Developing and Validating Short Form Protean and Boundaryless Career Attitudes Scales

Abstract

The Protean and Boundaryless Career Attitudes scales developed by Briscoe and colleagues have facilitated research on career attitudes and mindsets, but they are unnecessarily lengthy and somewhat redundant in their content. To address these concerns, this article presents three studies that develop and validate short forms of the Protean and Boundaryless Career Attitudes scales (i.e., PCA-SF and BCA-SF). Study findings suggest that the PCA-SF and BCA-SF provide a more efficient assessment of the protean and boundaryless career attitudes constructs than the full-length measures and exhibit partial measurement equivalence across U.S. and Korean populations.

Keywords

protean career attitudes boundaryless career attitudes scale refinement measurement equivalence personality organizational commitment career satisfaction

The traditional organizational career, where an individual proceeds linearly up the hierarchy of one or two organizations in order to achieve extrinsic rewards, is often contrasted with alternative career arrangements such as protean and/or boundaryless career approaches (Arthur & Rousseau, 1996; Hall, 2002; Sullivan & Baruch, 2009). Both protean and boundaryless career approaches are characterized by self-reliance in the pursuit of career development and employability. More specifically, protean careers are characterized by “self-direction in the pursuit of psychological success in one’s work” (Hall, 2002, p. 23) and the adherence to personal standards or values when making career choices rather than to the expectations of others (e.g., an employing organization). Boundaryless careers are characterized by psychologically or physically crossing organizational boundaries for work (Sullivan & Arthur, 2006). Although these “non-traditional” career approaches have been discussed in the careers literature for decades, empirical research is just beginning to illuminate their psychological origins and practical implications for individuals and organizations.

To facilitate the rigorous and systematic study of protean and boundaryless careers, Briscoe, Hall, and DeMuth (2006) developed and validated the Protean and Boundaryless Career Attitudes scales (items are listed in the Appendix). The Protean Career Attitudes (PCA) scale is composed of two subscales, that is, Self-Directed and Values-Driven. The Self-Directed (PCSD) subscale assesses the belief that one is “in charge” of his or her own career. The Values-Driven (PDVD) subscale measures whether one tends to follow his or her own values when making career decisions as opposed to organizationally or socially sanctioned values. The Boundaryless Career Attitudes (BCA) scale is also composed of two subscales, that is, Boundaryless Mindset and Organizational Mobility Preference. The Boundaryless Mindset (BCBM) subscale captures one’s desires and inclinations to cross organizational boundaries (e.g., functional units and organizations) to accomplish work, and the Organizational Mobility Preference (BCOMP) subscale assesses whether one favors changing employers over staying with one employer for a longer time.

Briscoe et al.’s (2006) scale development work signifies a crucial step in establishing the consistent measurement of protean and boundaryless career attitudes. Moving forward, we see two reasons why further refinement (or reduction) of these scales may be beneficial. First, the content match between the original scales and the corresponding theoretical constructs is somewhat questionable: Some items included in the measures are fairly repetitive or overly broad in their content coverage. Second, the scales are fairly long (13–14 items in each scale), which may discourage researchers with limited time and resources from using the scales and/or result in respondent fatigue.

To remedy issues of scale validity and length, we conducted three studies to develop the PCA and BCA short forms (PCA-SF and BCA-SF). In the first study, we reduce the number of items in the original PCA and BCA scales to remove redundant or irrelevant content. In the second study, we present two types of validity evidence for the PCA-SF and BCA-SF as assessments of protean and boundaryless career attitudes: the internal factor structures of each shortened scale and the empirical relationships of the short forms with a set of theoretically relevant personality antecedents and work and career outcomes. Moreover, we evaluate the construct validity of the PCA-SF and BCA-SF by comparing these findings with the findings of previous research that used the original PCA and BCA scales. In the third study, we evaluate the measurement equivalence of the PCA-SF and the BCA-SF across U.S. and South Korean working adults, thereby extending the potential use of the short forms to a different language/cultural population. Taken together, these studies aim to facilitate future protean and boundaryless careers research by addressing the psychometric and practical issues associated with the use of the original PCA and BCA scales as well as providing an initial foundation for broadening the scope of protean and boundaryless careers research across national boundaries.

Study 1: Developing PCA-SF and BCA-SF

As a first step in refining Briscoe and colleagues’ PCA and BCA scales, this study shortens the original PCA and BCA by removing irrelevant and/or redundant items, yielding the PCA-SF and BCA-SF. We drew from protean and boundaryless careers research to determine which items were theoretically relevant to the protean and boundaryless career attitudes constructs and therefore, which items should be retained in the PCA-SF and BCA-SF (e.g., Arthur & Rousseau, 1996; Briscoe, Hall, & DeMuth, 2006; Defillippi & Arthur, 1994; Hall, 2002; Hall & Moss, 1998; Inkson, 2006; Mirvis & Hall, 1994; Sullivan & Arthur, 2006).

Method

Procedure and Participants

Using Amazon Mechanical Turk (MTurk),¹ we administered Briscoe et al.’s (2006) measures of PCA and BCA via an anonymous online survey to 370 working adults residing in the United States. On average, participants were 36 years old (SD = 12.5) and had an organizational tenure of 5 years (SD = 6). Sixty-five percent of respondents were female; 77% were White, 8% were Black, 8% were Asian, and 4% were Hispanic. About 34% of participants had a bachelor’s degree, 30% had some college or an associate’s degree, 24% had graduate or professional training, and 10% had a high school diploma. Participants were from a variety of industries including information technology (14%), health care (13%), retail (11%), and financial services (7%).

Results and Discussion

To identify items that were candidates for removal, we conducted two separate sets of exploratory factor analyses (with maximum likelihood estimation and a promax rotation), one for the PCA and one for the BCA. The number of factors for each scale was determined by the number of eigenvalues greater than 1.0 and the number of factors present before the final steep decline of the scree plot; these criteria were consistent with one another. In addition, we evaluated the theoretical interpretability of each solution. For the PCA scale items, a three-factor solution initially emerged (as opposed to the expected two factor solution) that explained 58% of the variance in the PCA items. Upon investigating the factor pattern matrix, we found that Items 2 and 3 of the PCVD subscale split from the other PCVD items, creating two factors that were conceptually indistinguishable from one another. Given that the three-factor structure did not provide new insights into the protean career attitudes construct, we decided to refine the scales to obtain a two-factor solution, as originally theorized by Briscoe et al. (2006). For the BCA items, a two-factor solution initially emerged that explained 65% of the variance in the BCA items; all items loaded on their theoretically expected factors.

We continued conducting two separate factor analyses for the PCA and BCA scales in which we removed items from each scale one at a time based upon a combination of four criteria (Hinkin, 1998; Stanton, Sinar, Balzer, & Smith, 2002): (1) whether items exhibited loadings on more than one factor; (2) whether items demonstrated unacceptably low intercorrelations with items of the same scale (.40 or lower); (3) whether items exhibited unacceptably low communality levels (.40 or lower); and (4) whether item content was overlapping with other items or inconsistent with the construct. We explain our rationale for the removal of each item in the Appendix. Throughout this scale refinement process, we sought to keep the item content consistent with theoretical conceptualizations of protean and boundaryless career attitudes.

