Journal o f Youth and Adolescence, Vol. 10, No.

6, 1981

From Trust to Intimacy: A New Inventory

for Examining Erikson's Stages of
Psychosociai Development

Doreen A. Rosenthal, 1.4 Ross M. Gurney, 2 and Susan M. Moore 3

Received August 31, 1981

A new inventory f o r examining the first six ofErikson's psychosocial stages

is described. The self-report questionnaire, developed in a pilot study o f 97
adolescents and tested in a study o f 622 adolescents, has 12 items f o r each
subscale. Measures of reliability and validity are reported. It is concluded
that the Erikson Psychosocial Stage Inventory (EPSI) is a useful measure
for researchers interested in development from early adolescence and in
mapping changes as a function o f life events.

INTRODUCTION

This report describes the development of a new inventory for examin-

ing the first six of Erikson's psychosocial stages. Erikson's life-span devel-
opmental theory is well known and will not be described in detail; a brief

This research was supported by a grant from the Education Research and Development Com-
mittee.
*Lecturer, Department of Psychology, University of Melbourne. Received Ph.D. from Uni-
versity of Melbourne. Current research interests are adolescent adjustment, ethnicity, and
sex-role psychology.
2Research Assistant, Department of Psychology, University of Melbourne. Received Ph.D.
from University of Melbourne. Current research interests are effects of unemployment and
adolescent adjustment.
~Senior Lecturer, Department of Psychology, Melbourne State College. Received Ph.D. from
Florida State University. Current research interests are sex-role psychology and adolescence.
4Correspondence should be sent to D. A. Rosenthal, Department of Psychology, University of
Melbourne, Parkville, Victoria 3052, Australia.
525
0047-2891/81/ 1200.0525503.00/0~ 1981PlenumPublishingCorporation
526, Rosenthal, Gurney, and Moore

summary follows. The theory is cast in eight relatively separate "epigenetic"

stages (Erikson, 1959, 1963, 1968), each of which has its own distinctive
goal to be attained if "healthy" personality is to be achieved. Each stage rep-
resents a critical period of conflict and possible crisis for the emergence of
an ego quality such as trust, initiative, or identity. Each stage arises because
a new dimension of social interaction becomes possible with increasing
maturity. Infancy and childhood span the first four core conflicts: trust
versus mistrust, autonomy versus shame and doubt, initiative versus guilt,
and industry versus inferiority. In the adolescent period the conflict is
between identity and identity confusion, and in adulthood the core conflicts
include intimacy versus isolation, generativity versus self-absorption, and
integrity versus despair. Adolescence is regarded by Erikson (1968) as
central to his theory because when the individual reaches this stage, the use-
fulness of identification as a mode of adjustment ends and identity forma-
tion proper begins. If the adolescent does not succeed in forming a strong
identity-rooted in family, race, or ideology-adulthood becomes very dif-
ficult, with genuine intimacy being almost impossible and stable long-term
relationships unlikely.
Erikson is not without his critics, and one of the obvious drawbacks
from an experimental point of view has been that the theory has relied for
validation more upon subjective clinical impression and logical argument
than upon empirical data. However, there have been several attempts to
measure development according to Erikson's theory. Prelinger and Zimet
(1964) attempted to fit ratings of qualitative clinical material into an Erik-
sonian framework of stage development. Such clinical judgments are un-
wieldy and make replication of research difficult; and, to our knowledge,
no reliability and validity data for their instrument are available. Wessman
and Ricks (1966) used a Q-sort technique, but again little appears to have
been done to establish reliability and validity. A questionnaire approach
was adopted by both Rasmussen (1974) and Constantinople (1969). In all
these studies the subjects were young adults, so data are not available on the
applicability of these scales to younger age groups. In addition, Constan-
tinople's scale was originally developed for a male college population and
may not be appropriate for females. Ciacco (1971), like Prelinger and
Zimet, used a clinical approach, stimulating children to tell stories which
were then classified by a panel of judges according to Erikson's first five
stages of psychosocial development. Although some validating evidence
was obtained, this type of research poses problems for those seeking
normative data.
A considerable amount of research activity, most notably by Marcia
(1966), has centered on the fifth stage, that is, adolescence, and the identity
versus identity confusion crisis. Marcia used an interview to establish
identity status by assessing an individual's degree of crisis and commitment
From Trust to Intimacy 527

