DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Engaged managers are not workaholics: Evidence from a longitudinal person-centered analysis

Anne Mäkikangas1, Wilmar Schaufeli2,

Asko Tolvanen1, & Taru Feldt1

Department of Psychology, University of Jyväskylä, Finland1

Utrecht University, The Netherlands2

3 July 2014, 11-12Session: The role of work engagement

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Circumplex Model of Employee Well-being

Adopted from Russell (1980) and Warr (1994) (see also Bakker & Oerlemans, 2011)

Activation

Deactivation

PleasureDispleasure

Burnout

Workaholism Work engagement

Job satisfaction

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

The Aims of the Study

1) To investigate the relation between work engagementand workaholism by utilizing both variable- and person-centered approaches

2) To explore whether and how experiences of workengagement and workaholism relate to job changeduring the study period

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Participants and Study variables

o Work Engagement; UWES (9-items;

Schaufeli et al., 2006), α = .87-91

o Workaholism; DUWAS (10-items;

Schaufeli et al., 2008), α = .68-.79

o Job change at T2 (1 = stayer, 2 = mover)

- mover n = 96

o Technical and commercial managers

(total n = 3000) were randomly

selected from the membership

registers of two Finnish national

labor unions

o n = 902 (34%) T1, and n = 491 (68%)

T2

o Two-wave panel data (n = 463)

o 47% top management, 53% middle

management

o 46 years old (25-68, SD = 9.15)

o Working hours 46 per week (SD =

7.08)

Data Measures

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Analysis Strategy

1) Investigating the associations between work engagement and workaholism; CFA

2) Identifying the work engagement-workaholism classes; GMM

3) Investigating the relationship between work engagement-workaholism classes and job change; Cross-tabulation

Phase 1. Association between Work Engagement and Workaholism

χ2(141) = 393.68, p < .001, RMSEA = .06, CFI = .94, TLI = .93 at T1

Work engagement

Workaholism

Vigor

Dedication

Absorption

Working

exessively

Working

compulsively

T1 = -.07 ns. / T2 = -.08 ns.

χ2(141) = 495.75, p < .001, RMSEA = .07, CFI = .92, TLI = .90 at T2

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

No. of

classes

Log L No. of free

parameters

AIC BIC Entropy Latent Class

proportions (%)

1 -4618.78 33 9303.56 9440.11 - 100

2 -4479.59 44 9047.15 9229.24 .89 17/83

3 -4427.16 55 8964.31 9191.89 .91 14/80/6

4 -4373.85 66 8879.69 9152.79 .92 18/7/6/68

5 -4320.83 77 8795.65 9114.26 .91 8/5/4/13/70

6 -4280.75 88 8737.51 9101.63 .82 2/6/10/7/34/41

7 -4246.80 99 8691.61 9101.24 .83 3/7/42/5/8/33/2

Phase 2. Fit Indices for the Seven Estimated Latent Class Solutions

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Phase 2. The Identified Classes For Work Engagement and Workaholism

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Class Stayers

n

adj. res

Movers

n

adj. res

Total

1. High decreasing WE -

low stable WH

73

1.7

12

-1.7

85

2. Low increasing WE -

average decreasing WH

17

-4.1

16

4.1

33

3. Low decreasing WE -

low stable WH

26

1.4

3

-1.4

29

4. High stable WE-

average stable WH

251

0.1

65

-0.1

316

Total 367 96 463

Phase 3. Interdependency between the Work Engagement-Workaholism Classes and Job Change

Χ2(3) = 19.64, p < .001

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Conclusions

1) Work engagement and workaholism are largely independent psychological states (both between individuals as within individuals)

2) Work engagement and workaholism are both stable and dynamic in nature

3) Job conditions had an impact on the levels of both work engagement and workaholism as participants in the class ”low increasing WE-average decreasing WH” typically changed their job during the study period

4) The fact that both work engagement and workaholism are sensitive to job change suggest that both psychological conditions depend – at least partly – on the individuals work situation

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

For more information, see:

Mäkikangas, A., Schaufeli, W., Tolvanen, A., & Feldt, T. (2013). Engaged managers are not workaholics: Evidence from a longitudinal person-centered analysis. Journal of Work and Organizational Psychology, 29, 135-143.

Happiness and well-being at work: A special issue

DEPARTM ENT OF PSYCHOLOGY l UNIVERSITY OF JYVÄSKYLÄ

Thank you (= kiitos)!

The study was supported by: