Validation of a Four-Factor Model of Career Indecision

Abstract

Two studies were designed to explore whether a meta-analytically derived four-factor model of career indecision (Brown & Rector, 2008) could be replicated at the primary and secondary data levels. In the first study, an initial pool of 167 items was written based on 35 different instruments whose scores had loaded saliently on at least one factor in the Brown and Rector meta-analysis. These items were then administered to a sample of undergraduate college students and the resultant inter-item correlation matrix was subjected to principal axis factoring with oblique rotations. A four-factor solution was uncovered that resembled the four-factor meta-analytically derived solution but with a few theoretically and practically interesting exceptions. A second study used two existing correlation matrices derived from Gati and colleagues’ cognitive and personality/emotional models of career indecision. Exploratory factor analyses of these matrices revealed that the current four-factor model could, in part, be uncovered from these matrices as well. The theoretical and counseling implications of the results are discussed and future research directions are articulated.

Keywords

career indecision chronic indecisiveness choice anxiety lack of readiness interpersonal conflict

Two meta-analyses of the career intervention outcome literature have converged on two inescapable findings; namely, that (a) career interventions for choice-making difficulties are demonstrably effective but (b) the magnitude of their effects is modest (i.e., clients receiving some form of career intervention achieve about a third of a standard deviation better outcome than persons who receive no formal career help; Brown & Ryan Krane, 2000; Whiston, Brecheisen, & Stephens, 2003). Several suggestions have also been provided about how the effectiveness of career interventions can be improved; some of these predated and anticipated the meta-analytic results (see Brown & McPartland, 2005; Miller & Brown, 2005; Whiston & Rahardja, 2008). For example, various writers have observed that the sources of clients’ decision-making difficulties are diverse (Crites, 1969; Holland & Holland, 1977; Salmone, 1982; Tyler, 1969; Williamson, 1965), interventions should be targeted to the sources of clients’ difficulties (Savickas, 1989), and this matching process would be facilitated, and outcomes improved, by clearly derived and empirically supported vocational problem diagnostic systems (Rounds & Tinsley, 1984; Savickas, 1989).

These earlier writings seemed to have stimulated a large body of research exploring correlates of career indecision, factor analyzing a small subset of possible correlates, and developing instruments to measure various aspects of career indecision. Brown and Rector (2008) identified over 50 variables that had been explored as possible correlates of indecision and at least five factor analytic studies that yielded from three to eight possible underlying sources of career decision-making difficulties. Extant instruments also provide scales to measure an equally diverse set of indecision difficulties, including chronic indecisiveness, career information needs, career choice anxiety, self-clarity, work role salience, lack of motivation, career myths, dysfunctional career thoughts, vocational identity, approach-approach conflict, internal and external barriers, and interpersonal conflicts.

Unfortunately, only a modicum of attention seems to have been devoted to integrating this literature in a way that would yield meaningful taxonomies of career choice-making difficulties. Gati and his colleagues have developed two such systems over the past 14 years—one based on a decision theory approach (Gati, Krausz, & Osipow, 1996) and the other on personality and emotional factors that might underlie career choice difficulties (Saka, Gati, & Kelly, 2008). The former system distinguishes between difficulties that may arise prior to decision making (i.e., lack of readiness due to low motivation, general indecisiveness, and dysfunctional beliefs) and those that occur as a person is in the process of deciding (i.e., lack of information about the process, self, and occupations; inconsistent information due to the receipt of unreliable information or internal and external conflicts). The latter system posits that personality and emotional factors interfere with career decision-making via three avenues—(a) pessimistic views about the process of decision making, the world of work, and personal control; (b) anxiety about the process, uncertainty, choice making, and outcomes; and (c) self and identity factors associated with generalized anxiety, self-esteem, uncrystallized vocational identity, and interpersonal conflicts. Each system has yielded a measure that can be used in research as well as in counseling and vocational guidance (Amir, Gati, & Kleiman, 2008; Gati et al., 1996; Gati, Osipow, Krausz, & Saka, 2000; Saka et al., 2008).

Brown and Rector (2008), more recently, took a different approach to identifying underlying sources of indecision, reasoning that variables that had been found to be related to career indecision in previous research represented excellent starting places to identify higher order latent variables that might account for substantial covariation among these measured variables. If a few higher order constructs could be uncovered via factor analysis, these might provide a comprehensive, theoretically meaningful, and clinically useful taxonomy of career decision-making difficulties. Thus, rather than taking a top-down rational/theoretic approach as did Gati and colleagues, Brown and Rector (2008) took a bottom-up, data-driven approach by identifying 28 published correlation matrices that included collectively at least one measure of variables that had been found in the prior literature to be related to career indecision. They then factored each of these matrices via principal axis factoring with oblique rotations, engaged in an iterative procedure to identify common factors, and meta-analytically combined factor loadings to arrive at a loading for each variable on each factor.

The result was a four-factor solution that is illustrated in Table 1 . The first factor was defined primarily by (a) high levels of anxiety (both trait and state), depressive affect, and trait neuroticism; (b) low levels of self-esteem, psychological hardiness, and general problem solving confidence; (c) a tendency to focus on and fear what will go wrong with decisions (fear of commitment), engage in avoidant coping efforts, and rely on others when making important decisions (dependent decision-making style); and (d) beliefs that life is under the control of chance, powerful others, or other external factors (external locus of control). The first factor was also marked by measures of chronic indecisiveness that asked people directly about the chronicity and generality of decision-making difficulties. Brown and Rector (2008) labeled this a chronic indecisiveness/negative affectivity factor because the variables that loaded saliently on it were consistent with prior literature on chronic indecisiveness (e.g., Salmone, 1982) and negative affectivity (Tellegen, 1985; Watson & Clark, 1989).

Table 1.

Factor Loading Matrix From Brown and Rector (2008)

	Factors
Manifest Variables	I	II	III	IV
Trait Anxiety^a (k = 12)	.90	.09	.08	.06
Hardiness (k = 2)	−.88	–	–	−.04
Depression (k = 4)	.85	–	–	.14
State Anxiety^b (k = 12)	.76	.10	.11	.11
Fear of Commitment (k = 6)	.69	–	–	.30
Self-Esteem (k = 7)	−.67	–	−.03	−.07
Indecisiveness^c (k = 3)	.63	.05	.06	.26
Trait Neuroticism (k = 1)	.61	–	–	.13
Self-Criticism (k = 2)	.60	–	–	–
Dependent Decision-Making Style (k = 1)	.52	–	–	.02
External Locus of Control^d (k = 7)	.48	–	.15	.06
Problem Solving Confidence (k = 2)	−.48	–	–	−.12
Positive (Behavioral) Coping Style (k = 4)	−.48	–	–	–
Finding Additional Information (k = 4)	.01	.91	.07	.09
Lack of Occupational Information (k = 9)	.07	.75	.05	.12
Lack of Self Information (k = 7)	.15	.66	.22	.33
Lack of Information on the Process (k = 4)	–	.64	.06	.23
Approach-Approach Conflict (k = 2)	.05	.53	.31	–
External Barriers (k = 3)	.17	.20	.68	–
Situational Constraints (k = 1)	.08	.13	.62	.28
Interpersonal Conflict (k = 6)	.10	.05	.52	.04
Unreliable Information (k = 4)	.01	.41	.50	.09
CDS Indecision (k = 8)	.05	.17	.18	.88
Vocational Exploration and Commitment (k = 2)	–	–	.09	−.88
Career Decision-Making Self-Efficacy (k = 5)	–	–	−.09	−.85
Vocational Identity (k = 4)	−.12	−.21	−.03	−.84
Rational Decision-Making Style (k = 1)	−.02	–	–	−.76
Lack of Motivation (k = 4)	–	.12	.14	.71
Lack of Self-Clarity (k = 8)	.10	.14	.11	.70
Goal Instability (k = 3)	–	–	.08	.67
Intuitive Decision-Making Style (k = 1)	.07	–	–	.67
Career Maturity Attitudes (k = 2)	−.05	−.29	–	−.66
Dysfunctional Thinking/Career Myths (k = 5)	.09	.01	.06	.62
Achieved Ego Identity Status (k = 2)	–	–	–	−.54
Trait Conscientiousness (k = 1)	−.11	–	–	−.46
Choice/Work Importance (k = 4)	–	.16	–	−.44
Approach-Avoidance Style (k = 1)	.13	–	–	.35

Note: Empty cells indicate that the row variable was never included with other markers of the column factor. Salient loadings of .30 or higher are in boldface. k = number of matrices that included a measure of this manifest variable. Factor I = Indecisiveness/Trait Negative Affect, Factor II = Information Deficits, Factor III = Interpersonal Conflicts and Barriers, and Factor IV = Lack of Readiness

^a Includes measures of trait and manifest anxiety.

