In this assignment, you are to critically read and evaluate a scholarly articles strengths, weaknesses, and contributions to the study field. Learning how to critique a journal article

Intrinsic and extrinsic motivation as predictors of work effort: The moderating role of achievement goals

Anders Dysvik∗ and Bård Kuvaas BI Norwegian Business School, Oslo, Norway

This research explored the roles of intrinsic motivation (IM) and extrinsic motivation (EM) and the 2 × 2 model of achievement goals as predictors of increased work effort (WE). A cross-lagged field study was conducted among 1,441 employees from three large Norwegian service organizations across a 10-month time span. The results showed that the relationship between IM and increased WE was more positive for employees with high levels of mastery-approach goals. This observation suggests that having congruent goals may accentuate the positive relationship between IM and WE.

Work in contemporary organizations has become increasingly complex, less routinized, unidimensional, and strictly defined (Cascio, 1998). Accordingly, organizations are increasingly dependent upon employees to uphold high levels of work effort (WE) on their own initiative (Hunter & Thatcher, 2007) in contrast to using more traditional work practices that attempt to standardize and control WE (Braverman, 1984). This raises the question as to why some employees exert more effort at work than others, which in turn may benefit the organization as a whole.

According to self-determination theory (SDT; Deci & Ryan, 2000), differences in WE exertion may be explained by the type of work motivation employees are driven by. SDT distinguishes between autonomous and controlled motivation (Gagné & Deci, 2005). The former describes acting based on perceived volition and choice, whereas the latter describes acting based on the perceived pressure of having to engage in actions. In SDT, intrinsic motivation (IM), formally defined as the motivation to perform an activity for its own sake in order to experience the pleasure and satisfaction inherent in the activity (Deci, Connell, & Ryan, 1989), represents autonomous motivation in its purest form (Gagné & Deci, 2005).1 Intrinsically motivated employees work on tasks because they find them enjoyable, interesting and that participation is its own reward, which

∗Correspondence should be addressed to Anders Dysvik, Department of Leadership and Organizational Management, BI Norwegian Business School, 0484 Oslo, Norway (e-mail: [email protected]). 1SDT also distinguishes between different forms of autonomous and controlled motivation, but as the focus of this paper is on intrinsic and extrinsic motivation in particular, readers are directed to Gagné and Deci (2005) for a more comprehensive presentation of the full SDT motivational continuum with its different sub-dimensions.

DOI:10.1111/j.2044-8309.2011.02090.x

British Journal of Social Psychology (2013), 52, 412–430

© 2012 The British Psychological Society

www.wileyonlinelibrary.com

412

in turn should accentuate their task-directed effort (Ryan & Deci, 2000). In contrast, extrinsic motivation (EM) focuses more on the consequences to which the activity leads than on the activity itself (Gagné & Deci, 2005). Being extrinsically motivated involves performing an activity with the intention of attaining some separable consequence, such as receiving an award, avoiding guilt, or gaining approval (Deci, Ryan, & Williams, 1996, p. 167). Employees who are extrinsically motivated work harder to attain a desired consequence or to avoid a threatened punishment (Deci & Ryan, 2000). While previous theorizing advocated additive effects from IM and EM (Porter & Lawler, 1968), recent research suggests that IM and EM vary with respect to their influence on employee outcomes (Gagné & Deci, 2005).

The purpose of the present study is to explore the interplay between IM and EM and achievement goals, also referred to as goal orientation.2 Achievement goals refer to the purpose3 or cognitive–dynamic focus of competence-related behaviour (Elliot & McGregor, 2001, p. 501). The achievement goal approach (AGA) delineates between mastery4 and performance goals. Mastery goals represent purposes for which an employee is concerned with developing their competence or mastering a task, while performance goals represent purposes for which an employee is concerned with demonstrating their competence relative to others (Elliot, 2005). A second distinction made by AGA is whether employees are directed towards the possibility of obtaining competence (approach), or away from the possibility of incompetence (avoidance; Elliot & Harackiewicz, 1996). These four dimensions underpin a 2 × 2 conceptualization of achievement goals that entails each combination of the mastery-performance and approach-avoidance distinctions (Elliot & McGregor, 2001). Mastery approach (MAP) oriented individuals strive to achieve self-referent task mastery by skill acquisition and by comparing their current effort with past effort. In contrast, performance approach (PAP) oriented individuals strive towards demonstrating task mastery compared to others. Mastery-avoidance (MAV) oriented individuals strive to avoid skill loss or not mastering a task, with a self-referenced orientation, and performance-avoidance (PAV) oriented individuals seek to avoid failure and looking incompetent relative to others (Van Yperen, 2003).