For the PCA-SF, 7 items were retained, 4 items representing the PCSD subscale and 3 items representing the PCVD subscale. About 68% of the variance in the items was explained by these two factors. For the BCA-SF, 6 items were retained, 3 representing the BCBM subscale and 3 representing the BCOMP subscale. About 79% of the variance in these items was accounted for by these two factors. Table 1 reports the retained items with factor loadings and the Cronbach’s α reliability coefficients of the shortened scales.

Table 1.

Short-Form Protean and Boundaryless Career Attitudes Scales With Standardized Factor Loadings and Reliability Coefficients.

		Factor
		1	2
Protean Career Attitudes
Self-Directed				α = .83
3	I am responsible for my success or failure in my career.	.74	−.08
4	Where my career is concerned, I am very much “my own person.”	.69	.07
5	Overall, I have a very independent, self-directed career.	.77	−.00
7	Freedom to choose my own career path is one of my most important values.	.81	.03
Values-Driven				α = .77
4	It doesn’t matter much to me how other people evaluate the choices I make in my career.	−.07	.75
5	I navigate my own career, based upon my personal priorities, as opposed to my employer’s priorities.	.08	.62
6	What’s most important to me is how I feel about my career success, not how other people feel.	−.00	.84

Boundaryless Career Attitudes
Boundaryless Mindset				α = .91
4	I like tasks at work that require me to work beyond my own department.	.76	.01
5	I would enjoy working on projects with people from across many organizations.	.96	−.01
6	I have sought opportunities in the past that allow me to work outside the organization.	.92	.01
Organizational Mobility Preference				α = .81
3	I would feel very lost if I couldn’t work for my current organization.^a	.04	.68
4	I like the predictability that comes with working continuously for the same organization.^a	−.07	.84
5	I prefer to stay in a company I am familiar with rather than look for employment elsewhere.^a	.04	.80

^aA reverse scored item.

Further, to evaluate whether the PCA-SF and BCA-SF assess similar constructs as the original measures, we examined the statistical relationships between the short and long form subscales. We found that the short forms and the long forms of the PCA and BCA subscales were very highly correlated, PCSD r(370) = .93, p < .01; PCVD r(370) = .86, p < .01; BCBM r(370) = .95, p < .01; BCOMP r(370) = .96, p < .01, which suggests that the short and long forms are comparable measures of protean and boundaryless career attitudes.

In summary, this study shortened Briscoe et al.’s (2006) original PCA and BCA scales and, in so doing, addressed concerns associated with the validity and length of the scales. Specifically, redundant and tangentially relevant items were removed from the scales, resulting in more efficient assessments of protean and boundaryless career attitudes that can more easily be incorporated into subsequent research.

Study 2: Scale Validation

Study 2 builds upon Study 1 by presenting two types of evidence for the validity of the PCA-SF and BCA-SF as measures of protean and boundaryless career attitudes. First, we evaluate whether the observed internal factor structures of PCA-SF and BCA-SF align with the original theories underlying the protean and boundaryless career attitudes constructs and their corresponding indicators (as opposed to other alternative measurement models) using confirmatory factor analysis. Second, we investigate the empirical relationships of the shortened measures with other variables (i.e., personality antecedents and work and career outcomes), and we evaluate whether PCA-SF and BCA-SF show similar patterns of relationships with these variables as previous research using the full-length scales. To provide a theoretical background for these relationships, we briefly review the literature on protean and boundaryless careers in relation to the variables included in our study: personality antecedents (Big Five personality factors and proactive personality) and work and career outcomes (job satisfaction, organizational commitment, job market cognitions, job search, and career satisfaction).

Personality Antecedents

Previous research has investigated factors that might contribute to nontraditional career mindsets (e.g., Sargent & Domberger, 2007; Segers, Inceoglu, Vloeberghs, Bartram, & Henderickx, 2008), but these studies have overlooked the potential role of personality in facilitating these attitudes. One notable exception is Briscoe et al.’s (2006) work, which found that all four subscales of PCA and BCA (PCSD, PCVD, BCBM, and BCOMP) were positively associated with measures of proactive personality (Bateman & Crant, 1993) and openness to experience (one of the Big Five personality factors; John & Srivastava, 1999; McCrae & Costa, 1987). This study contributes to this literature by linking protean and boundaryless career attitudes with all Big Five personality factors and proactive personality, which may provide insights into the dispositional roots of these attitudes (i.e., who is likely to hold nontraditional career attitudes).

Work and Career Outcomes

Research suggests that workers who hold self-directed (protean) career attitudes often express feelings of attachment to their organizations (i.e., organizational commitment), perceive themselves to be capable of maintaining or acquiring new employment, and may engage in more job search activities (Çakmak-Otluoğlu, 2012; De Vos & Soens, 2008; Waters, Briscoe, Hall, & Wang, 2014). Moreover, self-directed workers are more likely to be satisfied with both their jobs and their careers (Briscoe, Henagan, Burton, & Murphy, 2012; O’Shea, Monaghan, & Ritchie, 2014). In contrast, those who follow their own values when making career choices (i.e., values-driven) are less likely to exhibit these outcomes (Briscoe & Finkelstein, 2009; Çakmak-Otluoğlu, 2012).

Workers with a boundaryless mindset, who prefer psychological mobility, are neither particularly attached to nor detached from their organizations (Çakmak-Otluoğlu, 2012); however, they tend to view themselves as valuable to the external labor market, are more likely to engage in job search, and are more likely to be satisfied with their jobs and careers (Briscoe et al., 2012; McArdle, Waters, Briscoe, & Hall, 2007; Vansteenkiste, Verbruggen, & Sels, 2013; Verbruggen, 2012). Those with a preference for organizational mobility (i.e., physical boundarylessness), on the other hand, are expected to feel less attached to their organizations (i.e., organizational commitment) and are more likely to be dissatisfied with their jobs and careers (Briscoe & Finkelstein, 2009; Çakmak-Otluoğlu, 2012; O’Shea et al., 2014; Verbruggen, 2012).

Method

Procedure and Participants

Similar to Study 1, we collected 311 responses from working U.S. residents to a second anonymous online survey via MTurk; participants were rewarded with a nominal fee (U.S. $3.00). On average, participants were 33 years old (SD = 12) and had an organizational tenure of 4 years (SD = 4.5). Fifty-four percent of respondents were male; 74% were White, 10% were Asian, 7% were Black, and 5% were Hispanic. About 40% of respondents had some college or an associate’s degree, 31% had a bachelor’s degree, 16% had graduate or professional training, and 13% had a high school diploma.

