Identity development from adolescence to adulthood: An extension of theory and a review of research

Abstract

Examines, in the light of research findings, those aspects of E. H. Erickson's (1963, 1968) theory of psychosocial development that concern the formation of a sense of personal identity. When the Stage 5 component is considered as a bipolar dimension, the expectation is that the transition from adolescence to adulthood involves a progressive strengthening in the sense of identity. When the identity construct is expanded to include the processes by which an identity is formed, it is hypothesized that identity status will change from the identity diffusion status to the foreclosure and then the moratorium status, and from the moratorium to the identity achievement status. The circumstances associated with these changes, regressive changes, and reentry into an identity crisis are discussed. The review of the literature deals with the direction and timing of identity development, sex differences in development, and the identification of antecedent conditions relating to the choice of development pathways. (2 p ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)

Full text

Developmental Psychology
1982,
Vol.
18, No. 3,
341-358
Copyright 1982
by the
American Psychological Association, Inc.
0012-1649/82/1803-0341
J00.75
Identity
Development
From Adolescence
to
Adulthood:
An
Extension
of
Theory
and a
Review
of
Research
Alan
S.
Waterman
Trenton
State
College
Those aspects
of
Erikson's theory
of
psychosocial
development concerning
the
formation
of a
sense
of
personal identity
are
examined
in
light
of
research evi-
dence. When
the
Stage
5
component
is
considered
as a
bipolar
dimension,
the
expectation
is
that
the
transition
from
adolescence
to
adulthood involves
a
pro-
gressive
strengthening
in the
sense
of
identity. When
the
identity construct
is
expanded
to
include
the
processes
by
which
an
identity
is
formed,
it is
hypoth-
esized that over time there
will
be
progressive developmental
shifts
in
identity
status, that
is,
from
the
identity
diffusion
status into either
the
foreclosure
or
moratorium
status,
from
the
foreclosure into
the
moratorium
status,
and
from
the
moratorium into
the
identity achievement status.
The
circumstances asso-
ciated
with
progressive
shifts,
regressive changes,
and
reentry into
an
identity
crisis
are
discussed.
The
review
of the
literature
focuses
on
questions
of the
direction
and
timing
of
identity development,
sex
differences
in
development,
and
the
identification
of
antecedent conditions relating
to the
choice
of
developmental
pathways.
Erikson's
(1963,
1968) construct
of
iden-
tity
has
become
the
principal tool
for un-
derstanding
the
development
of
personality
from
adolescence into adulthood:
The
wholeness
to be
achieved
at
this stage
I
have called
a
sense
of
inner
identity.
The
young person,
in
order
to
experience
wholeness, must
feel
a
progressive continuity
between
that
which
he has
come
to be
during
the
long
years
of
childhood
and
that
which
he
promises
to
become
in
the
anticipated
future;
between that
which
he
con-
ceives himself
to be and
that
which
he
perceives
others
to see in him and to
expect
of
him.
Individually
speaking,
identity
includes,
but is
more than,
the sum of all the
successive
identifications
of
those earlier years when
the
child wanted
to be, and
often
was
forced
to
become, like
the
people
he
depended
on.
Identity
is a
unique product,
which
now
meets
a
crisis
to be
solved
only
in new
iden-
tifications
with
age
mates
and
with leader
figures
outside
of
the
family.
(Erikson,
1968,
p. 87)
Erikson
has
identified several develop-
mental
issues
that
may
become
the
focus
of
an
identity crisis.
There
is the
choice
of an
occupation that
is
both socially recognized
and
personally expressive. Another
is the
development
of an
ideological world view
that
is
worthy
of a
personal commitment.
Such
an
ideology
may be
either
religious
or
political
in
nature.
Other
questions involve
Requests
for
reprints should
be
sent
to
Alan
S.
Water-
man,
Department
of
Psychology,
Trenton
State
College,
Hillwood Lakes,
CN
550, Trenton,
New
Jersey 08625.
sex
role appropriate behavior
and
sexual
orientation. Here
too the
task
is to find
modes
of
expression that reflect intrinsic
in-
clinations
and for
which
sufficient
sources
of
social support exist within
the
individual's
cultural milieu.
Since
the
construct
of
identity
was
intro-
duced
for
purposes
of
clinical analysis,
whether
of
individuals
or of
cultures,
its
def-
inition
could remain richly associative with-
out
being precisely specified.
The
impact
of
Erikson's clinical interpretations rests,
in
part,
on his
capacity
to
persuade
the
reader
that
quite diverse forms
of
behavior
are ac-
countable
for in
terms
of a
limited number
of
concepts*
However,
the
absence
of a de-
finitive
statement
of the
boundaries
of the
identity construct
has
posed
a
problem
for
researchers
attempting
to
develop
opera-
tional definitions that cover
the
range
of as-
pects attributed
to it. The
assessment
in-
struments
that
have been developed have
usually
focused
on
some combination
of
the
following
facets
of
identity:
(a) a
clear sense
of
self-definition;
(b) the
presence
of
com-
mitments regarding goals, values,
and be-
liefs;
(c) the
existence
of
activity directed
toward
the
implementation
of
commitments;
(d) the
consideration
of a
range
of
identity
alternatives;
(e) the
extent
of
self-accep-
341

