Communicating more or less? Employees’ playful work design and their voice and feedback-seeking behaviors

The effect of PWD on job-based psychological ownership

Relying on the extended self-theory (Belk, 1988, 2000; Dittmar, 1992; Pierce et al., 2001, 2003), we posit that PWD is positively related to employees’ job-based PO (Scharp et al., 2022). The extended self-theory emphasizes the role of personal possessions in individuals’ self-definition (Belk, 1988, 2000; Pierce et al., 2003). When individuals develop possessive feelings toward an object, they may extend their sense of self to that object, incorporating it into their identity (Brown et al., 2005, 2014). This self-extension process occurs through three routes—it emerges when individuals “(a) invest themselves in the objects to create them, (b) control them, and/or (c) know them well” (Belk, 2000; Pierce et al., 2001; Wang et al., 2019: 230). Once individuals consider their possessions as parts of themselves, they tend to maintain, preserve, and consolidate these possessive feelings. Psychological ownership stems from the extended self-literature, capturing the state in which individuals feel they own something, making the object an extension of their self (Brown et al., 2005; Pierce et al., 2001, 2003). Job-based psychological ownership describes a state where individuals feel ownership over their job and consider it part of their extended self (Wang et al., 2019).

In light of the extended self-theory, PWD may foster job-based PO through three routes described above. First, psychological ownership arises from investing the self into the objects (Pierce et al., 2001; Wang et al., 2019). In PWD, employees invest themselves in their work, make it self-absorbing and self-revealing, thereby linking their sense of self to their work (Brown et al., 2014). Specifically, to make their work playful, employees may first identify which tasks can be approached playfully (e.g., certain tasks like preparing a financial report may require careful attention and are less suitable for playful integration). They then deliberate on how to infuse fun and challenge into their work and assess the feasibility of engaging in PWD without incurring risks (Parker et al., 2010). Such thoughtful consideration reflects personal discretion in determining workstyles, fosters the fusion of self and work, and encourages employees to see their job as their own (Brown et al., 2005; Pierce et al., 2009). Moreover, imbuing their work with humor and excitement, employees introduce surprises and complexities to their job, personalize their tasks, and “get into” their work (Brown et al., 2014; Pierce et al., 2009). These emotional resources invested in the job further strengthen employees’ sense of job-based PO.

Second, extended self-theory suggests that exploration and the ability to exercise control foster the sense of psychological ownership (Brown et al., 2005; Chen et al., 2023; Pierce et al., 2009). Because work typically lacks inherent playfulness, designing work to be playful is volitional and self-oriented, involving a process of exploration, change, and demonstration of one’s competence and autonomy (Bakker et al., 2020; Scharp et al., 2022). By incorporating joy and personal goals—aligned with their inner nature—into their job, employees gain greater control over the direction and focus of their tasks, fostering a sense of self-determination (Deci et al., 2001; Scharp et al., 2022). This leads employees to see themselves as responsible for their work outcomes (Brown et al., 2014); that is, whether performance is successful or unsuccessful, they attribute it to themselves. Therefore, they may develop a strong sense of ownership over their job.

Third, extended self-theory also posits that intimately knowing the target is a pathway to psychological ownership (Pierce et al., 2001; Wang et al., 2019). In PWD, employees infuse fun and challenge into their work (Scharp et al., 2023). These endeavors, driven by an intrinsic desire for play and aligned with one’s inner nature (Dishon-Berkovits et al., 2024; Huizinga, 1949), generate joy, create meaningfulness, and foster deep psychological attachment to the job (Brown et al., 2005; Scharp et al., 2022). In this way, job metaphorically becomes a place in which employees dwell (Brown et al., 2014; Pierce et al., 2009). A sense of connection between self and job emerges—employees and their work become intertwined, such that employees can communicate their self through job, and their job, in turn, constructs their self interactively (Brown et al., 2005, 2014). As self and job become interwoven, employees develop a deep understanding of their job, inhabit it fully, and see it as a reflection or extension of themselves.

This view aligns with self-perception theory (Bem, 1967, 1972) ¹ , which suggests that individuals sometimes infer their internal states and come to know themselves by observing their own behaviors much like an external observer would. When employees design their work playfully, they may interpret these behaviors as a manifestation of a strong connection between themselves and their job, thereby cultivating job-based PO. Therefore, we predict that:

The mediating effect of job-based psychological ownership

Extended self-theory suggests that individuals’ possessive relationships with objects inherently involve multiple parties, rather than being simply dyadic (Belk, 1988, 2000). Such relationships constitute not only a bond between self and object but also a bond between others and object, as external claims or interference can undermine ownership (Chen et al., 2023; Wang et al., 2019). According to extended self-theory, psychological ownership can activate a protective motive, prompting employees to safeguard what they own in order to protect their sense of self (Brown et al., 2014; Pierce et al., 2001, 2003). They can do this in two primary ways: maintaining or enhancing their ownership, and preventing others from encroaching upon it (Chen et al., 2023; Wang et al., 2019). Drawing upon extended self-theory and research on psychological ownership, we propose that job-based PO influences employees’ voice and feedback-seeking behaviors, mediating the relationship between PWD and these behaviors.

Specifically, employees with high job-based PO are motivated to control what they possess at work and restrict others’ access (Chen et al., 2023). In doing so, they preserve a sense of security and maintain the continuity of self (Pierce et al., 2003; Wang et al., 2019). Employees’ unique ideas, opinions, and job-related information are valuable and personalized resources that help distinguish them from others; as such, these resources become ownership targets and a part of employees’ extended self (Pierce et al., 2009; Wang et al., 2019). High job-based PO prompts employees to defend their territory (Brown et al., 2005; Chen et al., 2023) and, consequently, to withhold these ideas and opinions over which they feel ownership (Pierce et al., 2009). Because voice involves the discretionary expression of constructive work-related ideas, suggestions, opinions, and concerns (Detert and Burris, 2007; Liu et al., 2017), it requires employees to share these personalized thoughts. Doing so may make employees high in job-based PO feel that their possessive relationship with these thoughts is being encroached upon (Wang et al., 2019). Moreover, voice is an other-oriented proactive behavior—it aims to change how the team or organization functions (Griffin et al., 2007), which is contradictory to the self-oriented nature of job-based ownership—viewing one’s work as a personal possession. When voiced suggestions are implemented, others can access and use the ideas employees created and valued, increasing the possibility that their originality may be dampened or even overlooked. This may further reinforce the sense that their territory is being infringed upon and that their “personal possessions” are no longer entirely their own. Therefore, employees with higher job-based PO are less inclined to voice their ideas. By withholding them, they can prevent others from establishing a bond with their owned possession (Brown et al., 2005) and protect their sense of ownership over their work-related insights.

Accordingly, we predict that job-based PO is negatively related to employee voice. Building on the argument that PWD fosters job-based PO and job-based PO reduces voice, we further suggest that job-based PO mediates the negative indirect relationship between PWD and voice. For example, in designing fun and competition, employees invest heavily in creatively redesigning their work, viewing their work experience as their own possession. If they discover a way to improve work procedures and boost productivity through PWD, they may be inclined to keep this shortcut to themselves. This is because proposing the idea to managers would make it public, reducing the extent to which the idea remains a personal and valued possession. Therefore, to protect the sense of ownership and maintain distinctiveness, they would be less likely to voice their suggestions after engaging in PWD and developing job-based PO. Hence, we propose the following hypotheses:

Also drawing on extended self-theory, employees with high job-based PO are more likely to seek feedback from others. As we have explained, when employees see their job as their own and internalize it as part of their self-definition, they strive to protect and enhance their possessive relationship with their job, thereby maintaining the continuity of their self (Brown et al., 2014; Chen et al., 2023; Pierce et al., 2001, 2003). To achieve this, employees not only need to keep their job on the right track but also endeavor to excel in their performance (Pierce et al., 2009; Wang et al., 2019). Otherwise, they may face the risk of being replaced or the threats that others could interfere with their possessive relationship with their job. Because work represents the self, employees are also motivated to perform well in order to express a positive image and gain social recognition (Pierce et al., 2009; Wang et al., 2019).

Seeking feedback from others (i.e., supervisors and coworkers), though may involve the risk of losing face (Anseel et al., 2015), brings important resource that helps employees function effectively and perform well (Ashford, 1986; Lam et al., 2007), serving to strengthen the bond between self and job. In particular, feedback provides employees with vital information and knowledge to complete their work, such as evaluations of their work progress, advice on finishing specific tasks, and expectations regarding their work role (De Stobbeleir et al., 2011). By helping employees regulate their work pace and directing them toward the right path to achieving their valued end state (Ashford, 1986; Lam et al., 2007), feedback-seeking supports the maintenance of their possessive relationship with their job. Although employees may fear receiving negative feedback when they feel ownership over their work, this avoidant emotion does not necessarily prevent them from seeking feedback, because they cognitively understand that feedback is ultimately beneficial for improving their functioning and reinforcing their bond with their job.

Besides protecting one’s feelings of possession, feedback-seeking can even help expand one’s feelings of possession. Unlike other-oriented voice behavior, which can be regarded as an output of one’s views and a potential loss of personal possessions, feedback-seeking represents an input of others’ perspectives, and thus contributes to one’s own work role (Griffin et al., 2007), and is thus conducive to expanding ones’ personal possessions (Chen et al., 2023). By incorporating others’ feedback, employees enrich their work-related knowledge and feel that what they know, control, and own is growing. Therefore, they can recognize the benefits of feedback and engage more in feedback-seeking. Self-perception theory also helps explain the positive relationship between job-based PO and feedback-seeking behaviors. According to self-perception theory, individuals come to know themselves by observing their own behaviors, and this self-understanding, in turn, guides behaviors that are consistent with their self-perceptions (Bem, 1967, 1972). After developing a sense of ownership of their job through PWD, employees are driven to seek feedback to improve their work and reinforce their possessive relationship with their job.

We therefore propose that job-based PO is positively related to employees’ feedback-seeking behaviors. Building on our previous reasoning that PWD increases job-based PO, we suggest that PWD has a positive indirect relationship with feedback-seeking behavior through heightened job-based PO. For example, a playful work designer reconstructs their job in a fun and competitive manner, aligning closely with their need for hedonism. By generating a strong sense of job-based PO, they hope to perform well and take pride in their job and themselves. To ensure they are performing well and to seek suggestions that can improve their work, they are motivated to communicate with their supervisor and coworkers and solicit feedback. Therefore, we hypothesize that:

The moderating effect of performance goal orientations

After establishing the rationale behind the relationships between PWD, job-based PO, and voice/feedback-seeking, we aim to investigate when these relationships become more or less significant. Extended self-theory suggests that the effect of psychological ownership depends on individual differences (Pierce et al., 2003), such as individuals’ trait motivations (Chen et al., 2023; Wang et al., 2019). Since voice and feedback-seeking behaviors not only involve possession-protecting and -gaining but are also achievement-related interpersonal behaviors—voice can be a way for employees to demonstrate their ability and gain prestige and status among peers (Morrison, 2011), while feedback-seeking involves performance evaluation and the risks of being perceived as incompetent or losing face (Anseel et al., 2015; Ashford and Cummings, 1983)—we focus on employees’ goal orientations as important moderators in this process.

Goal orientations, which are motivational dispositions reflecting individual differences in goal preferences in achievement situations (VandeWalle and Cummings, 1997; VandeWalle et al., 1999), are particularly relevant. Specifically, we focus on two of the three types of goal orientations: performance-prove goal orientation (PPGO) and performance-avoid goal orientation (PAGO). Unlike learning goal orientation (LGO), which represents a mastery goal orientation, both PPGO and PAGO fall under achievement (or performance) goal orientations. They capture an other-referenced focus, that is, the extent to which employees care about self-presentation in front of others and strive for superiority over others (Marijn Poortvliet et al., 2007). As they reflect an attempt to avoid failure, gain others’ respect, and ultimately achieve esteem and affiliation (Watson et al., 2013), they are highly relevant in evaluative situations, particularly in the presence of others (Zeidner, 1998). This aligns with our research focus on interpersonal behaviors as outcome variables (Marijn Poortvliet et al., 2007). In contrast, LGO captures the extent to which employees emphasize self-improvement—that is, they attach importance to developing competence, acquiring skills, and doing their best (Dweck, 1986; Marijn Poortvliet et al., 2007). Being self-referenced in nature, LGO is more relevant in situations where employees compare themselves with their past performance rather than with others. Because LGO is less directly concerned with interpersonal behaviors (e.g., speaking up, exchanging information with others), we do not include it in this study.

We first propose that employees’ PPGO mitigates the negative effect of job-based PO on voice. PPGO describes individuals’ desire to demonstrate their competence and ability and to outperform others. Employees with high PPGO view others as evaluative referents and tend to compare themselves with others (Downes et al., 2021). As previously discussed, employees with high job-based PO are inclined to withhold their ideas and opinions to maintain their sense of distinctiveness, making them less likely to voice their thoughts. However, voice is also a valuable way for employees to display their capabilities, enhance their status, and improve their impressions on managers (Morrison, 2011). In other words, voice serves as a means to prove one’s performance, aligning with the inclinations of employees with high PPGO. Therefore, the desire to compare with others and display competitiveness associated with PPGO may, to some extent, counteract the effect of withholding ideas and perspectives inherent in job-based PO. Additionally, because employees high in PPGO are more approach-oriented (Elliot and Church, 1997), they are less likely to view speaking up (i.e., voice) as a threat to their job-based PO. For example, they may perceive more benefits in voicing and believe that, even when their suggestions are shared and implemented by others, their ownership will still be recognized. Thus, higher PPGO may weaken negative relationship between job-based PO and voice.

In contrast, employees with low PPGO have little motivation to prove their competence in front of others, making them more likely to keep their ideas and suggestions to themselves to maintain their sense of ownership, thus voicing less. Combining the mediating effect of job-based PO and the moderating effect of PPGO, we further propose a moderated mediating effect, such that PPGO lessens the negative indirect effect of PWD on voice via job-based PO. Specifically, as employees playfully design their work and develop a sense of ownership over their job, those with a higher orientation to prove their performance are less likely to withhold their voice. Therefore, we hypothesize:

Furthermore, we propose that employees’ PAGO mitigates the positive effect of job-based PO on feedback-seeking. PAGO reflects individuals’ desire to avoid demonstrating incompetence and receiving negative performance evaluations (Elliot and Church, 1997; VandeWalle and Cummings, 1997; VandeWalle et al., 1999). While PAGO is an other-focused disposition and is sometimes moderately and positively correlated with PPGO (Fisher et al., 2013; Payne et al., 2007), it is distinct from PPGO—employees with high PAGO seek to avoid appearing incompetent to others. As previously discussed, job-based PO encourages employees to seek feedback from others. However, feedback-seeking can involve negative performance evaluations and the risk of losing face (Anseel et al., 2015; Ashford and Cummings, 1983), precisely the situations employees with high PAGO tend to avoid. Therefore, the motive to avoid negative appraisals of their abilities associated with PAGO may, to some degree, counteract the intention to solicit feedback from others prompted by job-based PO. For example, although employees with high job-based PO may seek to obtain job resources and enrich their possessions, they are less likely to ask others for feedback if they have high PAGO. Instead, they may opt for alternative methods, such as participating in training or engaging in self-reflection, to enhance their work and ownership. Conversely, employees with low PAGO are less concerned about receiving negative performance appraisals. For these employees, job-based PO is more likely to drive feedback-seeking, which is a useful way for them to improve their work and expand their ownership.

We also propose a moderated mediating effect by combining the mediating effect of job-based PO and the moderating effect of PAGO. Specifically, PAGO reduces the positive indirect effect of PWD on feedback-seeking through job-based PO. In other words, as employees develop a sense of ownership over their job after playfully designing their work, those with a higher orientation to avoid negative performance evaluations are less likely to engage in feedback-seeking. Thus, we hypothesize:

Participants and procedures

We recruited 110 employees from various industries and occupations through Credamo, a Chinese online data collection platform widely used by scholars (e.g., Lei, 2024; Qin et al., 2024). Previous research has shown that data collected via Credamo is as reliable as data collected through traditional methods (e.g., Pan et al., 2024; Taneva and Peng, 2024). We specifically recruited participants with approval rates over 80% in China and provided a payment of 3.0 yuan (0.4 USD) per participant. We included an attention check item (e.g., “Please select ‘slightly disagree’ for this item”), and one participant failed this check and was removed from the sample, resulting in a final sample of 109 participants. The average age of participants was 40.9 years (SD = 9.85), and they had worked an average of 12.5 years in their current jobs (SD = 10.13). Among them, 67 (62.5%) were male. Regarding educational levels, all participants had at least a high school education, 68 (62.4%) had at least a bachelor’s degree, and 10 (9.2%) had at least a master’s degree.

We used a one-factorial (PWD: low and high) between-subjects design to examine Hypothesis 1. Participants were randomly assigned to one of two conditions: high PWD or low PWD. We created a manipulation for each condition, and participants were asked to read the manipulation, recall a related experience, and write information relevant to the manipulation. After completing their written response, participants finished a manipulation check and a measure of job-based PO. Finally, they were compensated, thanked, and debriefed.

Manipulations and measures

Results

A one-way ANOVA on participants’ reported PWD indicated that our manipulation was successful: participants in the high PWD condition reported significantly higher level of PWD (M = 5.85) than participants in the low PWD condition (M = 5.20), F(1, 107) = 14.14, p < 0.001. We then conducted another one-way ANOVA on participants’ job-based PO to examine whether the participants in the high PWD condition reported significantly higher job-based PO than the participants in the low PWD condition (Hypothesis 1). The results showed that the participants in the high PWD condition had a higher job-based PO (M = 5.37) than participants in the low PWD condition (M = 4.77), F(1, 107) = 5.74, p < 0.05. These results supported our Hypothesis 1.

Study 1 provided initial support for our Hypothesis that an employee’s PWD can increase job-based PO. We then designed two survey studies to examine other hypotheses and attempted to provide more external validity.

Sample and procedure

We conducted a two-wave, multi-source survey study in one small and medium-sized enterprise (SME) and one large enterprise in South Korea. The SME was a human resource management consulting firm focused on staffing, and the large enterprise was an information and telecommunication firm. We targeted a total of 354 employees and their 32 direct supervisors. At Time 1, we measured employees’ PWD and job-based PO. A total of 271 employees completed the first survey. One month later (Time 2), we measured employee voice and feedback-seeking from supervisors. Our final sample consisted of 221 employee–supervisor dyads (with 28 supervisors), resulting in a response rate of 62.4% for employees. Among these employees, 186 (84.2%) were male, and most were between 30 and 50 years old (144 employees, 65.2%). In terms of educational level and organizational tenure, the majority had at least a bachelor’s degree (168, 76.0%) and more than 6 years of tenure (131, 59.3%). Table 1 presents the means, standard deviations, and correlations of the variables.

OPEN IN VIEWER

Table 1.

Means, standard deviations, and correlations among variables (Study 2).

Variable	Mean	SE	1	2	3	4	5	6	7
1. Age	4.76	1.74
2. Gender	0.84	0.37	0.47**
3. Education	3.83	0.61	−0.30**	−0.23**
4. Organizational tenure	5.21	2.34	0.76**	0.36**	−0.31**
5. PWD (Time 1, self-rated)	4.88	0.85	0.29**	0.13 +	−0.06	0.15*
6. Job-based PO (Time 1, self-rated)	6.07	0.70	0.11	0.09	−0.07	0.06	0.54**
7. Voice (Time 2, supervisor-rated)	3.40	1.69	−0.06	−0.02	0.16*	−0.00	0.01	0.07
8. Feedback-seeking (Time 2, supervisor-rated)	2.38	1.18	−0.15*	−0.12 +	0.17*	−0.12 +	−0.02	0.04	0.73**

PWD: playful work design; PO: psychological ownership.

N = 221. Age coded: 1 = 20 years or more but less than 25 years, to 7 = 50 years or more; gender coded: 1 = male, 0 = female; education coded: 1 = below high school education, to 6 = doctoral degree; organizational tenure coded: 1 = less than 1 year, to 7 = 6 years or more.

+

p < 0.10; *p < 0.05; **p < 0.01.

Measures

To ensure our measures translated well into Korean, we followed the translation-back-translation procedure (Brislin, 1980). Unless otherwise indicated, all measures were rated on a 7-point Likert scale ranging from “1 = strongly disagree” to “7 = strongly agree.”

Analytical approach

Because our data may have some nestedness issues, we conducted a one-way analysis of variance on all study variables. The results for PWD (F = 1.66, p < 0.05), voice (F = 14.46, p < 0.001), and feedback-seeking (F = 17.93, p < 0.001) were statistically significant. The intraclass correlation coefficient (ICC (1)) for PWD was 0.08; for voice was 0.63; and for feedback-seeking was .68. Based on these results, we controlled for between-supervisor variance of PWD, voice, and feedback-seeking by allowing them to have a random intercept at Level-2 (Hofmann and Gavin, 1998). We implemented Mplus 8.3 (Muthén and Muthén, 2017) to examine Hypotheses 1–5, with a path model. The exogenous variable, PWD, was grand-mean-centered (Hofmann et al., 2000). And, we relied on 20,000 Monte Carlo resampling approach to examine our hypothesized mediation effects (Hypotheses 3 and 5).

Results

We conducted confirmatory factor analyses (CFAs) to evaluate the distinctiveness of our studied variables. The hypothesized four-factor model showed a good fit with the data (χ²(269) = 558.23, p < 0.001; CFI = 0.92; TLI = 0.91; RMSEA = 0.070; SRMR = 0.062), and all the items were significantly loaded on their intended factors (p < 0.001). The four-factor model was significantly superior to other possible three-factor models (combining PWD and job-based PO: χ²(272) = 1015.02, p < 0.001; CFI = 0.79; TLI = 0.77; RMSEA = 0.111; SRMR = 0.085; combining voice and feedback-seeking: χ²(272) = 621.01, p < 0.001; CFI = 0.90; TLI = 0.89; RMSEA = 0.076; SRMR = 0.064), two-factor model (combining PWD and job-based PO, and voice and feedback-seeking respectively: χ²(274) = 1077.74, p < 0.001; CFI = 0.77; TLI = 0.75; RMSEA = 0.115; SRMR = 0.086), and one-factor model (combining all variables: χ²(275) = 2237.55, p < 0.001; CFI = 0.44; TLI = 0.39; RMSEA = 0.180; SRMR = 0.177).

Table 2 shows the unstandardized path coefficients and standard errors in our analysis. Hypothesis 1 suggests that a positive relationship between PWD and job-based PO. The results showed that PWD had a positive and significant effect on job-based PO (B = 0.47, p < 0.001). Therefore, Hypothesis 1 was supported. Hypothesis 2 articulates that a negative relationship between job-based PO and voice. The results were inconsistent with our prediction: job-based PO had a positive and significant relationship with voice (B = 0.34, p < 0.01). Therefore, Hypothesis 2 was not supported. Hypothesis 3 predicts that job-based PO mediates the relationship between PWD and voice. The results also did not lend support for Hypothesis 3: the indirect effect was positive (indirect effect = 0.16, 95% Monte Carlo confidence interval CI (0.0123, 0.4196)). Hypothesis 4 describes the positive relationship between job-based PO and feedback-seeking. The results demonstrated that the effect of job-based PO on feedback-seeking was positively and marginally significantly related to feedback-seeking (B = 0.17, p = 0.06). Hypothesis 4 was therefore supported. Hypothesis 5 states that job-based PO mediated the positive and indirect relationship between PWD and feedback-seeking. The results were aligned with our prediction: the indirect effect was positive and significant (indirect effect = 0.08, 95% Monte Carlo CI (0.0047, 0.1892)), supporting Hypothesis 5 ³ .

OPEN IN VIEWER

Table 2.

Unstandardized path coefficients and standard errors at individual level (Study 2).

Variables	Job-based PO		Voice		Feedback-seeking
	B	SE	B	SE	B	SE
1. Age	−0.05	0.04	0.08	0.07	0.07	0.06
2. Gender	0.10	0.14	0.45*	0.20	0.07	0.10
3. Education	−0.06	0.07	0.08	0.11	0.04	0.04
4. Organizational tenure	0.01	0.03	0.02	0.05	−0.06	0.04
5. PWD	0.47**	0.05	−0.07	0.11	−0.03	0.06
6. Job-based PO			0.34**	0.12	0.17 +	0.09

N = 221.

PWD: playful work design; PO: psychological ownership.