Measures

Unless otherwise noted, participants were asked to rate the extent to which they agreed with statements based upon a 5-point scale (1 = strongly disagree; 5 = strongly agree). To assess protean and boundaryless career attitudes, we used the PCA-SF and BCA-SF developed in Study 1 (see Table 1 for items). We assessed the Big Five personality factors (extraversion, neuroticism, conscientious, agreeableness, and openness to experience) using the 44-item Big Five Inventory (John & Srivastava, 1999). Proactive personality was measured using the 17-item proactive personality scale (Bateman & Crant, 1993), which assesses the propensity to show initiative, take action, and persevere to bring about change. Job satisfaction was assessed using the 8-item Abridged Job-in-General Scale (Russell et al., 2004), which asks participants to rate the extent to which adjectives (e.g., “Good” and “Undesirable”) describe their job. Organizational commitment was assessed using Meyer, Allen, and Smith’s (1993) 6-item measure of affective commitment in which participants rated their emotional attachment to their organizations. Job market cognitions were assessed using the Employment Opportunity Index (Griffeth, Steel, Allen, & Bryan, 2005), a multidimensional measure of one’s perceived ability to acquire an alternative job based upon job market conditions and personal attributes. This 17-item measure has five components: ease of movement, desirability of movement, networking, crystallization of alternatives, and mobility. Job search was assessed using Blau’s (1993) job search behavior measure. Respondents were asked to rate the frequency with which they engaged in job search behaviors (e.g., “Filled out a job application”) over the previous 6-month period on the following scale: 1 = never (0 times), 2 = rarely (1 or 2 times), 3 = occasionally (3–5 times), 4 = frequently (6–9 times), and 5 = very frequently (at least 10 times). Finally, career satisfaction was assessed using Greenhaus, Parasuraman, and Wormley’s (1990) Career Satisfaction Scale, which is composed of 5 items that assess satisfaction with one’s career in general (e.g., I am satisfied with what I have achieved in my career). The Cronbach’s α reliability estimates for these scales are reported in Table 2.

Table 2.

Means, Standard Deviations, Reliability Coefficients, and Intercorrelations of Study 2 Variables.

	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19
1. Self-Directed	3.79	.76	(.80)
2. Values-Driven	3.80	.85	.39**	(.78)
3. Boundaryless Mindset	3.60	1.00	.37**	.20**	(.91)
4. Organizational Mobility Preference	3.00	1.01	−.14*	.04	.11*	(.76)
5. Extraversion	2.94	.80	.32**	.05	.43**	.00	(.87)
6. Agreeableness	3.82	.63	.24**	.14*	.31**	−.03	.35**	(.83)
7. Conscientiousness	3.87	.66	.28**	.29**	.09	−.08	.25**	.43**	(.85)
8. Neuroticism	2.74	.80	−.22**	.01	−.17**	−.04	−.40**	−.38**	−.46**	(.87)
9. Openness	3.69	.63	.30**	.37**	.31**	.14*	.25**	.25**	.20**	−.05	(.84)
10. Proactive Personality	3.57	.64	.48**	.32**	.46**	−.03	.49**	.42**	.46**	−.32**	.53**	(.92)
11. Job Satisfaction	3.56	.92	.40**	.03	.18**	−.30**	.29**	.34**	.32**	−.35**	.05	.27**	(.94)
12. Organizational Commitment	4.11	1.50	.28**	−.02	.11	−.36**	.31**	.27**	.23**	−.24**	−.01	.20**	.67**	(.89)
13. Ease of Movement	3.04	.97	.30**	.06	.18**	−.04	.20**	.07	.21**	−.19**	.06	.34**	.33**	.25**	(.78)
14. Desirability of Movement	3.12	.98	−.05	.10	.14*	.08	−.04	−.04	−.03	.15*	.05	.09	−.34**	−.26**	.20**	(.89)
15. Networking	2.86	1.07	.28**	.07	.31**	−.01	.37**	.16**	.12*	−.18**	.27**	.43**	.26**	.21**	.46**	.16**	(.87)
16. Crystallization of Alternatives	3.09	1.00	.05	.03	.13*	−.00	.14*	−.16**	−.15*	.02	.02	.13*	−.07	−.03	.25**	.34**	.41**	(.72)
17. Mobility	3.16	1.06	.09	.04	−.03	.17**	−.08	−.07	.07	.01	.00	−.04	−.01	.02	.07	.06	−.09	.08	(.70)
18. Job Search	2.45	1.26	.02	.04	.07	.15**	.08	−.16**	−.17**	.16**	.07	.02	−.32**	−.33**	−.16*	.27**	.04	.31**	.05	(.93)
19. Career Satisfaction	3.18	.97	.34**	−.01	.19**	−.25**	.33**	.21**	.19**	−.29**	.07	.38**	.57**	.45**	.38**	−.20**	.42**	.17**	−.09	−.13**	(.91)

*p < .05. **p < .01.

Results

Evaluating Internal Factor Structures

Using Mplus version 5.1, we conducted two sets of confirmatory factor analyses on the PCA-SF and BCA-SF separately. We evaluated model fit using four indices: the chi-square statistic, comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). CFI values of .90 or above indicate acceptable fit and values of .95 or greater indicate good fit; RMSEA and SRMR values of .08 or lower indicate acceptable fit and values of .05 or less indicate good fit (Hu & Bentler, 1999; McDonald & Ho, 2002). Further, we compared the fit of each model using the chi-square difference test.

For both the PCA-SF and the BCA-SF, we compared theoretically expected two-factor models (in which subscales were represented as separate factors) to one-factor models. Compared to the one-factor PCA model (χ ² = 194.89, df = 14; CFI = .76; RMSEA = .20; SRMR = .10), the two-factor model provided a better fit to the data (χ ² = 25.47, df = 13; CFI = .98; RMSEA = .05; SRMR = .04) as evidenced by the chi-square difference test (Δχ ² = 169.42, Δdf = 1, p < .001). Likewise, the two-factor BCA model (χ ² = 7.75, df = 8; CFI = 1.00; RMSEA = .00; SRMR = .02) provided a superior fit to the data compared to the one-factor model (χ ² = 423.54, df = 9; CFI = .75; RMSEA = .28; SRMR = .17), which was also supported by the chi-square difference test (Δχ ² = 415.79, Δdf = 1, p < .001).²

Evaluating Relationships with Other Constructs

To further evaluate the validity of the PCA-SF and BCA-SF in assessing protean and boundaryless career attitudes, we examined their bivariate relationships with personality antecedents (i.e., Big Five and proactive personality) and work and career outcomes (i.e., job satisfaction, organizational commitment, job cognitions, job search, and career satisfaction). Table 2 presents the zero-order (bivariate) correlations of study variables, along with their means, standard deviations, and Cronbach’s α reliability estimates. Moreover, to examine whether the PCA-SF and BCA-SF assess protean and boundaryless career attitudes in a similar manner as the original (long) scales, we used the Fisher’s z test to evaluate whether the relationships between the short forms and the aforementioned study variables were comparable to findings reported in previous research using the original PCA and BCA scales (see Table 3).

Table 3.