in relation to occupation and ideology. A number of studies have been

reported in which his interview has been used, of which Bourne (1978b) had
located more than 40. Much of the research using Marcia's interview has
been designed to examine the relation between ego identity status and
selected personality, cognitive, and behavioral variables. Bourne (1978a), in
reviewing these studies with respect to intelligence, concluded that "in some
samples and on some cognitive tasks subjects toward the identity achieve-
ment end of the continuum appear to perform better" (p. 236). This quali-
fied conclusion can be extended to the relation between identity status and
other domains. Thus, identity achievement appears to be related, for
example, to more satisfactory interpersonal relationships (Marcia, 1976),
academic achievement (Cross and Allen, 1970), and moral development
"(Podd, 1972); but not consistently to self-esteem (Marcia and Friedman,
1970; Orlofsky, I978) or locus of control (Matteson, 1974). Failures to rep-
licate earlier findings and the unexpected nature of some results raise ques-
tions about the validity of Marcia's interview-based measures.
Marcia's interview, limited to the stage of identity versus identity con-
fusion, may be useful for those seeking a more detailed examination of that
stage, but it obviously has less merit for those wishing to take a broader
approach. Indeed, since in Erikson's view the crisis typically is resolved
towards the end of a stage, it might be argued that the application of
Marcia's ego identity status approach to any but older adolescents is inap-
propriate. Other criticisms of the construct and external validity of Marcia's
identity statuses, as well as of the reliability of the interview have been
offered by Bourne (1978b), who concluded that "certain difficulties hamper
the conceptual-operational formulation of ego identity developed by
Marcia and his collaborators" (p. 382). Nevertheless, Bourne conceded that
while methodological criticisms can be made, the identity status interview is
no less adequate a measure than any other used in this research field.
Development of the present inventory arose from the need for a mea-
sure of Erikson's psychosocial stages suitable for administration to a large
sample of subjects in early and late adolescence. A study of adolescent ad-
justment in the light of ethnic differences, school attitudes, social networks,
and parental attitudes was to be undertaken; and while other inven-
tories were available--especially those of Offer (1969) and Greenberger and
Sorensen (1974)-one which adhered more closely to Erikson's conceptual
framework was wanted. While both Offer and Greenberger and Sorensen
drew on Erikson's ideas, they have developed their own models of
adolescence. These models have considerable appeal, but in the opinion of
the present authors each is limited by treating adolescence out of the context
of total life-span development. Moreover, the scales of Offer and Green-
berger and Sorensen were considered useful supplements to a scale based
more directly on Erikson's theory.
528 Rosenthal, Gurney, and Moore

A second justification for developing the present inventory was that in

an earlier study of the impact of unemployment on school leavers, Gurney
(1980) found that a modification of Constantinoples's measure revealed
some unexpected and interesting effects relating the experience of unem-
ployment to changes in scores for the industry versus inferiority, identity
versus identity confusion, and trust versus mistrust subscales of the ques-
tionnaire. An apparent "regression" in terms of resolution of the trust crisis
emphasized the importance of including measures of earlier psychosocial
resolutions, as well as of identity, in studies of adolescence. The internal
reliabilities of Gurney's subscales were lower than desirable, so in order to
investigate further the findings reported by him, a psychosocial stage
inventory with stronger reliability and validity was needed.

METHOD

Description of the EPS!