^b Includes measures of state, career-choice, and social anxiety.

^c Includes CFI and CDP Indecisiveness scales.

^d Includes measures of external, powerful others, and chance locus of control. Reprinted with permission of the authors.

The second factor clearly reflected a lack of information latent dimension with the only other variable showing a salient loading on this factor being a measure of approach-approach conflict. Brown and Rector (2008) labeled this a need for information factor and suggested that the salient loading of approach–approach conflict reflects the possibility that many people with information deficits may not be seeking out additional options but rather need additional information to decide between a couple of good options.

The third factor seemed to reflect an external barrier/interpersonal conflict latent dimension (with measures of external barriers and conflict with significant others having salient loadings), while the fourth factor appeared to be a bit more complex. It was marked saliently by measures of identity diffusion, lack of self-clarity, and low career decision-making self-efficacy beliefs. It was also defined by measures of immature career attitudes, unstable career goals, a lack of motivation to make and commit to a career choice, and low conscientiousness. High scorers on this factor were also less likely to be in an ego-achieved identity status. Brown and Rector (2008) suggested that the pattern of loadings was consistent with developmental conceptions of a lack of readiness to make a career decision (e.g., Peterson, Sampson, Lenz, & Reardon, 2002; Phillips, 1992). Thus, it may be that “high scorers” on this factor may not have yet developed the attitudes, self-knowledge, goal setting, and decision-making skills (and confidence) to make or commit to a vocational choice.

Although the results of the Brown and Rector (2008) meta-analysis offer a promising, comprehensive, and data-driven conception of sources of career decision-making difficulties, they require replication before their robustness can be established. The primary goal of this report was to provide further insight into the Brown and Rector (2008) four-factor model of career indecision. To accomplish this objective, we first (Study 1) created a 167-item inventory that comprehensively covered at the item level all variables that loaded saliently on at least one factor in the Brown and Rector (2008) meta-analysis. We then administered the measure to a sample of undergraduate students and subjected the resultant inter-item correlation matrix to a principal factor analysis with oblique rotations to explore whether the four- factor structure could be replicated using original rather than meta-analytic data.

Our second approach (Study 2) was to use published correlation matrices of Gati and colleagues’ alternative cognitive and personality/emotional models of career indecision to explore whether a similar four-factor model could be extracted from these data sets. Should we find that the four-factor model of career indecision can be uncovered via primary and secondary analyses as well as meta-analytically, this would provide strong evidence that the four-factor model represents an empirically and conceptually reasonable representation of the major sources of difficulty that people (in this case college undergraduates) may experience in their attempts to arrive at a career choice.

Study 1

The primary purpose of this study was to explore factor analytically whether Brown and Rector’s (2008) four-factor model of career indecision provides a reasonable representation of the major sources of career indecision among college students. A 167-item inventory was developed, administered to a representative sample of college students, and factor analyzed via exploratory factor analysis.

Method

Instrument Development

Table 1 displays the 37 variables measured by 35 different instruments that loaded saliently on at least one of the four factors in Brown and Rector’s (2008) meta-analytically derived factor analysis. In order to ensure comprehensive coverage of all relevant variables, we used the 35 instruments to create items for the questionnaire, termed the Career Indecision Profile-167 (CIP-167), used in this study. A list of instruments along with variables measured by each is available from the first author upon request. We studied the test manuals (when available) and items to write an initial set of 221 new items that conceptually covered all major variables (facets) associated with Brown and Rector’s (2008) four-factor model. After eliminating highly redundant items, we created a 167-item instrument. The final set of items was also written to ensure adequate factor and facet saturation: each of the 37 saliently loading variables (facets) identified in the Brown and Rector (2008) factor analysis was represented by 4-6 items. The CIP-167 also contained a page to collect demographic information (e.g., age, year in school, gender, and race/ethnicity) on each participant as well as questions about the participants’ current level of career decidedness (1 = very undecided, 6 = very decided) and the importance of making a career decision now (1= very unimportant, 6 = very important).

Participants and Procedures

The CIP-167 was administered in classrooms to 183 undergraduates from two Midwestern universities. The study was approved by the Institutional Review Boards of both universities and all participants signed, and turned in, informed consent forms before completing the CIP. In order to ensure that an adequate range of scores would be obtained, the inventory was administered to students enrolled in one-semester career development courses taken by undecided students as well as in psychology, education, research methodology, and statistics courses at the two universities. Inventories completed by eight respondents contained greater than 5% (k = 8) missing items and were eliminated from subsequent analyses. Missing values on the remaining inventories (those with fewer than 8 missing values) were imputed via mean substitution. Missing values on these inventories ranged from 1 (n = 34) to six (n = 1). A total of 114 (62%) of the initial 183 inventories contained no missing values.

The final sample was, therefore, composed of 175 participants with a mean age of 23.91 years (SD = 8.17, range 17-57) and mean level of career decidedness of 4.37 (SD = 1.37, range = 1-6) and decision importance of 5.02 (SD = 1.33, range = 1–6). Women comprised 77.1% (n = 135) and men comprised 22.9% (n = 40) of the sample. The majority of the sample was Caucasian (n = 116; 63.4%). The remainder of the sample included 33 (18.0%) African Americans, 10 (5.5%) Mexican Americans or others of Hispanic descent, 7 (3.8%) multiracial, 3 (1.6%) Asian Americans, and 1 (0.5%) Native American. A total of 3 (1.6%) participants indicated their race/ethnicity to be “other,” while 2 (1.1%) failed to answer this demographic question.

Although a total sample size of 175 may seem inadequate for a factor analysis of 167 measured variables, data on sample size requirements in exploratory factor analysis suggest that sample size requirements hinge much less on the number of measured variables than on the quality of the measured variables and the degree to which factors are saturated. Specifically, prior research has suggested that if initial communalities among measured variables are high (equal to or greater than .70) and the factors are highly saturated with 4–6 measured variables per factor, samples sizes as small as 100 may be adequate to uncover a meaningful and replicable solution (Fabrigar, Wegener, MacCallum, & Strahan, 1999; MacCallum, Widamen, Zhang, & Hong, 1999). The initial communality estimates obtained in our analysis ranged from .91 to .99 and we wrote 4–6 items to represent each facet (which could represent a separate factor). Thus, a sample size of 175 was adequate.

