The Self-Directed Search and Career Self-Efficacy.pdf

8/10/2019 The Self-Directed Search and Career Self-Efficacy.pdf

1/16

http://jca.sagepub.com

Journal of Career Assessment

DOI: 10.1177/1069072799007002041999; 7; 145Journal of Career Assessment

Patrack F. Feehan and Joseph A. JohnstonThe Self-Directed Search and Career Self-Efficacy

http://jca.sagepub.com/cgi/content/abstract/7/2/145The online version of this article can be found at:

Published by:

http://www.sagepublications.com

can be found at:Journal of Career AssessmentAdditional services and information for

http://jca.sagepub.com/cgi/alertsEmail Alerts:

http://jca.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://jca.sagepub.com/cgi/content/refs/7/2/145Citations

at University of Missouri-Columbia on March 13, 2009http://jca.sagepub.comDownloaded from
http://jca.sagepub.com/cgi/alertshttp://jca.sagepub.com/cgi/alertshttp://jca.sagepub.com/subscriptionshttp://jca.sagepub.com/subscriptionshttp://jca.sagepub.com/subscriptionshttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsPermissions.navhttp://jca.sagepub.com/cgi/content/refs/7/2/145http://jca.sagepub.com/http://jca.sagepub.com/http://jca.sagepub.com/http://jca.sagepub.com/http://jca.sagepub.com/cgi/content/refs/7/2/145http://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsReprints.navhttp://jca.sagepub.com/subscriptionshttp://jca.sagepub.com/cgi/alerts


2/16

The Self-Directed Search and

Career Self-EfficacyPatrack F. Feehan

JosephA. Johnston

University of MissouriColumbia

This paper presents the results of a study aimed at furthering

understanding of the meaning of responses to Hollands Self-DirectedSearch (SDS). Drawing on current social cognitive theory, this studyprovides some evidence that SDS Summary scores might bepredictive of respondents future self-efficacy expectations.Asignificant relationship was found between responses to the SDS andan instrument specifically designed to measure career self-efficacy,the short form of the Task-Specific Occupational Self-efficacy Scale(TSOSS). Evidence for the validity of the SDS in predicting careerself-efficacy is provided by the correlational patterns and genderdifferences shown in the responses to both the SDS and TSOSS.

Implications and suggestions for further research are discussed.

Journal of CareerAssessment

Volume 7/Number 2/Spring 1999/Pages 145-159

Bandura (1977) introduced the concept of self-efficacy expectations (i.e.,ones belief in their ability to successfully perform a given behavior) as afundamental component of his social learning theory. Renamed as socialcognitive theory (Bandura, 1986), this theory posits that self-efficacyexpectations determine whether or not a behavior will be initiated, how mucheffort will be expended, and how long the behavior will be sustained in theface of obstacles and aversive experiences. The theory differentiates betweenefficacy expectations (beliefs concerning the performance of a behavior)and outcome expectations (beliefs concerning the consequences of a behavior).Both types of expectations (efficacy and outcome) vary on the dimensions of

level, strength, and generality.

Career Self-Efficacy TheoryThis theory has led to the development of the applied construct of career

self-efficacy, which has emerged from a number of empirical studies basedon social cognitive theory (Betz & Hackett, 1981; Betz, Harmon, & Borgen,1996; Lenox & Subich, 1994; Osipow, Temple, & Rooney, 1993; Rooney &Osipow, 1992). The construct of career self-efficacy, as it has been defined

We are grateful to John Holland for his continued input and critique of results onearlier drafts.

Correspondence concerning this article and requests for offprints should be addressedto Patrick Feehan, 16 Hill Hall, University of Missouri-Columbia, Columbia, MO 65211.

Published and copyright 1999 by PsychologicalAssessment Resources, Inc.All rights reserved.

http://jca.sagepub.com/http://jca.sagepub.com/http://jca.sagepub.com/http://jca.sagepub.com/


3/16

146

in this research, generally takes the form of an individuals belief in theirability to successfully engage in specific work-related tasks or activities. Inthis translation of self-efficacy theory to career development, the theory hascome to

emphasizethe

interlocking processesof interest

development,choice, and performance. The theory attempts a &dquo;cognitive constructivist&dquo;approach to career development, where forethought, anticipation of outcomes,and active construction of meaning is emphasized. Individuals are seen inthis theory as proactive shapers of their environments, not merely asresponders (Lent, Brown, & Hackett, 1994).

