Deviant behavior in constrained environments: Sensation-Seeking

0 downloads 0 Views 333KB Size Report
of Sensation-Seeking and Deep Learning on workplace deviance depend on how ... particular, we found that in highly constrained workplaces, Sensation ...
Personality and Individual Differences 108 (2017) 20–25

Contents lists available at ScienceDirect

Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid

Deviant behavior in constrained environments: Sensation-Seeking predicts workplace deviance in shallow learners Peter J. O'Connor a, Sharon Stone a, Benjamin R. Walker b,⁎, Chris J. Jackson b a b

School of Management, Queensland University of Technology Business School, GPO Box 2434, Brisbane, Queensland 4001, Australia School of Management, UNSW Business School, University of New South Wales, Sydney, NSW 2052, Australia.

a r t i c l e

i n f o

Article history: Received 18 May 2016 Received in revised form 24 November 2016 Accepted 28 November 2016 Available online xxxx Keywords: Sensation-Seeking Workplace Deviance HLMP Interpersonal Deviance Organizational Deviance

a b s t r a c t Workplace deviance represents voluntary and intentional behavior that is harmful to organizations. In the current study, we examine workplace deviance using the Hybrid Model of Learning in Personality (HMLP), which has previously been shown to predict dysfunctional behavior in the workplace. Using a sample of part-time workers, we investigated whether dimensions of Rationality and Sensation-Seeking predict workplace deviance, when controlling for known predictors from the Big Five. More interestingly, we also assessed whether the effects of Sensation-Seeking and Deep Learning on workplace deviance depend on how constrained employees feel in their current position. Overall, our results indicate the unique importance of Rationality in the prediction of Interpersonal Deviance, and Sensation-Seeking and Deep Learning in the prediction of Organizational Deviance. In particular, we found that in highly constrained workplaces, Sensation Seekers tend to engage in deviant behavior when they have low levels of Deep Learning, but tend not to engage in deviant behavior when they have high levels of Deep Learning. Results are consistent with the HMLP and have implications for the management of deviant behavior in the workplace. © 2016 Published by Elsevier Ltd.

1. Introduction Workplace deviance can be defined as “voluntary behavior that violates significant organizational norms and in so doing threatens the well-being of an organization, its members, or both” (Robinson & Bennett, 1995, p. 556). Workplace deviance involves various negative work behaviors ranging from discrete demeanors such as taking unapproved breaks, to more destructive deeds such as aggression and violence. Research has identified several precursors to workplace deviance; these include highly constrained working conditions (Spector & Jex, 1998) and individual personality characteristics of employees (Diefendorff & Mehta, 2007). The current study examines workplace deviance from both a personality trait and situational perspective, in order to further our understanding of situations where particular individuals might be at a high risk of engaging in deviant workplace behavior. This research is necessary because despite the increasing incidence of workplace deviance (Chirayath, Eslinger, & De Zolt, 2002) researchers have not yet identified specific risk factors for sub-groups of individuals. Workplace deviance contains two dimensions: Organizational Deviance and Interpersonal Deviance. Organizational Deviance refers to deviant behavior directed at an organization or its systems, such as ⁎ Corresponding author. E-mail address: [email protected] (B.R. Walker).

http://dx.doi.org/10.1016/j.paid.2016.11.062 0191-8869/© 2016 Published by Elsevier Ltd.

stealing, taking long lunch breaks, or leaving early (Liao, Joshi, & Chuang, 2004). Interpersonal Deviance refers to deviant behaviors directed at other individuals in the organization, such political deviance, gossiping, and aggression (Bennett & Robinson, 2000). These variables are generally only moderately correlated with each other (Bennett & Robinson, 2000) and are likely predicted by different variables. In the present study, we focus on both forms of workplace deviance, but argue that they occur in different people and for different reasons. Previous research exploring the relationship between personality traits and workplace deviance has linked the Big Five model of personality to workplace deviance (Diefendorff & Mehta, 2007). The Big Five is the dominant framework for describing individual differences in personality, and consequently provides a good starting point for the investigation of the relationship between personality traits and workplace deviance. Research on the Big Five and workplace deviance has found that low Conscientiousness (i.e., low trait levels of organization, reliability) and low Agreeableness (i.e., low trait levels of empathy, compassion) have been the most consistent predictors of workplace deviant behavior (e.g., Bolton, Becker, & Barber, 2010). In addition to the Big Five, specific motivational traits have also been linked to workplace deviance. Motivational traits represent stable individual differences in dimensions of motivation, thought to underlie the basic dimensions of personality (Gray & McNaughton, 2000) and therefore provide a theoretical platform that is helpful in describing why individuals are motivated to partake in certain work behaviors (Bennett &