Data Analysis

The distribution of scores for each item was first inspected to identify items that yielded an insufficient range of responses or substantial skew (skew > 2.00) or kurtosis (kurtosis > 7.00). Four items were eliminated from the 167-item measure. The inter-item correlation matrix of the remaining 163 items was then subjected to a principal factor analysis with an oblique (direct oblimin) rotation. Four major criteria were used to determine how many factors to extract before rotation. The first three criteria were (a) the scree test (Cattell, 1966), (b) parallel analysis (Horn, 1965; O’Connor, 2000), and (c) factor interpretability. As a fourth criterion, we also factored the inter-item correlation matrix with a maximum likelihood (ML; Browne, Cudeck, Tateneni, & Mels, 2008) procedure and inspected the Root Mean Square Errors of Approximation (RMSEA) to explore the incremental fit of solutions with an increasing number of factors (see Fabrigar et al., 1999). The items were sufficiently univariate normal (skew < 2.00 and kurtosis <.7.00; see Fabrigar et al., 1999) to justify a ML solution.

Results and Discussion

Exploratory Factor Analyses

The 163-item inter-correlation matrix was first subjected to principal factor and ML analyses. The scree test suggested from 4 to 6 meaningful and reproducible latent variables in the reduced correlation matrix (with squared multiple correlations in the diagonal), with 4 factors suggested by the ML solution and 6 suggested by the parallel analysis criterion.

We thus extracted four-, five-, and six-factor solutions and rotated each obliquely via the direct oblimin rotation method (we assumed, from the results of the Brown and Rector meta-analysis, that the factors would be correlated). The sixth factor in the six-factor solution was small (6 items with loadings equal or greater to .40) with substantial cross-loadings (4 of the 6 items had loadings of .30 or greater on other factors). On the other hand, both the four- and the five-factor solutions yielded interpretable factors (with minimal cross-loadings) that substantially reproduced Brown and Rector’s (2008) solution. The main difference between the four- and five-factor solutions was that in the latter solution, items developed to measure career decision-making self-efficacy beliefs split from the lack of readiness factor to form a separate career decision-making self-efficacy factor.

However, when we eliminated items from the four- and five-factor solutions that did not load saliently on any factor (loadings < .40) or cross-loaded saliently (loading > .30) on more than one factor and re-extracted and rotated to four- and five-factor solutions, the five-factor solution deteriorated markedly, yielding a fifth factor with only two saliently loading items. On the other hand, the four-factor solution obtained after eliminating cross-loading and non-saliently loading items continued to yield an interpretable solution that was similar to the solution obtained with the larger correlation matrix. The four-factor solution accounted for 44.91% variance in the reduced correlation matrix. Thus, the four-factor solution appeared to reasonably represent the data and is reproduced in Table 2 .

Table 2.

Four-Factor Matrix for the CFI-167 after Non-Saliently and Saliently Cross-Loading Items were Removed (Upper Panel) and Loadings of Indecisiveness Items (Lower Panel)

Item	Factor I	Factor II	Factor III	Factor IV
Often feel fearful and anxious	.78	−.08	−.07	−.02
Frequently feel overwhelmed	.76	−.05	−.12	.08
Take setbacks harder than other people	.75	.07	−.10	−.05
Takes a long time to feel good after a setback	.71	.09	.02	.02
Often feel insecure	.71	−.05	.15	−.08
Hard time forgetting when something goes wrong	.71	−.13	.03	.01
Often feel tired and worn out	.71	.00	−.11	.00
I am a worrier	.67	−.16	−.17	−.14
Easily embarrassed	.66	.06	.08	−.10
Feel ashamed of self	.65	.06	.27	.09
Sleeping more or less lately	.63	.11	−.09	.12
So sad it’s hard to go on	.63	.17	.13	.15
Stress makes me ill	.62	.09	−.01	.08
Worry about what others think of me	.62	−.01	−.04	−.17
Often feel like crying	.61	.03	−.10	.03
Focus on what will go wrong with decisions	.61	−.18	.04	−.07
Don’t have energy to pursue ideas	.61	−.01	.18	.08
Trouble concentrating	.60	−.01	.11	.13
Hard to get things done when upset	.58	−.09	.04	.06
Eating more or less lately	.58	.14	−.10	.13
Can’t measure up to own standards	.58	−.14	.06	.07
Hard to cope with more than one or two problems at a time	.57	−.05	.08	.00
Lack energy to do anything	.57	.11	.19	.19
Need encouragement and support from others when making a decision	.56	−.24	−.06	.04
Think about alternatives didn’t choose after a decision	.55	−.17	.01	.07
Hard to overcome shyness	.55	−.13	.10	−.05
Worry about whether others like me	.53	−.14	.08	−.12
Hope problems will just go away	.53	−.10	−.05	.10
So many ideas are overwhelming	.52	.18	.01	.11
Hard deciding without help	.52	−.28	.16	.08
Hope problems will take care of themselves	.50	−.10	.11	.00
No confidence in decisions unless friends support them	.50	−.22	.16	.20
Feel good about myself as a person (R)	.50	.10	.35	.00
Keep thoughts to self so friends won’t disapprove	.47	−.15	.07	.12
Often hope for a miracle	.47	.01	−.10	.18
Problems too complex to solve	.44	−.01	.31	.10
Feel relieved if someone decides for me.	.42	−.25	.24	−.04
Depend on others to make my decisions	.36	−.21	.30	.04
Hard time narrowing my interests	−.15	.80	.16	.11
Need to learn more about my interests	−.03	.76	.00	.00
Feel stuck because don’t know enough about occupations	.09	.75	.15	−.04
Need to learn how to make a good career decision	.19	.74	.12	.00
Conflicted because of a number of appealing careers	−.12	.74	−.13	.10
Nervous when thinking about picking a career	.18	.74	.03	.01
Not enough occupational information	.14	.72	.10	.00
Can’t commit because don’t know other available options	.06	.72	.13	−.07
Need information on occupations would succeed in	.09	.72	−.05	−.11
Difficult because like so many things	−.11	.72	−.06	.08
Need to learn more about what I want	.07	.71	.01	−.08
My interests change too much	−.05	.71	.09	.02
Concerned my goals may change after deciding	.06	.69	−.02	.05
Need information on careers I might like	.10	.69	−.03	−.08
Need to learn more about myself	.23	.68	.07	−.07
Need to know more about job opportunities before deciding	−.03	.68	−.14	.01
Feel discouraged about making a career decision	.21	.68	.09	.03
Concerned interests might change after deciding	.16	.67	−.01	.02
Need a clearer picture of my abilities	.13	.64	.01	−.09
Don’t know much about occupations I’m considering	.10	.63	.16	.11
Hard time deciding between a couple of good options	−.04	.62	−.21	.00
Sometimes feel directionless	.15	.61	.14	.05
Don’t know where to get information	.12	.59	.15	.02
I like to keep my options open	−.16	.59	−.02	.10
Uncomfortable committing myself	−.05	.58	.15	−.06
Familiar with options but not ready to commit	−.07	.57	−.02	.12
Hard to find career that will allow time with friends or family	.23	.48	−.09	.17
Breathing gets tight when thinking about deciding	.24	.42	.10	.08
Put off making decisions	.28	.42	.26	.03
Hard to find a career that gives time for non-work interests	.21	.42	.13	.20
I try to excel at everything (R)	−.12	−.05	.69	.08
Work productively to get job done (R)	.03	.04	.65	.02
Strive hard to achieve goals (R)	−.17	−.08	.63	−.10
With enough effort I can solve anything (R)	.16	.23	.61	.09
Able to carry out my plans (R)	.04	−.07	.58	.00
Verify information before deciding (R)	−.02	.01	.56	−.07
Think carefully about decisions (R)	−.15	−.11	.55	.11
Plan ahead for important decisions (R)	.02	−.04	.54	−.18
I’m a worthwhile person (R)	.18	.13	.53	.06
Confident I’ll achieve my goals (R)	.18	−.26	.53	.06
When bad things happen I keep going (R)	.19	−.02	.51	.04
Confident will find career perform well in (R)	.05	−.24	.51	.07
Be able to find a career that fits interests (R)	.11	−.18	.47	.07
Confident will overcome obstacles (R)	.25	−.24	.47	.03
Thoroughly consider consequences before deciding (R)	−.06	−.01	.43	.17
Think of creative solutions to problems (R)	.11	−.03	.42	−.04
Trust my ability to solve problems (R)	.39	.04	.41	.03
Trust feelings when deciding	−.19	−.14	.40	.15
Confident my choice will be acceptable to important others (R)	.00	−.03	.36	.30
Liking a career is necessary for a fulfilling life	.20	−.13	−.34	−.20
Important people disagree with my choice	.04	.01	−.14	.85
Important people do not support my choice	.07	.11	.07	.80
Important people discourage my choice	.12	−.08	−.11	.72
Important people give me conflicting information	.08	−.17	−.05	.67
Go against wishes of others with my choice	.17	−.03	−.11	.62
Only one occupation for each person	.11	.03	.03	.37
Racial/ethnic discrimination makes it hard to pursue what I want	−.07	−.17	.04	.30
A career is a life-time commitment	−.01	−.01	.04	.28
I don’t plan because what happens is a matter of luck	.06	−.20	.19	.28
Luck is why good things happen	.20	−.03	.21	.23
Don’t have to think about decisions because right answer pops into my head	−.01	−.03	.02	.19
Chronic Indecisiveness Items
Worry I’ll make the wrong decision	.53	.37	−.01	.05
Have trouble making decisions	.46	.45	.11	−.07
Feel trapped when making a decision	.47	.36	.09	.14
Decision-making feels unclear	.41	.39	.24	.07
Uncertain about my decisions	.34	.53	.12	.06
Decision-making is frustrating	.46	.51	−.08	−.09
I am an indecisive person	.38	.42	.21	−.10
Friends tell me I have trouble deciding	.33	.41	.21	−.10
Change my mind a lot	.33	.44	.17	.12
Put off making decisions	.35	.44	.26	.09