Both AGA and SDT emphasize the importance of individual perceptions of autonomy, that is, feeling like the source of one’s own behaviours (Ryan & Deci, 2002, p. 8) and competence, or feeling effective in one’s interactions with the social environment and experiencing opportunities to exercise one’s capacities (Ryan & Deci, 2002, p. 7). AGA scholars (e.g., Dweck, 1985; Nicholls, 1984) propose that individuals high in mastery goals and involved in a task based on self-oriented behaviour are also intrinsically motivated, which contributes to initiating and sustaining the activity. In turn, this involvement may be experienced as rewarding and developmental when task mastery and feelings of competence emerge. As such, the concept of mastery goals aligns well with IM (Deci & Ryan, 2000).

2We are adhering to Elliot’s (2005) call to refer to goal orientation as achievement goals in order to move towards a more specific and contextual level of analysis. 3Achievement goals are also used in different operational levels such as a combination of reason or aim (Dweck, 1986) or overarching orientation (Ames & Archer, 1988). 4We use mastery goal and performance goal as labels in this paper. In contrast, other researchers refer to mastery goals as task goals (Nicholls, 1984) or learning goals (Dweck, 1999). Performance goals are often referred to as ego goals (Nicholls, 1984).

Work motivation and work effort 413

Still, SDT and AGA differ with respect to the motives held by individuals when engaged in goal-directed behaviour. AGA is mainly concerned with the purpose for employees’ behaviour and argues that dispositional goals influence cognition, affect, and behaviour in achievement contexts. SDT, in contrast, focuses on the inherent pleasure and satisfaction derived from the activity based on the fulfilment of innate needs (e.g., Deci & Ryan, 2000; Elliot & Dweck, 2005; Elliot & Harackiewicz, 1996; Ntoumanis, 2001; Rawsthorne & Elliot, 1999) or universal necessities that are essential for human development and integrity (Gagné & Deci, 2005). In SDT, the satisfaction of the need is more important than whether there are individual differences in need strength. To say that a need is universal implies that there should not be high variation in need strength, and that individuals are likely to suffer more or less equally from need thwarting. Accordingly, goals/motives and traits/dispositions are likely to vary between persons, whereas needs are assumed to be universal across persons (Sheldon, Cheng, & Hilpert, 2011). Therefore, SDT research does not focus on the consequences of the strength of those needs for different individuals, but rather on the consequences of the extent to which individuals are able to satisfy the needs within social environments. Also, SDT describes the concept of competence unidimensionally, while AGA underscores the differences in competence perception, and that such perceptions may be self- or other-referenced (Elliot, McGregor, & Thrash, 2002). In sum, SDT places more emphasis on underlying needs and perceptions of need fulfilment, and AGA focuses on what makes individuals feel successful (Marsh, Craven, Hinkley, & Debus, 2003).

Although AGA and SDT can both explain variation in the motivation to exert WE, we do not know how the interplay between the different motives predicted by AGA and SDT influences WE since surprisingly few studies combine these two theories (Pulfrey, Buchs, & Butera, 2011). This may be an unfortunate oversight, given the likelihood that employees are subject to different motivational sources. Accordingly, we aim to contribute to our understanding of how employee motivation predicts WE by investigating the interaction between IM and EM and achievement goals. Furthermore, both SDT (Gagné & Deci, 2005) and AGA (DeShon & Gillespie, 2005; Fryer & Elliot, 2007; Payne, Youngcourt, & Beaubien, 2007; Yeo, Loft, & Xiao, 2009) stress the dynamic nature of employee motivation. Still, prior research relating both achievement goals and facets of work performance (including WE; Payne et al., 2007) and IM and facets of work performance (Gagné & Deci, 2005) is predominantly cross-sectional. Accordingly, by investigating the interplay between IM and EM and achievement goals over time, we contribute to SDT and AGA by capturing the dynamism of employee work motivation.

Theory and hypotheses According to SDT, IM requires the fulfilment of three innate, psychological needs: the need for autonomy, competence, and relatedness. The fulfilment of these needs predicts the influence of social contextual factors on individual growth-oriented processes and well-being (Deci & Ryan, 2000). When the needs are being met in a specific environment, individuals will be more likely to engage in activities for personal enjoyment rather than because they feel coerced into them (Ryan & Deci, 2006). Furthermore, the review by Gagné and Deci (2005) and more recent research, convincingly demonstrates how intrinsically motivated employees are more involved in their jobs and demonstrate greater effort and goal attainment than those less intrinsically motivated (e.g., Dysvik & Kuvaas, 2011; Grant, 2008; Piccolo & Colquitt, 2006; Zapata-Phelan, Colquitt, Scott, & Livingston, 2009).

