A Psychometric Evaluation of the Career Decision Self-Efficacy Scale–Short Form With Turkish University Students

Abstract

The current study aimed to examine the psychometric properties of the Career Decision Self-Efficacy scale–Short Form (CDSES-SF) in a sample of 695 Turkish university students. Accordingly, internal reliability, test–retest reliability, convergent validity, and factor structure of the CDSES-SF were examined. The results indicated high internal and test–retest reliability for total scores. The evidence for the convergent validity was provided by a relatively high correlation between career decision-making self-efficacy and general self-efficacy. To estimate the factor structure of the scale, Betz, Klein, and Taylor’s five-factor model of the CDSES-SF, along with a number of empirically derived measurement models of the CDSES-SF were tested via confirmatory factor analyses. Only Gaudron’s four-factor model exhibited good model fit for the Turkish sample. The findings of this study are discussed in accordance with previous studies and the current cultural context of Turkey.

Keywords

Career Decision Self-Efficacy scale reliability validity confirmatory factor analysis Turkish university students

Making a career decision is one of the significant and inevitable tasks of life. In this regard, a considerable body of literature has focused on factors that play important role on individuals’ career decisions. Career decision-making self-efficacy has become one of the widely investigated construct since its introduction with the work of Taylor and Betz (1983) on the development of Career Decision Self-Efficacy scale (CDSES).

The CDSES is primarily founded on Bandura’s (1977, 1997) concept of self-efficacy expectations, which suggests that individuals’ beliefs about their ability to successfully perform a given task are the major mediators of behavior and behavior change and Hackett and Betz’s (1981) original suggestion. In addition, the five career choice competencies postulated in Crites’ (1961, 1978) model of career maturity were used as a framework by the authors (Taylor & Betz, 1983) of the scale to decide on required skills in career decision making.

The CDSES (Taylor & Betz, 1983) aims to measure an individual’s degree of belief that he or she can successfully complete tasks necessary to making career decisions. The CDSES composed of 50 items designed to measure five domains of career decision-making self-efficacy on the basis of Crites’ (1978) model of career maturity. These five domains (subscales) were accurate self-appraisal, gathering occupational information, goal selection, making plans for the future, and problem solving. Respondents are asked to rate their confidence about performing each task on a 10-point Likert-type scale ranging from 0 (no confidence) to 9 (complete confidence).

Due to the length of the original scale, a short form of the scale was developed by eliminating 5 of the 10 items from each of the five subscales (Betz, Klein, & Taylor, 1996). Thus, the Career Decision Self-Efficacy scale–Short Form (CDSES-SF) contains 25 items taken from the original CDSES. It is again rated on a 10-point Likert-type scale ranging from 0 (no confidence) to 9 (complete confidence). Betz et al. (1996) reported the internal consistency reliability of the short form ranged from .73 (self-appraisal) to .83 (goal selection) for the subscales and .94 for the total score. The result of the factor analysis of the short form revealed a five-factor solution; however, items did not load on their respective factor consistent with proposed model.

Recently, Betz, Hammond, and Multon (2005) examined the internal consistency reliability and criterion-related validity of 5-level response continuum ranging from 1 (no confidence at all) to 5 (complete confidence). Betz et al. (2005) reported coefficient α values ranging from .78 to .87 for the five subscales and .95 for the total score. Criterion-related validity correlations with career indecision and vocational identity were similar for the two response continua. Accordingly, the authors concluded that 5-level response continuum, like 10-level response continuum provided reliable and valid scores.

More recently, Hartman and Betz (2007) made a minor revision on the CDSES-SF. The authors replaced the original item “Find information in the library about occupations you are interested in” with a new item “Use the internet to find information about occupations that interest you” to keep up with technological changes. Hartman and Betz (2007) reported item-total correlations for the new and original items as .54 and .50, respectively, and Cronbach’s α for the CDSES-SF as .96. Based on these results, the authors added the “Internet” item and deleted the “library” item in the last version of 25-item scale.

Currently, the CDSES is one of the most frequently used and popular instruments in the field of career psychology and counseling (O’Brien, 2003). CDSES has received a good deal of research attention. Thus, there is extensive evidence for the reliability and validity of both the original and the short forms of the scale. For example, Nilsson, Schmidt, and Meek (2002) reviewed 56 studies including published journal articles and dissertations and found that the α reliability coefficients of the original and short forms of the scale ranged from .83 to .97 for total scores. Additionally, Luzzo (1993) and Mau (2000) both computed a test–retest reliability coefficient of .83 based on a 6-week and a 4-week interval, respectively. Gaudron (2011) provided a test–retest reliability coefficient of .81.

