A Multilevel Model of Employee Innovation

Thriving and Individual Innovation

Thriving is defined as the “psychological state in which individuals experience both a sense of vitality and learning at work” (Spreitzer et al., 2005: 538; see also Porath, Spreitzer, Gibson, & Garnett, 2012). Vitality, or feelings of energy and aliveness and a zest for work (Nix, Ryan, Manly, & Deci, 1999), and learning, referring to the acquisition of skills and knowledge to build confidence and capability (Edmondson, 1999), represent the affective and cognitive components of thriving. Together, the two dimensions are viewed as reflecting self-regulation in the workplace by providing internal cues that help employees assess their forward progress (Porath et al., 2012; Spreitzer et al., 2005; Spreitzer & Sutcliffe, 2007). Moreover, thriving serves an adaptive function by helping employees adjust to their work context and promoting personal development and growth. When thriving, employees are better able to gauge their own development in order to improve short-term effectiveness as well as long-term adaptability to their work context (Spreitzer et al., 2005). Importantly, some of the tenets of thriving—specifically vitality at work—closely align with the idea of intrinsic motivation, which has often been studied as a key element in relation to creativity (Amabile, 1996) and innovation (Anderson et al., 2014). However, according to Sonenshein, Dutton, Grant, Spreitzer, and Sutcliffe (2006), thriving is distinct from intrinsic motivation in that although thriving may at times derive from a desire to perform a behavior based on the affective enjoyment of the behavior itself (i.e., vitality), at other times the impetus for thriving has been described by emphasizing personally important achievement and recognition goals (i.e., learning).

Carmeli and Spreitzer (2009) offer three ways by which thriving sets the stage for creativity and innovation to occur. First, when employees are learning and developing at work, they are in an ideal position to recognize and implement improvement opportunities. Learning is required to obtain expertise, which then drives creative behavior (Amabile, 1998) and ensures the success of creative efforts (Carmeli & Spreitzer, 2009). Second, when individuals are thriving, they have more energy and motivation to devote toward investigating and implementing new work processes. Third, thriving encapsulates positive moods and emotions, both of which facilitate expansive cognitive thinking and creative problem solving (Hirt, Levine, McDonald, Melton, & Martin, 1997). Furthermore, the experience of positive emotions builds psychological and social resources (Fredrickson, 2001), which allows individuals to be more innovative. Previous empirical work supports this link between thriving and innovative work behaviors (i.e., Carmeli & Spreitzer, 2009). Thus, we expect to replicate this relationship in our hypothesized model:

Regulatory Focus and Thriving

Because thriving is viewed as a type of self-regulatory psychological state, it is malleable and, therefore, may be influenced by stable individual characteristics, such as regulatory focus. Regulatory focus, typically conceptualized as a chronic, individual disposition, represents two distinct forms of approach motivation: promotion focus and prevention focus (Higgins, 1997, 2000; Higgins, Friedman, Harlow, Idson, Ayduk, & Taylor, 2001; Higgins, Roney, Crowe, & Hymes, 1994). A promotion focus is associated with growth and developmental needs and involves striving for ideals, aspirations, and rewards through accomplishment, whereas a prevention focus is associated with needs for security and safety and involves fulfilling duties and obligations through responsible behaviors. With a promotion focus, salient goals are perceived as “gains” or “nongains” and involve construal of achievement goals as aspirations such that, when accomplished, they yield the highest expected utility and result in the emotional pleasure of cheerfulness (Higgins, 1997, 2000). Employees utilizing a prevention focus perceive salient goals as a “nonloss” or “loss,” leading them to focus on avoiding negative outcomes by fulfilling the basic needs and requirements of the job. In this way, a prevention focus involves the construal of goals as responsibilities rather than aspirations. When those responsibilities are accomplished, losses are prevented and utility is gained, which results in the emotional pleasure of quiescence (Shah & Higgins, 1997). Furthermore, values consistent with an “ideal” self—valuing hopes, desires, and aspirations—are manifested in a promotion focus, whereas a prevention focus is consistent with an “ought” self—valuing duty and responsibility (Higgins, 1997). Prior research supports a positive relationship between promotion focus and innovation and a negative relationship between prevention focus and innovation (cf. Lanaj et al., 2012). Although important, we are not primarily concerned with replicating these direct relationships. Rather, our focus is on expanding on these previously demonstrated direct relationships by explaining them via the mediating effect of thriving and identifying a context that enhances their effects.

