“I know it’s not your job but…”: Extra-role tasks, communication, and leader-member exchange relationships

Abstract

Employees are sometimes assigned tasks that lie outside their official role. Employees may perceive such extra-role tasks (ERTs) as unreasonable or illegitimate. Grounded in role theory and leader-member exchange (LMX) theory, we conducted two studies that examined how ERTs are associated with supervisor-employee relationship quality and how the ERT assignment is communicated. While supervisor-subordinate relationship quality was unrelated to the likelihood and frequency with which supervisors assign ERTs to employees, employees with low-quality relationships evaluated ERTs as more unreasonable than those in high-quality relationships. Study 1 showed that ERT messages that included a request (vs. command), acknowledgement, explanation, and appreciation were associated with higher quality LMX relationships. Study 2 found that ERT message characteristics influenced the perceived unreasonableness of the task for employees in high-quality relationships, suggesting employees in such relationships are particularly vulnerable to “job creep” and role expansion.

Keywords:

• employer-employee
• leader-member exchange theory
• organizational communication
• supervisor-subordinate communication

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. In both studies, all scales were subjected to confirmatory factor analyses, which confirmed their unidimensional structures. Details of the CFAs are unreported to save space but are available upon request.

2. Prior to primary analyses, we conducted analyses to identify potential covariates. Results indicated that supervisory level was positively associated with LMX quality (r = 0.25, R² = 0.05, p < 0.01), and unreasonable ERT frequency (r = 0.23, R² = 0.05, p < 0.01). Job stress was significantly associated with whether or not the respondent’s supervisor had actually made the request they reported in response to the prompt, t(1, 128) =2.48, p< 0.01. Accordingly, supervisory level and whether or not the supervisor had made the request were included as a covariate in relevant statistical testing.

3. Levene’s test indicated equality of error variance in the LMX scores, F= 1.45, p< 0.17.

4. Prior to primary analyses, analyses was conducted to identify potential covariates. Correlations indicated LMX was positively associated with supervisory level (r= 0.19, p< 0.03), organization size was positively related to perceived unreasonableness of the personal ERT (r= 0.17, p< 0.04), and frequency of receiving ERT assignments was positively associated with the perceived unreasonableness of personal ERT assignments (r= 0.20, p< 0.02). Respondent race was also significantly related to the perceived reasonableness of the work related task (F= 2.69, p< 0.03). Accordingly, supervisory level, organization size, frequency of receiving ERTs, and race were included as covariates in relevant statistical tests.