Identity formation

Published on January 2017 | Categories: Documents | Downloads: 33 | Comments: 0 | Views: 195
of 16
Download PDF   Embed   Report

Comments

Content


The process of exploration in identity formation:
the role of style and competence
ALAN M. BERMAN, SETH J. SCHWARTZ, WILLIAM M. KURTINES
AND STEVEN L. BERMAN
This study drew on two approaches to identity formation, each postulating differing
but complementary exploration components (style and competence), in an effort to
better understand the identity exploration process. The sample for this study (n=215)
was socio-demographically diverse with respect to gender, age, ethnicity and setting,
with participants drawn from two universities in two different settings (rural
residential, and urban non-residential), with differing ethnicities (primarily non-
Hispanic White vs. primarily Hispanic) and varying ages. Overall patterns of results
revealed by hierarchical regression analyses were consistent with a process model of
exploration and with the view that exploration is a multi-dimensional process
comprized of multiple components. The findings also provide evidence for the utility of
both of the critical constructivist and co-constructivist approaches to identity
formation used in this study, and for the validity of the rich clinical/qualitative
literature that gave rise to the construct of exploration.
# 2001 The Association for Professionals in Services for Adolescents
Introduction
Beginning with the theoretical writings of Erikson (1950; 1982), the process of exploration
has been viewed as central to the formation of an identity. The recognition of the importance
of exploration is evident in the considerable empirical work (e.g. Grotevant and Cooper,
1981; Marcia and Archer, 1993) generated by Marcia’s (1966; 1980) pioneering work on the
identity status paradigm. In adapting Erikson’s concept of identity for empirical research,
Marcia identified exploration and commitment as the two basic dimensions for defining the
individual’s status with respect to achieving an identity. By simultaneously considering an
individual’s levels of exploration and commitment, Marcia was able to derive four statuses for
characterizing an individual’s development toward a mature identity. These statuses are
Achievement, Moratorium, Foreclosure, and Diffusion (for further information on the
statuses see Marcia, 1980; 1993; Waterman, 1993). Consistent with the psychosocial
developmental theoretical tradition, Marcia defined exploration as the search for a revised
and updated sense of self, whereas commitment was taken to symbolize ‘‘the [adherence] to a
course of action, a kind of settling down’’ (1988, p. 213). Exploration is thus a process of
examination and discovery of who and what one might be, with commitment to an identity
being a consolidation of this process. As such, exploration might be seen as a basic process
underlying the formation of an identity.
Recent work has begun to investigate exploration and to articulate more fully
its component processes. Grotevant (1987) for example, has proposed a process model
of identity formation in which ‘‘the ‘work’ of identity is seen as the process of exploration. . .
Reprint requests and correspondence should be addressed to William M. Kurtines, Department of Psychology,
Florida International University, University Park Campus, Miami, FL 33199. (E-mail: [email protected]).
0140-1971/01/040513+16 $35Á00/0
# 2001 The Association for Professionals in Services for Adolescents
Journal of Adolescence 2001, 24, 513–528
doi:10.1006/jado.2001.0386, available online at http://www.idealibrary.com on
[where] identity exploration may be defined as problem-solving behavior aimed at eliciting
information about oneself or one’s environment in order to make a decision about an
important life choice’’ (p. 204). Moreover, Grotevant’s process model includes, ‘‘those abilities
and orientations that individuals bring to bear on the identity formation process’’ (pp. 204–
205, emphasis added).
Subsequent work has helped to advance our understanding of the process of exploration
by examining the respective roles that abilities and orientations play in that process.
Berzonsky (1989) and Kurtines (2001), for example, have each proposed theoretical
approaches to identity formation. Both approaches draw on the constructivist tradition
(e.g. Kelly, 1955) and include efforts to conceptualize and operationalize components of the
exploration process. This tradition emphasizes the individual as an intentional agent
proactively participating in the construction of her/his world. These two approaches,
however, focus on differing components of the process of the constructive of the self (i.e. the
ego synthesis) that is the outcome of exploration process. Berzonsky’s critical constructivist
approach focuses on the contribution of personal orientations to identity formation whereas
Kurtines’s co-constructivist approach focuses on the abilities that individuals bring to the
identity formation process. More specifically, the critical constructivist approach of Berzonsky
focuses on the role of information processing style whereas the co-constructivist approach of
Kurtines focuses on the contribution of problem-solving competence to the process of identity
formation. The critical constructivist and co-constructivist approaches thus draw on a similar
theoretical background but propose differing (though perhaps complementary) components
of the process of exploration.
Social-cognitive style: a critical constructivist approach
Berzonsky (1990) has proposed that each individual adopts a characteristic information-
processing style when forming an identity. According to this framework, each individual
adopts one of three styles as his/her characteristic orientation to making identity-relevant
decisions during the process of identity formation. The most facilitative of these, the
Informational style, is characterized by active exploration of alternatives, information seeking,
and flexible commitments. Alternatively, the Normative style displays little exploration of
alternatives and is characterized by attitudes such as subservience, deference to authority
and dogmatic, inflexible commitments. This style represents conformity to social and familial
expectations. Lastly, the Diffuse/Avoidant style is characterized by the avoidance of
exploration or exploration that is unsystematic and characterized by procrastination,
avoidance and the unwillingness to confront the decision-making process.
While each style is associated with a characteristic identity status (i.e. Informational with
Achievement and Moratorium; Normative with Foreclosure, and Diffuse/Avoidant with
Diffusion; Berzonsky, 1989), identity styles tend to be characterized as less developmental
than identity statuses. That is, whereas identity statuses are generally seen as developmental
pathways for resolving the issue of identity vs. role confusion (an issue with a phase-specific
time of ascendance; Waterman and Archer, 1990), styles are more process and individual-
difference oriented in late adolescent and adult populations. Conceptually it has been
postulated that virtually all normal 18-year-olds are capable of utilizing all three social-
cognitive strategies. Style is thus the type of social-cognitive strategy that individuals prefer
in seeking a successful identity resolution (Berzonsky, 1990). Berzonsky (1989; 1997) has
developed a style measure that specifically taps content relevant to identity issues. This self
514 A. M. Berman et al.
report measure asks participants to provide ratings (on a five-point Likert rating scale) of
how they generally tend to handle identity-related problems when they arise.
