Best Practices and Industry Benchmarks

Sample and Procedure

We examined our hypotheses using data collected from two enterprises in mainland China. Specifically, one enterprise is a large joint venture brand automobile sales and service enterprise, whose branches cover all provincial capitals in China; the other is a large manufacturing enterprise. After obtaining the consent of the human resources departments of the two enterprises, we chose the sales department from the first enterprise and employees who engaged in technical research and development and professional knowledge positions from the second enterprise. For this study, we selected a sales group as our participants, mainly because they had experience with divided sales territories, which could predispose them to feelings of territoriality. Furthermore, employees in technical research and development and professional knowledge positions have more obvious knowledge territoriality. Similar to the existing research methods, we prepared two scales. Two sets of questionnaires were distributed in our survey: One for the subordinates and the other for their team supervisors. A total of 350 supervisor–subordinate matching surveys from 85 work teams participated in the survey in July 2018. Among them, 291 matches were returned and completed, representing a response rate of 83.142%. All of the participants completed this study on a voluntary basis. These 291 employees were nested within 80 teams, each guided by a supervisor. Of the participants, 51.2% (SD = 0.501) were women; the mean age was 32.96 years (SD = 5.908); approximately 41.9% (SD = 1.090) of the participants held a university or higher educational degree; and their annual average income from their organization was 71.1 thousand yuan (SD = 4.749).

Measures

To control the common variance, a method combining self-assessment and other evaluations was adopted for data collection. High performance expectations were evaluated by direct supervisors, whereas stress, territoriality, and task autonomy were evaluated by the subordinates. The English measurements involved were all translated into Chinese following the back-translation procedure because the survey was conducted in China. All items used a 5-point Likert-type scale (1 = strongly disagree; 5 = strongly agree). The reliability of all the scales was estimated by Cronbach's alpha.

Territoriality

To assess territoriality, we used the scale of Avey et al. (2009), which includes four items. The scale has been widely used in academic circles and has been proved to be applicable to Chinese situations. Sample items are "I feel that people I work with in my organization should not invade my workplace" and "I feel I need to protect my ideas from being used by others in my organization". The reliability coefficient (Cronbach's alpha) for this scale was 0.823.

High Performance Expectations

High performance expectations were measured using the three-item scale of Podsakoff et al. The scale has been proved to be applicable to Chinese situations. Sample items are "I insist on only the best performance" and "I show employees that I expect a lot from them". The reliability coefficient (Cronbach's alpha) of this scale was 0.765.

Stress

We asked the employees to describe their work stress using the 13-item scale of Zhang et al., which is widely used in Chinese contexts and is of good measurement validity. We specifically asked them about "the amount of work that must be accomplished in the time limit" and to respond to the statement, "I need to go through too much red tape to get my job done" (1 = "strongly disagree", 5 = "strongly agree"). The reliability coefficient (Cronbach's alpha) for this scale was 0.803.

Task Autonomy

We used the 3-item scale of Liu et al. to measure task autonomy. Sample items are "I decide how to complete the task by myself" and "This job provides me with quite a lot of opportunities to complete work tasks independently and freely". The reliability coefficient (Cronbach's alpha) for this scale was 0.811.

Control Variables

We selected the potential control variables based on theoretical and methodological considerations. Previous studies have indicated that differences between individuals may affect employees' territorial behavior. We controlled for educational level, age (in years), income, and gender in our analyses. Scholars have previously found that gender influences territoriality and that gender, educational level, and income affect stress. Thus, it is important to control such factors that may potentially affect stress and territoriality.

Analysis Strategy

Although our proposed model operates at the individual level, our data structure is nested because one supervisor rated several employees on their high performance expectations. In this study, to account for the nested effects, we adopted the multilevel structural equation modeling (MSEM) approach, and first calculated the ICC (1) for the independent variable. We tested the moderating effect of task autonomy on the relationship between stress and territoriality using the method of LMS proposed by Preacher, Zhang, and Zyphur and used the MSEM method to test the proposed hypotheses with Mplus 7.4. For the estimated path coefficients, we followed Liu, Zhang, Wang, and Lee's parameter bootstrapping technique to test the moderated indirect effects using Mplus 7.4.