Hollands TheoryHolland (1997) describes Staats (1981) theory of social behaviorism as

most consistent with his theory of how types develop.As described byHolland, a childs biology and experience lead to preferences for some

activities over others. Engagement in these activities subsequently leads tothe development of long-term interests, competencies, and dispositions(Holland, 1997). In Hollands theory, the choice of a vocation is an expressionof individual personality based to some degree upon self-concept. Hollandcites other theorists noting of the same sentiment: &dquo;Interest inventoryscores are measures of self-concept&dquo; (Bordin, 1943); &dquo;vocational choice...isthe implementation of a persons self-concept&dquo; (Super, 1972). Hollandmaintains that &dquo;these orientations consistently imply that peoples vocationalinterests flow from their life history and personality&dquo; (p. 8). Individuals hereare seen as developing a &dquo;characteristic repertoire of attitudes and skills forcoping with environmental problems and tasks&dquo; (p. 2). Similar to self-efficacytheory, individuals are viewed as &dquo;active rather than passive recipients ofenvironmental influence&dquo; (p. 2).

The Intersection of the Theories

The Holland themes as articulated within the SDS have provided aframework for several studies that have examined the relationship betweeninventoried vocational interests and career self-efficacy (Betz, Harmon, &Borgen, 1996; Lenox & Subich, 1994; Lent, Larkin, & Brown, 1989; Matsui &Tsukamoto, 1991). The development of research in current social cognitivetheory as it applies to career decision-making appears to be providing

increasing evidence that many forms of self-knowledge (i.e., known interestsand competencies, favored activities, and self-estimates of ability) maycomprise a persons self-efficacy beliefs.While the two theories outlined above originate in different schools of

thought, they appear to share a great deal ofcommon purpose. Most relevantto this study is the acknowledgement by both theories of the importance ofself-efficacy or self-concept as a mediator or basis for the development ofsubsequent vocational behavior. Based on this theoretical linkage, it appearsreasonable to ask how effective Hollands SDS might be in predicting futureself-efficacy expectations.

This study was specifically undertaken in an attempt to determine ifresponses to Hollands SDS (Holland, 1990) might be predictive of self-efficacy expectations as embodied in responses to an instrument specificallydesigned to measure career self-efficacy expectations as defined in a social



4/16

147

cognitive context. The specific instrument chosen for comparison was theshort form of the Task-Specific Occupational Self-Efficacy Scale (TSOSS;Osipow et al., 1993).

SDS Summary scores, which are each comprised of five SDS subscalescores (Activities, Competencies, Occupations, and two Self-Estimates)were compared with TSOSS Factor scores (see Measures section). The

Summary scores were chosen because of their clinical availability,applicability (the Holland three-letter code is a proxy for the Summaryscores), and demonstrated reliability.

Hypothesized RelationshipAmong the Variables

Based on the face validity of the factor structure of the TSOSS, it washypothesized that TSOSS scores for Factor 1 (verbal, interpersonal skills)would be positively correlated with SDS Social and Enterprising Summaryscores; that TSOSS scores for Factor 2 (quantitative, logical, business,scientific skills) would be positively correlated with SDS Conventional and

Investigative Summary scores; that TSOSS Factor 3 (physical strengthand agility) would be positively correlated with SDS Realistic Summaryscores; and that TSOSS Factor 4 (aesthetic skills) would be positivelycorrelated with SDSArtistic Summary scores.

Gender Differences

Career self-efficacy has from its origin been conceptualized as anexplanatory variable useful for predicting the career behavior of women

(Betz&

Hackett, 1981; Hackett&

Betz, 1981).Since that

time, manyinvestigators have found significant differences in response patterns for menand women in assessments specifically designed to measure this construct(Betz & Hackett, 1981; Betz et al., 1996; Matsui & Tsukamoto, 1991; Osipow,et al., 1993; Rooney & Osipow, 1992).

In an attempt to build upon these outcomes, another intent of this studywas to examine the gender differences between womens and mens SDSSummary scores, and to compare them for similarity with womens andmens TSOSS Factor scores. For the high school and young college agestudents used for this study, scores were hypothesized to follow the patternfound by Betz et al. (1996) for college students responding to the SkillsConfidence Inventory (SCI): the men in this sample were expected todemonstrate more perceived self-efficacy than women for tasks classified inHolland terms as Realistic, Investigative, Enterprising, and Conventional;the women in this sample were expected to demonstrate more self-efficacythan men for tasks classified as Social; and both the men and women in this

sample were expected to demonstrate similar levels of self-efficacy for tasksclassified asArtistic.