Comparisons of Study 2 Findings With Previous Research Results Using Fisher’s z Test.

Previous Study	Subscale	Current Study Correlations	Previous Study Correlations	Fisher’s z test
Previous Study	Subscale	Proactive Personality		Fisher’s z test
Briscoe, et al. (2006), N = 493 (Study 3)	PCSD	.48	.40	1.37
	PCVD	.32	.28	.60
	BCBM	.46	.41	.85
	BCOMP	−.03	.11	−1.11
		Openness to Experience
Briscoe et al. (2006), N = 493 (Study 3)	PCSD	.30	.21	1.32
	PCVD	.37	.18	2.84**
	BCBM	.31	.32	.15
	BCOMP	.14	.16	.28
		Job Satisfaction
O’Shea, et al. (2014), N = 202	PCSD	.40	.19	2.54*
O’Shea, et al. (2014), N = 202	BCOMP	−.30	−.33	−.04
Verbruggen (2012), N = 357	BCBM	.18	.07	1.44
	BCOMP	−.30	−.10	2.68
		Organizational Commitment
Briscoe and Finkelstein (2009), N = 212	PCSD	.28	.06	2.54*
	PCVD	−.02	−.03	−.50
	BCBM	.11	.00	1.23
	BCOMP	−.36	−.33	−.35
Çakmak-Otluoğlu (2012), N = 373	PCSD	.28	.11	2.30*
	PCVD	−.02	−.08	−.78
	BCBM	.11	.06	.65
	BCOMP	−.36	−.44	−1.24
		Job Search
Briscoe, et al. (2012), N = 362	PCSD	.02	.06	−.05
Briscoe, et al. (2012), N = 362	BCBM	.07	.14	−.91
McArdle et al. (2007), N = 416	BCBM	.07	.26	−2.60**
Vansteenskiste et al. (2013), N = 1840	BCBM	.07	.27	−3.36
		Career Satisfaction
Briscoe et al. (2012), N = 362	PCSD	.34	.28	.86
Briscoe et al. (2012), N = 362	BCBM	.19	.45	−3.76**
O’Shea et al. (2014), N = 202	PCSD	.34	.27	.85
O’Shea et al. (2014), N = 202	BCOMP	−.25	−.22	.35
Verbruggen (2012), N = 357	BCBM	.19	.10	1.18
Verbruggen (2012), N = 357	BCOMP	−.25	−.03	2.89**

Note. PCSD = Protean Career Self-Directed; PCVD = Protean Career Values-Driven; BCBM = Boundaryless Career Boundaryless Mindset; BCOMP = Boundaryless Career Organizational Mobility Preference.

*p < .05. **p < .01.

Personality

The short forms of the PCSD, PCVD, and BCBM subscales (PCSD-SF, PCVD-SF, and BCBM-SF, respectively) were positively associated with openness to experience and proactive personality. The short form of the BCOMP subscale (i.e., BCOMP-SF), on the other hand, was only associated with openness. While the magnitude of these findings was largely consistent with previous research (see Table 3), we note that the PCVD-SF exhibited a stronger relationship with the theoretically relevant variable of openness to experience than has been reported in previous research using the original PCVD subscale (Briscoe et al., 2006).

Regarding the relationships between the short forms and the remaining Big Five personality factors, the PCSD-SF was positively related to extraversion, agreeableness, and conscientiousness, and negatively related to neuroticism, whereas the PCVD-SF was positively related to agreeableness and conscientiousness. The BCBM-SF was positively associated with extraversion and agreeableness, and negatively associated with neuroticism; the BCOMP-SF was unrelated to the remaining Big Five personality factors.

Work and Career Outcomes

The PCSD-SF was associated with greater job satisfaction and organizational commitment, whereas the PCVD-SF was unrelated to these work attitudes. The PCSD-SF was also positively related to perceptions of alternative employment opportunities (i.e., ease of movement) and maintaining a network of contacts with access to alternative job information (i.e., networking); the PCVD-SF was unrelated to job market cognitions. Finally, both the PCSD-SF and PCVD-SF were positively related to career satisfaction. In general, these findings were consistent with previous research with two exceptions; the Fisher’s z test revealed that the PCSD-SF showed stronger relationships with theoretically relevant variables of job satisfaction and organizational commitment than the original subscale.

The BCBM-SF was positively related to job satisfaction and unrelated to organizational commitment, whereas the BCOMP-SF was negatively associated with these outcomes. The BCBM-SF was positively related to job market cognitions: those with a boundaryless mindset thought that they could find an alternative job if they were looking (i.e., ease of movement), had a network of contacts to whom they could turn for job leads (i.e., networking), believed that better jobs were available to them (i.e., desirability of movement), and were more likely to have an alternative job offer in hand (i.e., crystallization of alternatives). On the other hand, the BCOMP-SF was only related to mobility, the perception that one could easily leave his or her current organization. Finally, the BCBM-SF was positively associated with career satisfaction, whereas the BCOMP-SF was negatively related to this outcome.

While relationships between the BCA-SF subscales and openness, proactive personality, and organizational commitment were of consistent magnitudes as prior research, relationships between the BCA-SF subscales and job satisfaction, job search, and career satisfaction were only partially consistent with previous research using the original BCA subscales. In one study comparison, the BCBM-SF predicted job search equally as well as the original scale; in two other study comparisons, the short form had a weaker relationship with job search than the long form. In addition, the relationship between the BCBM-SF and career satisfaction was equivalent to the relationship reported in prior research in one study comparison, but weaker in a second study comparison. Finally, in one study comparison, the BCOMP-SF had similar relationships with job satisfaction and career satisfaction as research using the long form, but in another study comparison, the short form exhibited stronger relationships with these variables.

Discussion

These results provide evidence for the validity of the PCA-SF and BCA-SF as assessments of protean and boundaryless career attitudes in two ways. First, the short form items adhere to the PCA and BCA dimensions proposed in previous research (Briscoe et al., 2006). More specifically, compared to a one-factor model, we found that a two-factor model provided superior fit to both the PCA-SF and BCA-SF data, which suggests that the scale items are assessing their respective constructs as theoretically expected. Second, this study also provided evidence that the PCA-SF and BCA-SF relate to personality and work and career constructs as theoretically expected. Further, we found that the magnitudes of the relationships between the shortened PCA and BCA scales and the examined antecedents and outcomes were often similar to the relationships of the original PCA and BCA scales and study variables. However, we note that in some cases, the short forms were more strongly related to theoretically relevant variables than the long forms (e.g., PCSD-SF with job satisfaction and organizational commitment), whereas in other cases, they were more weakly related (e.g., BCBM-SF with job search). It is possible that these somewhat mixed findings are a function of the equivocal results presented in prior research: While some studies report moderate relationships between theoretically relevant constructs, other studies report weak relationships between the same constructs. These differences may also be a consequence of different sampling strategies or operationalizations of study variables employed across studies. Given these possibilities, future research may evaluate relationships of the PCA-SF and BCA-SF scales with theoretically related variables in different samples or with different operationalizations of those constructs and compare these findings with previous research to strengthen inferences that they adequately assess protean and boundaryless career attitudes.

