Mackey, L., M.S., Northern Kentucky University; Sellers, J., M.S., Northern Kentucky University; Stegbauer, C., M.S., Northern Kentucky University; Syroney, G., M.S., Northern Kentucky University; Moberg, P. J., Ph.D., Northern Kentucky University
Purpose/Objectives
The present study explored three objectives, (1) to hypothesize potential types of work anxiety, (2) to develop a measure assessing anxious reactions to the work, working, and the workplace, and (3) to evaluate evidence of factor structure and generate relations with nine external validity constructs.
Background
In 2009, the American Psychological Association reported that 40% of Americans feel anxious during the workday and 72% experience anxiety that interferes with daily work. Work anxiety has been found to lead to dysfunctional actions such as unethical work behavior (Kouchaki & Desai, 2015), elevated risk-taking (Mannor et al., 2016), and reduced job performance.
Despite considerable research investigating stressor-stress-strain processes, comparatively less effort has been invested to explore the affective component of strain, or anxiety. Bar-Haim et al., (2007) defined work anxiety as a strain response to workplace stressors that are perceived as threatening or challenging and that exceed one’s capacity to respond. Although work anxiety is thought to reflect feelings of nervousness and tension about job related performance, prior research has proposed five dimensions of work anxiety. Thus, we hypothesized dimensions representing individual differences in anticipatory unease and tension stemming from one’s capabilities to perform, the physical workplace, coworkers, and supervisor.
Methods
We constructed an initial item pool of 120 items (e.g., “I worry over mistakes I make at work,” “I get tense with I think about going to work?”) and distilled to 40 to represent the five proposed dimensions. We administered the scale digitally using a frequency response format ranging from 1 (almost never) to 5 (almost always) to 216 employed or previously employed undergraduate students attending an urban, midwestern university who received course credit in exchange for participation.
To illuminate the meaning of scale dimensions and evaluate inferences of construct validity, we included nine previously validated measures of potentially related constructs, job satisfaction (Minnesota Job Satisfaction Scale; Weiss, Gibson, Davis, & Lofquist, 1977), absenteeism (Employee Absenteeism Scale; Paget, Lang, & Schulz, 1998), physical symptoms (Physical Symptoms Inventory; Spector & Jex, 1997), interpersonal conflict (Interpersonal Conflict Scale; Spector & Jex, 1998), workload (Quantitative Workload Inventory; Spector & Jex, 1997), organizational constraints (Organizational Constraints Scale; Spector & Jex, 1997), job anxiety (Job Anxiety Scale; Muschalla & Linden, 2009), and turnover intention (Turnover Intention Scale; Roodt, 2004).
Findings
To examine the dimensionality embedded in responses to the proposed work anxiety scale, we conducted an exploratory factor analysis using principal axis factoring with oblique rotation to allow for correlated factors. Three interrelated factors emerged. Factor one was comprised of eight items (α= .92), explained 43.1 percent of scale variance, and was labeled, “Global Worry.” Factor two was comprised of four items (α= .79), explained 7.5 percent of scale variance, and was labeled, |Affective Discomfort.” Factor three was comprised of eight items (α= .93), explained 5.0 percent of scale variance, and was labeled, “Performance Unease.” Factor intercorrelations ranged from r = .19 to .72.
To generate evidence supporting construct validity, we examined factor correlations with nine external scales. Five job satisfaction facets correlated in the expected direction with Global Worry (r = -.21 to -.49), Affective Discomfort (r = -.26 to -.44), and Performance Unease (r = -.16 to -.43). Job engagement related negatively to Global Worry (r = -.33), Affective Discomfort (r = -.36), and Performance Unease (r = -.29). As predicted, absenteeism, physical symptoms, workload, organizational constraints, engagement, turnover intentions, and job anxiety related positively with Global Worry (r = .12 to .49), Affective Discomfort (r = .12 to .41), and Performance Unease (r = .27 to .68; all r significant at p < .01, two-tailed).
Discussion
We constructed the work anxiety scale to examine five dimensions of work anxiety proposed in prior research, anticipatory unease and tension stemming from doubts about one’s capabilities to perform, the physical workplace, coworkers, and supervisor. Examination of the emerging factor structure revealed three dimensions and provided partial support for our hypothesis.
This evidence, while preliminary, suggests that work anxiety may be comprised of three dimensions, a global or generalized sense of worry, affective unease associated with work and work interactions, and tension about the adequacy of one’s capabilities and job performance. Findings from the present study necessarily are constrained by the undergraduate student respondents who likely had accumulated limited work experience. Future scale development efforts will be needed to examine the proposed dimensions of anxiety in samples of working employees.
Conclusions
Although continued efforts are required, the current study provides insights into the construct of work anxiety and supports continued development of an anxiety scale. Work anxiety appears to be comprised of (a) dispositional tendency to worry, (b) general unease about the workplace, and (c) doubts about performance capabilities. Elevated standing on these dimensions suggests potential tertiary coping interventions (e.g., stress reduction, efforts to enhance self-efficacy, identification of threatening work characteristics) to minimize the experience of anxiety and enhance employee mental health.