Note: Salient loadings (greater than .40) and cross-loadings (greater than .30) are in bold. R = Reverse scored items. Factor I = Neuroticism/Negative Affectivity, Factor II = Choice/Commitment Anxiety, Factor III = Lack of Readiness, Factor IV = Interpersonal Conflict.

An inspection of Table 2 reveals that the four-factor solution uncovered in this study was similar to Brown and Rector’s (2008) four-factor meta-analytically derived solution. Four factors were obtained that contained items tapping into high levels of trait anxiety, depressive affect, fear of commitment, dependent decision-making styles, and low levels of psychological hardiness (Factor I); high needs for self- and occupational information and approach-approach conflicts (Factor II); low goal directedness, career decision-making self-efficacy beliefs, and planfulness/conscientiousness (Factor III); and high levels of interpersonal conflict (Factor IV). There were, however, some interesting differences from the Brown and Rector meta-analysis that may shed new light on sources of career indecision that could have implications for future research and career practice.

One difference was that Factor I, while highly similar to Brown and Rector’s Indecisiveness/Trait Neuroticism factor, now more closely resembled a trait neuroticism/negative affectivity latent dimension in two important ways. First, the items remaining on this factor represent at least four of the six major facets of neuroticism as identified in past research (e.g., Costa & McCrae, 1992), including trait anxiety (i.e., nervousness, apprehensiveness, and a tendency to dwell on things that might go wrong), depressive affect (e.g., feelings of hopelessness and discouragement), self-consciousness (e.g., embarrassment, feelings of inferiority, and shyness), and vulnerability (e.g., difficulty coping with stress and dependency). Second, items representing an external locus of control, which has been found to be a correlate rather than a defining facet of trait neuroticism (Costa & McCrae, 1992), failed to load saliently on this or any other factor in the current analysis.

A second, potentially more important difference was found on Factor II. All information-related items and approach-approach conflict items continued to load on this factor. However, additional items reflecting an inability to commit to, and anxiety about making, a decision also loaded highly on the second factor along with related items signifying a difficulty narrowing interests, concerns that interests and goals may change, and conflicts among a number of appealing options. We, therefore, relabeled this as a Choice/Commitment Anxiety factor. Interestingly, when considered at the total score level, measures of vocational exploration and commitment (e.g., Blustein, Devenis, & Kidney, 1989) loaded saliently on the Lack of Readiness factor in the Brown and Rector (2008) meta-analytic model. However, when considered at the item level, items that we wrote to reflect a difficulty committing to a choice loaded saliently only on the second factor in the present analysis.

A third finding with potentially useful implications is revealed in the bottom panel of Table 2; namely, that the 10 items initially written to measure respondents’ levels of chronic indecisiveness showed substantial cross-loadings on both Factors I and II. In addition to describing themselves as indecisive, high scorers on both factors indicated that they were often uncertain about their decisions, put off making decisions, felt frustrated with, and worried about, the decision-making process, and often felt trapped. These results, if replicated, would suggest that chronic indecisiveness is not a unitary construct but rather may have two different (but related) underlying causes—the emotional and behavioral characteristics of trait neuroticism (i.e., the tendency to dwell on what will go wrong with a decision and the concomitant emotional reactions) or an inability to commit to a decision because of the availability of a number of appealing options and a concern about what might be given up or what might change. Persons scoring high on the first factor may have chronic problems with decision making (or at least see themselves as indecisive) because they tend to focus on the negative aspects of different options, be relatively dissatisfied with available options, and, as a consequence, see few good options. On the other hand, persons scoring high on Factor II may be chronically frustrated in their decision-making efforts because they perceive a number of attractive options and feel unable to commit to any one of them for fear that they may change or may eliminate an option that they will regret later.

Factors III and IV in the present analysis are mostly consistent with Brown and Rector’s (2008) Lack of Readiness and External Barriers/Interpersonal Conflict factors. The former reflects a lack of goal-directedness, planning, and confidence in career decision-making abilities, and a less than rational, more intuitive, decision-making style. The fourth factor in the present analysis, however, represents more of an interpersonal conflict factor, with most external barriers items (e.g., discrimination and lack of resources) failing to load saliently on any factor. The latter findings may be attributable to our college student sample or to the possibility that various forms of barriers do not share enough variance among themselves at the item level to coalesce into a single latent dimension. For example, persons experiencing one form of discrimination (e.g., discrimination on the basis of race) may not concomitantly experience other forms of discrimination (e.g., discrimination due to age or disability status) with nearly equal frequency. As a result, items asking about each type of discrimination may share insufficient common variance to form a single latent variable in factor analysis.

Study 2

The purpose of this study was to explore whether the four-factor model identified in Study 1 could be uncovered from Gati and colleagues’ three-factor cognitive and personality/emotional models of career decision-making difficulties. Gati et al.’s (1996) cognitive model hypothesizes that career decision-making difficulties are associated with three major cognitive sources: (a) Lack of Readiness, (b) Lack of Information, and (c) Inconsistent Information. Each of the three major sources is further hypothesized to be influenced by a series of lower order facets. For Lack of Readiness, these include lack of motivation, indecisiveness, dysfunctional myths, and lack of knowledge about the process of career decision making (i.e., persons who are not ready to make a decision may lack motivation, be generally indecisive, adhere to dysfunctional myths about careers, and/or lack knowledge of how to make decisions). People who have difficulty due to a lack of information may need information about the self, occupations, and ways of obtaining further information. Finally, people may receive inconsistent information due to internal conflicts, external conflicts, or because they receive unreliable information from others. The Career Decision-Making Difficulties Questionnaire (CDDQ; Gati et al., 1996) contains 10 scales to measure each of the facets as well as the higher order decision-making difficulty sources.