Spreitzer et al.’s (2005) conceptual work on thriving acknowledged that regulatory focus might serve as an important trait that enables some individuals to thrive in the workplace more than others. Furthermore, Spreitzer et al. suggested that it is crucial for thriving that employees act agentically (i.e., being active and purposeful at work), which involve behaviors that are closely aligned with a promotion focus. Because of the salience of growth, accomplishment, and development to promotion-focused individuals, we argue that employees with a high (rather than low) promotion focus are more motivated to exert effort toward engaging in agentic work behaviors (e.g., concentrating on exceeding work expectations, exploring alternative task procedures, assisting team members and being attentive to their needs and engaging in risk taking and experimentation on the job) that are associated with thriving.

Because they are driven by aspirations, promotion-focused employees are likely to concentrate their behaviors on achieving and exceeding their assigned work responsibilities in order to feel a sense of accomplishment and increased energy. Förster, Grant, Idson, and Higgins (2001) found that successful completion of tasks generated higher levels of motivation for promotion-focused individuals. Focusing on attainment also relates to learning (Niessen et al., 2012; Orvis, Fisher, & Wasserman, 2009). In a meta-analysis of regulatory focus, Lanaj and colleagues (2012) found that learning goal orientation, an approach temperament, positively related to promotion focus. This suggests that a promotion focus may facilitate learning by directing behavior toward achieving new knowledge salient to mastering work tasks. Learning and mastery are also heavily reliant on social learning (Bandura, 1977), and social learning in the workplace often occurs through assisting other coworkers. Employees with a high promotion focus are also more likely to engage in exploratory behaviors because they are open to novel experiences that have the potential for rewards (Friedman & Förster, 2001; Wallace & Chen, 2006). Thus, promotion-focused employees view these behaviors as an appealing opportunity to fulfill experimentation and personal growth strivings. In turn, when employees engage in exploratory behaviors, they are more likely to exhibit both increased vitality and learning (Spreitzer et al., 2005). Exploration allows employees to generate, encounter, and implement novel ideas, information, and strategies for accomplishing work, which can provide and restore energy (e.g., R. Kaplan & Kaplan, 1995; S. Kaplan & Berman, 2010). Exploration also increases learning because the new ideas and strategies employees encounter expands their repository of knowledge and skills that can be applied in the workplace.

Because a high prevention focus manifests itself as strategies that concentrate on fulfilling basic requirements and duties while simultaneously guarding against errors and avoiding behaviors that move one toward mismatches and risks, behaviors are less agentic in nature. For example, by focusing only on basic duties and avoiding behaviors that may lead to negative outcomes, prevention-focused employees are far less likely to seek out opportunities to develop and change routines to make their work more efficient and effective. Engaging in the development and refinement of work routines to increase effectiveness is indicative of agentic behaviors that contribute to thriving at work (Spreitzer et al., 2005). Furthermore, because their salient outcomes are viewed as losses and nonlosses, prevention-focused employees are not likely to engage in the exploratory behaviors necessary to grow and learn at work. Prevention-focused individuals perceive that risks and the potential for negative outcomes outweigh the prospects for performance rewards and personal development. In turn, these employees are more likely to display a quiescent emotional state and a performance avoidance goal orientation (Lanaj et al., 2012), which inhibits the agentic behaviors necessary for thriving, suggesting that there is a negative relationship between prevention focus and thriving. Therefore, we propose,

Based on the aforementioned arguments and by linking Hypothesis 1 with Hypotheses 2 and 3, respectively, we propose the following additional hypotheses:

The Role of Employee Involvement Climate

Although individual traits may predispose some employees to have the needed motivation to thrive more than others, understanding features of the workplace environment that provide the requisite conditions for employees to thrive may contribute more to theory development and practice (Spreitzer et al., 2005). For example, even though promotion-focused employees may be more motivated to engage in highly agentic behaviors that result in experienced psychological states of vitality and learning, the opportunity to pursue such behaviors depends on the workplace context in which they act. Thus, without the right context that provides opportunities to thrive in the workplace, employees’ motivational predispositions may be of little circumstance.

The benefits of workplace context to employee thriving are suggested by tenets of SDT (cf. Gagné & Deci, 2005). SDT is a motivational framework that rests on the assumption that individuals possess an innate desire for personal growth, a sense of choice, and personal initiative in their lives (Deci, Connell, & Ryan, 1989; Deci & Ryan, 2000). Furthermore, SDT distinguishes autonomous motivation from controlled motivation such that the former encompasses both intrinsic motivation and valued activities internalized within the self (i.e., integrated extrinsic motivation) and the latter reflect purely extrinsic motives. In this way, autonomous motivation aligns with the self-regulatory, motivational aspects of thriving. SDT also posits that when employees are more able to satisfy their basic human needs for autonomy, competence, and relatedness, they are more likely to engage in agentic behaviors that ultimately result in greater psychological growth and development (Ryan & Deci, 2000), and, thus, a higher likelihood of thriving at work. However, Gagné and Deci (2005) and Deci et al. (1989) argue that the degree to which employees are autonomously motivated hinges on the extent to which the interpersonal work climate and support created by managers satisfies employees’ basic human needs. As such, we suggest that the fulfillment of employees’ basic human needs is embodied by a high employee involvement climate that provides essential prerequisites for enhancing the likelihood of thriving.