414 Anders Dysvik and Bård Kuvaas

Extrinsically motivated behaviours depend upon the perception of a contingency between the behaviour and attaining a desired consequence such as implicit approval or tangible rewards or avoiding a negative consequence such as punishment (Gagné & Deci, 2005). The effectiveness of extrinsic motivators for increasing WE remains a controversial issue within motivational research, for instance, with respect to variable pay systems (e.g., Gerhart & Rynes, 2003; Kuvaas, 2006; Weibel, Rost, & Osterloh, 2010). Among the available research, meta-analytical evidence is supportive of a positive relationship between variable pay systems and increased performance quantity, but not quality of work (Jenkins, Mitra, Gupta, & Shaw, 1998). Furthermore, a recent meta-analysis reports a strong positive relationship between extrinsic motivators and performance for less- interesting tasks (Weibel et al., 2010). Both meta-analyses are therefore supportive of a positive relationship between EM and WE.

The moderating role of achievement goals SDT proposes that IM may emerge or be sustained universally as the need for autonomy, competence, and relatedness are basic to all individuals (Gagné, 2009). This approach, which focuses on the current and situational-specific perceptions of need satisfaction (DeShon & Gillespie, 2005; Elliot et al., 2002) differs slightly from AGA, which focuses on more general and less situational-dependent mid-level trait-type dispositions. In addition, the main focus of SDT is whether individuals feel coerced to perform activities or choose to engage in them based on the satisfaction derived from the activity itself. AGA, on the other hand, focuses more on purposes for engaging in performance-related behaviours (self- vs. other-regulated; directed at improvement vs. avoiding loss of competence). Consequently, IM and achievement goals should be regarded as conceptually separate (Elliot et al., 2002; Ntoumanis, 2001). Nevertheless, the two theories share considerable similarities, such as the importance of competence-supportive work environments, and that extrinsic rewards, social comparisons, and normatively based standards may impede individual outcomes (Deci & Ryan, 2000; DeShon & Gillespie, 2005; Gagné, 2009). In what follows, we argue that achievement goals will influence the relationship between IM and EM and WE depending on whether the goals pursued are congruent with the two types of motivation.

Prior studies have found MAP-oriented individuals to direct their achievement strivings towards personal improvement and skill development with an internal locus of perceived control and causality (see Elliot, 2005 for a review). In work settings, MAP- oriented individuals regard their skills as being more malleable and exhibit effort not only to achieve current tasks, but also to develop the ability to master future tasks. This drive should, in turn, facilitate higher levels of WE (Dragoni, Tesluk, Russell, & Oh, 2009; Paparoidamis, 2005; VandeWalle, Brown, Cron, & Slocum, 1999) and interest for the task at hand (Rawsthorne & Elliot, 1999). In support of this, prior studies have found positive relationships between MAP goals and WE (e.g., Porath & Bateman, 2006; VandeWalle et al., 1999). Furthermore, research on the self-concordance of individual goal systems, or the degree to which stated goals express enduring interests and values (Sheldon & Elliot, 1999), shows that individuals pursuing self-concordant goals based on IM put more effort into their work. Therefore, in addition to the motivation to work hard stemming from inherent satisfaction with the work, MAP goal orientation should explain additional effort arising from the motivation to improve one’s self. This resembles the suggestion that the self-referent motivation to improve and the pleasure-based motivation stemming

Work motivation and work effort 415

from the activity are congruent (Deci & Ryan, 2000). Consequently, MAP goals should accentuate the relationship between IM and WE.

Hypothesis 1: The relationship between IM and increased WE is moderated by MAP goals. The higher the MAP goals, the more positive the relationship.

As for the remaining three achievement goal dimensions, none of these focus on the development of skill or the interesting aspects of the task itself; therefore, they may be said to be incongruent with interest in general (Van Yperen, 2003) and IM in particular (Deci & Ryan, 2000).

In contrast to MAP goals, PAP goals are more normatively oriented towards demon- strating competence relative to that of others (Van Yperen, 2006). Such concerns may distract individuals away from the activity itself and instead towards assessing the individual’s performance relative to that of others. As such, extrinsically motivated employees whose behaviours are controlled by specific external contingencies should exert more effort when high in PAP or PAV goals, given the congruence between EM and the normative dimension of performance goals. As for the MAV dimension, employees with high levels of such goals focus on trying to avoid self-referent negative outcomes, which may evoke feelings of risk when facing challenging tasks or feelings of worry and apprehension about not meeting one’s own standards of competence and success (e.g., Baranik, Stanley, Bynum, & Lance, 2010; Elliot & McGregor, 2001; Sideris, 2007). Consequently, no interactions between IM or EM and MAV goals should occur. We therefore hypothesize:

Hypothesis 2: The relationship between EM and increased WE is moderated by PAP goals. The higher the PAP goals, the more positive the relationship.