The Erikson Psychosocial Inventory Scale (EPSI) has six subscales

based on the first six of Erikson's stages. Each subscale has 12 items, half of
which reflect successful and half unsuccessful resolution of the "crisis" of
the stage. The items are randomly ordered and presented in a questionnaire
format suitable for group or individual administration to respondents of
about 13 years of age and above. Time required to complete the inventory is
approximately 20 minutes. Respondents are asked to tick one of five
positions from "almost always true" (5) to "hardly ever true" (1) on a Likert
rating scale for each item. Scores on each subscale are used to yield a profile
of scores for each respondent. This is considered more meaningful than
computing an overall "psychosocial maturity" score because the notion of a
unitary concept of maturity is inconsistent with Erikson's theory. The com-
plete inventory is presented in the Appendix, and a manual containing
further details may be obtained from the senior author.
In the first phase of scale development Erikson's theoretical state-
ments (especially 1959, 1963, 1968) were used to derive key words and
phrases denoting characteristics of each stage up to and including intimacy
versus isolation. These were then used in selecting items from previous
measures and in generating additional items. The items were screened for
ambiguity and face validity by the three authors until there were approxi-
mately 20 simple statements for each stage, balanced in number for success
or failure to resolve the relevant crisis. The language level was simplified as
much as possible and the items randomly ordered. During piloting the scales
were a combination of a 5-point Likert scale with a sixth category labeled
From Trust to Intimacy 529

"don't understand." This latter category was added to identify items causing
difficulty of comprehension.
The questionnaire was administered to a sample of secondary school
students in two Melbourne schools, as part of a pilot program leading to an
ongoing study of adolescent adjustment. The sample consisted of 58
students in year 9 (26 males, 32 females) and 44 students in year 10 (17
males, 27 females). Greenberger and Sorensen's (1974) Psychosocial
Maturity Inventory (PSM), Form D, was also administered to this sample
for comparison purposes.
The data were analyzed first in terms of the "don't understand"
category to see whether any items had an unduly high number of responses
in this category and whether any students had used this category excessively.
As a result, 2 items were removed from the questionnaire and 5 respondents
were excluded from further analyses. Items which showed poor item-total
correlations for each subscale were eliminated in a procedure to maximize
internal reliability (alpha coefficients). The 12 items showing highest item-
total correlations within each subscale were retained. Correlations between
subscales of the EPSI and PSM were computed as a check of construct
validity.
The EPSI was then administered to a test sample of 622 adolescents,
320 from year 9 (14I males, 179 females) and 302 from year 11 (146 males,
156 females). The sample was drawn from nine Melbourne metropolitan
high schools. Alpha coefficients were recalculated using this test sample,
and comparisons between grade levels and sexes on each of the subscale
scores were made using a two-way analysis of variance.

Psychometric Properties of the EPSI

Reliability. Table I presents, for both the pilot and test samples, the
mean scores based on item means (possible range 1 to 5), standard devia-
tions, and alpha coefficients for each of the subscales of the EPSI. For the
pilot sample, alpha coefficients were uniformly high across all subscales.
This finding gives encouraging support for the conceptual basis underlying
item selection, given that the maximizing procedure used in elimination of
unreliable items influences the level of alpha coefficients.
For the test sample, alpha coefficients were somewhat lower, but
reached an adequate level. In three cases, removal of one item from a sub-
scale would raise the alpha to a substantially higher level (Autonomy: ot =
0.69 if item 62 is removed; Initiative: a -- 0.64 if item 46 is removed; Trust:
= 0.68 if item 20 is removed). Further work with the EPSI may need to
include the testing of reworded versions of these possibly ambiguous items.
For example, item 62 ("I can stand on my own two feet") was checked
530 Rosenthal, Gurney, and Moore

Table 1. Mean Scores, Stand~ird Deviations, and Alpha

Reliability Coefficients Obtained from Responses to the
EPSI from Two Samples
Pilot sample Test sample
(N = 97) (N = 622)
Subscale M SD c~ M SD ct

Trust 3.40 0 . 6 6 0.77 3.40 0.55 0.63

Autonomy 3.65 0 . 5 9 0.74 3.66 0.50 0.62
Initiative 3.59 0.71 0.81 3.61 0.50 0.57
Industry 3.35 0 . 7 0 0.79 3.63 0.62 0.75
Identity 3.38 0 . 6 6 0.78 3.67 0.61 0.71
Intimacy 3.43 0 . 6 4 0.73 3.46 0.58 0.63