P.J. O'Connor et al. / Personality and Individual Differences 108 (2017) 20–25

Robinson, 2000). Research has indicated that individuals who are high in approach motivation and high in avoidance motivation are likely to engage in deviant behaviors (Diefendorff & Mehta, 2007). In the current study, we build on research investigating the trait basis of workplace deviance, by examining dysfunctional workplace behavior through the lens of the Hybrid Model of Learning in Personality (HMLP; Jackson, 2008). We also consider how traits interact with situational variables to produce deviant workplace behavior. As outlined in more detail later, the HMLP is a model of learning and personality designed to explain performance, counterproductive behavior, and learning, based on the idea that all individuals have an underlying, biologically based drive termed ‘Sensation-Seeking’ that motivates them to learn and explore their environment. According to the model, Sensation-Seeking will result in functional or dysfunctional behavior, depending on whether individuals learn to re-express or redirect their instinctively driven Sensation-Seeking approach tendencies with more complex conscious, socio-cognitive based cognitive styles, namely: Mastery, Rationality, Deep Learning, and Conscientiousness (Jackson, 2008). In the present research, we focus specifically on Sensation-Seeking, Rationality, and Deep Learning in the prediction of workplace deviance, as we believe these dimensions play important roles in the prediction of deviant behavior. 2. Rationality and Interpersonal Deviance Rationality has received very little attention in personality psychology but is well understood in clinical psychology from the perspective of Rational Emotive Behavior Therapy (Ellis, 2004; Jackson, Izadikhah, & Oei, 2012), in which rational beliefs are defined as beliefs that are logical and consistent with reality, whereas irrational beliefs are defined as unstable, illogical, and not consistent with reality. At the trait level, Rationality can be defined as the tendency to hold rational beliefs, and is characterized by flexible, non-dogmatic thinking, and emotional independence (Jackson, 2008). Rationality is therefore associated with functional cognitions (flexible thinking, emotional independence) whereas low Rationality is associated with dysfunctional cognitions and behaviors (fixed, dogmatic, unreasonable, and emotionally dependent beliefs associated with high demandingness). We suggest that Rationality will be a direct predictor of interpersonal workplace deviance (i.e., deviance characterized by aggression, interpersonal conflict, bullying, etc.). Individuals with irrational beliefs are known to have problems with interpersonal conflict in general and are more likely than people with rational beliefs to have poor mental health (Ellis, 2004). Additionally, the dogmatic nature of individuals low in Rationality indicates they will poorly handle individuals who disagree with their perspective. It follows therefore that low Rationality will be associated with workplace behavior characterized by aggressive, rude, or antisocial behavior directed at other individuals within the organization (Interpersonal Deviance). H1. There will be a negative relationship between Rationality and Interpersonal Deviance.

3. Sensation-Seeking, Deep Learning, and Organizational Deviance We also argue that Deep Learning will moderate Sensation-Seeking in the prediction of Organizational Deviance (i.e., deviant behavior directed at the organization) in highly constrained workplaces. The HMLP claims that a common biological foundation exists for positive and negative behavior within the workplace (Jackson, 2008; O'Connor & Jackson, 2008). Specifically, the model posits that both functional and dysfunctional learners have an underlying instinctive urge to learn and explore their environment, which manifests at the trait level as Sensation-Seeking (O'Connor & Jackson, 2008). Functional learners are understood to adaptively utilize Sensation-Seeking through socio-