At the facet level, there are clear parallels between this model and the four-factor model uncovered in Study 1. For example, both models posit that chronic indecisiveness, information needs, and interpersonal conflicts are significant sources of choice-making difficulties. However, our four-factor model would reconfigure these facets by considering chronic indecisiveness as a facet of career choice anxiety/commitment and, therefore, hypothesize that chronic indecisiveness would load more strongly with information needs than with a lack of readiness. Both models would expect an interpersonal conflict factor to emerge (called an Inconsistent Information factor by Gati and colleagues). The Gati model contains two facets (lack of motivation and dysfunctional beliefs) that do not have parallels in the present four-factor model and also does not include facets that are clearly affectivity related. Thus, we posited that a reconfigured choice anxiety/commitment factor (that includes indecisiveness and information needs) and an interpersonal conflict factor would emerge from our secondary analysis. We had no hypotheses about the loadings of the lack of motivation and career myths facets.

Saka et al.’s (2008) personality/emotional model of career decision-making difficulties also hypothesizes that three major personality/emotional factors can underlie career decision-making difficulties: (a) Pessimistic Views, (b) Anxiety, and (c) Self and Identity factors. Each of these higher order factors can be influenced by a series of lower order facets associated with pessimistic views about the process, the world of work, and one’s personal control (Pessimistic Views); anxiety associated with the process of decision making, uncertainty, the choice itself, and its outcomes (Anxiety); and general anxiety, self-esteem, uncrystallized identity, and conflictual attachments (Self and Identity). The Emotional and Personality Career Difficulties Scale (EPCD; Saka et al., 2008) was designed to provide a measure of each of the 11 facets and three higher order constructs.

There are again parallels between this model and our four-factor model. Both posit that anxiety can significantly affect the decision-making process as can pessimistic views and identity factors. However, our four-factor model would reconfigure some of these facets. First, our four-factor model would hypothesize that general anxiety and self-esteem, as central features of neuroticism, would move from Saka et al.’s Self and Identity factor and form their own factor, while the conflictual attachment and separation facet would stand alone as its own factor (Interpersonal Conflict). The items on the pessimistic view of the process facet subscale (e.g., “I can’t find out enough about all occupations to make the right choice” and “I can’t take all relevant considerations into account when choosing a career”) could alternatively be viewed as reflecting choice/commitment anxiety (see Table 2) and thus could be hypothesized to load more strongly on an anxiety than on a pessimistic views factor. Similarly, items on the uncrystallized identity scale seem also to reflect commitment problems (e.g., “My estimates of my skills and abilities change often” and “I still don’t know what my vocational interests are”). These could also load on an anxiety factor (rather than on the self and identity factor). Thus, we posited that a reconfigured anxiety factor (paralleling our Choice Anxiety/Commitment factor) would emerge that includes pessimistic views about the process and uncrystalized identity along with the other anxiety facets of the Saka et al. model. We further hypothesized that a neuroticism/negative affectivity factor composed of the generalized anxiety and self-esteem facets would emerge from our secondary analysis as well as a single facet factor (Interpersonal Conflicts) marked only by Saka et al.’s Conflictual Attachment and Separation facet. There were, again, no clear parallels between Saka et al.’s pessimistic views of the world of work and personal control facets so we had no hypotheses about how these might fit with our four-factor model, although they might together suggest some degree of lack of readiness and thus form a factor of their own.

In sum, we sought to explore whether aspects of the present four-factor model could be uncovered via secondary analysis of published correlation matrices. Specifically, we hypothesized that the Indecisiveness facet of Gati et al.’s (1996) cognitive model would load more strongly with their Lack of Information than with their Lack of Readiness factor (reflecting Choice/Commitment Anxiety in our model) and that an interpersonal conflicts factor would emerge that is identical to Gati and colleagues’ Inconsistent Information factor. We had no a priori notions about the likely loadings of the CDDQ’s lack of motivation and dysfunctional myths facets. In the case of the personality/emotional model, we hypothesized that a reconfigured choice anxiety/commitment factor would emerge as well as a single facet interpersonal conflict factor and a two-facet neuroticism (Generalized Anxiety and Self-Esteem) factor. We had no hypotheses about the loadings of the pessimistic views of work and personal control facets.

Method

We used two published inter-facet correlation matrices for our secondary analyses. In order to explore our notions about how Gati et al.’s cognitive model might be reconfigured on the basis of our four-factor model, we used the matrix provided by Gati et al. (1996) for 304 U.S. college students (lower diagonal of Gati et al., 1996, Table 2, p. 516). For our reanalysis of the Personality/Emotional model, we used the matrix for 276 U.S. college students published as Appendix A (lower diagonal) in Saka et al. (2008, p. 421).

Because we could not fully specify either model (i.e., no hypotheses could be made about loadings for lack of motivation, dysfunctional myths, pessimistic views of work, and personal control facets), we subjected the two correlation matrices to exploratory factor analyses via principal axis factoring, extracted three and four factors, and rotated the solutions obliquely (direct oblimin).

Results and Discussion

Results of our secondary factor analysis of the CDDQ (cognitive model) inter-facet correlation matrix yielded an interpretable three-factor solution that largely supported our a priori expectations. As shown in Table 3 , saliently loading facets on the first factor were consistent with our four factor model—all information facets loaded saliently (.73–.92) on this factor along with the indecisiveness facet (.46). The third factor, as hypothesized by both models, was marked by the unreliable information, internal conflict, and external conflict facets. Lack of motivation did not load saliently on any of the three factors, and dysfunctional beliefs seemed to form its own factor. Thus, it appears, as hypothesized, that at least two factors of our four-factor model are extractable from this correlation matrix. In addition, the composition of the first factor is more consistent with our model than with Gati et al.’s cognitive model in that both general indecisiveness and information needs load on a single factor.

Table 3.

Factor Loading Matrix From Reanalysis of Gati et al.’s (1996) Cognitive Model

Facets From Gati et al.’s Model	Factor I	Factor II	Factor III
Lack of Motivation	.29	−.15	.29
General Indecisiveness	.46	.31	.07
Dysfunctional Career Beliefs	−.03	.47	−.01
Information on the Decision-Making Process	.92	.02	−.07
Information on the Self	.71	−.08	.20
Information on Occupations	.73	−.01	.11
How to Get Additional Information	.73	−.01	.11
Receipt of Unreliable Information from Others	.10	.01	.79
Internal Conflict	.02	−.12	.87
External Conflict	−.03	.07	.60

Note: Salient loadings are in boldface.

The results of our factor analysis of the inter-correlations of the EPCD (personality/emotional model) facets yielded an interpretable four-factor solution that is presented in Table 4 . As hypothesized, the first factor was marked by pessimistic views of the process and uncrystallized identity along with the other four anxiety facets, while the fourth factor was a single facet conflictual attachment and separation (interpersonal conflict) factor. Both generalized anxiety and self-esteem, as hypothesized, loaded together on the second factor, while the two facets about which we had no hypotheses (pessimistic views of the world of work and personal control) formed the third factor. Thus, it appears that Gati and colleagues’ cognitive and personality/emotional models can be recast, in part, in terms of our four-factor model. First, a choice/commitment anxiety factor potentially emerged from both analyses, with indecisiveness and information facets loading together in the analysis of the cognitive model and pessimistic views of the process and uncrystallized identity loading with anxiety facets in the analysis of the personality/emotional model.

Table 4.