"Almost always true" by 84°7o o f the sample, suggesting that it was inter-
preted in a literal rather than metaphoric sense. A more direct statement o f
this aspect o f autonomy would be desirable in further uses o f the EPSI.
Interscale correlations were expected to be moderate, given the the-
oretical relationship between developmental stages. That is, those who have
achieved satisfactory resolution o f later stages would be expected to have
satisfactorily resolved earlier crises, although the resolution o f earlier stage
crises does not necessarily imply later stage resolution. The correlations are
reported in Table II. For all subscales there were moderate and significant
correlations of each scale with its predecessors, the only exception for the
pilot sample being a nonsignificant industry-intimacy correlation. This
relationship was significant, however, for the larger test sample.
Validity. The construct validity o f the EPSI was examined in two
ways.
1. Subscale scores were correlated with the subscales o f Greenberger
and Sorensen's PSM, Form D. The PSM was administered on the same day
to the pilot sample which completed the EPSI. The PSM is a self-report
attitude inventory with three major scales, each o f which has three sub-
scales. The first scale is labeled "Individual Adequacy" and its three sub-
scales are self-reliance, work orientation, and identity. The second scale is
"Interpersonal Adequacy," with the subscales communication skills, en-
lightened trust, and knowledge o f m a j o r roles. The final scale, "Social
Adequacy," has the subscales social commitment, openness to sociopolitical
change, and tolerance o f individual and cultural differences. The PSM
subscales are derived from Greenberger and Sorensen's model of the
dimensions o f psychosocial maturity.
Even though there was not always a clear one-to-one correspondence
between the notions o f maturity expressed by Erikson and by Greenberger
and Sorensen, the conceptual links between the EPSI subscales and the
From Trust to Intimacy 531

Table Ii. Intercorrelations between Subscales of the EPSI for

Two Samples*
Subscales
Subscale Sample 2 3 4 5 6
1. Trust Pilot 0.32 0.65 0.42 0.46 0.40
Test 0.53 0.50 0.48 0.62 0.41
2. Autonomy Pilot 0.57 0.56 0.56 0.23
Test 0.63 0.53 0.67 0.31
3. Initiative Pilot 0.64 0.46 0.37
Test 0.55 0.59 0.37
4. Industry Pilot 0.36 0.14
Test 0.56 0.28
5. Identity Pilot 0.29
Test 0.41
6. Intimacy Pilot
Test
=For pilot sample, correlations of 0.20 and 0.26 are significant
at p < 0.05 and p < 0.01, respectively, for 95 dr. For test
sample, correlations of 0.08 and 0.11 are significant at p <
0.05 and p < 0.01, respectively, for 620 dr.

P S M subscales seemed to be strong e n o u g h to predict relationships between

the following sets:

EPSI PSM
Trust vs. mistrust Enlightened trust
Autonomy vs. shame, doubt Self-reliance
Initiative vs. guilt Self-reliance, work orientation
Industry vs. inferiority Work orientation
Identity vs. identity confusion Identity
Intimacy vs. isolation Communication skills, tolerance of individual
differences
T h e complete matrix o f correlations is shown in Table III. It can be seen
that in the m a j o r i t y o f cases where predictions were m a d e , subscales o f the
E P S I s h o w e d encouragingly high correlations with relevant subscales o f the
P S M , providing s o m e measure o f construct validity. T h e lack o f correlation
between ti'ust versus mistrust a n d enlightened trust m a y be explained by the
observation that a l t h o u g h the scales are similarly n a m e d , they appear, on
the basis o f their items, to be measuring substantially different constructs.
Similarly, the lack o f correlation between the identity ( E P S I ) and social
c o m m i t m e n t (PSM) subscales, despite Greenberger and Sorensen's claims
a b o u t a conceptual resemblance, can be explained by the dissimilarity
between the individual items o f the two scales.
2. A n examination was m a d e , using the test sample, o f the differences
between sexes a n d older and y o u n g e r respondents (years 9 and 11) on the
E P S I . It might be expected f r o m Erikson's t h e o r y that, f o r each o f the sub-
532 Rosenthal, Gurney, and Moore

Table !il. lntercorrdations Between EPSI and PSM Subscale Scores*

EPSI
PSM Trust Autonomy Initiative Industry Identity Intimacy
Self-reliance 0.47 0.35 0.65 0.44 0.24 0.46
Work 0.46 0.28 0.50 0.64 0.21 0.26
Identity 0.69 0.36 0.59 0.51 0.56 0.36
Communication 0.35 0.31 0.50 0.34 0.35 0.39
Role 0.10 0.06 0.19 0.05 -0.02 0.11
Trust 0.08 -0.02 0.23 0.02 -0.03 0.09
Social commitment 0.28 0.12 0.45 0.35 0.02 0.27
Tolerance 0.25 0.23 0.35 0.24 0.04 0.48
Change 0. ! 2 0. ! 5 0.22 0.09 0.06 0.38
*Correlations of 0.20 and 0.26 are significant at p < 0.05 and p < 0.01, respectively, for 95 df.

scales, older students would achieve higher "adjustment" scores, simply be-
cause they have had more time to work through each o f the crises. Erikson
emphasizes that earlier crises do not cease to influence an individual just
because the optimum ages for resolution have passed. Remnants of these
crises, substantially resolved or not, [Link] through subsequent stages.
Through maturity and the influence o f auspicious environmental events, the
progress toward greater resolution of each crises occurs. Comparison of
year 9 and year 11 groups showed, as expected, that older students scored
higher in the positive direction on each o f the subscales: Trust: F(I, 618) =
6.78, p < 0.01; Autonomy: F(1,618) = 10.98, p < 0.001; Initiative: F(1,
618) = 5.93, p < 0.05; Industry: F(I, 618) = 6.44, p < 0.05; Identity:
F(1,618) = 10.60, p < 0.001; Intimacy: F(I, 618) = 22.23, p < 0.001.
Table IV shows means and standard deviations as a function of grade level
and sex for each subscale.
Most personality inventories (e.g., Edwards, 1954) show sex differ-
ences consistent with male and female stereotypes. Males, on the whole,
show higher scores on instrumental dimensions such as autonomy, asser-

Table IV. Mean Scores and Standard Deviations Obtained from Test Sample
Responses to the EPSI as a Function of Grade Level and Sex (N = 622)
Grade 9 Grade 11
Males Females Males Females
Subscale M SD M SD M SD M SD
Trust 3.39 0.56 3.32 0.54 3.44 0.51 3.49 0.57
Autonomy 3.67 0.54 3.53 0.50 3.77 0.47 3.70 0.48
Initiative 3.65 0.55 3.49 0.52 3.71 0.43 3.61 0.46
Industry 3.63 0.70 3.51 0.61 3.73 0.60 3.67 0.58
Identity 3.69 0.60 3.50 0.61 3.76 0.59 3.73 0.62
Intimacy 3.29 0.57 3.41 0.55 3.45 0.58 3.67 0.55
From Trus! to Intimacy 533

tion, and ambition; while females are relatively high on expressive traits
such as warmth, empathy, and gentleness. Sex differences of this nature
were evident for the EPSI subscales, with males scoring higher on auton-
omy (F(l, 618) -- 7.37, p < 0.01) and initiative (F(1,618) = 10.88, p <
0.001) and females scoring higher on intimacy (F(1, 618) = 14.38, p <
0.001). The findings for autonomy are consistent with those reported by
Enright et al., (1980) and for intimacy by Hodgson and Fischer (1979).
Another sex difference was that males scored higher on the identity
subscale (F(1, 618) = 5.33, p < 0.05). Most studies of identity achievement
fail to shed light on the issue of possible sex differences, either because only
one sex is used in the study or because the focus is on the different relation-
ships for males and females between identity status and other variables
(Bourne, 1978a). Studies that have specifically looked at sex differences in
resolution of the identity crisis have conflicting findings. Thus, Hodgson
and Fischer (1979) found in their study of college students that males were
more advanced in issues relating to intrapersonal aspects of identity
achievement, while females were more advanced in issues relating to inter-
personal aspects. La Voie (1976), however, found no differences between
male and female high school students on level of identity achievement. With
respect to the present finding, a number o f speculative explanations may be
posited. For example, items on the identity subscale express feelings about
one's sense of well-being and self-esteem, areas where females are more
prone than males to admit problems. Further, in these times of change, the
role of women is diffuse and less clear than the male role; this may cause un-
certainty in this adolescent female population as to what they, as women,
are meant to be. Such a sex difference is not inconsistent with the construct
validity of the identity subscale.
There were no significant interactions between grade level and sex on
any subscales o f the EPSI.
In summary, it can be reasonably claimed that in terms of the tests
carried out to date, the EPSI exhibits psychometric properties of reliability
and validity which are acceptable.