We also note that this study extends prior research by linking protean and boundaryless career attitudes with previously uninvestigated antecedents (i.e., Big Five personality factors) and outcomes (i.e., job market cognitions). By doing so, this study presents initial insights into the tendencies and mindsets of those who hold protean and boundaryless career attitudes. For instance, we found that tendencies to be extraverted, agreeable, and emotionally stable (i.e., not neurotic) were associated with workers who are self-directed and hold boundaryless mindsets; and these same workers are also likely to have positive evaluations of their abilities to acquire alternative employment.

Study 3: Measurement Invariance Between U.S. and Korean Samples

Building upon the prior studies that developed and provided initial validity evidence for the PCA-SF and BCA-SF, the goal of Study 3 is to investigate whether it is appropriate to use the PCA-SF and BCA-SF in a different cultural population, that is, Korean. Before the PCA-SF and BCA-SF are used to assess protean and boundaryless career attitudes in Korean samples, researchers should have evidence that the translated scales assess the same construct as the English scales or evidence of measurement equivalence (or invariance) across groups. Without measurement equivalence evidence, researchers are at risk of drawing erroneous conclusions about the constructs assessed in each group, such as invalid group differences in the mean-levels of a construct or inflated relationships among constructs in one group or another (see Chen, 2008). With this in mind, Study 3 evaluates the measurement equivalence of the PCA-SF and BCA-SF across U.S. and Korean samples to evaluate whether these scales adequately assess protean and boundaryless career attitudes in another cultural context.

Method

Participants and Procedure

Data from 295 U.S. adult workers were collected via MTurk; participants completed the survey for a nominal reward (U.S. $3.00). On average, participants were 33 years old (SD = 11.8) and had an organizational tenure of 4.5 years (SD = 5.5). Fifty-eight percent of respondents were male; 76% were White, 9% were Asian, 8% were Black, and 5% were Hispanic. About 38% of respondents had some college or an associate’s degree, 31% had a bachelor’s degree, 13% had graduate or professional training, and 10% had a high school diploma.

Data from 419 Korean adult workers were collected using an online panel survey firm; participants earned cash points for completing the survey. On average, participants were 37 years old (SD = 10.4) and had an organizational tenure of 9.7 years (SD = 8.0). Sixty-four percent of respondents were female. About 8% had a master degree, 68% had a bachelor’s degree, and about 24% had a high school diploma.

Measures

To assess protean and boundaryless career attitudes in English, we used the PCA-SF and BCA-SF developed in Studies 1 and 2. For the Korean sample, the items were first translated from English to Korean by the third author, who is proficient in both languages; the second author, who is bilingual, compared and examined the two versions. The two authors discussed any discrepancies between the two versions and revised the items until no more modifications were necessary. Throughout the translation process, the authors took into account (subtle, yet important) linguistic and sociocultural differences between the two populations, following the International Test Commission’s Test Adaptation Guidelines (Coyne, 2000).

Analysis and Results

To assess the measurement equivalence of the PCA-SF and BCA-SF across U.S. and Korean groups, we used multigroup confirmatory factor analysis with maximum likelihood estimation in MPlus version 5.1 (French & Finch, 2006; Raju, Laffitte, & Byrne, 2002; Reise, Widaman, & Pugh, 1993; Vandenberg & Lance, 2000). We assessed the fit of each measurement model using the chi-square statistic, the comparative fit index (CFI), the root mean square error of approximation (RMSEA), and the standardized root mean square residual (SRMR). In line with recommended practices (Cheung & Rensvold, 2002), we also compared “nested” models by evaluating the change in the chi-square statistic and CFI values; a change in a CFI value of .01 signifies a substantial difference in fit. Table 4 reports the results of these analyses.

Table 4.

Evaluation of the Measurement Equivalence of the PCA-SF and BCA-SF Across U.S. and Korean Samples.

	χ ²	Δχ ²	df	Δdf	CFI	ΔCFI	RMSEA	SRMR
Protean Career Attitudes
Test of Invariance of Covariance Matrices	344.68		29		.77		.17	.53
Baseline Model (Test of Configural Invariance)	77.04		26		.96		.07	.04
Model 1PC (Test of Metric Invariance)	81.69	4.65	31	5	.96	.00	.07	.04
Model 2PC (Test of Scalar Invariance)	243.51	166.47**	45	19	.86	.10	.11	.13
Boundaryless Career Attitudes
Test of Invariance of Covariance Matrices	224.50		21		.88		.16	.27
Baseline Model (Test of Configural Invariance)	42.83		16		.98		.07	.04
Model 1BC (Test of Metric Invariance)	59.69	16.86**	20	4	.98	−.01	.07	.07
Model 2BC (Test of Partial Metric Invariance—OMP-SF Item 1 allowed to vary)	43.65	0.82	19	3	.99	.01	.06	.04
Model 3BC (Test of Scalar Invariance)	372.5	329.17**	31	15	.80	−.19	.17	.20

Note. CFI = comparative fit index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual; PCA-SF = Protean Career Attitudes Short Form; BCA-SF = Boundaryless Career Attitudes Short Form; OMP-SF = Organizational Mobility Preference Short Form.

**p < .01.

Single-Group CFA

Before testing for measurement equivalence, we first investigated whether a two-factor model adequately modeled the protean career and the boundaryless career data in each sample (U.S. and Korean) separately using single-group CFA (Byrne, Shavelson, & Muthen, 1989). First, we evaluated the PCA-SF in the U.S. sample and the Korean sample, separately. The two-factor model fit the U.S. PCA-SF data well (χ ² = 50.04, df = 13; CFI = .95; RMSEA = .10; SRMR = .04). In the Korean sample, the two-factor model fit the PCA-SF data very well (χ ² = 27.00, df = 13; CFI = .98; RMSEA = .05; SRMR = .03). Second, we evaluated the BCA-SF in the U.S. sample and the Korean samples. The two-factor model fit the BCA-SF data very well in the U.S. sample (χ ² = 15.72, df = 8; CFI = .99; RMSEA = .06; SRMR = .04) and in the Korean sample (χ ² = 27.11, df = 8; CFI = .98; RMSEA = .08; SRMR = .04). As the PCA-SF and BCA-SF adhered to the theoretically expected factor structure in both U.S. and Korean samples, we continued testing for measurement equivalence between the English and Korean versions of both scales.

Invariance of Covariance Matrices

We conducted two omnibus tests of the invariance of the covariance matrices for the PCA-SF and the BCA-SF (separately) across U.S. and Korean samples simultaneously. When covariance matrices are equivalent, groups are considered equivalent on a measure and no further tests of invariance are necessary (Vandenberg & Lance, 2000). However, the test of equivalence of covariance matrices is very strict and often untenable (Byrne et al., 1989). Not surprisingly, we failed to find invariance across matrices for the PCA-SF and the BCA-SF. Therefore, it was necessary to continue subsequent tests of measurement equivalence to identify the source of variance.