+

p < 0.10; *p < 0.05; **p < 0.01.

Supplementary analysis

As we do not focus on the specific effect of each dimension of PWD (i.e., designing fun and designing competition), and we contend that these two dimensions do not differ theoretically in our proposed relationships, we did not separate them when testing our hypotheses. Nevertheless, we also conducted supplemental analyses by separating the two dimensions and testing the hypothesized relationships. The results showed that the relationship between designing fun and job-based PO was positive and marginally significant (B = 0.14, p < 0.10), whereas the relationship between designing competition and job-based PO was positive and significant (B = 0.32, p < 0.001). These findings were generally consistent with Hypothesis 1. We then examined the indirect effects of designing fun and designing competition on voice via job-based PO (for designing fun: indirect effect = 0.05, 95% Monte Carlo CI (0.0054, 0.0982); for designing competition: indirect effect = 0.11, 95% Monte Carlo CI (−0.0219, 0.3625)). The pattern of results was consistent with those obtained when combining the two dimensions to test Hypothesis 3. Similarly, we tested the indirect effects of designing fun and designing competition on feedback-seeking via job-based PO (for designing fun: indirect effect = 0.02, 95% Monte Carlo CI (0.0006, 0.0623); for designing competition: indirect effect = 0.05, 95% Monte Carlo CI (−0.0152, 0.1553)). Again, the pattern of results was generally consistent with those obtained when combining the two dimensions to test Hypothesis 5.

Study 2 mainly focused on examining Hypotheses 1–5. The results showed that two of the hypotheses (Hypotheses 2 and 3) were not supported. To further examine these unsupported hypotheses and investigate the moderating effects of PPGO and PAGO, we designed Study 3.

Sample and procedure

This study was conducted in a state-owned engineering company located in Southern China. The company’s main business involves building roads, houses, and infrastructure for the city. Since blue-collar workers may have limited autonomy to integrate play and work, we invited all white-collar employees (202 employees) and their direct supervisors to participate in our survey. Three waves of surveys were created through a website designed specifically for the creation and distribution of academic surveys in China. Each time we sent our survey links to participants, they received a message on their mobile phones. They could then click on the link and complete our questionnaires. The first-wave questionnaire included measures of PWD and control variables, the second-wave questionnaire (administered 1 month after the first wave) included measures of job-based PO, PPGO, and PAGO, and the third-wave survey (administered 1 month after the second wave) had two questionnaires: one for supervisors (measuring employee voice) and one for employees (measuring feedback-seeking). All participants received 6 CNY (about 1 USD) as a token of appreciation.

The final sample comprised 194 employee–supervisor dyads, resulting in a 96.0% response rate. The employees were from various departments, such as project management, quality inspection, safety supervision, finance, and administration. Among them, 136 (70.1%) were male, and 58 (29.9%) were female. On average, they were 36.4 years old and had 7.6 years of organizational tenure. In terms of educational level, almost all had at least a high school education (189 employees, 97.4%), and most had at least a bachelor’s degree (142 employees, 73.2%). Table 3 presents the means, standard deviations, and correlations of the variables.

OPEN IN VIEWER

Table 3.

Means, standard deviations, and correlations among variables (Study 3).

Variable	Mean	SE	1	2	3	4	5	6	7	8	9
1. Age	36.39	7.78
2. Gender	0.70	0.46	0.05
3. Education	3.82	0.88	−0.25**	−0.13 +
4. Organizational tenure	7.62	5.59	0.65**	0.00	−0.25**
5. PWD (Time 1, self-rated)	4.98	1.00	−0.04	0.15*	0.21**	−0.12
6. Job-based PO (Time 2, self-rated)	4.39	1.03	−0.08	−0.05	0.03	−0.06	0.32**
7. PPGO (Time 2, self-rated)	4.44	0.90	0.04	0.11	0.21**	−0.02	0.41**	0.30**
8. PAGO (Time 2, self-rated)	3.98	1.03	0.04	−0.05	0.02	0.03	−0.08	0.05	0.41**
9. Voice (Time 3, supervisor-rated)	4.55	0.91	−0.12	0.08	0.22**	−0.13 +	0.07	−0.08	−0.09	−0.19**
10. Feedback-seeking (Time 3, self-rated)	4.11	0.81	−0.01	0.02	−0.01	−0.06	0.26**	0.38**	0.36**	0.43**	−0.12

PWD: playful work design; PO: psychological ownership; PPGO: performance-prove goal orientation; PAGO: performance-avoid goal orientation.

N = 194. Gender coded: 1 = male, 0 = female; education coded: 1 = below high school education, to 7 = doctoral degree; organizational tenure was coded in years.

+

p < 0.10; *p < 0.05; **p < 0.01.

Measures

To ensure our participants clearly understand our measures, we followed the translation-back-translation procedure (Brislin, 1980). Unless otherwise indicated, all measures were anchored on a 7-point Likert scale (from “1 = strong disagree” to “7 = strongly agree”).

Analytical approach

To test our hypothesized relationships, we used path analysis. Specifically, we conducted a one-way analysis of variance on all study variables because our employee data were nested within supervisors, and our observations were not independent. The results for PWD (F = 1.61, p < 0.05) and voice (F = 4.56, p < 0.001) were statistically significant. The intraclass correlation coefficient (ICC (1)) for PWD was 0.10; and for voice was 0.38. These results suggested the necessity of controlling for between-supervisor variance (Hofmann and Gavin, 1998). Therefore, we integrated the “Type = Complex” syntax in Mplus 8.3 (Muthén and Muthén, 2017) with the path analysis to address the non-independence issue and simultaneously examine all our hypothesized relationships in one model. We grand-mean-centered the exogenous variables (Hofmann et al., 2000) and allowed PAGO to predict voice and PPGO to predict feedback-seeking. A 20,000 Monte Carlo resampling approach was used to examine our hypothesized mediation effects and moderated mediation effect.