An employee involvement climate exists when employees within some defined unit (e.g., organization, department, or team) mutually understand that they (a) possess the power to make decisions and act on them, (b) may access and share the informational resources needed to undertake those actions effectively, (c) have opportunities to update their knowledge in order to continually develop their effectiveness, and (d) are rewarded for improving the effectiveness of their work unit and organization (Lawler, 1996; Richardson & Vandenberg, 2005; Riordan, Vandenberg, & Richardson, 2005; Vandenberg, Richardson, & Eastman, 1999). The primary benefit of a high employee involvement climate is that it facilitates a deeper cognitive understanding of the workplace as well as allows employees the freedom to work autonomously rather than being constrained (Blau, 1987; Butts, Vandenberg, DeJoy, Schaffer, & Wilson, 2009; Fuller, Marler, & Hester, 2006; Lawler, 1986; Mowday & Sutton, 1993; Spreitzer, 1996). Importantly, employee involvement climate likely operates most effectively at lower organizational levels because unit managers often can provide closely supervised functional directives for employee involvement within the work unit (Bowen & Ostroff, 2004; Ostroff & Bowen, 2000; Richardson & Vandenberg, 2005). As such, the immediate manager serves as an integral part in forming the group’s employee involvement climate, allowing for meaningful variation across units (Bowen & Ostroff, 2004; Richardson & Vandenberg, 2005).

We suggest that a high employee involvement climate allows for activities and behaviors that provide psychological “nutriments” (i.e., autonomy, relatedness, and competence) to satisfy fundamental human needs (Deci & Ryan, 2000). These psychological nutriments are supplied in a high employee involvement climate by offering opportunities for participation in decision making among groups, providing avenues for training and development, and, perhaps most important, allowing employees freedom to work autonomously through encouraged self-initiation. However, motivational benefits of the psychological nutriments provided by a high employee involvement climate may depend on the individual traits of employees in the organization, such as regulatory focus, which makes employees more or less prone to thrive in a high employee involvement climate. Defining this occurrence in terms of the “match hypothesis,” SDT proposes that people who are development-focused are the ones who are more motivated in climates that are more autonomy-supportive, resulting in positive performance and better well-being (Gagné & Deci, 2005).

For employees with a high promotion focus, a high employee involvement climate provides opportunities for autonomous motivation gains and work achievements that align with the promotion focus mindset, thereby creating a better regulatory fit between the work context and these employees’ desire to align their ideal selves with their work roles in order to feel fully present in their jobs (Kahn, 1990; Lanaj et al., 2012). That is, high employee involvement climates provide opportunities that motivate promotion-focused employees to fulfill their utmost potential—a “regulatory fit” (Higgins, 1997, 2000, 2005) that is maximized by features of the person (i.e., promotion focus) matching a context that supports fulfillment of fundamental human needs (i.e., a high employee involvement climate). Thus, it is possible that thriving may be enhanced when the motivational opportunities provided by a high employee involvement climate correspond with employees’ regulatory focus (Aaker & Lee, 2006; Agrawal, Menon, & Aaker, 2005; Dimotakis, Davison, & Hollenbeck, 2012; Higgins, Cesario, Hagiwara, Spiegel, & Pittman, 2010; Spiegel, Grant-Pillow, & Higgins, 2004). Accordingly, we expect that a high employee involvement climate allows increased autonomous motivation for those individuals with a high promotion focus, leading to higher levels of thriving and, subsequently, higher levels of innovation. This corresponds to what Edwards and Lambert (2007) define as first stage moderated mediation, but the proposed moderating effect occurs across levels of analysis.

Because of the conceptually distinct nature of prevention focus, as well as the demonstrated small correlation with promotion focus (Lanaj et al., 2012), we do not expect employee involvement climate to affect the prevention focus–thriving relationship. The combination of a high employee involvement climate and a high prevention focus does not provide a regulatory fit between features of the person (i.e., prevention focus) and the context of high involvement. Thus, we do not propose a cross-level moderated effect with regard to prevention focus and employee involvement climate. As demonstrated in previous research, mismatches in regulatory fit, such as we expect here, are often found to produce null effects (Dimotakis et al., 2012; Freitas & Higgins, 2002; Van Dijk & Kluger, 2011). Based on the arguments above and by building on our previous hypotheses, we propose the following:

Participants and Procedure

A total of 346 employees, comprising 75 work groups led by 75 unique supervisors, of two physical facilities organizations participated in the study. Employees were repair generalists whose jobs involved a wide variety of building and maintenance tasks including electrical, plumbing, and mechanical work as well as operating heavy machinery and boiler room work. While innovation needs may not be apparent in this type of work, the industry is routinely looking for more efficient methods to complete work as well as looking for “greener” methods to improve practices—methods that speak well to generating and implementing new and useful ideas in the workplace (i.e., innovation). The recruited sample across the two organizations consisted of 700 full-time employees, of whom 354 provided useable self-report data (response rate = 51%). Innovation performance ratings were unavailable for 8 participants, resulting in a total useable sample of self-report employee data and supervisor innovation performance ratings for 346 participants (79.2% male; final response rate = 49%). The work groups consisted of 2 to 18 employees (average size = 4.8; SD = 4.7) reporting to a single supervisor. Average participant age was 41.3 years (SD = 14.8), and employees had an average organizational tenure of 12.8 years (SD = 9.3), with 89.4% Caucasian, 6.1% African American, 3% Hispanic, and 1% Asian participants, with the remaining 1% either Other or not reported.

Measures

Measures were administered over a lagged, 3-month time period with employees initially completing measures of regulatory focus and employee involvement climate. One month later, participants completed a measure of thriving. Approximately one month after the completion of the thriving measure, supervisors completed ratings of each participant’s innovation-related performance.

Confirmatory Factor Analysis

We conducted a confirmatory factor analysis on our study measures to examine construct distinctiveness. Specifically, a multilevel measurement model was tested with the items for employee involvement climate, promotion focus, prevention focus, thriving, and innovation. In the measurement model, employee involvement climate items were loaded onto a factor at both the individual level (Level 1) and the group level (Level 2), allowing for random intercepts at the individual level that are free to vary across units at the group level (Muthén & Muthén, 2010). All other measures were modeled at the individual level only, while taking into account the lack of independence within units. Furthermore, following past factor-analytic work on the construct (Carmeli & Spreitzer, 2009; Porath et al., 2012) we modeled thriving as two distinct, but related factors representing learning and vitality.

The measurement model fit the data reasonably well, χ²(532) = 1475.73, comparative fit index (CFI) = .91, Tucker–Lewis index (TLI) = .90, root mean square error of approximation (RMSEA) = .07, within-units standardized root mean square residual (SRMR_Within) = .06, and between-units SRMR (SRMR_Between) = .06. All items loaded significantly on their respective factors at the a priori level of analysis and were above .76. Furthermore, the interfactor correlation between learning and vitality was high (.87) and within the range of correlations found in prior construct validation work on thriving (Porath et al., 2012). Because learning and vitality have been theoretically defined as elements of a higher-order thriving construct and our primary focus is on the effects of thriving as a whole rather than its constituent parts, we chose to operationalize thriving in a way that mimics its higher-order structure by combining scores on the two composites for learning and vitality (after group-mean centering each variable). In summary, these results provide evidence that our study measures capture distinct constructs at their expected level of analysis, and they substantiate our decision to combine the learning and vitality composite scores into a single measure of thriving in subsequent analyses.

Aggregation of Employee Involvement Climate

Researchers (e.g., Bliese, 2000) have proposed three steps to determine the viability of aggregation: sufficient within-group homogeneity, sufficient between-group heterogeneity, and that the group is naturally occurring. Because the unit of analysis (i.e., work groups) naturally occurs, establishing sufficient within-group homogeneity and between-group heterogeneity was necessary to justify aggregation for employee involvement climate. To assess within-group homogeneity, we used the r_wg(j) statistic (James, Demaree, & Wolf, 1993). Using uniform null and normal distributions, the average r_wg(j)values found were .93 and .87, respectively (range = .78-.99 and .63-.99; median = .89 and .85). In addition, the intraclass correlation and reliability of the mean, or ICC(1) and ICC(2) respectively (James, 1982; Shrout & Fleiss, 1979), were used to further assess homogeneity. The ICC(1) value is interpreted as the proportion of variance explained by group membership; an ICC(2) value greater than or equal to .70 represents that the group means are reliably different (Bliese, 2000; Bryk & Raudenbush, 1992). The intraclass correlations for employee involvement climate, calculated using the Bartko formula (see Bliese, 2002), were ICC(1) = .30 and ICC(2) = .72, with significant between-groups variance, F(74, 302) = 3.05, p < .05. These results supported aggregating employee involvement climate for the group-level analyses.