21

cognitive skills to accomplish productive outcomes (e.g. self-reported work performance, school performance). Dysfunctional learners on the other hand maladaptively utilize their Sensation-Seeking, based on poorly developed socio-cognitive skills, and consequently engage in risky or counterproductive behavior. This aspect of the HMLP is supported (Gardiner & Jackson, 2015; Jackson, Baguma, & Furnham, 2009; Jackson, Hobman, Jimmieson, & Martin, 2009; Jackson et al., 2012; O'Connor & Jackson, 2008) and extends research claiming that Sensation-Seeking increases the tendency to engage in primarily dysfunctional behavior such as risk-taking behaviors (Ball & Zuckerman, 1990). Deep Learning is associated with deep processing and critical thinking. Individuals high in Deep Learning have both the tendency and ability to devote their cognitive resources to reflecting on experiences and integrating new information (Jackson, 2008). According to the HMLP (Jackson, 2008), Deep Learners are effective experiential learners (see Kolb, 1984), in that they seek out concrete experiences and adaptively use such experiences to reflect and learn. Consistent with Jackson (2008), we argue that Sensation Seekers with low Deep Learning will struggle to learn from their experiences, and will be motivated to move from one concrete experience to another. On the contrary, we argue that Sensation Seekers with high Deep Learning will learn from their experiences, and only seek out new experiences once they have reflected and integrated knowledge based on their prior experiences. In the context of work, we argue that Sensation-Seeking and Deep Learning will have predictable relationships with Organizational Deviance under certain conditions. In particular, we focus on high levels of Organizational Constraints (i.e., constrained working conditions, restrictive rules/ procedures, inadequate facilities) because previous research has illustrated the importance of this variable in the prediction of workplace deviance (Spector & Jex, 1998). Specifically, we suggest that Sensation Seekers with low levels of Deep Learning will have difficulty when working under high levels of Organizational Constraints, because the restrictive rules and procedures limit their need for new experiences. Frustration with such conditions will likely result in Sensation Seekers engaging in Organizational Deviance behaviors such as breaking rules and leaving early. On the other hand, we suggest that Sensation Seekers with high Deep Learning will have less difficulty working under high levels of Organizational Constraints. It is likely they will quickly learn that Organizational Deviance behavior is inappropriate (based on their effective experiential learning) and also develop a complex understanding of why such restrictions might be necessary in their environment. Furthermore, their ability to reflect on, and learn from difficult experiences, as opposed to simply seeking out new experiences, might mean that such individuals will respond adaptively to limited resources. Indeed such individuals may recognize innovative and creative ways to make the best of a difficult situation. Consistent with this, Sensation-Seeking and related constructs (e.g. extraversion) have been found to predict creativity in constrained conditions (see O'Connor, Gardiner, & Watson, 2016; Zuckerman, 2014). Consequently, it follows that Sensation Seekers with high Deep Learning will be less likely to feel frustrated in high levels of Organizational Constraints, will more likely be engaged in their work, and be less likely to engage in Organizational Deviant behavior overall. H2. Deep Learning will moderate the relationship between SensationSeeking and Organizational Deviance under conditions of high Organizational Constraints. In such conditions, Sensation-Seeking will lead to Organizational Deviance at low Deep Learning, but will not predict deviance at high levels of Deep Learning.

4. Method 4.1. Participants Participants were part-time workers from various occupations and organizations who were concurrently undertaking university study