342
ALAN
S.
WATERMAN
tance;
(f) a
sense
of
personal uniqueness;
and
(g)
confidence
in
one's personal future.
Among
the
global
paper-and-pencil
instru-
ments
with items tapping
different
combi-
nations
of
these themes
are
Rasmussen's
(1961)
Ego
Identity
Scale,
Dignan's
(1965)
Ego
Identity
Scale,
the Ego
Identity-Incom-
plete Sentences Blank (Marcia, 1966),
the
Inventory
of
Psychosocial Development
(Constantinople, 1969),
and the
Identity
Achievement
Scale
(Simmons, 1970).
Such
a
global approach
to
assessment per-
mits
a
determination
of the
extent
to
which
a
person
has
established
a
clear sense
of
identity
but
does
not
provide information
on
the
processes
by
which
identity
elements
were
developed.
In
writing
on the
processes
of
development, Erikson
(1958)
used Wil-
liam
James's
distinction between those
who
are
once born
and
those
who go
through
a
second birth
or
growth crisis
in the
shaping
of
their identity.
The
former experience
no
difficulties
in fitting
themselves into their
society. They translate their childhood iden-
tifications
into adult experience without
se-
rious
doubt
or
conflict.
In
contrast,
the
twice
born
undergo
a
crisis
of
purpose
or
faith
in
which
their adult commitments
are
formed
as an act of
personal choice.
Marcia (1966) constructed
a
semistruc-
tured interview
by
which
to
assess
both
the
clarity
of
personal identity
and the
process
by
which
it
develops.
The
classification sys-
tem
used
is
composed
of
four
ego
identity
statuses:
(a)
identity achievement,
(b)
mor-
atorium,
(c)
foreclosure,
and (d)
identity
diffusion.
These
statuses
are
defined
in
terms
of
two
dimensions: crisis
and
commitment.
"Crisis"
refers
to a
period
of
struggle
or ac-
tive
questioning
in
arriving
at
such
aspects
of
personal identity
as
vocational
choice
and
ideological beliefs.
"Commitment"
involves
making
a firm,
unwavering decision
in
such
areas
and
engaging
in
appropriate imple-
menting
activities.
An
identity achiever
is
someone
who has
gone
through
a
period
of
crisis
and has de-
veloped relatively
firm
commitments.
The
term
moratorium
is
used
to
refer
to a
person
who
is
currently
in a
state
of
crisis
and is
actively seeking among alternatives
in an
attempt
to
arrive
at a
choice.
A
person
is
classified
as a
foreclosure
if he or she has
never
experienced
a
crisis
but is
nevertheless
committed
to
particular goals, values,
or
beliefs.
The
commitments
that
foreclosures
have
made generally reflect
the
wishes
of
parents
or
other authority
figures. The
cat-
egory
of
identity
diffusion
includes individ-
uals
who do not
have
firm
commitments
and
who
are not
actively trying
to
form
them.
They
may
never have been
in
crisis,
or
they
may
have
had a
period
of
questioning
and
been unable
to
resolve
it,
subsequently
emerging without having made
a
decision.
There
have been several recent reviews
of
the
research literature
on
identity (Bourne,
1978a, 1978b;
Marcia,
1980,
Note
1)
eval-
uating
the
validity
of the
various
assessment
techniques
and
documenting
the
broad range
of
variables
to
which identity
is
related.
The
more limited focus here
is on
theory
and re-
search pertaining
to
questions
of
identity for-
mation. Specifically,
the
issues
to be
dis-
cussed
are (a) the
direction
and
timing
of
identity
development,
(b) sex
differences
in
development,
and (c)
antecedent conditions
relating
to the
choice
of
developmental path-
ways.
Developmental
Hypotheses
Direction
and
Timing
of
Development
At its
simplest,
the
basic
hypothesis
of
identity
development
is
that
the
transition
from
adolescence
to
adulthood involves
a
progressive strengthening
in the
sense
of
identity.
It is
this hypothesis that
has
been
tested
in
numerous studies using paper-and-
pencil
measures
of
identity.
Because
Marcia's
identity interview pro-
cedure
involves
categorizing individuals
ac-
cording
to
their strategy
for
handling
the
task
of
identity formation
and
does
not
yield
a
single continuous measure,
a
more complex
set of
developmental
patterns
can be
iden-
tified.
A
schematic presentation
of the
path-
ways
of
identity status formation consistent
with
Erikson's theory
is
given
in
Figure
1.
A
person
who is in the
identity
diffusion
status
may (a)
become
a
moratorium
by
beginning
to
seriously explore
a
variety
of
identity alternatives
(D
—»
M); (b)
become
a
foreclosure
by
latching
on to the first
real
possibility
that
is
presented
without ever
evaluating
other
courses
of
action
(D
—>
F);