Hypothesis 3: The relationship between EM and increased WE is moderated by PAV goals. The higher the PAV goals, the more positive the relationship.

Method Participants The participants in our study were employees in three large Norwegian service organizations from different industries (670 within power supply and maintenance, 643 within auditing and consulting services, and 1,665 within banking and finance). Representatives of the three organizations distributed questionnaires to their employees by use of a web-based tool (Confirmit). The first data collection was conducted between September and November 2008. The second data collection was conducted between August and October 2009. This resulted in complete data sets from 1,441 employees and a response rate of 48%. The participants were informed that their responses would be treated confidentially when responding to the survey, in order to reduce the presence of response distortion (Chan, 2009). Of the respondents, 39.8% were women and 60.2% were men; 71% held a university degree of 3 years’ study or more; and average tenure was 11 years.

416 Anders Dysvik and Bård Kuvaas

Materials and procedure All the items were placed on a 5-point Likert response scale (1 = strongly disagree and 5 = strongly agree). The items can be consulted in Appendix. Cronbach’s alphas for each scale are presented in Table 1.

IM was measured at time one by means of six items previously developed and used in a Norwegian setting by Kuvaas and Dysvik (2009).

EM was measured at time one by means of four items previously developed and used in Norwegian settings (Kuvaas & Dysvik, 2011).

Achievement goals MAP, PAP, and PAV goals were measured at time one by the 13-item scale validated by VandeWalle (1997), and previously used in a Norwegian context by Dysvik and Kuvaas (2010). The MAV goal dimension was measured at time one by the six-item scale validated by Baranik et al. (2007).

WE was measured at time one and time two by five items that capture how much effort employees put in their jobs. This scale has previously been used by Kuvaas and Dysvik (2009).

To control for potential socio-demographic and organizational differences in the predictor, the dependent variables education (measured by six categories where 1 represented “primary and lower secondary school” and 6 represented “master’s degree of five years’ study or more”), gender (measured by two categories where 1 represented “women” and 2 represented “men”), organizational tenure (in years), and dummy variables for organizational affiliation were included as controls in the analyses. We included the measure of WE at time one as a control variable in order to unveil the incremental validity of our independent variables on WE at time two.

Initially, an exploratory principal component analysis with promax rotation was performed on all the multiple-scale items to determine item retention (Farrell, 2010). In order to avoid confounded measures, we applied relatively stringent rules of thumb and retained only items with a strong loading of .50 or higher on the target construct (Nunnally & Bernstein, 2007), a cross loading of less than .35 on other included factors (Kiffin-Petersen & Cordery, 2003), and a differential of .20 or more between included factors (Van Dyne, Graham, & Dienesch, 1994).

To test for moderation, we used hierarchical moderated regression (Cohen, Cohen, West, & Aiken, 2003) and the computer software SPSS 19.0. Interaction terms often create multicollinearity problems because of their correlations with main effects. We thus computed the interaction terms by centering the variables before multiplying them with each other. In the first step, the control variables were regressed on WE, followed by IM and EM (Step 2), the four achievement goals (Step 3), and finally, the interaction terms between IM and EM and each of the four achievement goal dimensions (Step 4).

Results The principal component analysis revealed that all items met our inclusion criteria (see Appendix for details). The final scales were computed by averaging the items. All scales demonstrated acceptable reliability estimates, ranging from .76 to .89. The means, standard deviations, bivariate correlations, and reliability estimates are reported in Table 1. Pairwise and multiple variable collinearity were inspected by collinearity

Work motivation and work effort 417

Ta bl

e 1.

D es

cr ip

tiv e

st at

is tic

s, co

rr el

at io

ns ,a

nd sc

al e

re lia

bi lit

ie s

M ea

n SD

1. 2.

3. 4.

5. 6.

7. 8.

9. 10

. 11

. 12

. 13

. 14

.

1. O

rg an

iz at

io n

1 0.

28 –

– 2.

O rg

an iz

at io

n 2

0. 22

– −.

33 ∗∗

– 3.

O rg

an iz

at io

n 3

0. 50

– −.

62 ∗∗

−. 54

∗∗ –

4. G

en de

r 1.