22

P.J. O'Connor et al. / Personality and Individual Differences 108 (2017) 20–25

(N = 86, 71 women, 15 men, Mage = 26.48 years, SDage = 10.85, age range = 17 to 60 years). Participants received course credit in return for participation and completed the measures online. 4.2. Measures 4.2.1. Hybrid Model of Learning in Personality (HMLP) The HMLP, measured using the Learning Styles Profiler (for details see Jackson, 2008), consists of 75 items distributed across five subscales: Sensation-Seeking (α = 0.75), Mastery (α = 0.80), Rationality (α = 0.81), Deep Learning (α = 0.74), and Conscientiousness (α = 0.75). Participants responded on a 3-point scale (0 = false, 1 = can't decide, 2 = true). Several articles by Jackson and colleagues support the psychometrics of the HMLP and its claims of validity in predicting functional and dysfunctional performance (Jackson et al., 2012). The items for these scales remain the same but the labels have been updated. 4.2.2. Interpersonal and Organizational Deviance (IOD; Bennett & Robinson, 2000) This widely used measure consists of seven items that assessed deviant behavior directed at individuals (α = 0.81) and 12 items that assessed deviant behavior directed at organizations (α = 0.81). Participants indicated the extent to which they engaged in a variety of behaviors in the past year on a 5-point scale (1 = never, 2 = once or twice a year, 3 = several times a year, 4 = once or twice a month, 5 = weekly). 4.2.3. Organizational Constraints scale (OCS; Spector & Jex, 1998) This 11-item scale assesses the degree participant's job performance is constrained by work issues (α = 0.83). Participants responded on a 5point scale (1 = strongly disagree, 5 = strongly agree). 4.2.4. Agreeableness, Conscientiousness (Big Five Inventory, BFI; John, Donahue, & Kentle, 1991) Agreeable individuals tend to be compassionate and cooperative (α = 0.78). Conscientious individuals tend to organized, disciplined, and dutiful (α = 0.81). We administered the full 44-item BFI but only used Agreeableness and Conscientiousness as they are the most relevant for this study. Participants responded on a 5-point scale (1 = strongly disagree, 5 = strongly agree). Prior to data collection, the study was granted full ethics approval by the lead author's university human research ethics committee. 5. Results Means, standard deviations, Cronbach Alphas, and variable correlations with Interpersonal Deviance and Organizational Deviance are presented in Table 1. The correlation between Interpersonal and Organizational Deviance (r = 0.48, p b 0.001) showed a strong

relationship but sufficient differentiation that dimensions of workplace deviance can be considered as two separate constructs. In support of H1, only Rationality in the HMLP had a moderate negative relationship with Interpersonal Deviance (r = −0.26, p = 0.014). Further support for H1 was found by regression. In order to assess whether Rationality could account for variance in Interpersonal Deviance when controlling for known trait predictors of Deviance, we conducted a hierarchical multiple regression, controlling for Conscientiousness and Agreeableness from the Big Five Inventory in the first step (these variables have been shown to predict workplace deviance previously, see Bolton et al., 2010). The overall model with all predictors entered was significant R2 = 0.09, F(3, 82) = 2.75, p = 0.048. Importantly, the addition of Rationality at step 2 resulted in a significant R2 change (R2 change = 0.05, F(1, 82) = 4.57, p = 0.03), indicating that Rationality could explain an additional 5% of the variance in Interpersonal Deviance, when controlling for the variance accounted for by Conscientiousness and Agreeablness in the Big Five Inventory. H2 proposed that Deep Learning would moderate the relationship between Sensation-Seeking and Organizational Deviance under conditions of high Organizational Constraints. To test this hypothesis, we ran a 3-way moderated regression, with Sensation-Seeking as the IV, Organizational Deviance as the DV, and Deep Learning and Organizational Constraints as the two moderators. Prior to running this analysis, we standardized the IV's and moderator variables, in order to create mean centered variables and eliminate problems with multicollinearity. We also created three 2-way interaction terms by multiplying all combinations of the IV and moderators, so that we could control for potential 2-way interactions in our test of the hypothesized 3-way interaction. Finally, we calculated the 3-way interaction term by calculating the product of Sensation-Seeking, Deep Learning, and Organizational Constraints. We then ran a hierarchical multiple regression with the IV and moderator variables at step 1, the three 2-way interactions at step 2 and the 3-way interaction at step 3. Results are summarized in Table 2. They demonstrate that the IV's and interaction terms combined explain a significant 22% of variance in Organizational Deviance (R2 = 0.22, F(7, 78) = 3.05, p = 0.007). Furthermore, the results demonstrate that consistent with H2, the 3-way interaction between Sensation-Seeking, Deep Learning, and Organizational Constraints was significant at step 3 of the model. Additionally, the two way interaction term between Sensation-Seeking and Deep Learning, was also significant at step 3. The 2-way and 3-way interactions were followed up with simple slopes analyses. Results are illustrated in Figs. 1 and 2. Fig. 1 indicates that, broadly consistent with the HMLP, SensationSeeking is positively associated with Organizational Deviance at low Deep Learning (β = 0.23, p = 0.11) and negatively associated with high Organizational Deviance at high Deep Learning (β = −0.31, p = 0.08). We note however that although the significant interaction term

Table 1 Means, Standard Deviations, Cronbach Alphas and Correlations of the Hybrid Model of Learning in Personality, Deviance, Organizational Constraints and Control Variables. Variable 1. Sensation-Seeking 2. Mastery 3. Rationality 4. Deep Learning 5. Conscientiousness 6. Interpersonal Deviance 7. Organizational Deviance 8. Organizational Constraints 9. Agreeableness 10. Conscientiousness ⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.