Factor Loading Matrix From Reanalysis of Saka et al.’s (2008) Personality/Emotional Model

Facets From Saka et al.’s Model	Factor I	Factor II	Factor III	Factor IV
Pessimistic Views on the Process	.50	−.08	.44	−.16
Pessimistic Views on World of Work	−.10	.04	.68	−.02
Pessimistic Views of Personal Control	.07	.04	.58	.12
Anxiety about the Process	.79	−.03	−.04	.10
Anxiety Due to Uncertainty	.83	.07	.00	−.03
Anxiety about Choosing	.92	−.03	−.06	.08
Anxiety about the Outcome	.68	.07	.04	.00
General Anxiety	.15	.49	.00	.30
Self Esteem	−.06	.98	.06	−.04
Uncrystallized Identity	.51	.39	.10	−.27
Conflictual Attachments	.18	.15	.28	.45

Note: Salient loadings are in boldface.

Second, an interpersonal conflict factor emerged from both analyses. Although this factor was consistent with Gati and colleagues cognitive model, it may not be just another facet of self and identity as hypothesized by the personality/emotional model. As hypothesized by our four-factor model, it may instead be a stand-alone factor that is as important to choice-making problems as is neuroticism/negative affectivity, choice/commitment anxiety, and lack of readiness. Third, in the matrix that included neuroticism facets, both generalized anxiety and self-esteem formed a single factor that was more consistent with our four-factor model than with Saka et al.’s personality/emotional model (i.e., self-esteem and anxiety did not load together with uncrystallized identity and conflictual attachment and separation). Together, these results, which are consistent with our four-factor model, suggest that both neuroticism/negative affectivity (or at least generalized anxiety and self-esteem) and interpersonal conflict should be considered as major sources of career indecision in their own right and not simply as facets of something else. Finally, in both analyses there were two facets that did not seem to have parallels in our four-factor model and these, in both cases, emerged as separate factors. For the cognitive model, these were lack of motivation and dysfunctional beliefs, while for the personality/emotional model they were pessimistic views of the world of work and personal control. Whether these two factors represent additional stand-alone factors or are facets of one of our four factors (e.g., lack of readiness) awaits further research.

General Discussion

Taken collectively, the results of our two studies suggest that a somewhat modified version of Brown and Rector’s (2008) four-factor model of career indecision may represent a comprehensive, empirically meaningful, and practically useful way of conceptualizing career decision-making difficulties. Factor analytic results of our original data suggested that a four-factor model could represent adequately the covariation among the items on the CIP-167. These factor analytic results also pointed out some modifications to Brown and Rector’s model that found additional support when used to reconfigure Gati and colleagues cognitive and personality/emotional models of career indecision. We will in this discussion first highlight what we consider to be the most important findings from our studies, focusing specifically on each factor. In the process, we will also attempt to embed each of the four factors in a beginning nomological network to direct future research and counseling efforts. We will then note some limitations of the present research and finish by outlining future research that is necessary to establish the robustness of the current model.

Neuroticism/Negative Affectivity (Factor I)

The first and potentially most important finding of our initial study was that chronic indecisiveness might not be a unidimensional construct but rather may be influenced by (or at least associated with) two different underlying causes. One appears to be the emotional and behavioral characteristics of trait neuroticism (e.g., the tendency to dwell on and worry about the potential negative consequences of available options and the affective concomitants of this negative cognitive/perceptual style).

Persons scoring high on the Neuroticism/Negative Affectivity factor may have chronic problems with decision making (or at least see themselves as indecisive) because they tend to focus on the negative aspects of different options, be relatively dissatisfied with available options, and, consequently, see few good options. If we are correct that this factor represents trait neuroticism, we would further hypothesize based on prior research on trait neuroticism, that the stress associated with decision making may lead to various avoidance options, including prematurely foreclosing on an available option despite being less than satisfied with it. Thus, future research on trait neuroticism and its relation to career indecision might explore the degree to which scores on the first factor (or other measures of trait neuroticism) relate to a tendency to foreclose and to satisfaction with the selected option. Indeed, others (e.g., Blustein & Phillips, 1990) have found a positive relation between dependent decision-making style (a facet of trait neuroticism) and foreclosed identity statuses (Marcia, 1980). Further, prior cluster analytic studies (see Brown & Rector, 2008) have all identified a cluster of college students and career center clients with characteristics that would identify them according to our model as relatively high in trait neuroticism (high trait anxiety and low self-esteem, self-reported indecisiveness, and dependent decision-making styles). It is noteworthy that, in one of the cluster analytic studies, Wanberg and Muchinsky (1992) identified two clusters of decided and two clusters of undecided students. One of the decided clusters (labeled as “concerned decided individuals”), comprising 40% of the sample, displayed a pattern of scores on other measures that would identify them as high in trait neuroticism (i.e., low self-esteem and high trait anxiety and self-reported indecisiveness), despite the fact that they also considered themselves to be career decided.

Taken together, these results suggest the possibility that persons scoring high on neuroticism/negative affectivity may prematurely foreclose on available options (as an avoidance strategy) by perhaps relying excessively on the input from others. At the same time, they may not be highly satisfied with their choice. An area for future research would be to examine more closely the relations of trait neuroticism/negative affectivity to levels of decidedness, satisfaction, and commitment. The relation of neuroticism/negative affectivity to identity status also merits attention. We would hypothesize that neuroticism/negative affectivity will relate more strongly to foreclosed identity statuses than to achieved, moratorium, or diffuse identity status.

Another avenue for future research would be to explore the relations between scores on the CIP-167 Neuroticism/Negative Affectivity scale and scores on separate measures of neuroticism and negative affectivity. There is still some disagreement in the literature about whether Neuroticism (N) and Negative Affectivity (NA) represent the same (e.g., Tellegen, 1985), related (Watson & Clark, 1989), or hierarchically structured constructs (Nemanick & Munz, 1997), with most of the evidence suggesting that they are not synonymous. Correlations between measures of each are relatively large (e.g., r = .58; Watson, Weise, Vaidya, & Tellegen, 1999), but N rarely has been found to predict significant additional variance in various measures of satisfaction (e.g., job and life) over and above NA. Assuming that our Neuroticism/Negative Affectivity scale is accurately named (i.e., that the affectivity dimension is adequately represented) we would further hypothesize that persons scoring high versus low on this scale may display persistent self-doubt (Watson & Pennebacker, 1989), less ambitious career and educational goals (Cook, Vance, & Spector, 2000), create more performance constraints for themselves (Spector & Jex, 1998), and demonstrate low levels of task (e.g., career decision making) motivation (Kaplan, Bradley, Luchman, & Haynes, 2009).

Choice/Commitment Anxiety (Factor II)

The second potential underlying dimension of chronic indecisiveness may be an inability to commit to a decision because of a number of appealing options and a concern about what might be sacrificed in selecting among them (choice/commitment anxiety). Persons scoring high on the Choice/Commitment Anxiety factor may perceive themselves to be indecisive and to feel frustrated and trapped in the decision-making process, but for different reasons than those scoring high on the Neuroticism/Negative Affectivity factor. In particular, they may perceive a number of good options available to them but feel unable to commit for fear that they may change or may eliminate options that they will later regret. They also seem to feel that they lack information and that (perhaps) further information may allow them to bring closure to their decisional dilemma.

This pattern of characteristics also finds some parallels in the cluster analytic literature (see Brown & Rector, 2008). Specifically, Larson, Heppner, Ham, and Dugan (1988) identified a cluster (Informed Indecisives) of undergraduates who concomitantly displayed rather high levels of anxiety about the process and high needs for career information, despite having more information than others in the sample. Larson et al. further noted that career planning specialists indicated that these clients could be frustrating. They seemed to be highly motivated and informed, but their counseling sessions seemed to be unsuccessful and the counselors’ patience was often tested. Our discussions with college career counseling center staff echo this sentiment and also provide another interesting possibility for research—that the current economy is exacerbating the tendency to try to keep all options open and to resist committing to any option for as long as possible. Thus, we might suggest that future research explore the relations of scores on the Choice/Commitment Anxiety scale to counseling outcome and to students’ concerns about their futures in the current economic climate.