A substantial body of empirical research has provided criterion related and construct validity of the CDSES and the CDSES-SF. Both forms of the scale have been found to be significantly correlated with career indecision (Betz et al., 1996; Betz & Luzzo, 1996; Taylor & Popma, 1990), vocational identity (Gushue, Scanlan-Kolone, Pantzer, & Clark, 2006; Munson & Savickas, 1998), career commitment (Betz & Sterling, 1993; Chung, 2002), career maturity (Luzzo, 1993), patterns of career choice (Gianakos, 1999), career exploratory behaviors (Blustein, 1989; Gushue et al., 2006), general self-efficacy and global self-esteem (Betz & Klein, 1996), academic and career outcome expectations, and exploratory intentions (Betz & Klein Voyten, 1997). Additionally, some researchers investigated demographic variables related to career decision self-efficacy. Among demographics, gender is the most frequently investigated variable. Studies that evaluated gender differences have generally reported no difference on career decision-making self-efficacy (e.g., Betz et al., 1996; Hampton, 2006; Luzzo, 1993).

After the development of the CDSES (Taylor & Betz, 1983), various factor analytic studies were conducted for both the original and the short forms of the scale (e.g., Betz, Hammond, & Multon, 2005; Creed, Patton, & Watson, 2002; Peterson & delMas, 1998). In these studies, several factorial analyses were employed including the principal component analysis by varimax rotation as in the original study of Taylor and Betz (1983), principal axis factoring with direct oblimin rotation (e.g., Creed et al., 2002), and confirmatory factor analysis (CFA) with maximum likelihood estimation (e.g., Miller, Sendrowitz Roy, Brown, Thomas, & McDaniel, 2009; Watson, Brand, Stead, & Ellis, 2001). However, most of them did not confirm the proposed five-factor model of the scale (e.g., Gaudron, 2011; Hampton, 2005; Peterson & delMas, 1998). Therefore, some authors suggested that the CDSES-SF could be regarded as a generalized career decision-making self-efficacy measure (i.e., measuring a general factor; Creed et al., 2002; Taylor & Popma, 1990; Watson et al., 2001).

Creed, Patton, and Watson (2002) identified two different three-factor 23-item solutions for Australian and South African high school students. Labels of the factors were information gathering, decision making, and problem solving. Likewise, Hampton (2005) identified a three-factor solution with 13 items for Chinese college students and labeled the factors same as Creed et al. (2002). In another study, Hampton (2006) investigated four-factor (empirically derived) and five-factor (theoretically derived) solutions for Chinese high school students. Since the distribution of the items was complex, Hampton (2006) decided to use CDSES-SF as a general measure of career decision-making behavior. Recently, Chaney, Hammond, Betz, and Multon (2007) proposed a four-factor solution for African American college students. The authors suggested that the first factor that contained 10 items from all five theoretical subscales was a general measure of career self-efficacy. The second factor was related to self-appraisal and determining ideal job or career. The third factor seemed to focus on knowing career options and the job search process. The fourth factor included 2 problem-solving items. In a more recent study, Gaudron (2011) identified a four-factor solution with 18 items for French university students. These factors were labeled as goal selection, problem solving, information gathering, and goal pursuit management.

To date, a few studies were conducted to examine the five-factor model of the CDSES-SF by Betz et al. (1996) with CFA. Overall, the results seem contradictory across the studies. For instance, Watson, Brand, Stead, and Ellis (2001) found the five-factor model to demonstrate an inadequate fit to the data, χ²(265, n = 364) = 807.53, comparative fit index (CFI) = .83, root mean square error of approximation (RMSEA) = .075, with South African college students. Hampton (2005) also reported unfavorable fit to the data, χ²(265, n = 256) = 784.95, CFI = .79, RMSEA = .08, with Chinese college students. In similar vein, Gaudron (2011) indicated that the five-factor model to display an inadequate fit to the data, χ²(265, n = 650) = 1,050.78, CFI = .78, RMSEA = .075, with French university students. On the other hand, Miller, Sendrowitz Roy, Brown, Thomas, and McDaniel (2009) provided a good fit to the data with Asian American university students, χ²(265, n = 267) = 521.57, CFI = .97, RMSEA = .06 and with European American university students, χ²(265, n = 239) = 593.71, CFI = .96, RMSEA = .07. As a result of the high correlations among the factors, Miller et al. (2009) decided to test a one-factor model (i.e., a measurement model that fixed each of the 25 CDSES-SF items to load on one factor). The authors, however, concluded that the five-factor model showed a significantly better model fit than the one-factor model.

Although studies that use the CDSES scales have been conducted mostly in the United States, there are a few studies with a number of diverse samples outside the United States (Abdalla, 1995; Creed et al., 2002; Gati, Osipow, & Givon, 1995; Gaudron, 2011; Hampton, 2005, 2006; Mau, 2000; Rowland, 2004; Watson et al., 2001). Especially, four studies assessed the reliability and the validity of the CDSES-SF (Creed et al., 2002; Gaudron, 2011; Hampton, 2005; Watson et al., 2001). Aforementioned, none of them confirmed the proposed five-factor model of Betz et al. (1996).