Problem-solving competence: a co-constructivist approach
Recently, Kurtines (1998) has investigated a competence component in order to further
understand the exploration process. Although it appears highly plausible that higher forms of
cognitive competence might facilitate exploration, research on the role of competence in
identity formation has been scant and inconsistent. As part of a developing program of
research, Kurtines (2001) has proposed that during the developmental process, individuals
acquire a complex set of cognitive and communicative competencies, including the capacity
for critical thinking and discussion. This approach defines intraindividual change as
developmentally normative only up through childhood, emphasizing instead the self-directed
nature of the developmental process in adolescence and adulthood (Brandstadter and Lerner,
1999; Lerner and Busch-Rossnagel, 1981). Thus, as with Berzonsky’s approach, after
adolescence, competencies are conceptualized as process and individual-difference oriented,
with the use of critical competencies hypothesized to contribute uniquely to the exploration
process and to identity formation as a whole.
Drawing on this work, Kurtines (2001) has identified three problem-solving processes
hypothesized to facilitate the exploration process during identity formation. Creativity is the
degree to which the individual is innovative or inventive in generating alternatives for life
choices encountered during the exploration process. Suspension of judgment represents the
degree to which the individual is capable of adopting multiple perspectives with respect to
life choices, that is, considering positive and negative aspects (i.e. ‘‘pros’’ and ‘‘cons’’) for
each alternative. Critical evaluation represents the degree to which the individual is capable
of questioning or challenging alternatives and willing to change one’s original choice in the
context of ‘‘a more viable alternative’’.
Accordingly, a cognitive competence measure that specifically taps content relevant to
identity issues has been developed. This performance-based measure requires participantsto
generate as many potential alternatives as possible for solving life-choice dilemmas,to suspend
judgment by providing justifications for alternatives they disagree with, and to indicate, with
supportive reasoning, the choice that they would make following critical thinking.
Theoretical implications and empirical questions
The availability of two identity perspectives, each postulating differing but complementary
exploration components (and with distinct component-appropriate measurement techni-
ques), opens up considerable potential for knowledge development concerning the
exploration process. As noted, for example, Grotevant’s process model conceptualizes
exploration as a multi-dimensional process comprized of multiple components, a claim that is
consistent with the richness of the clinical/qualitative literature that gave rise to the
construct of exploration (Erikson, 1982). Although the view that the exploration process is
comprized of multiple components appears highly plausible, the focus of research in the
identity formation literature has historically been mainly on the contribution of identity style.
A sizable literature has emerged, for example, establishing a link between identity style and
identity status. Informational style has been found to predict Moratorium and Achieve-
ment (Berzonsky, 1989), Normative style has been found to predict Foreclosure (Berzonsky
and Neimeyer, 1994), and Diffuse/Avoidant style to predict Diffusion (Berzonsky, 1989;
Streitmatter, 1993).
Style and competence 515
A link between cognitive competence (as a component of the exploration process) and
identity status has proved more difficult to establish than has a link between identity style and
identity status. Despite the paucity of evidence for the contribution of higher forms of cognitive
competence to the exploration process, it appears highly unlikely that the use of higher forms
of cognitive competence would not contribute to achieving an identity. Therefore, examining
alternative explanations for this knowledge gap may help to shed light on this issue.
The greater difficulty in establishing a link for competence than for style, for example, may
partially stem from the fact that historically, these constructs have been operationalized using
differing methods of measurement. More specifically, in the identity-formation research
literature, the constructs of identity status and identity style have historically tended to be
measured by the same type of methods, i.e. group administered self-report scales most often
using Likert type rating scales (e.g. Bennion and Adams, 1986; Berzonsky, 1989; Balistreri
et al., 1995). Moreover, measures of these two constructs tend to share similar content with
respect to identity issues, including overlapping items. Such measurement overlap may
inadvertently capitalize on shared method variance and, consequently, overestimate the
linkages between these variables.
Cognitive competence constructs, on the other hand, have historically been measured
mainly using performance-based ‘‘tests of limits’’ (e.g. Flavell and Markman, 1983) rather
than self-report methods, and these performance based measures often lack content
relevance for identity issues. The small research literature that does exist lends some
credibility to the possibility that the lack of shared method variance such as common
measurement source (e.g. similarity of measurement format and/or content) may result in
underestimating the link between cognitive competence and identity status. Reviews of the
literature (e.g. Marcia, 1993) have revealed only moderate success establishing a link
between cognitive competence and identity status, with studies using more socially oriented
measures of cognitive sophistication such as integrative complexity (Slugoski et al., 1984)
and skeptical doubt (Boyes and Chandler, 1992) tending to appear more promising than
studies using more performance-based indices of formal operational thought (e.g. Leiper,
1981; Rowe and Marcia, 1981; Wagner, 1987). Additional research using measures that are
more format and content appropriate thus has the potential to contribute to the classification
of the link between cognitive competence and identity status.
The current study
The aim of the current study was to draw on both the critical constructivist and co-
constructivist approaches to identity formation (with process-appropriate measures) to
empirically investigate whether competence variables make a contribution to the exploration
process beyond that of style variables and, if so, the nature of that contribution. In addition
to drawing on differing conceptual frameworks in an effort to ensure theoretical diversity, an
effort was also made to obtain a sample that was socio-demographically diverse with respect
to gender, age, ethnicity, and setting.