Another direction for future research involves the relation between choice/commitment anxiety and decision-making tendencies. Schwartz et al. (2002) expanded on Simon’s (1955, 1956, 1957) classic work on choice making under conditions of uncertainty and found that persons who tend to use maximizing versus satisficing strategies were less satisfied with their choices. Maximizers, according to Schwartz et al. (2002), seek out the best possible option and engage in exhaustive search efforts, while satisficers seek good enough options and stop searching when that option is found. Subsequent research has also shown that maximizers tend to become more fixated on options, explore more options, and experience greater levels of anxiety than do satisficers. For example, in a study of the job search process of graduating college seniors, Iyengar, Wells, and Schwartz (2006) found that, compared to satisficers, maximizers planned to apply for more jobs (20 vs. 10), experienced more anxiety in the process, sought out more information, were less satisfied with their choices, and retrospectively wished they had applied for more jobs. Thus, it would be interesting to explore the relations of maximizing and satisficing tendencies to scores on the CIP-167, especially the Choice/Commitment Anxiety scale, and to test whether such tendencies might add uniquely to the prediction of Factor II scores over and above such contextual factors as concern about the current economy.

Lack of Readiness (Factor III)

In addition to the multisource influence on chronic indecisiveness, this study replicated the Brown and Rector (2008) meta-analysis by identifying a factor that might be related to career decision readiness. This factor was marked primarily by a lack of planfulness and goal directedness and by low career decision-making self-efficacy beliefs. As Brown and Rector (2008) noted, such a constellation of characteristics may not be problematic until the time to make a career decision becomes imminent and may provide excellent targets for career development activities in educational contexts, with a focus on acquiring more rationally focused decision-making skills, facilitating career decision-making self-efficacy belief development, and increasing goal-directedness and planfulness.

Interestingly, recent research (e.g., Guay, Ratelle, Senecal, Larose, & Deschenes, 2006; Guay, Senecal, Gauthier, & Fernet, 2003) that has explored adolescent career decision status longitudinally has generated two findings of relevance to the present discussion. First, Guay et al. (2006) found that changes in career decision-making self-efficacy beliefs were the strongest predictors of change in decision status overtime (i.e., changes in levels of career indecision). Thus, as suggested by Guay et al. (2006), increasing career decision-making self-efficacy beliefs may be a very important component of early intervention efforts.

Second, Guay et al. (2003, 2006) found that parental autonomy support had both direct and indirect effects (via self-efficacy belief development) on decisional status over time. In other words, adolescents whose parents encouraged autonomy tended to become less undecided and developed more robust career decision-making self-efficacy beliefs over the course of the study than did adolescents whose parents were less autonomy supporting.

The findings concerning parent autonomy support are consistent both with Brown and Rector’s (2008) meta-analysis and with suggestions we have received in our presentations to practicing career counselors. Brown and Rector’s meta-analysis included measures of adult attachment styles (Collins & Read, 1990) and psychological separation (Hoffman, 1984). Two scales from these instruments (conflictual and functional independence from parents) loaded, albeit non-saliently, on the Lack of Readiness factor. These results suggest that the degree to which college students have achieved functional and conflictual independence from their parents (as well as autonomy support) may be associated with college students’ levels of decisional readiness. The college career counselors with whom we have spoken have also consistently indicated that many students they see display the pattern of characteristics associated with our lack of readiness factor and also noted that the career decision seems to be one of the first major decisions that these students have ever had to make on their own. Thus, we would suggest that future research on the role of family attachment and separation in career development might profit by taking a more fine-grained approach and focusing on the relation of family variables to lack of readiness rather than to global measures of career decidedness or development (see Whiston & Keller, 2004).

Interpersonal Conflicts (Factor IV)

Finally, the fourth factor that emerged from this study appeared uniquely to be an interpersonal conflicts factor. These findings suggest a number of questions for future research on the CIP. The first is to explore why items associated with external barriers and discrimination did not load as highly on this factor as scales measuring these loaded in the Brown and Rector (2008) meta-analysis. We have suggested two possibilities: that barriers and discrimination are not perceived as particularly career limiting by college students (we found no race/ethnicity difference on the mean scores obtained by our sample on this scale) or that they covary with insufficient frequency to form a homogeneous factor. Thus, future research needs to use the CIP-167 with more diverse (in terms of age, race, ethnicity, socioeconomic status, and sexual orientation) samples to explore whether a separate external barriers/discrimination factor might emerge or whether external barrier- and discrimination-oriented items might load saliently on the interpersonal conflicts factor (as found in the Brown and Rector meta-analysis).

It would also be interesting to compare the relations of interpersonal conflict and career indecision in collectivist versus more individualist cultures and to explore whether level of acculturation might moderate the relation between interpersonal conflict and indecision. For example, past research (e.g., Mau, 2004; Ma & Yeh, 2005) has suggested that family conflict may play a significant role in the career decision status of Asian American high school and college students, especially when students are more acculturated to mainstream U.S. culture than are their parents. It is possible that level of acculturation may moderate the relations between scores on the CIP-167 Interpersonal Conflict scale and career status for Asian American college students and students from other recent immigrant groups whose cultures of origin are more collectivistic than individualistic.

Limitations and Future Directions

Although the findings of our studies may have important implications for future research and practice, they are currently limited, as noted earlier, by the homogeneity of our sample. Our findings obviously need to be replicated in more diverse U.S. and international samples. Such research may not only provide greater clarity on the major factors related to career decision-making difficulties but also allow us to identify sources of difficulty that may be universal versus others that are more culturally specific.

Footnotes

Acknowledgments

The authors thank Michelle Johnson and Tom Sak for assistance with this research and Dr. Robert Lent, Theresa Chan, Anneliese Kranz, Colleen Martin, and Meaghan Rowe-Johnson for reading and commenting on earlier drafts of this article.

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

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Jason Hacker, Mathew Abrams, Andrea Carr, Kristen Lamp, and Kyle Telander were supported by Graduate Assistantships provided by the Graduate School and School of Education of Loyola University Chicago.

References

Amir

Gati

Kleiman

(2008). Understanding career decision-making difficulties. Journal of Career Assessment, 16, 281–309.

Blustein

D. L.

Devenis

L. E.

Kidney

(1989). Relationship between the identity formation process and career development. Journal of Counseling Psychology, 36, 196–202.

Blustein

D. L.

Phillips

S. D.

(1990). Relation of ego identity status and decision-making styles. Journal of Counseling Psychology, 37, 160–168.

Brown

S. D.

McPartland

E. B.

(2005). Career interventions: Current status and future directions. In Walsh

W. B.

Savickas

M. L.

(Eds.), Handbook of vocational psychology (3rd ed., pp. 195–226). Mahwah, NJ: Erlbaum.

Brown

S. D.

Rector

C. C.

(2008). Conceptualizing and diagnosing problems in career decision-making. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (4th ed., pp. 392–407). New York, NY: Wiley.

Brown

S. D.

Ryan Krane

N. E.

(2000). Four (or five) sessions and a cloud of dust: Old assumptions and new observations about career counseling. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (3rd ed., pp. 740–766). New York, NY: Wiley.

Browne

M. W

Cudeck

Tateneni

Mels

(2008). CEFA: A comprehensive Exploratory factor analysis, Version 3.02 [Computer software and manual]. Retrieved from http://faculty.psy.ohio-state.edu/browne/

Cattell

R. B.

(1966). The scree test for the number of factors. Multivariate Behavioral Research, 1, 245–276.