These inconsistent results might indicate that the originally proposed five-factor model may not be the best representation of the underlying factor solution (Miller et al., 2009) or possibly can be attributed to cross-cultural differences on the career development and decision-making processes (Lindley, 2006). In response to calls for further studies on the equivalence of the CDSES-SF across other U.S. racial/ethnic groups and non-English speaking countries (Creed et al., 2002; Miller et al., 2009), the aim of the current study was to examine the reliability and the validity of the CDSES-SF in a sample of Turkish university students.

Method

Participants

Participants of the current study were 695 university students enrolled in five different faculties at Middle East Technical University in Turkey. Of them, 326 (46.9%) were female, 368 (52.9%) were male, and 1 (0.1%) did not indicate gender. Participants of the study were 212 (30.5%) freshmen, 155 (22.3%) sophomores, 164 (23.6%) juniors, 163 (23.5%) seniors, and 1 (0.1%) did not report any class. Their age ranged from 17 to 27, with a mean of 21.39 (SD = 1.5).

Measures

CDSES-SF (Betz et al., 1996)

The scale was developed to measure “an individual’s degree of belief that he or she can successfully complete tasks necessary to making career decisions” (Betz et al., 1996, p. 48). The short form consisted of 25 items representing Crites’ (1978) five career choice competencies in his model of career maturity. Accordingly, self-appraisal, gathering occupational information, goal selection, making plans for the future, and problem solving are the subscales of the CDSES-SF. Items are rated on a 5-point Likert-type scale ranging from 1 (no confidence at all) to 5 (complete confidence). Thus, the possible total scores changed between 25 and 125 with higher scores on CDSES-SF indicating greater levels of career decision self-efficacy (Betz et al., 2005). Betz et al. (1996) reported the internal consistency reliability of the short form ranged from .73 (self-appraisal) to .83 (goal selection) for the subscales and .94 for the total score.

The CDSES-SF was administered in Turkish. The steps that were followed throughout the translation process were as follows. First, the scale was given to four experts (two advanced doctoral level counseling students and two English language experts) independently for translation into Turkish. Second, the translations made by four experts were compared and best translation for each item was picked. Third, the Turkish and original English versions of the scale were given to two professors of psychological counseling and guidance and a professor of measurement and evaluation to evaluate the correctness, clarity, wording of the items, and cultural relevancy of the Turkish-translated version of the scale. According to the feedback received by these three faculty members, some minor changes were made on the scale. Fourth, a separate focus group was conducted to check the understandability, clarity, and cultural appropriateness of the items of the Turkish version of the CDSES-SF with university students. Finally, few changes suggested by the students regarding the wording of the items in the scale were made.

To gather basic demographics of the participants, the researcher developed a short demographic information form that was placed before the CDSES-SF. The form included questions regarding gender, age, class, faculty, and department.

Generalized Self-Efficacy Scale (GSES; Schwarzer & Jerusalem, 1995)

The scale was used to measure general perceived self-efficacy. It was originally developed by Jerusalem and Schwarzer in 1981, in Germany, and in the meantime has been translated to various languages (Schwarzer & Jerusalem, 1995). It is a 10-item scale with a response format ranging from 1 (not at all true) to 4 (completely true) and yielding a total score between 10 and 40. Higher scores indicate stronger belief in self-efficacy. The scale was adapted to Turkish by Yesilay, Schwarzer, and Jerusalem (1996), and the internal reliability coefficient was reported as .92 for the Turkish sample.

Procedure

The data of the study were collected during 2009–2010 spring semester. Approval for conducting the study was obtained from the Human Subjects Ethics Committee of the researcher’s institution. All data were collected in the classroom settings by the researcher with the permission of the course instructors. Students voluntarily participated in this study. No identifying information was requested from the participants such as name, surname, and student id number to ensure the confidentiality and anonymity of the participants except test–retest applications.

Results

Descriptive statistics as means and standard deviations along with coefficient α values were calculated for the total score of CDSES-SF and for its five 5-item subscales by means of SPSS 17. The mean scores represent the average response over the 25 items of the total scale (the cumulative score over all 25 items is divided by 25), and over the 5 items of the subscales (the sum of the response scores for the subscale items are divided by 5; Betz et al., 2005). For the total CDSES-SF, scores ranged from 1.60 to 5.00, with a mean score of 3.50 (SD = .59). Table 1 provides the means and standard deviations of the subscales that ranged from 3.22 (problem solving) to 3.65 (self-appraisal). In addition, a t-test analysis was performed to examine the possible gender difference on career decision self-efficacy. Accordingly, result of the analysis revealed no significant difference between female (M = 3.49, SD = .61), male (M = 3.53, SD = .54), and students, t(692) = 1.01, p = .32.

Table 1.

Means, Standard Deviations, and α Coefficients of the CDSES-SF.

CDSES-SF	M	SD	α Coefficient
Total	3.51	.57	.92
Self-appraisal	3.66	.66	.74
Occupational information	3.60	.61	.61
Goal selection	3.56	.72	.81
Planning	3.51	.70	.72
Problem solving	3.22	.73	.68

Note. CDSES-SF = Career Decision Self-Efficacy scale–Short Form; M = mean; SD = standard deviation.