Method
Participants
A total of 215 psychology students (60 males, 155 females) were recruited for this study,
ranging in age from 18 to 25 years. The sample consisted of 103 Whites, 15 African-
516 A. M. Berman et al.
Americans, 75 Hispanics, five Asians, six Native Americans, and 12 Others. Participants
were recruited from two universities: 100 (33 males, 67 females) from a small, private
university in upstate New York; and 115 (30 males, 85 females) from a large, urban public
university in southern Florida.
The majority (n=74) of the Hispanic participants were recruited from the Florida sample,
whereas the majority (n=83) of the non-Hispanic White participants were recruited from
the New York sample. The mean age of the Florida sample was 22?2 years, with a standard
deviation of 7?16, whereas the mean age of the New York sample was 19?5 years, with a
standard deviation of 1?56. Because these age differences were statistically significant
[t (219?21)=3?83, p50?001], age was used as a covariate in all subsequent analyses.
Measures
Demographics. Participants provided their age, gender, year in school, and ethnicity on
a brief demographic form. Because Hispanics and non-Hispanic Whites together comprized
83% of the sample, ethnicity was coded as Hispanic, Non-Hispanic White, or Other.
Identity status. The Ego Identity Process Questionnaire (EIPQ; Balistreri et al., 1995)
was used to identify participants’ identity status. The EIPQ assesses exploration and
commitment within four ideological domains (politics, religion, occupation, and values) and
within four interpersonal domains (friendships, dating, sex roles, and family). This 32-item
self-report measure uses a six-point Likert scale and consists of two exploration items and two
commitment items per domain. The EIPQ yields a total score for exploration and a total
score for commitment. The Cronbach’s alpha value for the overall exploration scale has been
reported as 0?76, and the test–retest reliability coefficient for this scale as 0?90 (Balistreri
et al., 1995). The commitment scale was used in combination with the exploration scale to
assign participants to identity statuses. The Cronbach’s alpha value for the commitment scale
has been reported as 0?75 with a test–retest reliability coefficient of 0?76.
The EIPQ was developed using a six-point Likert-type rating scale. However, due to
constraints involving the standard response sheets available to the authors, a five-point scale
was used to collect the data for this questionnaire. For purposes of analyses, total scores on
exploration and commitment were prorated to their original metric using the following
quantitative transformation, Prorated Score: x/80=y/96, where x is the obtained score, y is
the prorated score, 80 the maximum score on the five-point scale, and 96 the maximum
score on the six-point scale. The converted medians obtained in this study (61?0 for
exploration, 58?0 for commitment) are comparable to those obtained by Balistreri et al.
(1995) in their study validating the EIPQ (66?5 for exploration, 62?0 for commitment).
The EIPQ assigns participants to one of four status categories by means of a median split
technique. For both exploration and commitment, scores falling on or above the median
were classified as ‘‘high’’, the remainder, ‘‘low’’. Identity status categories are assigned based
on these median splits, using medians provided by Balistreri et al. (1995), adjusted for the
five-point Likert scale used in this study. A person who is high on both the exploration and
commitment scales was considered to be in the Achievement status. Someone who is high
on exploration and low on commitment is judged to be in the Moratorium status. A
respondent with the reverse pattern is considered to be Foreclosed. Someone low on both
scales is placed in the Diffusion category.
Style and competence 517
Identity style. The Identity Style Inventory (ISI; Berzonsky, 1997) was used to assess
participants’ identity style. The ISI is a 40-item, group-administered questionnaire, for which
participants rate each item on a five-point Likert scale. The ISI contains 11 items measuring
the informational style, nine items measuring the normative style, 10 items measuring the
diffuse/avoidant style, and 10 items measuring commitment. The commitment scale was not
analyzed for this report.
Cronbach’s alpha coefficients for the ISI scales have been reported as follows:
Informational style, 0?70; Normative style, 66; and Diffuse/avoidant style, 0?76, with good
test-retest reliabilities (Berzonsky, 1997). The alpha coefficients for our sample were:
Informational style, 0?68; Normative style, 70; and Diffuse/avoidant style, 0?72.
Critical problem solving. The Critical Problem Solving Scale (Ferrer-Wreder et al.,
in press) was used to assess participants’ problem solving competence. The CPSS is a group-
administered performance based measure in which participants’ performances on four tasks
are elicited in response to two hypothetical life-choice dilemmas with identity content
relevance (a personal and an interpersonal dilemma) and used to index three problem solving
processes. The responses are used to obtain the following four scores tapping the three problem-
solving processes. Creativity is tapped by the Generation of Alternatives (GA) score, which is
the total number of different and distinct alternatives generated for both dilemmas. Suspension of
judgment is tapped by the Decentering, Positive Alternatives (DPA) and the Decentering,
Negative Alternatives (DNA) scores (the total number of ‘‘cons’’ and ‘‘pros’’ each participant
generated for her/his own ‘‘best’’ and ‘‘worst’’ alternatives). Critical evaluation is tapped by the
Modification (MO) score (the number times a participant changes (modifies) his/her original
choice and provides a ‘‘justification’’ for the change based on his/her previous pros and cons for
that alternative). Responses are coded and scores are assigned by raters who are trained in
coding responses to the CPSS. In the current study, 10 per cent of the responses were coded by
all 10 raters to obtain reliability with the remainder of the responses scored individually.
Interrater reliability in this study, for all four scores and across 10 raters, was 84 per cent. For the
individual scores, the interrater reliability was 82 per cent for the GA score, 86 per cent for the
DPA score, 81 per cent for the DNA score, and 87 per cent for the MO score.
Procedure
Participants were group administered the EIPQ, ISI, CPSS, and an informed consent form in
a classroom setting. Demographic information was obtained as part of the EIPQ.