Analysis Strategy

Our data contained a hierarchical structure in which individuals were nested within their groups/supervisors. Furthermore, certain hypotheses were multilevel in nature in that they involved testing relationships between group-level variables (Level 2) and individual-level variables (Level 1). To appropriately model this, we used multilevel path analysis with Mplus 6.12 (Muthén & Muthén, 2010) and robust full maximum likelihood estimation while following recommendations by Preacher, Zyphur, and Zhang (2010) on how to model multilevel mediation. Preacher et al.’s (2010) multilevel mediation approach in the context of path analysis is best suited for the current study because it allows estimation of covariances for Level 1 random effects and indirect effects, and the multiple paths that are components of these indirect effects, without conflating the individual-level and group-level relationships. Furthermore, this approach allows us to incorporate Preacher and Hayes’s (2004) simultaneous estimation method of testing indirect effects in mediation rather than relying on step-wise procedures to test mediation (i.e., R. M. Baron & Kenny, 1986) or piecemeal estimation techniques, such as hierarchical linear modeling, that may potentially bias results because they do not allow for simultaneous estimation of all parameters. To test for the hypothesized moderated mediation effect of employee involvement climate, we adapted Bauer, Preacher, and Gil’s (2006) simultaneous multilevel regression procedure and applied it within Preacher et al.’s (2010) approach to examine the indirect effect of promotion focus on innovation (via thriving) at different levels of employee involvement climate (i.e., conditional effects). Combining the approaches of Preacher et al. and Bauer et al. was necessary because the former does not address the case of moderated mediation using path analysis and the latter only addresses moderated mediation in the context of hierarchical linear modeling and not path analysis. Using these two approaches together allowed us to employ the most statistically appropriate framework to test our study hypotheses.

Consistent with our proposed relationships at the individual level, all variables except employee involvement climate were group-mean centered. Also, learning and vitality were group-mean centered before they were combined into a single composite score. The group-mean centering approach ensures that there is no conflation of the individual and group-level effects in our path analytic results and allows for more unbiased estimates of the slopes and effects (Preacher et al., 2010; Z. Zhang, Zyphur, & Preacher, 2009). Employee involvement climate was grand-mean centered because it was specified as a group-level (Level 2) variable only and aggregation tests supported this level of analysis, and thus group-level effects were of primary interest.

Model and Hypotheses Testing

Means, standard deviations, and bivariate correlations are presented in Table 1. Before testing the multilevel moderated mediation model, we first had to examine relationships at the individual level of analysis. To accomplish this, we followed Preacher et al.’s (2010) guidelines and tested a path model specifying indirect effects of regulatory focus (promotion and prevention) on innovation through thriving (X → M → Y) while simultaneously taking into account direct effects and the nesting of individuals within groups/supervisors (i.e., the inclusion of random intercepts and slopes). In addition, age and work hours were included as controls with fixed effects on thriving and innovation. The purpose of this analysis was to (a) test the significance of the direct and indirect effects from X to Y through M and (b) estimate the significance of variability in the indirect effects over Level 2 units, or random effects, in order to establish the need for investigating moderation at Level 2 (Bacharach, Bamberger, & Doveh, 2008; Bauer et al., 2006). Assuming a significant expected indirect effect is found and variability exists at Level 2, we proceeded with the multilevel moderated mediation analysis taking into consideration effects for employee involvement climate.

OPEN IN VIEWER

Table 1

Means, Standard Deviations, and Bivariate Correlations Among Study Variables

Variable	M	SD	1	2	3	4	5	6	7
1. Age	41.31	14.81	—
2. Work hours	40.08	4.87	−.06	—
3. Promotion focus	3.78	0.92	.07	.12*	(.84)
4. Prevention focus	3.26	0.87	−.03	−.06	.35*	(.82)
5. Thriving	3.05	1.04	.01	.13*	.39*	−.13*	(.91)
6. Innovation	3.76	1.13	−.02	−.03	.04	−.06	.30*	(.91)
7. Employee involvement climate	3.12	1.12	.03	.15*	.22*	−.13*	.49*	.05	(.84)

Note: N = 346. Employee involvement climate values are for individual perceptions before aggregation to the group level. Internal consistency reliabilities are in parentheses on the diagonal. Work hours = hours worked per week.

*

p < .05.

Thriving was hypothesized to relate to innovation (Hypothesis 1), while promotion and prevention focus were hypothesized to relate to thriving (Hypotheses 2 and 3, respectively). Furthermore, we proposed that thriving mediates the relationship between (a) promotion focus and innovation (Hypothesis 4) and (b) prevention focus and innovation (Hypothesis 5). Results from our analysis at the individual level support these hypothesized relationships. Specifically, the path model results indicated that thriving was positively related to innovation (γ = .31, p < .01). Furthermore, promotion focus was positively related to thriving (γ = .17, p < .01) and prevention focus was negatively related to thriving (γ = –.08, p < .01). To provide a test of the indirect effects (Hypotheses 4 and 5), we used a parametric bootstrap procedure that employed 20,000 Monte Carlo replications to estimate a confidence interval (CI) around the indirect effects (Preacher et al., 2010). Results showed a positive indirect effect of promotion focus on innovation via thriving (estimate = .054, 95% CI = .009, .105) and a negative indirect effect of prevention focus on innovation via thriving (estimate = –.030, 95% CI –.052, –.004). Taken together, these results provide support for Hypotheses 1 through 5.