In order to check the reliability of the CDSES-SF, the internal consistency and test–retest methods were used. The internal consistency reliability of the scale was .92 for the total score that can be regarded as high. The reliability of the subscales ranged from .61 (occupational information) to .81 (goal selection). Additionally, the test–retest reliability of the scale was calculated in a sample of 52 university students based on a 2-week interval. The reliability coefficient (stability coefficient) was .91 for the total score between these two administrations.

With the intention of providing evidence for the validity of the CDSES-SF, convergent validity was examined by calculating a Pearson correlation coefficient between the total score of CDSES-SF and total score of GSES (Schwarzer & Jerusalem, 1995) in a group of 125 (41 male, 84 female) university students. Accordingly, the correlation between the total CDSES-SF score and GSES scale was found to be .65 (p ≤ .01). As expected, a significant positive correlation was observed between two different self-efficacy measures in a way that higher levels of career decision making self-efficacy are associated with higher levels of general self-efficacy.

Moreover, a series of CFA was conducted to determine the factor structure of the CDSES-SF for Turkish university students. First, the originally proposed five-factor model of the CDSES-SF by Betz et al. (1996) was tested. Subsequently, five alternative models identified in the published literature were examined with CFA. Several recommended fit indices were used to assess model fit (Hu & Bentler, 1998; Kline, 2005; McDonald & Ho, 2002). χ² is the most commonly used fit indices to assess how well a model fits the observed data (Weston & Gore, 2006). A nonsignificant χ² suggests that the proposed model is consistent with the observed data (Weston & Gore, 2006). However, χ² is sensitive to sample size and it yields significant values with large sample size (Schumacker & Lomax, 2004). As suggested by several authors (e.g., McDonald & Ho, 2002), the ratio of χ² to its degrees of freedom (χ²/df) were used to deal with this limitation. The common recommended cutoff value for this ratio is 3 (Kline, 2005). RMSEA is an index that corrects for a model’s complexity (Weston & Gore, 2006). For the RMSEA, a value of 0 suggests an exact fit, a value of .06 or less indicates good fit and a value above .08 indicates poor fit (Hu & Bentler, 1998). The CFI compares the improvement of the fit of the researcher’s model over a null model (Weston & Gore, 2006). The CFI ranges from 0 to1 and values closer to 1 representing good fit. Specifically, a value greater than .95 was suggested (Hu & Bentler, 1998). The goodness-of-fit index (GFI) is a measure of the relative amount of variance and covariance in a sample covariance matrix (Byrne, 2010). Values of GFI range from 0 to 1, with .90 or greater indicating a good fit (Byrne, 2010; Kline, 2005).

The statistical software AMOS 18 (Arbuckle, 2009) was used to perform CFA with maximum likelihood estimation. First, Betz et al.’s five-factor model of CDSES-SF that fixed each of the 25 items to load only on their respective factor in a way consistent with Betz et al.’s (1996) model was tested. While the values of RMSEA and GFI seemed to be acceptable, values of CFI and the χ²/df indicated inadequate fit. The fit indices of the aforementioned measurement models are shown in Table 2. Similar to Miller et al. (2009), an examination of factor covariances revealed high-correlation coefficients among all five factors, ranged from .78 (occupational information and problem solving) to .98 (occupational information and planning). Accordingly, the one-factor model (Miller et al., 2009), which fixed each of the 25 items to load on one latent factor was tested. However, results of the CFA indicated that the one-factor model (Miller et al., 2009) of CDSES-SF did not provide acceptable fit to the data.

Table 2.

Fit Indices of the Measurement Models.

Model	χ²	df	p	χ²/df	RMSEA	CFI	GFI
25-Item scale
Betz et al. (1996) five-factor model	913.96	265	.001	3.4	.059	.89	.90
Miller et al. (2010) one-factor model	1,038.99	275	.001	3.8	.063	.87	.88
3-Factor model
Creed et al. (2002) 23-item scale (A)	798.83	227	.001	3.5	.060	.89	.90
Creed et al. (2002) 23-item scale (SA)	783.98	227	.001	3.5	.059	.89	.91
Hampton (2005) 13-item scale	333.91	62	.001	5.4	.079	.86	.93
4-Factor model
Gaudron (2011) 18-item scale	338.48	129	.001	2.6	.048	.94	.95

Note. CFI = comparative fit index; df = degrees of freedom; GFI = goodness-of-fit index; RMSEA = root mean square error of approximation; (A) = Australian sample; (SA) = South African sample.