CONCLUSION

The EPSI measures respondents' resolution of the conficts associated

with the first six psychological stages described by Erikson. Further modifi-
cation can extend the EPSI to include the last two stages of adulthood, pro-
viding a life-span approach to personality development consistent with cur-
rent theory.
While there are undoubted disadvantages to the self-report method-
ology, this is true of any alternatives yet devised. Given the problems
 534 Rosenthal, Gurney, and Moore

associated with other methodologies, including interviews, it can be argued

that a simple, easily administered and short paper-and-pencil inventory
such as the EPSI has its place as a research tool. In providing informa-
tion about the first six developmental stages, it can be used in studies
o f age groups ranging from childhood to young adults. It can be used
longitudinally to map developmental changes and to clarify the relation-
ships between developmental stages. Thus, changes can be linked to major
life events such as the transition from school, loss o f employment, or
divorce, where "regressive" changes, or a reemerging of previously resolved
conflicts may result (Gurney, 1980). Conversely, it may be possible to study
the effects o f "growth-promoting" experiences such as employment or
parenting.
In focusing on only one stage, as many measures do, researchers fail
to do justice to Erikson's theoretical formulation, which clearly implies that
resolution o f the core conflicts of earlier developmental stages can be
influenced in part by crises characteristic o f a late developmental period. A
research instrument which enables such changes to be mapped over a range
o f developmental stages is useful for the psychologist interested in life-span
developmental psychology.
Erikson's theoretical propositions have had wide influences on under-
standing and further theorizing in areas o f child development, adolescence,
adulthood, and aging. Much child-rearing, educational, and clinical prac-
tice has also been influenced, either explicitly or implicitly, by his stage
model o f development. Despite this, relatively little research has been done
on many aspects of Erikson's theory. One reason for this scarcity has un-
doubtedly been the difficulties in operationalizing and measuring his
theoretical constructs, which are often complex, vague, overlapping, yet
meaningful in that they encapsulate some essence o f human experience. It is
hoped that the EPSI has captured to some extent both the complexity and
meaningfulness o f Erikson's stages. Though further validating evidence for
the scale is desirable, the validity and reliability evidence summarized here
points in the direction o f a promising instrument for future research.

REFERENCES

Bourne, E. (1978a). The state of research on ego identity: A review and appraisal. Part 1.
J. Youth Adoles. 7: 223-251.
Bourne. E. (1978b). The state of research on ego identity: A review and appraisal. Part If.
J. Youth Adoles. 7: 371-392.
Ciacco, N. V. (1971). A test of Erikson's theory of ego epigenesis. Dev. Psychol. 4:306-311.
Constantinople, A. (1969). An Eriksonian measure of personality development in college
students. Dev. Psycho/. 1: 357-372.
Cross, H., and Allen, J. (1970). Ego identity status adjustment and academic achievement.
J. Consult. Clin. Psychol. 34: 288.
From Trust to Intimacy 535

Edwards, A. L. (1954). Edwards manual for the Personal Preference Schedule, Psychological
Corporation, New York.
Enright, R. D., Lapsley, D. K., Drivas, A. E., and Fehr, L. A. (1980). Parental influences on
the development of adolescent autonomy and identity. J. Youth Adoles. 9: 529-545.
Erikson, E. H. (1959). identity and the life cycle. PsychoL lss. 1: 1-71.
Erikson, E. H. (1963). Childhood and Society, Norton, New York.
Erikson, E. H. (1968). Identity: Youth andCrisis, Norton, New York.
Greenberger, E., and Sorensen, A. B. (1974). Toward a concept of psychosocial maturity.
J. Youth Adoles. 3: 329-358.
Gurney, R. M. (1980). The effects of unemployment on the psychosocial development of
school-leavers. J. Occupat. Psychol. 53: 205-213.
Hodgson, J. W., and Fischer, J. L. (1979). Sex differences in identity and intimacy develop-
ment in college youth. J. Youth Adoles. 8: 37-50.
La Vole, J. C. (1976). Ego identity formation in middle adolescence. J. Youth Adoles. 5:
371-385.
Marcia, J. E. (1966). Development and validation of ego identity status. J. Personal Soc.
Psychol. 3: 551-558.
Marcia, J. E. (1976). Identity six years after: A follow-up study. J. Youth Adoles. 5: 145-150.
Marcia, J. E., and Friedman, M. L. (1970). Ego identity status in college women. J. Personal.
38: 149-263.
Matteson, D. R. (1974). Alienation vs. Exploration and Commitment: Personality and Family
Correlaries of Adolescent Identity Statuses, Ungdom, Sforskning, Copenhagen.
Offer, D. (1969). The Psychological World of the Teenager, Basic Books, New York.
Orlofsky, J. L. (1978). Identity formation, achievement and fear of success in college men and
women. J. Youth Adoles. 7: 49-62.
Podd, M. H. (1972). Ego identity status and morality: The relationship between the two con-
structs. Dev. PsychoL 6: 497-507.
Prelinger, E., and Zimet, C. N. (1964). An Ego-Psychological Approach to Character
Assessment, Free Press, London.
Rasmussen, J. E. (1964). Relationship of ego identity to psychosocial effectiveness. PsychoL
Rep. 15: 815-825.
Wessman, A. E., and Ricks, D. F. (1966). Mood and Personality, Holt, Rinehart and Winston,
New York.