Results
Descriptive statistics
All continuous scores approximated a normal distribution except for the MO scale, which
was positively skewed (Kolmogorov-Smirnoff Z=6?35, p50?0001)
1
.
The identity status distribution in this sample was as follows: 39 participants were
classified as Diffused (18?1% of the sample), 48 as Foreclosed (22?3%), 48 as Moratorium
(22?3%), and 80 as Achieved (37?2%). There was no significant effect of age [F(3,
211)=0?65, ns] or year in school [F(3, 211)=0?57, ns] on identity status in this sample. Both
1
The MO scale was not used in subsequent analyses because it was not normally distributed.
518 A. M. Berman et al.
gender [w
2
(3, 215)=8?61, p50?04] and ethnicity [w
2
(6, 215)=16?31, p50?02] were
related to identity status. With respect to gender, 25?3 per cent of females, as opposed to
10?2 per cent of males, were classified into the Foreclosed status, whereas 27?1 per cent of
males, as opposed to 14?8 per cent of females, were classified into the Diffused status. With
respect to ethnicity, 31?0 per cent of Hispanics, as opposed to 19?5 per cent of non-Hispanic
Whites, were classified as Foreclosed, whereas 26?5 per cent of non-Hispanic Whites, as
opposed to 14?7 per cent of Hispanics, were classified as Moratorium. Gender and ethnicity
2
were thus entered as control variables in subsequent analyses.
Interrelationships between style and competence variables
Correlational analyses were used to investigate the interrelationship between the style and
competence variables (and among the variables within each group). Table 1 presents the
correlation matrix. The EIPQ exploration and commitment scores were also included in the
matrix because: (a) exploration scores were used as the dependent variable in some analyses;
and (b) exploration and commitment scores from the EIPQ were used to derive status
assignments.
3
The contribution of style and competence components to exploration
The analyses described in this section provided the opportunity to empirically investigate
whether competence variables make a contribution to the exploration process beyond that of
style variables and the nature of the contribution across all statuses. To do so, hierarchical
multiple regression analyses were used to estimate the unique contribution attributable to
the variables associated with each of the hypothesized exploration components (style and
competence). The data analytic strategy was to conduct hierarchical multiple regression
analyses using the EIPQ total exploration score as the dependent variable and with the
demographic variables (age, gender, year in school, ethnicity and data source), CPSS
variables (GA, DPA, and DNA), and ISI variables (Informational, Normative, and Diffuse/
Avoidant) as the predictor variables.
Demographic variables (age, gender, year in school, ethnicity and data source) were
entered into the regression hierarchy as the first block to serve as control variables, followed
by the block of CPSS variables and the block of ISI variables. The order of entry of the blocks
of variables was selected based on the expected relative strength of the competence and style
variables as predictor variables. As noted previously, substantial empirical research has
2
Because the majority of participants in this study were either Hispanic or non-Hispanic White, ethnicity was
dummy coded into three dichotomous variables: White, Hispanic, and Other. Of these three dummy coded
variables, only White and Hispanic were entered into the regression models as predictors. As Other as an ethnicity
variable was comprised of a small group of extremely diverse and ethnically heterogeneous participants, their
inclusion in the analyses would have rendered the results conceptually difficult to interpret.
3
In addition to the correlational analyses, three exploratory factor analyses (principal components, varimax
rotation) were also conducted to begin examine the factor structure of the two measures. The results provided
preliminary evidence that the measures assess constructs that are empirically distinct. When factor analysed, the
three ISI variables yielded a single factor (eigenvalue=1?30, 43?3% explained variance) with all three style scales
loading on this factor (Informational, 0?75, Normative 0?37, Diffuse/Avoidant, 70?78); the three CPSS variables
yielded a single factor (eigenvalue=1?68, 56?0% explained variance), with all three competence scales loading on
this factor (GA, 0?77; DPA, 0?78; and DNA, 0?69); the six ISI and CPSS variables, when factor analysed together,
yielded two factors: Factor 1 (eigenvalue=1?91, 31?9% explained variance) with positive loadings for all three CPSS
scales (GA, 0?76; DPA, 0?71; and DNA, 0?63) and a negative loading for the Normative Style scale (70?57); Factor
2 (eigenvalue=1?23, 20?5% explained variance) with a positive loading for the Diffuse/Avoidant style scale (0?79)
and a negative loading for the Informational Style scale (70?79). A complete description is available from the first
author.
Style and competence 519
established the contribution of style to the exploration process, whereas the contribution of
competence has been less clearly established in the literature. Consequently, the style
variables were expected to account for the largest amount of explained variance, with the
contribution of the competence variables less clearly specifiable in advance. The analytic
strategy was therefore to enter the competence variables into the regression hierarchy first
after the control variables to provide an estimate of what (if any) contribution the
competence variables made beyond the control variables. The block of style variables were
then entered to provide an estimate of the contribution of the style variables and what
(if any) of the remaining contribution the competence variables was uniquely attributable to
them. Table 2 presents the results of this analysis.
When the block of the control variables was entered into the regression hierarchy, the
overall regression equation was not significant, and none of the variables in that block were
significant predictors of exploration. With the addition of the block of CPSS variables, the
overall equation indicated a trend [adjusted R
2
=0?08, F(8, 177)=1?73, p50?09; R
2
change=0?06, F(3, 172)=3?18, p50?03], with the Generation of Alternatives score
emerging as a significant positive predictor of exploration (b=0?19, t=2?58, p50?02). With
the addition of the block of ISI variables, the overall equation became significant [adjusted
R
2
=0?47, F(11, 174)=13?90, p50?001; R
2
change=0?37, F(3, 169)=39?68, p50?001],
with the Generation of Alternatives score (b=0?16, t=2?12, p50?04) remaining a
significant predictor. Finally, both the Informational (b =0?59, t=9?83, p50?001) and
Normative (b =70?23, t=73?71, p50?001) styles emerged as significant predictors of
exploration.