To estimate the best model for Turkish university students, other previous factor analytic studies including Creed et al.’s (2002) three-factor model for Australian and South African, Hampton’s (2005) three-factor model, and Gaudron’s (2011) four-factor model were also tested with CFA. Creed et al.’s (2002) three-factor model based on Australian sample, which composed of 23 items of CDSES-SF to load information gathering, decision making, and problem-solving factors, exhibited poor model fit. In addition, the three-factor model of Creed et al.’s (2002) based on South African sample, which consisted of 23 items of CDSES-SF to load on information gathering, decision making, and problem-solving factors, indicated an inadequate fit. Hampton’s (2005) three-factor model, which fixed 13 CDSES-SF items to load on decision making, information gathering, and problem solving latent factors, did not fit the data properly (see Table 2, second section). Finally, Gaudron’s (2011) four-factor model, which fixed 18 CDSES-SF items to load on goal selection, problem solving, information gathering, and goal pursuit management latent factors, was tested and demonstrated good fit to the data (see Table 2, last section).

According to the results of CFAs, the best model in this sample was Gaudron’s (2011) four-factor model with 18 items of CDSES-SF. In this model, the first factor (goal selection) composed of 5 items including 3 goal selection items (Items 6, 11, and 16), and 2 accurate self-appraisal items (Items 9 and 22). The second factor (problem solving) contained 3 items from problem-solving subscales (Items 13, 17, and 25). The third factor (information gathering) consisted of 5 items including 4 occupational information items (Items 1, 15, 16, and 23), and 1 planning item (Item 21). The last factor (goal pursuit management) formed 5 items including 2 problem-solving items (Items 6 and 11), 2 planning items (Items 9 and 22), and 1 accurate self-appraisal item (Item 18). This four-factor model accounted for 38.6%, 45.7%, 10.3%, and 47.8% of the variance in goal selection (GS), problem solving (PS), information gathering (IG), and goal pursuit management (GPM) items, respectively.

Means, standard deviations, and values of coefficient α values were calculated for the four subscales and for the total 18-item CDSES-SF. As presented in Table 3, the reliability of the total score was .88, which can be considered high. The reliabilities of the subscales scores ranged between .64 and .77. The correlations among the four factors of the 18-item CDSES-SF ranged from moderate to high (goal selection–problem solving = .73, problem solving–information gathering = .67, information gathering–goal pursuit management = .73, goal selection–information gathering = .82, problem solving–goal pursuit management = .66, goal selection–goal pursuit management = .80). In addition, all of the estimated model parameters (n = 42) were statistically significant (p < .05), and they ranged from .34 to .75 (see Table 4).

Table 3.

Means, Standard Deviations, and α Coefficients of the Four-Factor 18-Item Scale.

Four-factor 18-item scale (Gaudron, 2011)	M	SD	α
Total	3.49	.57	.88
Goal selection (5 items)	3.58	.70	.77
Problem solving (3 items)	3.14	.88	.64
Information gathering (5 items)	3.70	.63	.67
Goal pursuit management (5 items)	3.42	.72	.75

Note. M = mean; SD = standard deviation.

Table 4.

Parameter Estimates for the 18-Item CDSES-SF for Turkish University Students.

Items		Parameter estimates
Factor 1 (goal selection)
Q9	Determine what your ideal job would be	.67
Q16	Make a career decision and then not worry about whether it was right or wrong	.60
Q11	Choose a career that will fit your preferred lifestyle	.65
Q6	Select one occupation from a list of potential occupations you are considering	.70
Q22	Define the type of lifestyle you would like to live	.54
Factor 2 (problem solving)
Q13	Change majors if you did not like your first choice	.52
Q17	Change occupations if you are not satisfied with the one you enter	.64
Q25	Identify some reasonable major or career alternatives if you are unable to get your first choice	.66
Factor 3 (information gathering)
Q23	Find information about graduate or professional schools	.58
Q15	Find out about the average yearly earnings of people in an occupation	.50
Q21	Identify employers, firms, and institutions relevant to your career possibilities	.75
Q19	Talk with a person already employed in a field you are interested in	.52
Q1	Use the Internet to find information about occupations that interest you	.34
Factor 4 (goal pursuit management)
Q8	Persistently work at your major or career goal even when you get frustrated	.63
Q7	Determine the steps you need to take to successfully complete your chosen major	.63
Q3	Make a plan of your goals for the next 5 years	.58
Q4	Determine the steps to take if you are having academic trouble with an aspect of your chosen major	.63
Q18	Figure out what you are and are not ready to sacrifice to achieve your career goals	.58

Note. CDSES-SF = Career Decision Self-Efficacy scale–Short Form. The question numbers relate to the original five subscales from the CDSES-SF (Betz et al., 1996). Q1, Q10, Q15, Q19, and Q23 = items for occupational information subscale; Q5, Q9, Q14, Q18, and Q22 = self-appraisal; Q2, Q6, Q11, Q16, and Q20 = goal selection; Q3, Q7, Q12, Q21, and Q24 = planning; and Q4, Q8, Q13, Q17, and Q25 = problem solving.

Discussion

The purpose of the study was to examine the reliability and validity of the CDSES-SF among Turkish university students. Consistent with previous studies, the result of this study supported a high α reliability coefficient for the total scale (.92) and moderate to high α reliability coefficients (.61 to .81) for the five subscales (e.g., Betz & Klein Voyten, 1997; Creed et al., 2002; Hampton, 2005). In addition, the test–retest reliability coefficient (r = .91) was higher than those reported in the past studies (Gaudron, 2011; Luzzo, 1993; Mau, 2000). This may be a result of short retest interval than those previously conducted studies.