M 21.30 21.77 21.06 16.72 23.09 11.00 20.53 20.69 34.17 34.57

SD

2

3

4

5

6

7

8

9

10

5.39 6.39 6.25 5.53 4.55 3.70 5.99 5.91 5.37 5.72

0.53⁎⁎⁎

0.12 0.28⁎⁎

0.13 −03 −0.04

−0.01 0.30⁎⁎ 0.29⁎⁎ 0.12

0.09 0.05 −0.26⁎⁎ −0.05 −0.18

0.06 −0.01 −0.04 0.00 −0.21 0.48⁎⁎

0.02 0.06 −0.34⁎⁎ −0.03 −0.04 0.32⁎⁎ 0.29⁎⁎

−0.03 0.03 0.22⁎ 0.15 0.22⁎

0.15 0.43⁎⁎ 0.28⁎⁎ 0.04 0.38⁎⁎

−0.20 −0.22⁎ 0.05

−0.05 −0.32⁎⁎ 0.03 0.10

P.J. O'Connor et al. / Personality and Individual Differences 108 (2017) 20–25 Table 2 Results from the 3-way Moderated Regression of Sensation-Seeking Predicting Organizational Deviance Moderated by Deep Learning and Organizational Constraints. Variable Step 1 Constant Sensation-Seeking (SS) Deep Learning (DL) Organizational Constraints (OC) Step 2 Constant Sensation-Seeking Deep Learning Organizational Constraints SS × DL SS × OC DL × OC Step 3 Constant Sensation-Seeking Deep Learning Organizational Constraints SS × DL SS × OC DL × OC SS × DL × OC

B

sr2

R2

R2 change

0.00 0.06 0.00 0.29⁎

0.00 0.00 0.08

0.09

0.09

0.04 −0.02 −0.03 0.38⁎⁎ −0.28⁎ 0.01 0.06

0.00 0.00 0.13 0.08 0.00 0.00

0.16⁎

0.07

0.09 −0.05 0.03 0.50⁎⁎ −0.26⁎ −0.14 0.015 −0.20⁎

0.00 0.00 0.17 0.05 0.01 0.00 0.06

0.22⁎

0.06⁎

⁎ p b 0.05. ⁎⁎ p b 001.

obtained in the hierarchical regression reveals that the slopes are different from each other the simple slopes analyses reveals that neither slope is significantly different from zero. Fig. 2 provides support for H2, in that a 2-way interaction effect was found between Sensation-Seeking and Deep Learning, only in employees who reported high levels of Organizational Constraints. Indeed, this interaction was highly significant at high (+1SD) levels of Organizational Constraints (conditional interaction = −0.46, p b 0.001), but not at low (−1SD) levels of Organizational Constraints (conditional interaction = −0.06, p = 0.67). 1The test of simple slopes in the significant interaction, revealed a non-significant positive relationship between Sensation-Seeking and Organizational Deviance at low levels of Deep Learning (β = 0.27, p = 0.11), and a significant negative relationship between Sensation-Seeking and Organizational Deviance at high levels of Deep Learning (β = − 64, p = 0.02). It seems therefore Deep Learning reduces deviant behavior in Sensation Seekers, but only in workplaces characterized by high Organizational Constraints. 6. Discussion

Organizational Deviance

This study investigated whether HMLP (Jackson, 2008) could be used to predict variation in interpersonal and organizational workplace deviance. We formulated hypotheses regarding whether specific scales of the HMLP could predict interpersonal and Organizational Deviance