The contribution of specific style and competence components
to identity status
The data analyses described in this section provided an opportunity to extend the findings of
the previous section as well as to begin to investigate in more depth the relative contribution
of style and competence to identity status. The results reported in the previous section
contributed to our understanding of role of style and competence to exploration as a general
process but not to our understanding of the role of style and competence as exploration
processes specific to each particular identity status. To this end, hierarchical multiple
regression analyses were again used, only in this case to investigate the unique contribution
of the variables associated with each of the specific exploration components (style and
competence) to particular identity statuses. As in the previous analyses, the demographic
Table 1 Intercorrelation matrix among style and competence variables
2 3 4 5 6 7 8
1. Informational Style 0?1370?27**70?08 70?01 70?06 0?57** 0?17*
2. Normative Style 70?07 70?32**70?14* 70?22**70?16* 0?57**
3. Diffuse/Avoidant Style 0?09 0?16* 0?08 70?15* 70?23**
4. Generation of Alternatives 0?33** 0?43** 0?12 70?37**
5. Decentering, Negative Alternatives 0?33** 0?13* 70?17*
6. Decentering, Positive Alternatives 0?09 70?29**
7. Exploration 70?05
8. Commitment
*p50?05; **p50?01; ***p50?001.
520 A. M. Berman et al.
variables, CPSS variables, and ISI variables were entered, in that order, as blocks. For
purposes of these analyses, a dichotomous variable was created for each identity status to
serve as the dependent variable for that status. These dichotomous variables were created for
each status by assigning a score of 1 to each participant who was assigned to that status and a
score of 0 to all other participants. The analyses described in this section thus provided the
opportunity to estimate the unique contribution of style and competence to each identity
status relative to all other statuses combined. Table 3 presents the results of these analyses.
Diffused status. When Diffused status was used as a dependent variable, the addition of
the control and CPSS block variables did not make a significant contribution to accounting
for membership in that status. When the ISI block variables were added to the regression
equation, the result was a significant equation [adjusted R
2
=0?15, F(14, 200)=3?25,
p50?001; R
2
change=0?14, F(3, 169)=9?98, p50?001]. The beta weights for the equation
indicated that Informational style score was a significant negative predictor of Diffused status
(b=70?36, t=74?75, p50?001).
Foreclosed status. When membership in the Foreclosed status was used as a dependent
variable, the addition of the control variables did not yield a significant equation. However,
the beta weight for gender emerged as a significant predictor of Foreclosure (b=0?14,
t=2?07, p50?04), with females more likely to be Foreclosed than males. The addition of the
block of CPSS variables yielded a significant overall equation [adjusted R
2
=0?10, F (8,
172)=2?29, p50?025; R
2
change=0?05, F(3, 172)=3?71, p50?02] with none of the
individual CPSS beta weights reaching significance. The addition of the ISI variables yielded
a significant equation [adjusted R
2
=0?18, F(14, 200)=3?91, p50?001; R
2
change=0?15,
F(3, 169)=10?95, p50?001]. The beta weights for the equation indicated that scores on
both the Informational (b=70?28, t=73?79, p50?001) and Normative (B=0?32, t=4?40,
p50?001) styles were significant predictors of Foreclosure, with Informational in a negative
direction and Normative in a positive direction. The beta weight for gender remained
significant and positive (b=0?14, t=2?04, p50?05).
Table 2 Regression of exploration by style and competence
Block Variables entered Partial R
2
b (final step) DR
2
Control 0?02
Gender 0?01 0?03
Age 0?00 70?01
Year in school 0?01 0?08
Ethnicity (White) 0?01 0?16
Ethnicity (Hispanic) 0?00 0?07
Data source 0?00 0?09
Competence 0?08
Generation of Alternatives 0?02 0?16*
Decentering Negative Alternatives 0?00 0?04
Decentering Positive Alternatives 0?01 0?11
Style 0?37***
Informational 0?30 0?59***
Normative 0?04 70?22***
Diffuse/Avoidant 0?00 70?04
*p50?05; **p50?01; ***p50?001.