63 –

−. 06

∗ .0

6∗ .0

1 –

5. Ed

uc at

io na

ll ev

el 4.

53 1.

34 .4

7∗∗ −.

19 ∗∗

−. 26

∗∗ .0

7∗ –

6. Te

nu re

10 .2

7 9.

60 −.

32 ∗∗

.1 1∗∗

.2 0∗∗

.0 1

−. 42

∗∗ –

7. W

or k

ef fo

rt (t

im e

1) 4.

18 0.

48 .1

4∗∗ −.

11 ∗∗

−. 03

−. 08

∗∗ .1

0∗∗ −.

10 ∗∗

– (.8

2) 8.

In tr

in si

c m

ot iv

at io

n (t

im e

1) 3.

81 0.

64 −.

12 ∗∗

−. 04

.1 4∗∗

.0 0

−. 02

.0 6∗

.3 9∗∗

– (.8

2)

9. Ex

tr in

si c

m ot

iv at

io n

(t im

e 1)

3. 23

0. 76

.0 9∗∗

−. 11

∗∗ .0

2 .1

0∗∗ .0

5∗ −.

07 ∗∗

.0 3

−. 07

∗∗ –

(.7 6)

10 .

M as

te ry

-a pp

ro ac

h go

al s

(t im

e 1)

3. 93

0. 56

.0 3

−. 07

∗ .0

3 .0

2 .2

1∗∗ −.

17 ∗∗

.4 0∗∗

.3 2∗∗

.0 5

– (.7

7)

11 .

M as

te ry

-a vo

id an

ce go

al s

(t im

e 1)

3. 74

0. 62

.0 6∗

−. 08

∗∗ .0

1 −.

13 ∗∗

−. 15

∗∗ .0

3 .2

2∗∗ .0

3 .1

4∗∗ −.

05 ∗

– (.7

9)

12 .

Pe rf

or m

an ce

-a pp

ro ac

h go

al s

(t im

e 1)

3. 22

0. 67

−. 07

∗ −.

09 ∗∗

.1 3∗∗

−. 02

.1 0∗∗

−. 10

∗∗ .2

0∗∗ .0

9∗∗ .2

4∗∗ .2

2∗∗ .1

9∗∗ –

(.8 4)

13 .

Pe rf

or m

an ce

-a vo

id an

ce go

al s

(t im

e 1)

2. 09

0. 65

−. 10

∗∗ .0

3 .0

7∗∗ .0

6∗ −.

13 ∗∗

.0 4

−. 25

∗∗ −.

16 ∗∗

.1 6∗∗

−. 38

∗∗ .0

8∗∗ .2

4∗∗ –

(.8 9)

14 .

W or

k ef

fo rt

(t im

e 2)

4. 16

0. 50

.1 6∗∗

−. 14

∗∗ −.

02 −.

10 ∗∗

.1 0∗∗

−. 08

∗∗ .6

1∗∗ .3

2∗∗ .0

5 .3

1∗∗ .1

6∗∗ .1

6∗∗ −.

17 ∗∗

– (.8

4)

N ot

e. N

= 1,

44 1;

co ef

fic ie

nt al

ph as

in di

ca tin

g sc

al e

re lia

bi lit

ie s

ar e

in pa

re nt

he se

s; ∗ p

� .0

5; ∗∗

p �

.0 1.

418 Anders Dysvik and Bård Kuvaas

Table 2. Regression analyses of the direct and moderated relationships

Work effort (time 2)

Step 1 Step 2 Step 3 Step 4

Organization 2 −.10 ∗ ∗ ∗ −.11

∗ ∗ ∗ −.12 ∗ ∗ ∗ −.11

∗ ∗ ∗

Organization 3 −.06 ∗ −.09

∗ ∗ −.09 ∗ ∗ −.09

∗ ∗

Gender −.05 ∗ −.05

∗ −.05 ∗ −.05

∗

Educational level .02 .01 .00 .00 Tenure .01 .01 .01 .01 Work effort (time 1) .59

∗ ∗ ∗ .54

Collepals.com Plagiarism Free Papers

Are you looking for custom essay writing service or even dissertation writing services? Just request for our write my paper service, and we'll match you with the best essay writer in your subject! With an exceptional team of professional academic experts in a wide range of subjects, we can guarantee you an unrivaled quality of custom-written papers.

Get ZERO PLAGIARISM, HUMAN WRITTEN ESSAYS

Why Hire Collepals.com writers to do your paper?

Quality- We are experienced and have access to ample research materials.

We write plagiarism Free Content

Confidential- We never share or sell your personal information to third parties.

Support-Chat with us today! We are always waiting to answer all your questions.