Configural Invariance

Next, we tested for configural invariance: an evaluation of whether items load on the same factor across samples or whether the pattern of factor loadings is similar across samples. Conceptually, evidence of configural invariance is support for the assumption that the same construct was assessed in both groups. To conduct this test, we fit the two-factor measurement model for both samples simultaneously (rather than sequentially as in the single group CFA). The configural invariance measurement model serves as the baseline model to which subsequent (more restrictive) tests of measurement equivalence are compared (Vandenberg & Lance, 2000). Thus, configural invariance must be established before conducting further tests of measurement invariance.

For both the PCA-SF and the BCA-SF data, configural invariance between U.S. and Korean groups was established: RMSEA values indicated acceptable fit and the CFI and SRMR values indicated that the models fit the data well (see Table 4). Thus, further tests of measurement equivalence were necessary to identify the source of variance between the two groups. The factor loadings of the PCA-SF and BCA-SF scale items for the U.S. and Korean samples are reported in Tables 5 and 6.

Table 5.

Standardized Factor Loadings, Item and Scale Means, and Reliability Coefficients for the Two-Factor Model of PCA-SF Across U.S. and Korean Samples.

Item	Unites States		Korean
Self-Directed	FL	Item Means	FL	Item Means
I am responsible for my success or failure in my career.	.78	3.50	.66	3.39
Where my career is concerned, I am very much “my own person.”	.78	3.72	.66	3.69
Overall, I have a very independent, self-directed career.	.78	3.86	.66	3.95
Freedom to choose my own career path is one of my most important values.	.63	3.81	.56	3.92
	α = .81	m = 3.73	α = .72	m = 3.73

Value Driven
It doesn’t matter much to me how other people evaluate the choices I make in my career.	.68	3.98	.69	3.62
I navigate my own career, based upon my personal priorities, as opposed to my employer’s priorities.	.70	3.77	.57	3.57
What’s most important to me is how I feel about my career success, now how other people feel.	.79	3.58	.64	3.26
	α = .76	m = 3.78	α = .67	m = 3.48

Note. FL = factor loadings; α = internal consistency (Cronbach’s α reliability) estimate; m = scale mean; PCA-SF = Protean Career Attitudes Short Form.

Table 6.

Standardized Factor Loadings, Item and Scale Means, and Reliability Coefficients for the Two-Factor Model of BCA-SF Across U.S. and Korean Samples.

Item	United States		Korean
Boundaryless Mindset	FL	Item Means	FL	Item Means
I like tasks at work that require me to work beyond my own department.	.84	3.54	.73	2.89
I would enjoy working on projects with people from across many organizations.	.93	3.43	.77	2.72
I have sought opportunities in the past that allow me to work outside the organization.	.89	3.41	.71	3.14
	α = .92	m = 3.44	α = .78	m = 2.94

Organizational Mobility Preference
I would feel very lost if I couldn’t work for my current organization.^a	.70	2.93	.56	3.60
I like the predictability that comes with working continuously for the same organization.^a	.79	3.16	.91	3.31
I prefer to stay in a company I am familiar with rather than look for employment elsewhere.^a	.67	2.96	.78	3.05
	α = .74	m = 3.01	α = .78	m = 3.32

Note. FL = factor loadings; α = internal consistency (Cronbach’s α reliability) estimate; m = scale mean; BCA-SF = Boundaryless Career Attitudes Short Form.

^aA reverse scored item.

Metric Invariance

Next, we tested for metric invariance—whether item factor loadings are invariant (equal) across groups—by constraining factor loadings to be equal across groups. Conceptually, evidence of metric invariance supports the assumption that the items are interpreted in a similar manner across groups. As suggested by Vandenberg (2002), we used exploratory factor analytic techniques (maximum likelihood estimation and a promax rotation) to identify referent indicators for each factor that were invariant across the U.S. and Korean samples (Schmitt & Kuljanin, 2008); the choice of referent indicator was important because referent indicators that vary across groups can bias results when identifying items that function differently across groups (Johnson, Meade, & DuVernet, 2009).

For the PCA-SF data, we found evidence of full metric invariance. The chi-square difference test revealed that there was no significant difference in fit between Model 1PC and the baseline (configural invariance) model (Δχ ² = 4.65, p > .10), and the change in CFI was negligible (ΔCFI = .000). For the BCA-SF data, full metric invariance was not supported. Compared to the baseline model, Model 1BC showed a decrement in fit as evidenced by a significant increase in the chi-square value (Δχ ² = 16.86, df = 4, p < .01) and a decrease in the CFI value (−.007), indicating that the factor loadings of one or more items differ between U.S. and Korean samples.

To identify the variant factor loading and establish partial metric invariance (where some but not all factor loadings are invariant across groups), we conducted a series of measurement invariance tests in which each item was constrained across samples (Cheung & Rensvold, 1999); if constraining an item resulted in a decrement in fit compared to the baseline model, then we concluded that the item with the constrained factor loading was the source of the variance. When Item 1 of the BCOMP-SF was held invariant across groups, there was a significant decrement in fit (Δχ ² = 11.96, df = 1, p < .01), suggesting that this item varied across groups. Upon examining the factor loadings, we found that this item had a stronger factor loading in the U.S. sample (.70) than in the Korean sample (.56). Therefore, we retained a model (Model 2BC) where this parameter was allowed to vary across groups for subsequent tests of measurement invariance.

In sum, across U.S. and Korean samples, we found support for metric invariance for the PCA-SF and partial metric invariance for the BCA-SF. Evidence of full or partial metric invariance is sufficient to proceed to further tests measurement invariance (Vandenberg & Lance, 2000); therefore, we continued to test for scalar invariance.

Scalar Invariance

Scalar invariance is a test of whether item intercepts (i.e., item means) are equal across groups when the mean of the latent factor is zero. Conceptually, support for scalar invariance indicates that respondents from each group answer items similarly. Differences in item intercepts across groups may indicate either bias in how groups respond to items (i.e., measurement variance) or true differences in the mean level of the construct between groups (Vandenberg & Lance, 2000). As a theoretical reason for true group differences in protean or boundaryless career attitudes across U.S. and Korean samples has yet to be proposed, we considered it appropriate to proceed with tests of scalar invariance. Neither the PCA-SF (Model 2PC) nor the BCA-SF (Model 3BC) showed evidence for scalar invariance: when item intercepts were held constant across groups, both models showed a significant decrement in fit compared to the baseline model according to change in the chi-square and CFI values.

To further investigate why scalar invariance was not supported, we compared the item means for the U.S. and Korean groups using a series of t-tests (see Tables 5 and 6 for item means). We found that the Korean sample was less likely than the U.S. sample to agree with PCVD items (t-values ranging from 3.11 for PCVD Item 2 to 6.17 for PCVD Item 1), whereas there was no significant difference between the U.S. and Korean samples’ likelihood of agreeing with the PCSD items (t-values ranging from .26 for PCSD Item 2 to −1.80 for PCSD Item 4). In addition, we found that the Korean sample was less likely to agree with the BCBM items (t-values ranging from 3.51 for BCBM Item 3 to 9.49 for BCBM Item 2) and more likely to disagree with BCOMP Item 1, t(733) = −9.44, p < .01 (when not reverse coded).