Style and competence 521
Table 3 Regression of identity status by style and competence
Block Variables entered Partial R
2
b (final step) DR
2
Diffused
Control 0?05
Gender 0?00 70?07
Age 0?01 0?10
Year in school 0?00 70?08
Ethnicity (White) 0?00 0?08
Ethnicity (Hispanic) 0?00 0?08
Data source 0?02 70?22
Competence 0?02
Generation of Alternatives 0?01 70?11
Decentering Negative Alternatives 0?01 0?15
Decentering Positive Alternatives 0?00 0?01
Style 0?14***
Informational 0?11 70?36***
Normative 0?00 70?07
Diffuse/Avoidant 0?00 0?06
Foreclosed
Control 0?05
Gender 0?02 0?14*
Age 0?00 0?08
Year in school 0?00 0?01
Ethnicity (White) 0?00 70?09
Ethnicity (Hispanic) 0?01 70?12
Data source 0?00 0?00
Competence 0?05*
Generation of Alternatives 0?00 70?08
Decentering Negative Alternatives 0?01 70?11
Decentering Positive Alternatives 0?00 70?07
Style 0?15***
Informational 0?06 70?27***
Normative 0?09 0?32***
Diffuse/Avoidant 0?00 70?03
Moratorium
Control 0?04
Gender 0?00 70?07
Age 0?00 70?07
Year in school 0?01 0?09
Ethnicity (White) 0?00 70?01
Ethnicity (Hispanic) 0?00 0?05
Data source 0?01 70?18
Competence 0?05*
Generation of Alternatives 0?02 0?21*
Decentering Negative Alternatives 0?00 70?04
Decentering Positive Alternatives 0?00 70?03
Style 0?14***
Informational 0?03 0?20*
Normative 0?11 70?37***
Diffuse/Avoidant 0?00 70?01
Achieved
Control 0?05
Gender 0?00 0?01
Age 0?00 70?06
Year in school 0?01 70?11
522 A. M. Berman et al.
Moratorium status. When membership in the Moratorium status was used as a
dependent variable, the addition of the control variables did not make a significant
contribution to status. The addition of the CPSS variables produced a significant R
2
change
[R
2
change=0?05, F(3, 172)=2?87, p50?04], although the overall model remained non-
significant. When the CPSS variables were added, the Generation of Alternatives score was
a significant positive predictor of Moratorium status (b=0?20, t=2?54, p50?02). When the
ISI variables were added, both the Informational (b=0?20, t=2?50, p50?02) and Normative
(b=70?37, t=74?85, p50?001) styles were again significant predictors. The Generation of
Alternatives score (b=0?21, t=2?14, p50?04) remained a significant predictor, and the
overall equation remained significant [adjusted R
2
=0?15, F(14, 200)=3?32, p50?001;
R
2
change=0?14, F(3, 169)=10?35, p50?001].
Achieved status. When membership in the Achieved status was used as a dependent
variable, the addition of the control variables did not make a significant contribution to
status. When the CPSS variables were added, the R
2
change was not significant, but the
Generation of Alternatives score was a significant predictor of Achievement status, this time
in a negative direction (b=70?17, t=72?12, p50?04), although the overall model was not
significant. Finally, when the ISI variables were added, Informational (b=0?35, t=5?17,
p50?001) style was a significant predictor, Generation of Alternatives (b=70?23,
t=72?23, p50?03) remained a significant predictor, and the overall equation remained
significant [adjusted R
2
=0?11, F(14, 200)=2?67, p50?005; R
2
change=0?13,
F(3, 169)=8?64, p50?001].
Discussion
The aim of the current study was to begin to shed light on the ways that specific exploration
components are related to exploration in general and identity statuses in particular, especially
with respect to the competence component. It was noted that a sizable literature has
established a link between identity style and identity status, but the existence of a link
between cognitive competence and exploration has been less clear. Thus, although it would
Table 3 (Continued)
Block Variables entered Partial R
2
b (final step) DR
2
Ethnicity (White) 0?01 0?16
Ethnicity (Hispanic) 0?00 0?09
Data source 0?01 0?21
Competence 0?02
Generation of Alternatives 0?02 70?23*
Decentering Negative Alternatives 0?00 0?03
Decentering Positive Alternatives 0?00 0?07
Style 0?13***
Informational 0?10 0?35***
Normative 0?01 0?11
Diffuse/Avoidant 0?00 70?01
*p50?05; **p50?01; ***p50?001.
Style and competence 523
appear highly plausible that the process of exploration might be facilitated by higher forms of
cognitive competence, the few possible results that have been reported in the literature have
been less than promising and difficult to disentangle from potential measurement source
issues.
This study provided the opportunity to begin to investigate in a more systematic way than
previously possible whether or not competence variables make a contribution to the
exploration process beyond that of style variables and the nature of that contribution. By
drawing on two identity perspectives, each postulating differing but complementary
exploration components (and with distinct component-appropriate measurement techni-
ques), this study was able to empirically estimate (through the use of hierarchical regression
analyses) the unique contribution of style and competence to overall exploration and to each
identity status relative to all other statuses combined.
The relative contribution of style and competence to exploration
and identity status
Within the limited range of theoretical perspectives (and associated measures) used in this
study and limitations on sample size and diversity, the findings from the study indicate that
although style variables tended to account for more variance, competence variables
consistently accounted for a small but statistically significant proportion of unique variance.
Moreover, the overall pattern of results from these analyses provided a cogent and plausible
pattern indicating style and competence to be two distinct components of the exploration
processes that operated in ways that were consistent with the differing theoretical
expectations of both the identity approaches used in this study.
For example, with respect to the general process of identity exploration, the results
provided support for the expectations of both perspectives. Two of the style variables and one
of the competence variables contributed significantly to total score variance. Consistent with
the critical constructivist perspective, the Informational and Normative styles were
significantly related to total exploration score, with Informational in a positive direction
and Normative in a negative direction. As expected, style variables generally accounted for
more variance than competence variables. Similarly, consistent with the co-constructivist
perspective, Generation of Alternatives was significantly and positively related to
exploration.
With respect to the unique contribution of the variables associated with each of the
specific exploration components (style and competence) to particular identity statuses, the
results also provided support for the differing theoretical expectations of both perspectives.
For example, the results of these within-group analyses indicated that style variables were the
only measures significantly predictive of the diffusion and foreclosure statuses. Moreover,
consistent with the theoretical expectations of the critical constructivist perspective, the
pattern of results indicated that Informational style was significantly and negatively
associated with both the diffused and the foreclosed status while the Normative style was
associated with the foreclosed status in a positive direction. None of the competence
variables were significantly predictive of either diffused or foreclosed status. Although it was
neither clearly consistent nor inconsistent with the co-constructivist perspective, this finding
does suggest future theoretical considerations. More specifically, it suggests that in the future,
theoretical formulations of the role of competence in identity exploration might consider
articulating more fully the differential role of competence variables, particularly for the
diffused and foreclosed statuses, which are defined by a lack of exploration.
524 A. M. Berman et al.
The results of the within group analyses for the moratorium and achieved statuses similarly
provided support for the theoretical expectations of both perspectives, particularly with
respect to the differing roles that style and competence play in identity formation processes.
The results indicated that, again consistent with the critical constructivist perspective, the
Informational and Normative styles were significantly related to the moratorium status, with
Informational in a positive direction and Normative in a negative direction, while only
Informational was significantly related to the Achieved status. In contrast to the findings
with the diffused and foreclosed statuses, Generation of Alternatives was significantly related
to both the moratorium and achieved statuses, though generally accounting for less variance
than the style variables. In addition, Generation of Alternatives was positively related to
Moratorium status but negatively related to Achieved status.