Results

CFAs were conducted to evaluate the distinctiveness of our studied variables. We created parcels for some variables to obtain an optimal sample size ratio for the number of parameters that we needed to estimate (Landis et al., 2000). Specifically, we parceled the items of PWD according to its dimensions (two parcels), and parceled the items of job-based PO, PPGO, and PAGO following a random assignment approach (three parcels for job-based PO, and two parcels for PPGO and PAGO, respectively). Two outcome variables (voice and feedback-seeking) were not parceled. The hypothesized six-factor model showed a good fit with the data (χ²(104) = 201.60, p < 0.001; CFI = 0.97; TLI = 0.96; RMSEA = 0.070; SRMR = 0.046), and all the items were significantly loaded on their intended factors (p < 0.001). The six-factor model was significantly superior to other possible models (see Table 4).

OPEN IN VIEWER

Table 4.

Results from confirmatory factor analyses comparison of alternative measurement models (Study 3).

Model	χ2	χ2/df	CFI	TLI	RMSEA	SRMR
Six-factor model	201.60	1.94	0.97	0.96	0.070	0.046
Five-factor model (combined PPGO and PAGO)	472.08	4.33	0.88	0.85	0.131	0.092
Five-factor model (combined voice and feedback-seeking)	629.56	5.78	0.83	0.79	0.157	0.177
Four-factor model (combined job-based PO, PPGO, and PAGO)	1224.83	10.84	0.64	0.57	0.225	0.141
Three-factor model (combined job-based PO, PPGO, and PAGO, and combined voice and feedback-seeking)	1608.61	13.87	0.52	0.43	0.258	0.218
One-factor model (combined all variables)	2387.86	20.07	0.27	0.16	0.313	0.267

CFI: comparative fit index; TLI: Tucker-Lewis index; RMSEA: root mean square error of approximation; SRMR: standardized root mean squared residual; PO: psychological ownership; PPGO: performance-prove goal orientation; PAGO: performance-avoid goal orientation.

N = 194.

Figure 2 and Table 5 show the unstandardized path coefficients and standard errors in our analysis. Hypothesis 1 predicts that PWD is positively related to job-based PO. The results showed that the path from PWD to job-based PO was positive and significant (B = 0.36, p < 0.001). Therefore, Hypothesis 1 was supported. Hypothesis 2 suggests that job-based PO is negatively related to voice. The results showed that job-based PO negatively and significantly predicted voice (B = −0.08, p < 0.05), supporting Hypothesis 2. Hypothesis 3 stated that job-based PO mediates the relationship between PWD and voice. The results showed the indirect effect was negative but not significant (indirect effect = −0.03, 95% Monte Carlo CI (−0.0867, 0.0249)). Therefore, Hypothesis 3 was not supported. Hypothesis 4 suggests the positive effect of job-based PO on feedback-seeking. The results supported Hypothesis 4: the path from job-based PO was positively and significantly related to feedback-seeking (B = 0.20, p < 0.05). Hypothesis 5 describes that job-based PO mediates the relationship between PWD and feedback-seeking. The results showed that the indirect effect was positive and significant (indirect effect = 0.07, 95% Monte Carlo CI (0.0488, 0.1316)). Thus, Hypothesis 5 was supported.

OPEN IN VIEWER

Figure 2.

Results of path analysis (Study 3).

OPEN IN VIEWER

Table 5.

Unstandardized path coefficients and standard errors (Study 3).

Variables	Job-based PO		Voice		Feedback-seeking
	B	SE	B	SE	B	SE
Age	−0.01	0.01	−0.00	0.01	0.00	0.01
Gender	−0.24 +	0.12	0.19	0.18	0.02	0.07
Education	−0.10	0.10	0.25**	0.07	−0.10 +	0.05
Organizational Tenure	0.01	0.02	−0.01	0.02	−0.01	0.01
PWD	0.36**	0.09	0.06	0.08	0.16**	0.06
Job-based PO			−0.08*	0.04	0.20*	0.08
PPGO			−0.61**	0.21	0.04	0.10
PAGO			−0.12 +	0.07	0.91**	0.19
Job-Based PO×PPGO			0.11**	0.04
Job-Based PO×PAGO					−0.11**	0.03

PWD: playful work design; PO: psychological ownership; PPGO: performance-prove goal orientation; PAGO: performance-avoid goal orientation.

N = 194.

+

p < 0.10; *p < 0.05; **p < 0.01.

Hypothesis 6 predicts that the negative relationship between job-based PO and voice was mitigated by an employee’s PPGO. The results showed that the interaction term between job-based PO and PPGO was positive and significant (B = 0.11, p < 0.01), which meant that the negative effect of job-based PO on voice was buffered by PPGO. Hypothesis 6 was supported. Figure 3 illustrates that job-based PO was less negatively related to voice when PPGO was high (at +1 SD: simple slope = 0.01, n.s.) than when it was low (at −1 SD: simple slope = −0.18, p < 0.01). Hypothesis 7 articulates that the negative indirect effect of PWD on voice via job-based PO was mitigated by an employee’s PPGO. The results showed that the indirect effect was less negative when PPGO was high (indirect effect = 0.00, n.s.) than when it was low (indirect effect = −0.06, p < 0.05). And, the difference in the indirect effects was significant (diff = 0.07, p < 0.05, 95% Monte Carlo CI (0.0179, 0.1174)). Thus, Hypothesis 7 was supported. Hypothesis 8 suggested that the positive relationship between job-based PO and feedback-seeking was lessened by an employee’s PAGO. The results showed that the interaction term between job-based PO and PAGO was negative and significant (B = −0.11, p < 0.001). This meant that the positive effect of job-based PO on feedback-seeking was mitigated by PAGO, supporting Hypothesis 8. Figure 4 presents that job-based PO was less positively related to feedback-seeking when PAGO was high (at +1 SD: simple slope = 0.08, n.s.) than when it was low (at −1 SD: simple slope = 0.31, p < 0.001). Hypothesis 9 states that the positive indirect effect of PWD on feedback-seeking via job-based PO was buffered by an employee’s PAGO. The results showed that the indirect effect was significantly less positive when PAGO was high (indirect effect = 0.03, n.s.) than when it was low (indirect effect = 0.11, p < 0.05) (difference in indirect effects = −0.09, p < 0.01, 95% Monte Carlo CI (−0.1283, −0.0406)). Hence, Hypothesis 9 was supported.

OPEN IN VIEWER

Figure 3.

The moderating effect of PPGO on the relationship between job-based PO and voice (Study 3).

OPEN IN VIEWER

Figure 4.

The moderating effect of PAGO on the relationship between job-based PO and feedback-seeking (Study 3).