Discussion

This study provides evidence of the measurement invariance of the PCA-SF and BCA-SF across U.S. and Korean samples, suggesting that it is appropriate to use these scales to assess protean and boundaryless career attitudes in Korean samples. Specifically, full metric invariance was established for the PCA-SF, while partial metric invariance was established for the BCA-SF. Subsequently, we discuss these findings and their practical implications.

Evidence for metric invariance has two implications for the use of the PCA-SF and BCA-SF. First, based upon this evidence, one can infer that respondents from the U.S. and Korea perceive a similar psychological construct or hold a similar frame of reference when interpreting items (i.e., configural invariance). Second, it suggests that each item is being interpreted similarly across the U.S. and Korean samples (i.e., metric invariance). Based upon this study, the PCA-SF has both of these attributes. The BCA-SF, on the other hand, falls just short; U.S. and Korean respondents interpret items similarly with the exception of Item 1 of the BCOMP-SF subscale: “I would feel very lost if I couldn’t work for my current organization” [reverse coded]. This item was more strongly related to the latent BCOMP factor for U.S. respondents than Korean respondents, and Korean respondents were more likely to disagree with the item (when it was not reverse coded). It may be the case that, compared to U.S. respondents, Korean respondents may construe themselves as belonging to multiple social groups such that the loss of membership in one social group (e.g., organizational member) is compensated for by membership in other groups (e.g., family, church, and community). Although differences between samples exist, evidence of partial metric invariance is sufficient for using the BCA-SF in Korean samples (Byrne et al., 1989; Reise et al., 1993).

Additionally, we note that neither the PCA-SF nor the BCA-SF showed evidence of scalar invariance. As mentioned earlier, we tested for scalar invariance because researchers have yet to propose a theoretical rationale for why U.S. and Korean workers may approach their careers differently. However, the lack of scalar invariance in this study raises the question of whether there are sociocultural factors that influence career approaches in different countries (e.g., work values); on the other hand, noninvariant intercepts across U.S. and Korean groups may indicate measurement artifacts (Chen, 2008). As such, this study raises some issues for future research in the area of careers, particularly how careers may be manifested differently across national cultures or whether individuals from different cultures respond to items differently, in general (e.g., Tay, Woo, Klafehn, & Chiu, 2010). Finally, based upon this study, researchers may consider using these scales in organizational and careers research not only in the United States, but also in Korea. Thus, this study contributes to careers research by providing an additional resource for careers researchers interested in investigating career approaches in Korea.

Conclusion

In an effort to provide a more efficient assessment of protean and boundaryless career attitudes, the three studies presented here developed and offered a variety of validity evidence for the PCA-SF and BCA-SF in both U.S. and Korean populations. We developed the PCA-SF and BCA-SF and provided validity evidence for their use by confirming their internal factor structures and evaluating whether they exhibit theoretically expected relationships with personality antecedents and work and career outcomes. Moreover, we statistically compared the current findings with findings from previous research using the full length measures, which revealed that the PCA-SF and BCA-SF relate to personality antecedents and work and career outcomes in a similar (but not the same) manner as the full-length measures. Finally, we extended the use of the PCA-SF and BCA-SF to a new cultural context (Korean), enabling researchers to more confidently use these scales for future cross-cultural research. Taken together, these findings suggest that the PCA-SF and BCA-SF are useful assessments of protean and boundaryless career attitudes in both U.S. and Korean samples.

Footnotes

Appendix

Acknowledgments

Part of this research was supported by the Research Grant of Kwangwoon University in 2013.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Arthur

M. B.

Rousseau

D. M.

(1996). The boundaryless career: A new employment principle for a new organizational era. Oxford, UK: Oxford University Press.

Bateman

T. S.

Crant

J. M.

(1993). The proactive component of organizational behavior: A measure and correlates. Journal of Organizational Behavior, 14, 103–118. doi:10.1002/job.4030140202

Blau

(1993). Further exploring the relationship between job search and voluntary individual turnover. Personnel Psychology, 46, 313–330. doi:10.1111/j.1744-6570.1993.tb00876.x

Briscoe

J. P.

Finkelstein

L. M.

(2009). The “new career” and organizational commitment: Do boundaryless and protean attitudes make a difference? Career Development International, 14, 242–260. doi:10.1108/13620430910966424

Briscoe

J. P.

Hall

D. T.

DeMuth

R. L. F.

(2006). Protean and boundaryless careers: An empirical exploration. Journal of Vocational Behavior, 69, 30–47. doi:10.1016/j.jvb.2005.09.003

Briscoe

J. P.

Henagan

S. C.

Burton

J. P.

Murphy

W. M.

(2012). Coping with an insecure employment environment: The differing roles of protean and boundaryless career orientations. Journal of Vocational Behavior, 80, 308–316. doi:10.1016/j.jvb.2011.12.008

Buhrmester

Kwang

Gosling

S. D.

(2011). Amazon’s mechanical turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6, 3–5. doi:10.1177/1745691610393980

Byrne

B. M.

Shavelson

R. J.

Muthen

(1989). The equivalence of factor covariance and mean structures: The issue of partial measurement invariance. Psychological Bulletin, 105, 456–466. doi:10.1037/0033-2909.105.3.456

Çakmak-Otluoğlu

K. O.

(2012). Protean and boundaryless career attitudes and organizational commitment: The effects of perceived supervisor support. Journal of Vocational Behavior, 80, 638–646. doi:10.1016/j.jvb.2012.03.001

10.

Chen

F. F.

(2008). What if we compare chopsticks with forks? The impact of making inappropriate comparisons in cross-cultural research. Journal of Personality and Social Psychology, 95, 1005–1018. doi:10.1037/a0013193

11.

Cheung

G. W.

Rensvold

R. B.

(1999). Testing factorial invariance across groups: A reconceptualization and proposed new method. Journal of Management, 25, 1–27. doi:10.1177/014920639902500101

12.

Cheung

G. W.

Rensvold

R. B.

(2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 9, 233–255. doi:10.1207/S15328007SEM0902_5

13.

Coyne

(2000). International Test Commission Test Adaptation Guidelines. International Test Commission. Retrieved from http://www.intestcom.org/test_adaptation.htm

14.

Defillippi

R. J.

Arthur

M. B.

(1994). The boundaryless career: A competency-based perspective. Journal of Organizational Behavior, 15, 307–324. doi:10.1002/job.4030150403

15.

De Vos

Soens

(2008). Protean attitude and career success: The mediating role of self-management. Journal of Vocational Behavior, 73, 449–456. doi:10.1016/j.jvb.2008.08.007

16.