The availability of multiple identity perspectives with differing but complementary
components of the exploration process and component-appropriate measures thus facilitates
addressing theoretical issues that have been identified as important for the field.
Additionally, these perspectives also help point to other potentially productive areas for
fostering future theoretical and empirical advances concerning the process of identity
exploration. The fact that style contributes significantly to placement in all four statuses
while competence does not, for example, raises intriguing questions with respect to the
interaction between the basic qualities of style and competence variables and the qualities
that define the statuses. In this case, for example, the differences in the contribution of style
and competence may exist, in part, because although both style and competence are
conceptualized as process and individual-difference oriented variables, the concept of style
appears to imply a more stable orientation. Style variables, consequently, may tend to be
more trans-situational and generalize across statuses. Competence variables, particularly of
the problem solving type, on the other hand, tend to come into play in solving particular
problems and hence tend to be more situational. Consequently, competence variables may be
more likely to come into play for those statuses that involve more specific and active types of
problem solving processes, i.e. the moratorium and achieved statuses.
Measurement issues
The results of this study also help to disentangle potential measurement source issues.
Although the study was not designed as a direct test of the degree to which content overlap
between self-report measures of identity status and identity style may inadvertently capitalize
on shared method variance and, consequently, over-estimate the linkages between these two
constructs, it did provide a direct test of the contribution of the performance based
competence measures to identity exploration and status relative to the self-report measures
of style. Moreover, although the pattern of results indicated that the contribution of
competence was modest when compared to the style measure, the absence of shared method
variance between the self-report measure used to assess identity exploration and identity
status and the performance-based measure of cognitive competence increases the likelihood
that the obtained results accurately estimate (rather than over-estimate) the linkages
between these two constructs.
Gender and ethnicity
Concerns about the impact of gender and ethnicity have recently come to the forefront in
several literatures including the identity literature (e.g. Patterson et al., 1992; Phinney and
Rosenthal, 1992). The current study contributed to this ongoing discussion by investigating
Style and competence 525
the possible impact of gender and ethnicity on the variables used in the study. In the current
study, an effort was made to obtain a sample that was socio-demographically diverse with
respect to gender, age, ethnicity, and setting. Participants were drawn from two universities in
two different settings (rural, residential, and urban, non-residential) with differing ethnicities
(primarily non-Hispanic White vs. primarily Hispanic) and varying age means and ranges. All
analyses investigated the potential effects of these variables.
The results of the analyses offer suggestions for furture research, even though the effects of
these variables tended to be somewhat limited, at least for the variables and populations used
in this study. For example, the results indicated that both gender and ethnicity were
significantly related to identity status with more females foreclosed and more males diffused
and more Hispanics foreclosed and more non-Hispanic Whites in moratorium. The pattern
of greater foreclosure in Hispanics is consistent with other findings of more foreclosure in
ethnic minorities while the pattern of gender differences with respect to foreclosed in this
study are more similar to pre-1977 studies as reviewed by Marcia (1993). Perhaps this reflects
the traditional atmosphere of many Hispanic families. These findings are worthy of follow-up
in future research.
Additional findings indicated that, although both gender and ethnicity were
significant related to identity status classification, this effect carried over to the exploration
variables only for gender in the foreclosed status. Although the apparent robustness
of the gender effect for the foreclosed status might be limited to this particular sample’s
ethnic composition, it is finding that is also worthy of follow-up in future research.
Future research should thus continue to investigate the impact of culture and ethnicity
as well as gender by targeting samples with greater diversity in gender and ethnicity
to test whether the findings obtained in this study will generalize to more diverse
samples.
Practical implications
Empirical research that addresses such issues not only has theoretical and empirical
significance but considerable practical significance as well. Specifically, the emergence
of a growing interest in developing and evaluating intervention programs to facilitate the
identity formation process (e.g. Archer, 1994) has called attention to the need for
more detailed and specific knowledge of effective intervention strategies. Because
exploration is seen as the ‘‘work’’ of identity formation (Grotevant, 1987) then, for
example, increasing exploration would appear to be in many cases a useful intervention
strategy. Thus, the development of workable and effective identity interventions requires
knowledge concerning specific components of the exploration process (e.g. style,
competence, etc.) and how such processes are related to identity status in the
normal course of development. An understanding of exploration components and how they
operate would help to ensure treatment integrity and specificity in such interventions.
Moreover, an empirically derived knowledge base would be useful in addressing general and
practical questions regarding the development of effective interventions. Examples of such
questions include whether it is more useful to focus on developing ‘‘generic’’ intervention
strategies that can be used with any population, or more effective to target specific types of
processes (style, competence, etc.) when working with specific populations (diffused,
foreclosed, moratorium).
526 A. M. Berman et al.
Conclusion
The overall patterns of the findings from this study appear to be consistent with Grotevant’s
(1987) process model of exploration and with the claim that exploration is a multi-
dimensional process comprized of multiple components. These findings also provide support
for the utility of both of the theoretical approaches to identity formation used in this study.
The pattern of these quantitative results validates the richness of the experience reported in
the clinical/qualitative literature that gave rise to the construct of exploration.
References
Archer, S. L. (1994). (Ed.). Interventions for adolescent identity development. Thousand Oaks, CA: Sage.
Balistreri, E., Busch-Rossnagel, N. A. and Geisinger, K. F. (1995). Development and preliminary
validation of the Ego Identity Process Questionnaire. Journal of Adolescence, 18, 179–190.
Berzonsky, M. D. (1989). Identity style: conceptualization and measurement. Journal of Adolescent
Research, 4, 267–281.