Supplementary analysis

Similar to Study 2, we also separated the two dimensions of PWD and tested our hypothesized relationships. The results showed that the relationship between designing fun and job-based PO was positive and marginally significant (B = 0.35, p < 0.10), whereas the relationship between designing competition and job-based PO was not significant (B = 0.02, n.s.). We further examined the indirect effects of designing fun and designing competition on voice via job-based PO (for designing fun: indirect effect = −0.01, 95% Monte Carlo CI (−0.0489, 0.0160); for designing competition: indirect effect = −0.00, 95% Monte Carlo CI (−0.0240, 0.0120)), and their indirect effects on feedback-seeking via job-based PO (for designing fun: indirect effect = 0.02, 95% Monte Carlo CI (0.0033, 0.1320); for designing competition: indirect effect = 0.00, 95% Monte Carlo CI (−0.0510, 0.0637)). These results showed some divergence from those of our earlier analysis in which the two dimensions of PWD were combined into a composite measure. Finally, we examined the moderated mediating effects using the separate dimensions of PWD. The results indicated that PPGO did not significantly moderate the indirect effect of designing fun on voice via job-based PO (diff = 0.03, 95% Monte Carlo CI (−0.0051, 0.1055)) and the indirect effect of designing competition on voice via job-based PO (diff = 0.00, 95% Monte Carlo CI (−0.0373, 0.0277)). In contrast, PAGO significantly moderated the indirect effect of designing fun on feedback-seeking via job-based PO (diff = −0.08, 95% Monte Carlo CI (−0.1893, 0.0033)), but did not significantly moderate the indirect effect of designing competition on feedback-seeking via job-based PO (diff = −0.01, 95% Monte Carlo CI (−0.0734, 0.0700)). These findings suggest potential differences in the results when examining designing fun and designing competition separately versus combining them into a single composite in the analyses.

Theoretical implications

This research offers several theoretical contributions. First, we extend the research on PWD by exploring its impact on interpersonal behaviors. Existing research on PWD has predominantly focused on intrapersonal outcomes. However, in the process of task completion and performance evaluation, playful work designers inevitably interact with coworkers and supervisors. Although PWD is self-focused, supervisors and coworkers significantly influence the work environment of playful work designers—they provide reactions to PWD, shape how PWD is evaluated, and affect the extent to which PWD behaviors need to be adjusted (Fong et al., 2022). The benefits and costs of PWD are also transmitted through social interactions. To better capture the externality of PWD, it is necessary to embed it within a social-interactional context at work. This underscores the importance of investigating the communication between playful work designers and others. By focusing on voice and feedback-seeking—two communication-centric behaviors—we examine how PWD may influence employees’ engagement in sharing and soliciting ideas and suggestions. This investigation advances our understanding of how PWD affects interpersonal interactions, allowing for a more complete view of its emergence, transmission, and outcomes.

Second, we propose an important theoretical lens to provide a fuller understanding of PWD. Drawing on the extended self-theory (Belk, 2000; Dittmar, 1992; Pierce et al., 2001, 2003), we argue that job-based PO serves as a critical mediator between PWD and distal outcomes. The extended self-theory is particularly suitable for examining PWD due to the link between PWD and job-based PO—PWD provides employees with joy, discretion, understanding, and control over their work, all essential factors of job-based PO (Belk, 2000; Brown et al., 2005; Pierce et al., 2003, 2009). Through this lens, PWD can be understood as an extension and expression of self, which relates to employees’ proactive communication. This perspective moves beyond previous views that consider PWD merely as a work tactic. Moreover, as existing research has primarily explored the positive outcomes of PWD (e.g., Liu et al., 2023; Scharp et al., 2022, 2023), we offer a preliminary glimpse into its complexity by introducing job-based PO as a key mediator, which has been shown to have multifaceted effects (Baer and Brown, 2012; Chen et al., 2023; Dawkins et al., 2017; Pierce et al., 2009). Specifically, Study 3 finds a negative effect of PWD on voice via job-based PO, whereas our Study 2 shows a positive relationship between the two. This highlights the need for future studies to explore the potential negative effects of PWD. Furthermore, by integrating PWD into the formation process of job-based PO, we introduce an important antecedent that has been overlooked. Extended self-theory emphasizes the role of self-investment in developing psychological ownership (Belk, 2000; Pierce et al., 2009), and our findings indicate that PWD can act as a form of self-investment that fosters job-based PO. This expands the nomological network of psychological ownership and bridges the literature on extended self-theory and proactivity research.

Finally, by examining when the effects of PWD on voice and feedback-seeking (via job-based PO) are stronger or weaker, we provide more insights into for whom PWD may bring more benefits or costs. The extant research on PWD has primarily focused on dispositional traits that amplify PWD’s benefits, such as openness, playfulness, and flow proneness (Dishon-Berkovits et al., 2024; Liu et al., 2023; Scharp et al., 2019). Our study demonstrates that goal orientations determine whether PWD is translated into self-protective reticence or developmental communication. Specifically, the results in Study 3 show that when PPGO is high, the relationship between job-based PO and voice is non-significant, whereas it is negative when PPGO is low; similarly, when PAGO is high, the relationship between job-based PO and feedback-seeking is non-significant, whereas it is positive when PAGO is low. These findings suggest that PWD—which integrates play with work and appears to have minimal instrumental implications—may critically depend on employees’ goal preferences in achievement situations to realize its beneficial effects. We therefore reveal a previously overlooked motivational tension between ownership motives and evaluation concerns in shaping communicative behaviors at work. When achievement evaluation concerns are strong, they may neutralize ownership-driven motives. Accordingly, we bridge the literature on PWD, extended self-theory, and performance goal orientation theory.

Practical implications

This research provides several valuable insights for organizations seeking to maximize the benefits of PWD while managing its potential downsides. First, previous research has predominantly emphasized the positive effects of PWD and recommended that organizations offer PWD training for employees (Proyer et al., 2021; Scharp et al., 2021). Our findings indicate that while PWD may stimulate employees’ feedback-seeking behavior, it may also inadvertently reduce their willingness to voice in certain contexts (though it may not always be the case, as shown in Study 2), possibly because heightened job-based PO leads employees to be more protective of their unique insights. Therefore, organizations should not only provide PWD training, but also implement systems that encourage open idea sharing while protecting individual ownership. For example, organizations can establish formal mechanisms to track and attribute the origins of ideas, such as innovation logs or digital records, ensuring that employees are recognized and rewarded for their creative contributions even when their ideas are adopted by others. Regular cross-team innovation sessions can also create safe spaces for employees to present and discuss ideas generated through PWD, reinforcing a culture of psychological safety and knowledge sharing.