French

B. F.

Finch

W. H.

(2006). Confirmatory factor analytic procedures for the determination of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 13, 378–402. doi:10.1207/s15328007sem1303_3

17.

Goodman

J. K.

Cryder

C. E.

Cheema

(2013). Data collection in a flat world: The strengths and weaknesses of Mechanical Turk samples. Journal of Behavioral Decision Making, 26(3), 213–224.

18.

Greenhaus

J. H.

Parasuraman

Wormley

W. M.

(1990). Effects of race on organizational experiences, job performance evaluations, and career outcomes. Academy of management Journal, 33, 64–86. doi:10.2307/256352

19.

Griffeth

R. W.

Steel

R. P.

Allen

D. G.

Bryan

(2005). The development of a multidimensional measure of job market cognitions: The employment opportunity index (EOI). Journal of Applied Psychology, 90, 335–349. doi:10.1037/0021-9010.90.2.335

20.

Hall

D. T.

(2002). Careers in and out of organizations. Thousand Oaks, CA: Sage Publications.

21.

Hall

D. T.

Moss

J. E.

(1998). The new protean career contract: Helping organizations and employees adapt. Organizational Dynamics, 26, 22–37. doi:10.1016/S0090-2616(98)90012-2

22.

Hinkin

T. R.

(1998). A brief tutorial on the development of measures for use in survey questionnaires. Organizational Research Methods, 1, 104–121. doi:10.1177/109442819800100106

23.

Bentler

P. M.

(1999). Cutoff criteria for fit index in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1–55. doi:10.1080/10705519909540118

24.

Inkson

(2006). Protean and boundaryless careers as metaphors. Journal of Vocational Behavior, 68, 48–63. doi:10.1016/j.jvb.2005.09.004

25.

John

O. P.

Srivastava

(1999). The big five trait taxonomy: History, measurement, and theoretical perspectives. In Pervin

L. A.

John

O. P.

(Eds.), Handbook of personality: Theory and research (2nd ed., pp. 102–139). New York, NY: Guilford Press.

26.

Johnson

E. C.

Meade

A. W.

DuVernet

A. M.

(2009). The role of reference indicators in tests of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 16, 642–657. doi:10.1080/10705510903206014

27.

McArdle

Waters

Briscoe

J. P.

Hall

D. T.

(2007). Employability during unemployment: Adaptability, career identity and human and social capital. Journal of Vocational Behavior, 71, 247–264. doi:10.1016/j.jvb.2007.06.003

28.

McCrae

R. R.

Costa

P. T.

Jr (1987). Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology, 52, 81–90. doi:10.1037/0022-3514.52.1.81

29.

McDonald

R. P.

M.-H. R.

(2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7, 64–82. doi:10.1037//1082-989X.7.1.64

30.

Meyer

J. P.

Allen

N. J.

Smith

C. A.

(1993). Commitment to organizations and occupations: Extension and test of a three-component conceptualization. Journal of Applied Psychology, 78, 538–551. doi:10.1037/0021-9010.78.4.538

31.

Mirvis

P. H.

Hall

D. T.

(1994). Psychological success and the boundaryless career. Journal of Organizational Behavior, 15, 365–380. doi:10.1002/job.4030150406

32.

O’Shea

Monaghan

Ritchie

T. D.

(2014). Career attitudes and satisfaction during recession. Journal of Managerial Psychology, 29, 226–245. doi:10.1108/JMP-02-2013-0061

33.

Paolacci

Chandler

Ipeirotis

P. G.

(2010). Running experiments on amazon mechanical turk. Judgment and Decision Making, 5, 411–419.

34.

Raju

N. S.

Laffitte

L. J.

Byrne

B. M.

(2002). Measurement equivalence: A comparison of methods based on confirmatory factor analysis and item response theory. Journal of Applied Psychology, 87, 517–529. doi:10.1037//0021-9010.87.3.517

35.

Reise

S. P.

Widaman

K. F.

Pugh

R. H.

(1993). Confirmatory factor analysis and item response theory: Two approaches for exploring measurement invariance. Psychological Bulletin, 114, 552–566. doi:10.1037/0033-2909.114.3.552

36.

Russell

S. S.

Spitzmuller

Lin

L. F.

Stanton

J. M.

Smith

P. C.

Ironson

G. H.

(2004). Shorter can also be better: The abridged job in general scale. Educational and Psychological Measurement, 64, 878–893. doi:10.1177/0013164404264841

37.

Sargent

L. D.

Domberger

S. R.

(2007). Exploring the development of a protean career orientation: Values and image violations. Career Development International, 12, 545–564. doi:10.1108/13620430710822010

38.

Schmitt

Kuljanin

(2008). Measurement invariance: Review of practice and implications. Human Resource Management Review, 18, 210–222. doi:10.1016/j.hrmr.2008.03.003

39.

Segers

Inceoglu

Vloeberghs

Bartram

Henderickx

(2008). Boundaryless careers: A study on potential motivators. Journal of Vocational Behavior, 73, 212–230. doi:10.1016/j.jvb.2008.05.001

40.

Stanton

J. M.

Sinar

E. F.

Balzer

W. K.

Smith

P. C.

(2002). Issues and strategies for reducing the length of self-report scales. Personnel Psychology, 55, 167–194. doi:10.1111/j.1744-6570.2002.tb00108.x

41.

Sullivan

S. E.

Arthur

M. B.

(2006). The evolution of the boundaryless career concept: Examining physical and psychological mobility. Journal of Vocational Behavior, 69, 19–29. doi:10.1016/j.jvb.2005.09.001

42.

Sullivan

S. E.

Baruch

(2009). Advances in career theory and research: A critical review and agenda for future exploration. Journal of Management, 35, 1542–1571. doi:10.1177/0149206309350082

43.

Tay

Woo

S. E.

Klafehn

Chiu

C.-Y.

(2010). Conceptualizing and measuring culture: Problems and solutions. In Walford

Tucker

Viswanathan

(Eds.), The SAGE Handbook of Measurement (pp. 177–203). London, UK: Sage.

44.

Vandenberg

R. J.

(2002). Toward a further understanding of and improvement in measurement invariance methods and procedures. Organizational Research Methods, 5, 139–158. doi:10.1177/1094428102005002001

45.

Vandenberg

R. J.

Lance

C. E.

(2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3, 4–70. doi:10.1177/109442810031002

46.

Vansteenkiste

Verbruggen

Sels

(2013). Being unemployed in the boundaryless career era: Does psychological mobility pay off? Journal of Vocational Behavior, 82, 135–143. doi:10.1016/j.jvb.2012.11.007

47.

Verbruggen

(2012). Psychological mobility and career success in the ‘New’ career climate. Journal of Vocational Behavior, 81, 289–297. doi:10.1016/j.jvb.2011.10.010

48.

Waters

Briscoe

J. P.

Hall

D. T.

Wang

(2014). Protean career attitudes during unemployment and reemployment: A longitudinal perspective. Journal of Vocational Behavior, 84, 405–419. doi:10.1016/j.jvb.2014.03.003