Berzonsky, M. D. (1990). Self-construction over the lifespan: a process perspective on identity
formation. In Advances in personal construct theory, Volume 1, G. J. Neimeyer and R. A. Neimeyer
(Eds). Greenwich, CT: JAI Press, pp. 155–186.
Berzonsky, M. D. (1997). Identity development, control theory, and self-regulation: an individual
differences perspective. Journal of Adolescent Research, 12, 347–353.
Berzonsky, M. D. and Neimeyer, G. J. (1994). Ego identity status and identity processing orientation:
The mediating role of commitment. Journal of Research in Personality, 28, 425–435.
Bennion, L. D. and Adams, G. R. (1986). A revision of the extended version of the Objective Measure
of Ego Identity Status: an identity instrument for use with late adolescents. Journal of Adolescent
Research, 1, 183–198.
Boyes, M. C. and Chandler, M. J. (1992). Cognitive development, epistemic doubt, and identity
formation in adolescence. Journal of Youth and Adolescence, 21, 277–304.
Brandtstadter, J. and Lerner, R. M. (1999). Action and self-development: theory and research through the
life-span. Thousand Oaks, CA: Sage.
Erikson, E. H. (1950). Childhood and society. New York: Norton.
Erikson, E. H. (1982). The life cycle completed: a review. New York: Norton.
Ferrer-Wreder, L. A., Cass-Lorente, C., Kurtines, W. M., Briones, E., Bussell, J. R., Berman, S. L. and
Arrufat, O. Promoting identity development in marginalized youth: transformative education.
Journal of Adolescent Research. In press.
Flavell, J. H. and Markman, E. M. (1983). Handbook of child psychology: cognitive development (Vol. 3).
New York: Wiley.
Grotevant, H. D. (1987). Toward a process model of identity formation. Journal of Adolescent Research,
2, 203–222.
Grotevant, H. D. and Cooper, C. R. (1981). Assessing adolescent identity in the areas of occupation,
religion, politics, friendships, dating, and sex roles: manual for the administration and coding of the
interview. Journal Supplement Abstract Service Catalog of Selected Documents in Psychology, 11, 52–53
(Ms. No. 2295).
Kelly, G. A. (1995). A theory of personality: the psychology of personal constructs. New York: Norton.
Kurtines, W. M. (1998). Human behavior and development: a co-constructivist approach. Unpublished
research monograph, Florida International University, Miami.
Lerner, R. M. and Busch-Rossnagel, M. A. (1981). Individuals as producers of their development:
conceptual and empirical bases. In Individuals as producers of their development: a life span perspective,
R. M. Lerner and M. A. Busch-Rossnagel (Eds). New York: Academic Press, pp. 1–36.
Leiper, R. N. (1981). The relationship of cognitive developmental structures to the formation of ego identity
in young men. Unpublished doctoral dissertation. Simon Fraser University, Burnaby, British
Columbia, Canada.
Style and competence 527
Marcia, J. E. (1966). Development and validation of ego identity status. Journal of Personality and Social
Psychology, 5, 551–558.
Marcia, J. E. (1980). Identity in adolescence. In Handbook of adolescent psychology, J. Adelson (Ed.).
New York: Wiley, pp. 159–187.
Marcia, J. E. (1988). Common processes underlying ego identity, cognitive/moral development, and
individuation. In Self, ego, and identity Integrative approaches, D. K. Lapsley and F. C. Power (Eds).
New York: Springer-Verlag, pp. 211–266.
Marcia, J. E. (1993). The ego identity status approach to ego identity. In Ego identity: a handbook for
psychosocial research, J. E. Marcia, A. S. Waterman, D. R. Matteson, S. L. Archer and J. L. Orlofsky
(Eds). New York: Springer-Verlag, pp. 1–21.
Marcia, J. E. and Archer, S. L. (1993). The Identity Status Interview, late Adolescent College Form. In
Ego identity: a handbook for psychosocial research, J. E. Marcia, A. S. Waterman, D. R. Matteson,
S. L. Archer and J. L. Orlofsky (Eds). New York: Springer-Verlag, pp. 205–240.
Patterson, S. J., Sochting, I. and Marcia, J. B. (1992). The inner space and beyond: women and identity.
In Adolescent identity formation: advances in adolescent development, G. R. Adams, T. P. Gullotta
and R. Montemayor (Eds.). Newbury Park, CA: Sage pp. 9–24.
Phinney, J. S. and Rosenthal, D. A. (1992). Ethnic identity in adolescence: process, context, and
outcome. In Adolescent identity formation: advances in adolescent development, G. R. Adams,
T. P. Gullotta and R. Monetemayor (Eds.). Newbury Park, CA: Sage, pp. 145–172.
Rowe, I. and Marcia, J. E. (1980). Ego identity status, formal operations, and moral development.
Journal of Youth and Adolescence, 9, 87–99.
Slugoski, B. R., Marcia, J. E. and Koopman, R. F. (1984). Cognitive and social interactional
characteristics of ego identity statuses in college males. Journal of Personality and Social Psychology,
47, 646–661
Streitmatter, J. L. (1993). Identity status and identity style: a replication study. Journal of Adolescence,
16, 211–215.
Wagner, J. A. (1987). Formal operations and ego identity in adolescence. Adolescence, 22, 23–35.
Waterman, A. S. (1993). Developmental perspectives on identity formation: from adolescence to
adulthood. In Ego identity: a handbook for psychosocial research, J. E. Marcia, A. S. Waterman,
D. R. Matteson, S. L. Archer and J. L. Orlofsky (Eds). New York: Springer-Verlag, pp. 42–68.
Waterman, A. S. and Archer, S. L. (1990). A life-span perspective on identity formation: developments
in form, function, and process. In Life-span development and behaviour, vol. 10, P. B. Baltes,
D. L. Featherman and R. M. Lerner (Eds). Hillsdale, NJ: Erlbaum, pp. 29–57.
528 A. M. Berman et al.

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close