Low

High

23

and also specified the conditions upon which this would occur. In support of H1, Rationality was a significant unique predictor of Interpersonal Deviance such that high Rationality was associated with decreased deviance. This result makes theoretical sense, because Rationality is a known predictor of positive interpersonal relationships, and has also been found to predict other functional outcomes such as performance and Grade Point Average (Jackson, Baguma, et al. 2009). It was also hypothesized (H2) that Deep Learning would moderate the relationship between Sensation-Seeking and Organizational Deviance under conditions of high Organizational Constraints. We found support for this hypothesis, in that we found a highly significant 2way interaction between Sensation-Seeking and Deep Learning in the prediction of Organizational Deviance at high levels of Organizational Constraints. However, this hypothesis was only partially supported, because the positive relationship between Sensation-Seeking and Organizational Deviance at low levels of Deep Learning was not quite significant (p = 0.11). However the negative relationship between Sensation-Seeking and Organizational Deviance at high Deep Learning was strong and significant (β = −64, p = 0.02). It seems therefore that, as hypothesized, Deep Learning serves to inhibit Sensation Seekers from engaging in Organizational Deviance, and this effect is particularly strong in highly constrained workplaces. Interestingly, the results also indicated that Deep Learners who were low in Sensation-Seeking were the most likely to engage in Organizational Deviance (in highly constrained workplaces). The results indicated that as these individuals increased in Sensation-Seeking, Organizational Deviance decreased dramatically. An important question therefore, is why are Deep Learners who are low in Sensation-Seeking particularly prone to deviant behavior? According to the HMLP (Jackson, 2008), socio-cognitive attributes such as Deep Learning are most adaptive when individuals also have high levels of SensationSeeking. In the context of highly constrained workplaces, the beneficial effects of Sensation-Seeking make sense. Consistent with the HMLP (Jackson, 2008) and Kolb's (1984) model of experiential learning, we argue that Deep Learners require exposure to ongoing concrete experiences in order to learn and operate functionally. In highly constrained workplaces, it is likely that such individuals will not be exposed to many novel experiences, and consequently will suffer from job dissatisfaction, poor work engagement, frustration, and ultimately deviant behavior. When Deep Learners are also high in Sensation-Seeking on the other hand, they can theoretically generate their own concrete experiences, even within the tight constraints they operate. For example, as mentioned previously, a Deep Learner also high in Sensation-Seeking is likely to come up with creative ways to meet their objectives with limited resources (see Zuckerman, 2014). Overall therefore, we argue that the increased novelty that occurs as a direct result of being a Sensation Seeker, will serve to substitute the lack of novelty that occurs in highly constrained jobs. Although we found that high Sensation-Seeking reduces the negative impact of Organizational Constraints on Deep Learning, we emphasize that this does not mean that a constrained environment is good for Sensation Seekers high in Deep Learning. As Fig. 2 suggests, Organizational Constraints enhanced workplace deviance overall; our results simply demonstrate that high Deep Learners also high in SensationSeeking simply handle this environment better. Consistent with this, we found no significant difference between high Deep Learners with high levels of Sensation-Seeking in conditions of high versus low Organizational Constraints. However, we found a significant difference between high Deep Learners with low Sensation-Seeking in conditions of high versus low Organizational Constraints.

Sensation-Seeking

Fig. 1. Interaction effect of Sensation-Seeking and Deep Learning on organizational workplace deviance.

1 These two conditional interactions represent the conditional effect of the SensationSeeking × Deep Learning interaction at high (+1SD) and low (−1SD) values of the moderator. They were calculated using PROCESS (Hayes, 2013) in SPSS.

24

P.J. O'Connor et al. / Personality and Individual Differences 108 (2017) 20–25 High Organizational Constraints

Low Organizational Constraints 1.5

Low Deep Learning

Organisational Deviance

1.0

High Deep Learning

0.5 0.0 Low

High -0.5

Low

High

-1.0

Sensation-Seeking

-1.5

Sensation-Seeking

Fig. 2. Three-way interaction between Sensation-Seeking, Deep Learning, and Organizational Constraints in the prediction of Organizational Deviance.