Method

Participants

Participants were 112 male and 125 female (N = 237) high schoolstudents enrolled in a career planning home study course offered by a majormidwestern university.Age of the participants was as follows: Complete



5/16

148

sample M = 17.8 years, SD = 1.6 years; Mean male age = 17.9 years, SD =1.6years; Mean female age = 17.7 years, SD = 1.5 years.

In general, students take this course to learn more about themselves

and theircareer

options.The

typicalstudent

enrolled in thiscourse

hasfallen behind in their studies due to poor scholastic performance or absenceresulting from pregnancy or illness, and is attempting to make up creditsfor graduation. However, many students from rural high schools withlimited curricular options enroll in this course to round out their academicexperience.As a result, it is reasonable to suggest that the students havereasonably diverse levels of interest and a correspondingly diverse range ofcareer self-efficacy expectations.

Procedure

As part of the introduction to the course, the students were asked to fill

out a Self-Directed Search (SDS Form R; Holland, 1990). Near the end of thecourse work, students were asked to take the short form of the Task-SpecificOccupational Self-efficacy Scale (TSOSS; Osipow et al., 1993).

Students in the class were tracked for the duration of their enrollment inthe class, and data were accumulated by student. Since this was a homestudy course, the time lag between receipt of the SDS and of the TSOSSaveraged 95.5 days, with a standard deviation of 65.5 days.

Measures

The Task-Specific Self-Efficacy Scale

The TSOSS (Osipow et al., 1993) was developed as a task-specific measureof career self-efficacy in accordance with Banduras social cognitive theory(1977, 1986). Rooney and Osipow (1992) originally developed a 230-item scaleaimed at measuring the level, generality, and strength of respondents self-efficacy expectations by using items describing work activities of differingdifficulty at the task-specific level. The items in the original TSOSS aredrawn entirely from descriptions of work activities listed in partA of theSelected Characteristics of Occupations Defined in the Dictionary ofOccupational Titles (U. S. Department of Labor, 1981, 1991). Respondentsindicate, through placement of their response on a 5-point Likert scale,

their confidence in their ability to perform each of the listed activities.Responses are coded from &dquo;no confidence&dquo; to &dquo;absolute certainty.&dquo;Osipow et al. (1993) further refined this scale through factor analysis,

resulting in a more efficient 60-item test, comprised of four subscales of 15items each. Each subscale represents a specific factor as follows: Factor 1:Verbal, interpersonal skills; Factor 2: Quantitative, logical, business,scientific skills; Factor 3: Physical strength and agility; and Factor 4:

Aesthetic skills.

These factors appear to correspond to the Holland types: Factor 1 toboth Social and Enterprising types; Factor 2 to both Investigative and

Conventional types; and Factors 3 and 4 to Realistic andArtistic types,respectively. Judged on the basis of its face validity alone, the factorstructure of the short form of the TSOSS appears to hold promise as abasis for a comparison of its responses with those to the SDS.



6/16

149

The validity of the short form of the TSOSS has not yet been completelyestablished (Osipow et al., 1993), but the original long form was validatedby Rooney and Osipow (1992) using the Occupational Self-Efficacy Scale

(OSES;Betz & Hackett, 1981). Evidence for the

validityof the OSES has been

provided by the original research and by Zilber (1988). Rooney and Osipow(1992) produced mixed results, with weak (r = .17, p < .0162) to moderate(r = .66, p < .0001) correlations between TSOSS and OSES responses.

Osipow et al. (1993) reported Cronbach alpha reliability for the short formof the TSOSS ranging from .91 to .93 for each of the four factors.

The Self-Directed Search

The SDS (Holland, 1990) has six scales of 38 items each. Holland (1985,p. 49) reports Cronbachs alpha reliabilities for the 1985 revision of theSDS for respondents aged 14 to 18 years by subscale as follows, dependingon gender and personality type:Activities, .59 to .86; Competencies, .67 to.83; Occupations, .74 to .89; Summary, .86 to .92. The Summary scalereliabilities include theActivities (11 items), Competencies (11 items), andOccupations (14 items) subscales, as well as two Self-Estimates ratings.The SDS Manual (Holland, 1985), the Manual Supplement (Holland,

1987), a recent bibliography (Holland and Gottfredson, 1990), and Hollandsbook on his theory (Holland, 1997) provide extensive evidence for thevalidity and reliability of the SDS and for the theory behind its development.

