Associação Portuguesa De Sistemas De Informação

View full text Add to dashboard Cite

With the growing, pervasive use of information systems (IS) by most organizations, particularly companies in technology-intensive businesses like banking, questions arise as to the effective success of those systems, how to measure it and, most of all, what influences it. User satisfaction, even if not an end goal of IS organizations, is frequently used as a measure of that success and it has been empirically shown to correlate with other more direct measures. The research reported here looked at what influences user satisfaction, particularly what characteristics of information systems and their development processes tend to result in higher user satisfaction, in the specific case study of a Portuguese bank, which conducts a comprehensive survey every year to measure the satisfaction of its thousands of IS users regarding its numerous information systems. Additional primary data was collected to characterize the information systems and their development processes, which, once structured and standardized alongside the user satisfaction survey results provided the basis for a quantitative analysis of the relationships between IS success, measured by user satisfaction, and its antecedents, using correlation and multiple linear regression techniques. The significant relationships are analyzed and the conceptual model's results discussed in light of prior theory on the subject. Several implications for the practice and research communities are finally identified.

show abstract

Suppressor Variables in Social Work Research: Ways to Identify in Multiple Regression Models

Cited by 184 publications

References 35 publications

Exploring the relationship between work-related rumination, sleep quality, and work-related fatigue.

Exploring the relationship between work-related rumination, sleep quality, and work-related fatigue.

What residualizing predictors in regression analyses does (and what it does not do)

Information Systems Design and User Satisfaction: A Banking Case Study

Contact Info

Product

Resources

About