APPENDIX

Erikson Psychosocial Stage Inventory (EPSI)

Item
number Subscale
Trust
3. I wish I had more self-control"
12. I find the world a very confusing place*
19. i worry about losing control of my feelings a
20. I have few doubts about myself
24. Other people understand me
31. I find that good things never last long*
36. Things and people usually turn out well"for me
38. I think the world and people in it are basically good
42. People are out to get me a
47. 1 find myself expecting the worst to happen '~
53. I'm as good as other people
64. I trust people
536 Rosenlhal, Gurney, and Moore

Appendix continued.

Item
number Subscale

Autonomy
1. I am able to take things as they come
2. I can't make sense of my life"
5. I can't make up my own mind about things"
8. I'm never going to get on in this world*
13. I know when to please myself and when to please others
28. ! really believe in myself
39. I am ashamed of myself"
54. t like to make my own choices
55. I don't feel confident of my judgment"
62. I can stand on my own two feet
63. I find it hard to make up my mind"
65. I like my freedom and don't want l . be tied down

Init&tive
7, I am able to be first with new ideas
16. I don't seem to have the ability that most others have got"
21. I rely on other people to give me ideas '~
23. I think 1 must be basically bad"
26. I feel guilty about many things"
34. I'm an energetic person who does lots of things
46. ! can stop myself doing things I shouldn't be doing
50. I find myself denying things even though they are true"
57. I cope very well
61. I'm a follower rather than a leader"
66. 1 like new adventures
69. I like finding out about new things or places

hldustry
15. I don't seem to be able to achieve my ambitions"
22. I don't enjoy working"
25. l'm a hard worker
32. I feel I am a useful person to have around
35. I'm trying hard to achieve my goals
40. I'm good at my work
45. I can't stand lazy people
52. I waste a lot of my time messing about °
58. I'm not much good at things that need brains or skill"
60. I stick with things until they're finished
68. I don't get things finished"
70. I don't get much done"

Identity
6. I change my opinion of myself a lot a
10. I've got a clear idea of what ! want to be
11. I feel mixed up a
14. The important things in life are clear to me
17. I've got it together
18. I know what kind of person I am
29. I can't decide what ! want to do with my life"
37. I have a strong sense of what it means to be female/male
From Trust to Intimacy 537

Appendix continued.

item
number Subscale

43, I like myself and a m proud o f what I stand for

44. I don't really know what l'm on about ~
49. I find 1 have to keep up a front when I'm with people"
51. I don't really feel involved °

Intimacy
4. I get embarrassed when s o m e o n e begins to tell me personal things =
9. I'm ready to get involved with a special person
27. I'm warm and friendly
30. It's important to me to be completely open with my friends
33. I keep what I really think and feel to myself"
41. I think it's crazy to get too involved with people"
48. I care deeply for others
56. I'm basically a loner a
59. I have a close physical and emotional relationship with another person
67. I prefer not to show too much o f myself to others"
71. Being alone with other people makes me feel uncomfortable"
72. I find it easy to make close friends
"Reversed items.