Self-undermining behavior at work: Evidence of construct and predictive validity

Arnold B. Bakker, Yiqing Wang

Research output: Contribution to journalArticlepeer-review

35 Citations (Scopus)


In this article, we conceptualize self-undermining at work and validate a measurement instrument. In a first study with 5 samples (total N = 1256), we examine the factorial validity and reliability of the instrument in China, the United States, Chile, Romania, and the Netherlands. The results of exploratory and confirmatory factor analyses show that the scale has a 1-factor structure that fits well with the data of each country. In addition, the scale reliability is sufficient for each country (.70 ≤ α ≤.88). In Study 2, we used 2 additional samples from China (total N = 595) and tested the convergent, discriminant, and predictive validity of the Self-Undermining Scale. The results indicate that self-reports of self-undermining are positively related to self-handicapping and supervisor ratings of self-undermining. Furthermore, as hypothesized, self-undermining is negatively related to (supervisor ratings of) job performance and positively related to burnout. Self-underminers are slightly less likely to be proactive, craft their jobs, and experience work engagement. Our findings illustrate the potential usefulness of the self-undermining construct. In addition, we conclude that the newly developed instrument is valid and reliable and can be used to trace self-undermining behaviors in the workplace to prevent potential job strain and burnout.

Original languageEnglish
Pages (from-to)241-251
Number of pages11
JournalInternational Journal of Stress Management
Issue number3
Publication statusPublished - Aug 2020
Externally publishedYes


  • Burnout
  • JD-R theory
  • Job crafting
  • Job demands
  • Self-undermining

ASJC Scopus subject areas

  • Education
  • General Business,Management and Accounting
  • Applied Psychology
  • General Psychology


Dive into the research topics of 'Self-undermining behavior at work: Evidence of construct and predictive validity'. Together they form a unique fingerprint.

Cite this