Similar to previous studies, no significant gender differences were found on career decision self-efficacy (Betz et al., 1996; Chung, 2002; Creed et al., 2002; Gaudron, 2011). However, Mau (2000) found that among Taiwanese undergraduates, women had lower career decision self-efficacy than men. Thus, as suggested by Lindley (2006), the interaction between gender and culture is important to consider with respect to career decision self-efficacy.

The convergent validity of the CDSES-SF was provided by a strong relationship (r = .65, p < .001) between the total scores of the CDSES-SF and GSES (Schwarzer & Jerusalem, 1995). Using another GSES (Sherer et al., 1982), for example, Betz and Klein (1996) found correlation coefficients of .59 for males and .50 for females. Hampton (2006) also reported a moderate correlation (r = .35) between CDSES-SF and GSES (Sherer et al., 1982).

On the other hand, the results of this study failed to support the hypothesized five-factor structure of the CDSES-SF. Previous factor analytic studies have also largely failed to support the five-factor structure of the CDSES-SF (e.g., Betz et al., 1996; Betz & Luzzo, 1996; Chaney, Hammond, Betz, & Multon, 2007; Peterson & delMas, 1998). In addition, this finding was similar to other international studies (Gaudron, 2011; Hampton, 2005; Watson et al., 2001). This finding, however, was inconsistent with Miller et al.’s (2009) findings obtained in the groups of Asian American and European American university students. The authors claimed that the difference between studies that used American versus non-American samples might reflect the fact that the behaviors associated with career decision self-efficacy may vary across cultures (Miller et al., 2009).

Since Taylor and Betz (1983) initial work that introduced a measure of career decision-making self-efficacy, CDSES has gained considerable research attention. Especially, psychometric properties of the CDSES have been examined by several researchers. Almost all studies investigated factor structure of the CDSES-SF failed to support Betz et al.’s five-factor model, with the considerable exception of Miller et al.’s (2009) study. In accordance with that many alternative models with one to four factors have been proposed. Subsequently, the five empirically derived alternative measurement models (Creed et al., 2002; Gaudron, 2011; Hampton, 2005; Miller et al., 2009) were tested in the present study. Accordingly, Miller et al.’s (2009) one-factor model, Creed et al.’s (2002) two different three-factor model for Australian and South African, Hampton’s (2005) three-factor model, and Gaudron’s (2011) four-factor model were tested with CFA, respectively.

Only Gaudron’s (2011) four-factor structure demonstrated a good fit to the data among the tested alternative measurement models. As in Gaudron’s study, the findings of this study confirmed that the CDSES-SF reflected four factors of goal selection, problem solving, information gathering, and goal pursuit management. Regardless of the different composition of the all factors, Gaudron (2011) labeled the first three factors as in past studies (Chaney et al., 2007; Creed et al., 2002; Hampton, 2005). Since the items of the fourth factor appeared to be related one’s ability and persistence to organize, plan, and manage one’s workload toward achieving a career goal, this factor was labeled as goal pursuit management (Gaudron, 2011). Gaudron (2011) reported the following values of coefficient α for four subscales: goal selection (.69), problem solving (.73), information gathering (.67), and goal pursuit management (.69). The total score (18 items) value was reported as .83.

In the present study, reliability levels of the 18-item CDSES-SF for the total score and four subscales were, except problem solving and information gathering subscales, higher than those obtained from French university students. The reliability coefficient of the problem solving subscale was found lower than Gaudron’s (2011) reported value; however, information gathering subscale had the same reliability coefficient in both studies. Further, all fit indices of the four-factor model gathered from the Turkish sample (χ²/df = 2.6, RMSEA = .048, CFI = .94, GFI = .95) were also found to be similar to those obtained from French sample (χ²/df = 2.8, RMSEA = .054, CFI = .91, GFI = .94).

Even French and Turkish university students seem to have different cultural background; similarity of the results may be explained by the changing cultural context of Turkey. Since the 1980s, Turkey has been undergoing rapid economic and social changes parallel to the worldwide trends toward liberalization and globalization (Karakitapoğlu-Aygün & Imamoğlu, 2002). Thus, Turkey has been living through a transition period between Eastern and Western attitudes, values, and lifestyles. The rapid change in institutions, values, attitudes, and many more aspects increasingly reflects such Western values as independence, autonomy, and competition (Mocan-Aydın, 2000). In such a transition period, traditional values may exist side by side with new individualistic values. Several studies conducted in Turkey also provide evidence for such an integration of collectivistic traditional values with individualistic attitudes and values (e.g., Çileli, 2000; Göregenli, 1997; Karakitapoğlu-Aygün & Imamoğlu, 2002). Therefore, considering the changing cultural context of Turkey, confirmation of the French version of CDSES-SF by the Turkish sample may not be a surprise.