Second, given that job-based PO has been found to have complicated effects (e.g., enhanced engagement, increased knowledge hiding and territoriality) (Baer and Brown, 2012; Chen et al., 2023; Dawkins et al., 2017), managers should actively monitor for signs of excessive ownership, such as reluctance to share information or collaborate, and intervene through coaching or mediation as needed. It is also important to set clear expectations that both initiative and collaboration are valued, and to align incentives and performance evaluations accordingly. We recommend that organizations regularly gather feedback through surveys or open forums to monitor the dynamics of PWD, PO, voice, and feedback-seeking, allowing for timely adjustment of practices and policies. By taking these concrete steps, organizations can foster a balanced environment in which PWD and psychological ownership drive not only individual creativity, but also collective learning, open communication, and sustained. organizational growth.

Finally, our study shows that employees’ goal orientations (PPGO and PAGO) moderate the effects of job-based PO. Specifically, PWD is most functional for employees with high PPGO and low PAGO. Organizations can assess these orientations when designing team interventions or development programs, providing targeted support and training where needed. For example, training program or workshop that cultivate employees’ approach motivation while reducing avoidance motivation may be helpful. Managers can reframe feedback to emphasize skill development and provide encouragement, reinforcing employees’ PPGO through mastery-focused language. They can also design micro-tasks that are more likely to be completed on time. Employees, in turn, are encouraged to record their successful experiences to strengthen PPGO through self-persuasion and to motivate themselves in adverse situations, thereby reducing PAGO.

Strengths and limitations

This research has several strengths. We conduct multi-studies and employed different methods, which enhances the robustness and credibility of our findings and reduces the possibility of common method bias (e.g., method-specific variance) (Podsakoff et al., 2012; Rousseau and Fried, 2001). Moreover, our effort to collect samples in multiple Asian countries adds to the PWD literature, which has primarily focused on European countries (e.g., Scharp et al., 2022). In doing so, we examine the adaptability of theoretical tenets of PWD research across different cultural contexts, expanding their generalizability ⁴ .

Despite these strengths, several limitations suggest directions for future research. First, we do not consider alternative mechanisms linking PWD with voice and feedback-seeking. As noted earlier, other potential pathways may exist, such as relational energy or positive emotions (Scharp et al., 2022, 2023). Because our focus is on voice and feedback-seeking—achievement-related communicative behaviors that are instrumental for both employees and organizations—we do not include these affective mechanisms. Future studies should incorporate these alternative mediators to better capture the broader interpersonal effects of PWD.

Second, because we measured voice and feedback-seeking behaviors solely from supervisors’ perspectives, we did not incorporate coworkers’ viewpoints in examining the social and communicative impacts of PWD. This represents another limitation of our study. Future research should theorize and assess relevant outcomes from the perspectives of coworkers and other stakeholders.

Third, since the effect of job-based PO is contingent on various factors, we focus on second-stage moderators to identify conditions under which the distal outcomes of PWD are stronger or weaker. It is also valuable to examine first-stage moderator that may influence the relationship between PWD and job-based PO. For example, work centrality (Hirschfeld and Field, 2000) may strengthen the effect of PWD on job-based PO: employees who attach great importance to their work may experience stronger ownership feelings after playfully designing their job. We therefore encourage future research to investigate additional boundary conditions, particularly regarding the relationship between PWD and job-based PO, to provide a fuller picture of the PWD influence.

Fourth, research design of all three studies could be improved. Study 1 is designed only to examine the relationship between PWD and job-based PO. This could be enhanced by using a more complex scenario that allows participants to voice opinions and seek feedback in a simulated context. A field experiment is another option, in which researchers provide PWD training to a treatment group and compare behavioral and perceptual differences with a control group. Due to practical constraints, Study 2 does not include data on PPGO and PAGO, preventing us from testing the moderating and indirect effects. Study 2 is also limited to only two waves of data collection, measuring our predictor and mediator in the same wave, which restricts causal inference. To address this, we design a three-wave survey in Study 3. While Study 3 improves temporal separation, it is limited by its relatively small sample size, and we encourage future studies to replicate the model with larger samples.

Furthermore, in our supplementary analyses where we separated the two dimensions of PWD (i.e., designing fun and designing competition), we found both consistent and divergent results compared to the analyses using the aggregated PWD construct. These supplementary findings strengthen the integrity of our research and highlight the unique value of examining PWD as a holistic construct. That is, an overarching approach to PWD captures properties that are not inherent in its individual components. This reflects the well-known notion in Gestalt psychology that “the whole is other than the sum of its parts” (Koffka, 2013; Wagemans et al., 2012). Therefore, we call for future studies to examine our hypotheses and explore PWD using both separated and aggregated approaches, in order to better theorize and understand the complexity of this construct.

Future research directions

Based on our findings, we suggest several directions for future research. First, as our study indicates that PWD may be associated with both positive and negative interpersonal outcomes, future research should present a more balanced view of PWD by identifying its nuanced impacts not only on playful work designers but also on important others who observe or interact with them. For example, how others evaluate playful work designers and whether they intervene in response to PWD are important questions. It can be preliminarily inferred that PWD may elicit double-edged judgments from others, depending on observers’ work values or playfulness-related personality traits (Miller et al., 2002; Petelczyc et al., 2018). Second, we suggest that PWD may be related to employees’ longer-term career outcomes, which warrants further research attention. Addressing this issue would require a longer time horizon, including examinations of the relationship between PWD and career adaptability and success over time. Finally, better embedding PWD within organizational contexts represents another promising direction. On the one hand, PWD may help employees enhance person–environment or person–job fit (Tims et al., 2016). On the other hand, contextual factors may shape the effects of PWD. For instance, in organizational contexts where play is encouraged, playful work designers may flourish, whereas in contexts where play is discouraged, PWD may yield limited positive outcomes or even negative consequences.