7. Managerial implications We believe that the two main findings from this research have important implications for managers. First, our finding that Rationality provides incremental validity over known trait-based predictors of Interpersonal Deviance allows managers to potentially identify individuals who might be at risk for this form of deviance. Specifically, it seems that employees low in Rationality (i.e., individuals who hold a range of irrational beliefs, tend to be dogmatic, etc.) are more likely to engage in deviant behavior than their more rational colleagues. We suggest that knowing which individuals are at risk for deviant behavior will assist managers in helping regulate and manage this problematic behavior (e.g., during the performance planning and review process). Indeed, much research on Rational Emotive Behavior Therapy has demonstrated that irrational behavior can be managed through the use of a few relatively simple techniques such as setting goals and challenging dysfunctional beliefs (see Ellis, 2004) that we suggest can be applied effectively in the workplace. Second, our finding that Deep Learners with low Sensation-Seeking are at risk of engaging in Organizational Deviance in highly constrained conditions also has important managerial implications. Indeed, in contrast to low Rationality, high Deep Learning is not an intuitive predictor of deviant behavior in the workplace, and therefore would not usually be considered a risk factor for such behavior. Our results show that in low levels of Organizational Constraints, Deep Learning is virtually unrelated to deviance, whereas in high levels of Organizational Constraints, Deep Learning is a clear risk factor for Organizational Deviance, except in individuals who also have a high level of Sensation-Seeking. In particular, Deep Learners working in highly constrained conditions (i.e., many rules, few resources) are likely to act out against their organization, possibly due to frustration with the lack of learning opportunities and novelty inherent to their position. We therefore suggest that managers of employees who are high in Deep Learning, provide such employees with much needed learning opportunities and novelty. 8. Limitations The primary limitation of the current study was the relatively small sample size (N = 86). Due to this relatively small sample, we only had sufficient power to detect moderate or large direct effects and large conditional effects, which possibly explains why H2 was only partially supported. A further problem with our use of a relatively small sample size was the necessity to use an ad-hoc p b 0.05 criterion for significance. Using this liberal p-value increased our chances of incorrectly rejecting the null and consequently making a type I error. We believe however that our two key results were probably not due to chance; in particular the hypothesized interaction between Sensation-Seeking and Deep Learning at high levels of constraints received a p-value of b0.001.

This indicates that even in a sample of 86 participants, a finding of this effect size (or greater) would occur by chance b1 in 1000 times. Nevertheless we acknowledge that further research is needed to replicate these results with a larger sample. 9. Conclusion Our results showed that the HMLP can be used to understand and predict Interpersonal and Organizational Deviance in the workplace. In accordance with theory, we found that Rationality and SensationSeeking moderated by Deep Learning predicted workplace deviance. Results align with the theory underlying the HMLP (Jackson, 2008) in that Sensation-Seeking can predict functional and dysfunctional learning in conjunction with higher cognitions. This study contributed to the empirical evidence for the theory by demonstrating that Sensation-Seeking interacts with Deep Learning in the prediction of Organizational Deviance at high levels of Organizational Constraints. This demonstrates that Sensation Seekers are not inherently more likely to engage dysfunctional, risk-taking behaviors, and importantly that Sensation Seeking does not always lead to higher levels of workplace deviance. These results are therefore consistent with earlier research on Sensation-Seeking demonstrating that Sensation-Seeking can be functional (e.g., Ball & Zuckerman, 1990). Furthermore, we suggest that Deep Learning is relatively unexplored in the workplace literature and may be an important variable. A desire to gain deep knowledge and understanding of particular topics may be a risk factor for deviance in highly constrained work environments. In such situations, high levels of Sensation-Seeking seem to operate as a preventative factor for Deep Learners. References Ball, S. A., & Zuckerman, M. (1990). Sensation seeking, Eysenck's personality dimensions and reinforcement sensitivity in concept-formation. Personality and Individual Differences, 11, 343–353. http://dx.doi.org/10.1016/0191-8869(90)90216-e. Bennett, R. J., & Robinson, S. L. (2000). Development of a measure of workplace deviance. Journal of Applied Psychology, 85, 349–360. http://dx.doi.org/10.1037//0021-9010.85. 3.349. Bolton, L. R., Becker, L. K., & Barber, L. K. (2010). Big Five trait predictors of differential counterproductive work behavior dimensions. Personality and Individual Differences, 49, 537–541. http://dx.doi.org/10.1016/j.paid.2010.03.047. Chirayath, V., Eslinger, K., & De Zolt, E. (2002). Differential association, multiple normative standards, and the increasing incidence of corporate deviance in an era of globalization. Journal of Business Ethics, 41, 131–140. http://dx.doi.org/10.1023/a:1021358425403. Diefendorff, J. M., & Mehta, K. (2007). The relations of motivational traits with workplace deviance. Journal of Applied Psychology, 92, 967–977. http://dx.doi.org/10.1037/00219010.92.4.967. Ellis, A. (2004). Why rational emotive behavior therapy is the most comprehensive and effective form of behavior therapy. Journal of Rational-Emotive & Cognitive-Behavior Therapy, 22, 85–92. http://dx.doi.org/10.1023/b:jore.0000025439.78389.52. Gardiner, E., & Jackson, C. J. (2015). Personality and learning processing underlying maverickism. Journal of Managerial Psychology, 30, 726–740. http://dx.doi.org/10. 1108/JMP-07-2012-0230.