Collins

N. L.

Read

S. J.

(1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58, 644–663.

10.

Cook

K. W.

Vance

C. A.

Spector

P. E.

(2000). The relation of candidate personality with selection interview outcomes. Journal of Applied Psychology, 30, 867–885.

11.

Costa

P. T.

Jr. McCrae

R. R.

(1992). NEO PI-R professional manual. Odessa, FL: Psychological Assessment Resources.

12.

Crites

J. O.

(1969). Vocational psychology: The study of vocational behavior and its development. New York, NY: McGraw-Hill.

13.

Fabrigar

L. R.

Wegener

D. T.

MacCallum

R. C.

Strahan

E. J.

(1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4, 272–299.

14.

Gati

Krausz

Osipow

S. H.

(1996). A taxonomy of difficulties in career decision-making. Journal of Counseling Psychology, 43, 510–526.

15.

Gati

Osipow

S. H.

Krausz

Saka

(2000). Validity of the Career Decision-Making Difficulty Questionnaire: Counselee versus career counselor perceptions. Journal of Vocational Behavior, 56, 99–113.

16.

Guay

Ratelle

C. F.

Senecal

Larose

Deschenes

(2006). Distinguishing developmental from chronic indecision: Self-efficacy, autonomy, and social support. Journal of Career Assessment, 14, 235–251.

17.

Guay

Senecal

Gauthier

Fernet

(2003). Predicting career indecision: A self-determination theory perspective. Journal of Counseling Psychology, 50, 77–97.

18.

Hoffman

(1984). Psychological separation of late adolescents from their parents. Journal of Counseling Psychology, 31, 170–178.

19.

Holland

J. L.

Holland

J. E.

(1977). Vocational indecision: More evidence and speculation. Journal of Counseling Psychology, 24, 404–414.

20.

Horn

J. L.

(1965). A rational and technique for estimating the number of factors in factor analysis. Psychometrika, 30, 179–185.

21.

Iyengar

S. S.

Wells

R. E.

Schwartz

(2006). Doing better but feeling worse: Looking for the “best” job undermines satisfaction. Psychological Science, 17, 143–150.

22.

Kaplan

Bradley

J. C.

Luchman

J. N.

Haynes

(2009). On the role of positive and negative affectivity in job performance: A meta-analytic investigation. Journal of Applied Psychology, 94, 162–176.

23.

Larson

L. M.

Heppner

P. P.

Ham

Dugan

(1988). Investigating multiple subtypes of career indecision through cluster analysis. Journal of Counseling Psychology, 35, 439–446.

24.

P. W.

Yeh

C. J.

(2005). Factors influencing the career decision status of Chinese American youths. The Career Development Quarterly, 53, 337–347.

25.

MacCallum

R. C.

Widaman

K. F.

Zhang

Hong

(1999). Sample size in factor analysis. Psychological Methods, 4, 84–89.

26.

Marcia

J. E.

(1980). Identity development in adolescence. In Adelson

(Ed.), Handbook of adolescent psychology (pp. 159–187). New York, NY: Wiley.

27.

Mau

W. J.

(2004). Cultural dimensions of career decision-making difficulties. The Career Development Quarterly, 53, 67–77.

28.

Miller

M. J.

Brown

S. D.

(2005). Counseling for career choice: Implications for improving interventions and working with diverse populations. In Brown

S. D.

Lent

R. W.

(Eds.), Career development and counseling: Putting theory and research to work (pp. 441–465). New York, NY: Wiley.

29.

Nemanick

R. C.

Jr. Munz

D. C.

(1997). Extraversion and neuroticism, trait mood and state affect: A hierarchical relationship? Journal of Social Behavior and Personality, 12, 1079–1092.

30.

O’Connor

B. P.

(2000). SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behavior Research Methods, Instrumentation, and Computers, 32, 396–402.

31.

Peterson

G. W

Sampson

J. P

Jr. Lenz

J. G

Reardon

R. C.

(2002). A cognitive information approach career problem solving and decision making. In Brown

Associates (Eds.), Career choice and development (4th ed., pp. 312–369). San Francisco, CA: Jossey Bass.

32.

Phillips

S. D.

(1992). Career counseling: Choice and implementation. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (2nd ed., pp. 513–547). New York, NY: Wiley.

33.

Rounds

J. B.

Tinsley

H. E. A.

(1984). Diagnosis and treatment of vocational problems. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (pp. 137–177). New York, NY: Wiley.

34.

Saka

Gati

Kelly

K. R.

(2008). Emotional and personality-related aspects of career decision-making difficulties. Journal of Career Assessment, 16, 403–424.

35.

Salmone

P. R.

(1982). Difficult cases in career counseling: II. The indecisive client. Personnel and Guidance Journal, 60, 496–499.

36.

Savickas

M. L.

(1989). Annual review: Practice and research in career counseling and development, 1988. Career Development Quarterly, 38, 100–134.

37.

Schwartz

Ward

Montersso

Lyubomirsky

White

Lehman

D. R.

(2002). Maximizing versus satisficing: Happiness is a matter of choice. Journal of Personality and Social Psychology, 83, 1178–1197.

38.

Simon

H. A.

(1955). A behavioral model of rational choice. Quarterly Journal of Economics, 59, 99–118.

39.

Simon

H. A.

(1956). Rational choice and the structure of the environment. Psychological Review, 63, 129–138.

40.

Simon

H. A.

(1957). Models of man, social and rational: Mathematical essays on rational human behavior. New York, NY: Wiley.

41.

Spector

P. E.

Jex

S. M.

(1998). Development of four self-report measures of job stressors and strain: Interpersonal Conflict at Work Scale, Organizational Constrains Scale, Quantitative Work Overload Inventory, and Physical Symptoms Inventory. Journal of Occupational Health Psychology, 3, 356–367.

42.

Tellegen

(1985). Structures of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In Hussain

T. A.

Maser

J. D.

(Eds.), Anxiety and the anxiety disorders (pp. 681–706). Hillsdale, NJ: Erlbaum.

43.

Tyler

L. E.

(1969). The work of the counselor. (2nd ed.). New York, NY: Appleton-Century-Crofts.

44.

Wanberg

C. R.

Muchinsky

P. M.

(1992). A typology of career decision status: Validity of the vocational decision status model. Journal of Counseling Psychology, 39, 71–80.

45.

Watson

Clark

L. A.

(1985). Negative affectivity: The disposition to experience aversive emotional states. Psychological Bulletin, 96, 465–490.

46.

Watson

Pennebaker

J. W.

(1989). Health complaints, stress, and distress: Exploring the central role of negative affectivity. Psychological Review, 96, 234–254.

47.

Watson

Weise

Vaidya

Tellegen

(1999). The two general activation systems of affect: Structural findings, evolutionary considerations, and psychobiological evidence. Journal of Personality and Social Psychology, 76, 820–838.

48.

Whiston

S. C.

Brecheisen

B. K.

Stephens

(2003). Does treatment modality affect career counseling effectiveness? Journal of Vocational Behavior, 62, 390–410.

49.

Whiston

S. C.

Keller

B. K.

(2004). The influence of the family of origin on career development: A review and analysis. The Counseling Psychologist, 32, 493–568.

50.

Whiston

S. C.

Rahardja

(2008). Vocational counseling process and outcome. In Brown

S. D.

Lent

R. W.

(Eds.), Handbook of counseling psychology (4th ed., pp. 444–461). New York, NY: Wiley.

51.

Williamson

E. G.

(1965). Vocational counseling: Some historical, philosophical, and theoretical perspectives. New York, NY: McGraw-Hill.