Although the present study verified the one of the empirically derived measurement models of the CDSES-SF validated multidimensionality of the scale, the model does not reflect the all five domains of career choice competence proposed by Crites (1961). For example, the first factor of Gaudron’s (2011) four-factor model which is goal selection was a combination of the 2 items of self-appraisal and the 3 items of the goal selection factors of the CDSES-SF. This finding may be considered as a further evidence for the suggestion of Creed et al. (2002) that the self-appraisal and goal setting components may not be adequately reflected in the 25-item CDSES-SF when used with different cultural populations. Thus, some improvements may be required for the CDSES-SF scale, specifically for self-appraisal competency.

On the other hand, the findings of this study (high internal consistency, high test–retest reliability, evidence of convergent validity, and high inter-subscale correlations) indicated that the total score of CDSES-SF can be used as a generalized measure of career decision-making self-efficacy as recommended by several authors (e.g., Creed et al., 2002; Peterson & delMas, 1998; Taylor & Popma, 1990; Watson et al., 2001) with Turkish university students.

These findings should be considered within the limitations of the study. Participants of this study were comprised of students from one of the most high-ranking, prestigious, and competitive state universities in Turkey. Hence, obtained findings can only be generalized to the similar populations. Even if the sample of the current study represented all faculties and classes, it did not rely on one of the random sampling that limits the generalizability of the findings. It would be useful for future testing of reliability and validity of the CDSES-SF to include more diverse samples recruited from different age groups and different type of universities including state and private from different regions of Turkey. Consequently, the findings of this study need to be cross validated.

Footnotes

Authors’ Note

This article is partly based on author's doctoral dissertation.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the State Planning Organization as a part of author's doctoral dissertation. Grant no: BAP-08-11DPT-2002K120510.

References

Abdalla

I. A.

(1995). Sex, sex-role self-concepts and career decision-making self-efficacy among Arab students. Social Behavior and Personality: An International Journal, 23, 389–402.

Arbuckle

J. L.

(2009). AMOS 18 user’s guide. Chicago, IL: SPSS Inc.

Bandura

(1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215.

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: W.H. Freeman.

Betz

N. E.

Hammond

M. S.

Multon

K. D.

(2005). Reliability and validity of five-level response continua for the Career Decision Self-Efficacy Scale. Journal of Career Assessment, 13, 131–149.

Betz

N. E.

Klein

K. L.

(1996). Relationships among measures of career self-efficacy, generalized self-efficacy, and global self-esteem. Journal of Career Assessment, 4, 285–298.

Betz

N. E.

Klein Voyten

(1997). Efficacy and outcome expectations influence career exploration and decidedness. The Career Development Quarterly, 46, 179–189.

Betz

N. E.

Klein

K. L.

Taylor

K. M.

(1996). Evaluation of a short form of the career decision-making self-efficacy scale. Journal of Career Assessment, 4, 47–57.

Betz

N. E.

Luzzo

D. A.

(1996). Career assessment and the career decision-making self efficacy scale. Journal of Career Assessment, 4, 313–328.

10.

Betz

N. E.

Sterling

(1993). Criterion-related and construct validity of fear of commitment. Journal of Career Assessment, 1, 21–34.

11.

Blustein

D. L.

(1989). The role of the goal instability and career self-efficacy in the career exploration process. Journal of Vocational Behavior, 35, 194–203.

12.

Byrne

B. M.

(2010). Structural equation modeling with AMOS: Basic concepts, applications, and programming (2nd ed.). Madison, NY: Routledge.

13.

Chaney

Hammond

M. S.

Betz

N. E.

Multon

K. D.

(2007). The reliability and factor structure of the Career Decision Self-Efficacy Scale-SF with African Americans. Journal of Career Assessment, 15, 194–205.

14.

Chung

Y. B.

(2002). Career decision-making self-efficacy and career commitment: Gender and ethnic differences among college students. Journal of Career Development, 28, 277–284.

15.

Creed

P. A.

Patton

Watson

M. B.

(2002). Cross-cultural equivalence of the career decision-making self-efficacy scale-short form: An Australian and South African comparison. Journal of Career Assessment, 10, 327–342.

16.

Crites

J. O.

(1961). A model for the measurement of vocational maturity. Journal of Counseling Psychology, 8, 255–259.

17.

Crites

J. O.

(1978). Career maturity inventory. Monterey, CA: CTB/McGraw-Hill.

18.

Çileli

(2000). Change in value orientations of Turkish youth from 1989 to 1995. The Journal of Psychology, 134, 297–305.

19.

Gati

Osipow

S. H.

Givon

(1995). Gender differences in career decision making: The content and structure of preferences. Journal of Counseling Psychology, 42, 204–216.

20.

Gaudron

J.-P.

(2011). A psychometric evaluation of the Career Decision Self-Efficacy Scale Short-Form among French university students. Journal of Career Assessment, 19, 420–430.

21.

Gianakos

(1999). Patterns of career choice and career decision-making self-efficacy. Journal of Vocational Behavior, 54, 244–258.

22.