For the individual-level path model, we also found significant random (Level 2) effects for promotion and prevention focus (p < .05), suggesting possible moderators should be considered to explain this variability (Kenny, Korchmaros, & Bolger, 2003). Thus, we proceeded to test for multilevel moderated mediation in accord with Bauer et al. (2006). To estimate the cross-level relationships, we tested a model that included a direct relationship between employee involvement climate and thriving (i.e., a main effect) as well as a random slope between promotion focus and thriving predicted by employee involvement climate (i.e., a cross-level moderation effect). In an additional analysis, we tested a separate cross-level moderation model for prevention focus to explore the possibility that employee involvement climate has an effect on the prevention focus–thriving relationship. These model results indicated that employee involvement climate had a nonsignificant effect on the random slope between prevention focus and thriving (γ = .05, p > .05). Therefore, as expected, no support was found for cross-level moderation regarding prevention focus. Below, we turn our attention to promotion focus and discuss the model estimates obtained from our proposed multilevel moderated mediation model (see Figure 2).

OPEN IN VIEWER

Figure 2

Moderated Mediation Model Path Coefficients

As shown in Figure 2, all relationships in the proposed moderated mediation model were significant (p < .05). Furthermore, the observed relationships at the individual level did not change substantially after including employee involvement climate in the analysis. Promotion focus and prevention focus were related to thriving (γ = .17, p < .01, and γ = –.07, p < .01, respectively), while thriving was related to innovation (γ = .33, p < .01). We used Bryk and Raudenbush’s (1992) formulas to calculate pseudo-R² (~R²) for the model, which reflects the proportional reduction in Level 1 and Level 2 errors due to the inclusion of predictors in the model. Predictors accounted for 17% of the total variance in thriving and 15% of the total variance in innovation, suggesting promotion focus, prevention focus, and employee involvement climate were practically important in predicting thriving and innovation.

The model results (see Figure 2) provide support for Hypotheses 6 and 7. Our results indicated that employee involvement climate was positively related to thriving (γ = .24, p < .01) and the indirect effect of employee involvement climate on innovation via thriving was .082, with a Monte Carlo parametric bootstrap 95% CI of .018, .169.

Hypothesis 8a predicted that the effect of promotion focus on thriving would be moderated by employee involvement climate. The multilevel modeling results indicated a positive effect of employee involvement climate on the random slope between promotion focus and thriving (γ = .09, p < .05), often referred to as a cross-level interaction, and provided necessary initial support for the first stage moderated mediation (Edwards & Lambert, 2007) proposed in Hypothesis 8b. Following Aiken and West’s (1991) procedures, we plotted this interaction at higher and lower levels of employee involvement climate (1 SD above and below the mean). As shown in Figure 3, promotion focus related more positively to thriving when employee involvement climate was higher rather than when employee involvement climate was lower. Thus, Hypothesis 8a was supported.

OPEN IN VIEWER

Figure 3

The Moderating Effect of Employee Involvement Climate on the Relationship Between Promotion Focus and Thriving

Next, we used the method outlined by Bauer et al. (2006) for determining significance of conditional indirect effects in the context of multilevel regression by estimating the indirect effect of promotion focus on innovation via thriving at higher (+1 SD) and lower (–1 SD) levels of employee involvement climate. Results indicated that the indirect effect of promotion focus on innovation via thriving differed as a function of employee involvement climate. That is, the indirect effect was stronger when employee involvement climate was higher (estimate = .08, SE = .03, p < .05) and weaker when employee involvement climate was lower (estimate = .03, SE = .02, p > .05), supporting Hypothesis 8b. Furthermore, after including employee involvement climate as a moderator, the residual group-level variance in the slope of the relationship between promotion focus and thriving was no longer significant (σ² = .02, p > .05), indicating that the cross-level moderator significantly accounted for the between-group variance in that relationship and, subsequently, the variability in the indirect effect of promotion focus on innovation via thriving. In summary, results from our multilevel path analysis provided strong support for our hypothesized innovation process at the individual and group levels as well as the cross-level moderating effect of employee involvement climate.