Results

Results were

analyzed usingboth bivariate correlation and

multipleregression.

Correlation analysisPearson product-moment correlations were calculated between the raw

scores for each of the four TSOSS factors and six SDS Summary scores.Values were calculated separately for the male portion of the sample, thefemale portion of the sample, and the total sample. Table 1 shows thesevalues. Means, standard deviations, Pearson correlations, and coefficientalpha reliability estimates for males, females, and the total sample alsoappear in Table 1.

To manage the risk of making a Type I error with the 45 correlationsamong the 6 SDS and 4 TSOSS variables at the .05 familywise level, acrude Bonferroni correction was applied to the significance level, and athreshold alpha level of .001 was used to evaluate the statistical significanceof these correlations.

For the complete sample, the bivariate correlations revealed three SDSSummary scores that were significantly related to TSOSS Factor 1 (verbal,interpersonal skills): Social (r = .32), Enterprising (r = .38), and Conventional(r = .27). Three SDS Summary scores were also significantly related toTSOSS Factor 2 (quantitative, logical, business skills): Investigative (r = .27),

Enterprising (r=

.37), and Conventional (r=

.49). Onlyone

SDS Summaryscore was significantly related to TSOSS Factor 3 (physical strength andagility and Factor 4 (aesthetic skills) respectively: Realistic (r = .55), and

Artistic (r = .48).



7/16

150

m

h0

0e0

~-

GO

0CC~

a)

0o

CO

>~

#u!H 05

H S4~1-4OOOQP-4

Cd.t-) B~CL P?ccSacOQ

~

~cca)~~

.P.4~ 0) !-


8/16

151

OB ~

CdCd Vcj SoO

S9f

~~

Z%

!i#

p$CXniG

r&dquo; V i~ Og t6 uV

1E-4 ~ .0

0 cdr,4 bOCd Cd

mw / M ~

~ V &dquo;, E&dquo;~ V

/b. ~-uflcli cl .

,.q11

P6,cdr4

i

rr O 5,a1. 0

Pr4 Cded

czn ~ CQm


9/16

152

Multiple regression

Using simultaneous multiple regression, each of the four TSOSS Factorscores were regressed in turn on the linear combination of six (R,I,A,S,E,C)

SDS Summaryscores

(see Table 2). These four regressions accounted for19% of the variance in TSOSS Factor 1, F(6, 230) = 9.16, p < .0001, adjustedr2 = .17; 31% of the variance in TSOSS Factor 2, F(6, 230) = 17.04, p < .0001,adjusted r2 = .29; 34% of the variance in TSOSS Factor 3, F(6, 230) = 19.46,p < .0001, adjusted r2 = .32; and 24% of the variance in TSOSS Factor 4,F(6, 230) = 12.40, p < .0001, adjusted r2 = .22.

Uniqueness indices and beta weights (standardized multiple regressioncoefficients)were then calculated to assess the relative importance of the sixSDS Summary scores in their prediction of TSOSS Factor scores. For eachSummary score, the uniqueness index is the unique percent of variance inTSOSS Factor scores accounted for

bythat SDS

Summaryscore,

beyondthe variance accounted for by the other five SDS scores. The standardizedbeta weight for each SDS Summary score represents the average change inTSOSS Factor scores (in standard deviation units) associated with a 1-standarddeviation change in each respective Summary score, while holding theother five Summary scores constant. It is a measure of the relative strengthof the relationship between SDS Summary scores and predicted self-efficacybeliefs, as embodied in the TSOSS Factor scores. Taken together, uniquenessindices and beta weights provide complementary evidence of the relationshipbetween SDS Summary scores and predicted self-efficacy.

Table 2 shows that for this sample, self-efficacy in verbal or interpersonalskills was positively predicted by Enterprising and Social SDS Summaryscores. These Summary scores uniquely accounted for a statisticallysignificant portion (5% and 2%, respectively) of the variance in TSOSSFactor 1 scores. The SDS Enterprising Summary scores have the strongestrelationship with predicted self-efficacy, with a beta weight of .30 (p

Documents

The Self-Directed Search and Career Self-Efficacy.pdf