Göregenli

(1997). Individualist and collectivist tendencies in a Turkish sample. Journal of Cross-Cultural Psychology, 28, 787–794.

23.

Gushue

G. V.

Scanlan-Kolone

R. L.

Pantzer

K. M.

Clark

C. P.

(2006). The relationship of career decision-making self-efficacy, vocational identity, and career exploration behavior in African American high school students. Journal of Career Development, 33, 19–28.

24.

Hackett

Betz

N. E.

(1981). A self-efficacy approach to the career development of women. Journal of Vocational Behavior, 18, 326–339.

25.

Hampton

N. Z.

(2005). Testing for the structure of the career decision self-efficacy scale-short form among Chinese college students. Journal of Career Assessment, 13, 98–113.

26.

Hampton

N. Z.

(2006). A psychometric evaluation of the career decision self-efficacy scale-short form in Chinese high school students. Journal of Career Development, 33, 142–155.

27.

Hartman

R. O.

Betz

N. E.

(2007). The five-factor model and career self-efficacy: General and domain specific relationships. Journal of Career Assessment, 15, 145–161.

28.

Bentler

P. M.

(1998). Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods, 3, 424–453.

29.

Karakitapoğlu-Aygün

Imamoğlu

E. O.

(2002). Value dimensions of Turkish adults and university students. The Journal of Social Psychology, 142, 333–351.

30.

Kline

R. B.

(2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford.

31.

Lindley

L. D.

(2006). The paradox of self-efficacy: Research with diverse populations. Journal of Career Assessment, 14, 143–160.

32.

Luzzo

D. A.

(1993). Reliability and validity testing of the career decision-making self efficacy scale. Measurement and Evaluation in Counseling and Development, 26, 137–142.

33.

Mau

W. C.

(2000). Cultural differences in career decision-making styles and self-efficacy. Journal of Vocational Behavior, 57, 365–378.

34.

McDonald

R. P.

M.-H. R.

(2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7, 64–82.

35.

Miller

M. J.

Sendrowitz Roy

Brown

S. D.

Thomas

McDaniel

(2009). A confirmatory test of the factor structure of the short form of the career decision self-efficacy scale. Journal of Career Assessment, 17, 507–519.

36.

Mocan-Aydın

(2000). Western models of counseling and psychotherapy within Turkey: Crossing cultural boundaries. The Counseling Psychologist, 28, 281–298.

37.

Munson

W. W.

Savickas

M. L.

(1998). Relation between leisure and career development of college students. Journal of Vocational Behavior, 53, 243–253.

38.

Nilsson

J. E.

Schmidt

C. K.

Meek

W. D.

(2002). Reliability generalization: An examination of the Career Decision-Making Self-Efficacy Scale. Educational and Psychological Measurement, 62, 647–658.

39.

O’Brien

K. M.

(2003). Measuring career self-efficacy promoting confidence and happiness at work. In Lopez

S. J.

Snyder

C. R.

(Eds.), Positive psychological assessment: A handbook of models and measures (pp. 109–126). Washington, DC: American Psychological Association.

40.

Peterson

S. L.

delMas

R. C.

(1998). The component structure of career decision-making self-efficacy for underprepared college students. Journal of Career Development, 24, 209–225.

41.

Rowland

(2004). Career decision-making skills of high school students in the Bahamas. Journal of Career Development, 31, 1–13.

42.

Schumacker

R. E.

Lomax

R. G.

(2004). A beginner’s guide to structural equation modeling. Mahwah, NJ: Lawrence Erlbaum Associates.

43.

Schwarzer

Jerusalem

(1995). Generalized Self-Efficacy scale. In Weinman

Wright

Johnston

(Eds.), Measures in health psychology: A user’s portfolio. Causal and control beliefs (pp. 35–37). Windsor, England: NFER-Nelson.

44.

Sherer

Maddux

J. E.

Mercandante

Prentice-Dunn

Jacobs

Rogers

R. W.

(1982). The self-efficacy scale: Construction and validation. Psychological Reports, 51, 663–671.

45.

Taylor

K. M.

Betz

N. E.

(1983). Applications of self-efficacy theory to the understanding and treatment of career indecision. Journal of Vocational Behavior, 22, 63–81.

46.

Taylor

K. M.

Popma

(1990). Construct validity of the career decision-making self efficacy scale and the relationship of CDMSE to vocational indecision. Journal of Vocational Behavior, 37, 17–31.

47.

Watson

M. B.

Brand

H. J.

Stead

G. B.

Ellis

R. R.

(2001). Confirmatory factor analysis of the career decision-making self-efficacy scale among South African university students. Journal of Industrial Psychology, 27, 43–46.

48.

Weston

Gore

P. A.

(2006). A brief guide to structural equation modeling. The Counseling Psychologist, 34, 719–751.

49.

Yesilay

Schwarzer

Jerusalem

(1996). Turkish adaptation of the general perceived self efficacy scale. Retrieved from http://userpage.fuberlin.de/∼health/turk.htm