Theoretical Implications

Our findings have several theoretical implications. First, the current research is one of the first attempts to empirically demonstrate the joint effects of regulatory focus and employee involvement climate on innovation and to incorporate an explanatory mechanism for those effects. Because approaches to studying organizational behavior that are solely dispositional or situational in nature are both theoretically and statistically underspecified (Cappelli & Sherer, 1991; House, Rousseau, & Thomas-Hunt, 1995), an integrative, multilevel approach—as we adopted here—allows for a richer understanding of how the interplay between individuals and their work environment can elicit a more motivating state that is beneficial to innovation. This multilevel approach also extends the literature on SDT by providing insights into how employees can self-regulate in the workplace through staying attuned to their psychological state of thriving in order to influence their own innovative behaviors, as well as the complex nature of organizational and individual conditions that operate together to foster thriving. Shedding light on propositions from SDT that increased opportunities for growth and development at work are best realized by those that (a) possess the necessary disposition and (b) are embedded in work environments that allow them to better engage in such opportunities (Deci & Ryan, 2000, 2008), our results support that a work climate high in employee involvement provides promotion-focused employees with a heightened sense of self-determination, volition, and freedom from organizational constraints and pressures by helping them to meet their fundamental human needs at work. This, in turn, leads to enhanced thriving at work and subsequent innovation.

Second, our results have implications for the individual-level process that unfolds from regulatory focus to influence innovation. Although cumulative research shows that regulatory focus is an important predictor of innovation (Lanaj et al., 2012), we are not aware of any research that has been devoted to examining how this effect occurs. We addressed this void by showing that individual differences in regulatory focus (i.e., promotion and prevention focus) exhibit differential relationships with thriving and, thereby, differential indirect effects on innovation. We showed that promotion focus is positively related to thriving and innovation, suggesting that those possessing values consistent with their “ideal” self—striving to reach hopes, desires, and aspirations—are more likely to act agentically and experience positive emotions, a desire for progress, and higher levels of energy at work as characterized by thriving. Thriving, in turn, relates to innovation. Conversely, our findings suggest that those individuals who focus mainly on the values of duty and responsibility that reflect a prevention focus experience decrements in innovation as a result of less frequently engaging in agentic behaviors associated with thriving. Spreitzer et al.’s (2005: 540) seminal theoretical work on thriving gives only passing notice to the role of individual predispositions such as regulatory focus and how they may influence the agentic work behaviors which are the “engine of thriving.” Our study elaborates on this suggestion and provides empirical evidence that thriving serves as an important self-regulatory mechanism that operates as an internal gauge of personal development and growth through which employees’ regulatory focus predispositions translate into higher (lower) innovation. This bolsters prior theoretical work on thriving and substantially adds to the empirical interest in the construct.

Finally, our findings highlight the important role that contextual influences play in affecting the individual-level innovation process. Explaining how features of an employee involvement climate motivate employees by providing psychological nutriments that satisfy their fundamental human needs (Deci & Ryan, 2000), we showed not only that employee involvement climate cascades down to influence employee thriving but also that the individual-level relationship of promotion focus with thriving and subsequent innovation differs depending on the degree to which employees are situated in autonomy-supportive climates fostered by high employee involvement. SDT suggests that both work contexts and individual traits in unison should fuel employee motivation (Deci & Ryan, 1985), but little attention has been devoted to the interaction or match between these two motivational considerations. Our findings confirm the importance of this interaction as suggested by SDT (Gagné & Deci, 2005) and regulatory fit theory (Lanaj et al., 2012) by showing that even the most promotion-focused employee will find it difficult to optimally flourish (i.e., thrive) unless the right workplace climate is in place that delivers the appropriate fit for the employee. It is only through the presence of both a high promotion focus and a high employee involvement climate providing opportunities to meet needs for autonomous motivation that thriving is maximized, and this effect is carried through to innovation.

In summary, by delineating and supporting our complex, multilevel relationships, we have extended previous theory and research on SDT (Deci & Ryan, 1985, 2000) and the multilevel process leading to innovation (e.g., Carmeli & Spreitzer, 2009; Jung et al., 2003). In doing so, we provide a more comprehensive understanding of how regulatory focus and employee involvement climate relate to innovation and the important role that thriving serves as an explanatory mechanism. Moreover, our findings were obtained through a rigorous time-lagged design across three points in time with multiple data sources, which has seldom been achieved in prior work. This methodological strength of our study, coupled with our use of multilevel path analysis (in which all hypothesized relationships were estimated simultaneously), attests to the robustness of our findings and their relevance to helping understand the motivations for employee innovation within the setting of complex individual and contextual interactions.

Limitations and Directions for Future Research

Although our study makes important contributions toward understanding the integration of regulatory focus, employee involvement climate, and thriving in predicting innovation, it is not without limitations. First, although participants were drawn from two different organizations, employees engaged in similar types of jobs, which entailed primarily day-to-day facilities work. Thus, our findings may not generalize to more managerial, cognitively laden jobs. However, because all participants were generalists who performed a variety of tasks, which contributes to job complexity and has long been considered an important contributor to creative performance (Oldham & Cummings, 1996), we are confident that innovation is a relevant component of performance in this context. Also, research supports the importance of innovation for other nonmanagerial jobs that lack a high degree of cognitive demands (i.e., Janssen, 2000).