P.J. O'Connor et al. / Personality and Individual Differences 108 (2017) 20–25 Gray, J. A., & McNaughton, N. (2000). The neuropsychology of anxiety: An enquiry into the functions of the septo-hippocampal system (2nd ed.). Oxford, England: Oxford University Press. Hayes, A. F. (2013). An introduction to mediation, moderation, and conditional process analysis: A regression based approach. New York: Guilford Press. Jackson, C. J. (2008). Measurement issues concerning a personality model spanning temperament, character, and experience. In G. Boyle, G. Matthews, & D. Saklofske (Eds.), Personality theory and assessment (pp. 73–93). London, England: Sage Retrieved from https://www.researchgate.net/profile/Chris_Jackson4/publications Jackson, C. J., Baguma, P., & Furnham, A. (2009). Predicting grade point average from the hybrid model of learning in personality: Consistent findings from Ugandan and Australian students. Educational Psychology, 29, 747–759. http://dx.doi.org/10.1080/ 01443410903254583. Jackson, C. J., Hobman, E. V., Jimmieson, N. L., & Martin, R. (2009). Comparing different approach and avoidance models of learning and personality in the prediction of work, university, and leadership outcomes. British Journal of Psychology, 100, 283–312. http://dx.doi.org/10.1348/000712608x322900. Jackson, C. J., Izadikhah, Z., & Oei, T. P. S. (2012). Mechanisms underlying REBT in mood disordered patients: Predicting depression from the hybrid model of learning. Journal of Affective Disorders, 139, 30–39. http://dx.doi.org/10.1016/j.jad.2011.09.025. John, O. P., Donahue, R. E., & Kentle, R. L. (1991). The big five inventory: Versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality and Social Research.

25

Kolb, D. (1984). Experiential learning. Englewood Cliffs, NJ: Prentice Hall. Liao, H., Joshi, A., & Chuang, A. C. (2004). Sticking out like a sore thumb: Employee dissimilarity and deviance at work. Personnel Psychology, 57, 969–1000. http://dx.doi.org/10. 1111/j.1744-6570.2004.00012.x. O'Connor, P. J., Gardiner, E., & Watson, C. (2016). Learning to relax versus learning to ideate: Relaxation-focused creativity training benefits introverts more than extraverts. Thinking Skills and Creativity, 21, 91–108. http://dx.doi.org/10.1016/j.tsc.2016.05.008. O'Connor, P. J., & Jackson, C. J. (2008). Learning to be saints or sinners: The indirect pathway from sensation seeking behavior through mastery orientation. Journal of Personality, 76, 733–752. http://dx.doi.org/10.1111/j.1467-6494.2008.00502.x. Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant workplace behaviors: A multidimensional-scaling study. Academy of Management Journal, 38, 555–572. http://dx. doi.org/10.2307/256693. Spector, P. E., & Jex, S. M. (1998). Development of four self-report measures of job stressors and strain: Interpersonal conflict at work scale, organizational constraints scale, quantitative workload inventory, and physical symptoms inventory. Journal of Occupational Health Psychology, 3, 356–367. http://dx.doi.org/10.1037/1076-8998.3. 4.356. Zuckerman, M. (2014). Sensation seeking (psychology revivals): Beyond the optimal level of arousal. Psychology Press.