Development and Validation of the Chestnut Global Partners Workplace Outcome Suite

Abstract

This article describes the development and validation of a five-item scale Workplace Outcome Suite designed as an open access instrument aimed at facilitating empirical research on Employee Assistance Program (EAP) interventions. The suite contains five-item measures of absenteeism, presenteeism, work engagement, life satisfaction, and workplace distress. All but the absenteeism measures are effect-indicator structures derived from classical psychometric theory. The absenteeism measure used a formative model that captures the individual components of being away from the job site because of personal problems. These components are not thought to be internally consistent with one another but rather represent distinct manners that force time off. Two separate validation studies of the suite, one with a paper-and-pencil modality (N = 220) and another with a telephone interview modality (N = 228) tested the reliability of the scales, the structural validity of the items, and the construct validity of the unit-weighted scale scores. The effect-indicator scales were found to have moderate (in the range of .75) to excellent levels (in the range of .90) of internal consistency. Only two items in the Work Engagement scale produced low factor loading for the telephone interview study, but the low loading did not replicate in the paper-and-pencil study and thus was considered spuriously low for the time being. None of the items produced factor loading below .30 in the paper-and-pencil study. Correlations of the scale scores with self-reported measures of relevant behavior and emotions provided limited evidence of construct validity for all five scales. The results suggest support for the use of the Workplace Outcome Suite to evaluate EAP services and interventions.

KEYWORDS:

• EAP evaluation
• Employee Assistance Programs
• outcome measures
• workplace outcomes

Although the measures are copyrighted by Chestnut Global Partners, they are available for unqualified public use. We request that those interested contact us prior to using the measures so we can keep track of how the measures are being used by the industry. Questions regarding the use of the measures should be directed to the corresponding author.

Notes

Note. ^a Absenteeism is a cause indicator model for which coefficient alpha and item total correlations are irrelevant for estimating reliability.

Note. ^a Absenteeism is a cause indicator model for which factor analysis is inappropriate.

Note. AB = absenteeism; PR = presenteeism; WE = work engagement; LS = life satisfaction; WD = workplace distress.

**p < .01.