Second, although the current study is predictive and uses a rigorous time-lagged design over three time periods, it is not truly longitudinal. As such, we encourage researchers to build on our results in a more longitudinal fashion in order to examine the role of changes in employee involvement climate and thriving over time in relation to changes in innovation. Longitudinal investigations such as this would also provide an opportunity to examine the reciprocal nature of thriving and innovation to determine the role that successful innovation plays in subsequent thriving at work and vice versa, testing previous suggestions by Spreitzer et al. (2005) that thriving operates through a continuous feedback loop to resources that enable future thriving.

Third, although we examined important individual and contextual factors relevant to innovation, there are likely other individual and contextual antecedents (e.g., passion, and creative self-efficacy; see Zhou & Shalley, 2011, for a review), moderators (e.g., job complexity), and mediators (e.g., job satisfaction) that could account for important variance in innovation. For example, the level of stress and anxiety experienced in the workplace may influence the extent to which employee involvement climate interacts with promotion focus to increase thriving and innovation. Drawing on research that supports a two-dimensional model of work stressors (e.g., Wallace, Edwards, Arnold, Frazier, & Finch, 2009), the extent to which employees are able to enact their promotion focus and thrive at work likely depends on the types of stressors in the workplace. Whereas “challenge” stressors (e.g., high workload, time pressures, and high levels of responsibility) facilitate innovative performance, “hindrance” stressors (e.g., organizational politics, red tape, and role ambiguity) inhibit such performance. In addition, off-task factors originating from a variety of sources (e.g., family, friends, and coworkers) increase cognitive load and, quite possibly, affect the extent to which a promotion focus is positively related to innovation, even in the presence of an employee involvement climate and thriving.

Finally, leadership style is a particularly important contextual factor that could be explored to expand our findings. Just as transformational leadership has been related to subordinate creativity (Shin & Zhou, 2003) and employee need satisfaction (Kovjanic, Schuh, Jonas, Quaquebeke, & Dick, 2012), supportive leadership styles (e.g., authentic leadership, ethical leadership) might further enhance the motivational benefits stemming from a high employee involvement climate and, thereby, promote thriving and innovation in the workplace.

Practical Implications and Conclusion

Our results suggest that organizations interested in innovation may benefit from focusing on employees’ regulatory focus when crafting the work environment. Although possibly difficult in practice, the selection of individuals who are repeatedly characterized by a promotion focus would be a first step toward a more innovative workforce. More important, though, is the potential impact that the joint relationship between promotion focus and employee involvement climate has on innovation. Thus, rather than placing restrictions on employee selection, a more pragmatic approach is one where organizational leaders and managers design and oversee the workplace in a way that fosters employee involvement, thereby enabling employees with a high promotion focus to experience forward progress and development through thriving and, in turn, increase their innovation.

The integral role that thriving played as a linking mechanism by exhibiting relationships with employee involvement climate, both types of regulatory focus, and innovation also has practical implications. Because of its key mediating role in the innovation process, managers would be wise to devote considerable time and effort to ways in which thriving at work can be maximized. More and more employees are looking for work to be a place where they are given opportunities to grow and develop as well as feel energized by the work they do. By involving employees—perhaps most easily accomplished by managers establishing norms of support, autonomy, and cooperation by providing power, information, rewards, and knowledge to subordinates—a high employee involvement climate may become instilled in the workplace and provide employees with just the right opportunities needed to enable high levels of thriving. Furthermore, developing a high employee involvement climate that enables thriving may deliver much higher returns on investment than more expensive benefits and perks such as gourmet meals and in-house concierge services that some employers tout as necessary to create a workplace environment that motivates creative and innovative behaviors.

Increased workforce innovation has become a strategic focus of many organizations as they strive to achieve and maintain a competitive advantage, but there is not a singular mechanism, context, or individual characteristic that leads to innovation. Rather, a myriad of multilevel processes drive innovation (Anderson et al., 2014; Hennessey & Amabile, 2010; Woodman, Sawyer, & Griffin, 1993). The current study supports this notion and provides important insights into how regulatory focus and employee involvement climate, both alone and together, influence employee thriving, and, subsequently, innovation. Our results suggest that workplace climates characterized by high employee involvement facilitate employee thriving and innovation through a positive interaction with promotion focus. For organizational scholars, our findings provide an example of how applying SDT within a multilevel framework, giving more attention to broader organizational considerations by incorporating individual and contextual characteristics, affords a more nuanced view of the motivational process underlying employee innovation. Hopefully, organizations can also make use of these findings to leverage their employees’ characteristics and craft aspects of the work environment that foster employee involvement in order to increase thriving and innovation in the workplace.