IDENTITY DEVELOPMENT
343
or
(c)
continue indefinitely
in the
diffusion
status, never making
a
serious
effort
to
work
out
identity issues
(D
—»
D).
A
person
who is in the
foreclosure status
may
(a)
become
a
moratorium
if the
early
commitments
are
challenged
in a way
that
requires
the
consideration
of
alternative pos-
sibilities
(F
—*
M); (b)
continue
as a
fore-
closure, carrying into adulthood commit-
ments
to the
goals
and
values that were
developed prior
to or
during adolescence
(F
—»
F); or (c)
become
a
diffusion
if the
initial
commitments gradually become less
meaningful
without steps being taken
to
re-
vise
or
replace them
(F
—>
D).
A
person
who has
entered
the
moratorium
status
may (a)
become
an
identity achiever
by
establishing
firm,
meaningful
commit-
ments
to
specific
goals
and
values
(M
—*
A)
or (b)
become
an
identity
diffusion
by
giving
up
on the
efforts
to find
something worth-
while
to
which
to
become committed
(M^D).1
A
person
who has
become
an
identity
achiever
may (a)
continue
as an
achiever,
maintaining
the
commitments
to the
goals
and
values that were worked
out
during
the
identity
crisis
(A
—»A);
(b)
again become
a
moratorium, reentering crisis
if the
earlier
resolution proves unsatisfactory
in
some
way
(A
—»M);
or (c)
become
a
diffusion
if the
commitments
that were established gradu-
ally
lose their vitality without triggering
a
new
crisis
(A
—>
D).
Of
the
changes
in
status just described,
the
following
constitute progressive devel-
opmental
shifts:
from
the
identity
diffusion
status into either
the
foreclosure
or
mora-
torium
status,
from
the
foreclosure into
the
moratorium
status,
and
from
the
morato-
rium
into
the
identity achievement status.
Each represents movement
involving
either
the
initiation
of
reflective consideration
of
identity alternatives
or the
development
of
personally meaningful commitments,
A
change
into
the
identity
diffusion
status
from
any
of the
other
statuses
can be
viewed
as
developmentally
regressive, since
it
involves
a
putting aside
of
identity
concerns,
at
least
temporarily,
without
having
established
a
satisfactory resolution.
The
shift
from
the
identity achievement
to the
moratorium sta-
tus
might more
appropriately
be
considered
M
M
M
F
Figure
1. A
model
of the
sequential patterns
of ego
identity
development.
(D =
identity
diffusion;
F =
fore-
closure;
M =
moratorium;
A =
identity achievement.)
as
the
resumption
of a
crisis
rather
than
a
developmental
regression.
It
reflects
a
con-
tinuation
of the
identity formation process,
an
attempt
to
make more rewarding choices,
not
a
renunciation
of
identity concerns.
The
above constitutes
a
descriptive model
rather than
a
theory
of
development,
since
there
are
virtually
no
patterns
of
identity
status
change inconsistent with
it. The
utility
of
the
model rests
in the
opportunity
it af-
fords
to
study
the
relative
frequency
of the
different
developmental paths
and the
cir-
cumstances that influence
their
adoption.
It
thus
becomes possible
to
compare
the
pat-
terns
of
development among various groups
within
a
population (and between popula-
tions)
and to
identify
the
impact
of
socio-
historical
conditions
on
identity formation.
The
basic hypothesis
of
identity develop-
ment
may now be
phrased: Movement
from
adolescence
to
adulthood involves
a
prepon-
derance
of
changes
in
identity status which
can
be
characterized
as
progressive devel-
opmental
shifts.
There
are two
questions related
to the
direction
of
identity development about
which
little
a
priori theorizing
has
been done.
One
concerns
the
relative
stability
of the
various
identity statuses.
It can be
antici-
'
Continuing
in the
moratorium status
is not
included
here, since
indefinitely
maintaining
an
identity
crisis
is
highly
improbable (although theoretically possible).
Moving
from
the
moratorium
(or
achiever)
status
to the
foreclosure status
is not a
theoretical
possibility. Once
a
person
has
undergone
an
identity crisis,
he or she no
longer
can
meet
the
definition
of a
foreclosure, even
if
he
or she
goes back
to the
same goals
and
values
that
were
held
before
the
crisis.

344
ALAN
S.
WATERMAN
pated that
the
moratorium category will
be
the
least stable
of the
statuses,
since
it is
associated with
an
expressed desire
to
make
changes
in
one's life. Further
it is
difficult
for
individuals
to
sustain
the
subjective
dis-
comfort
usually associated with identity
crises.
If a
successful resolution cannot
be
achieved,
the
person
is
likely
to
eventually
renounce
the
task
as
unresolvable. However,
it is
unclear
as to
which
status
will
be the
most
stable.
While both
the
foreclosure
and
identity achievement statuses involve
com-
mitments,
the
stability
of
either type
of
com-
mitment
may be
influenced
by a
wide variety
of
life events.
The
lack
of
commitment char-
acteristic
of the
identity
diffusion
status
may
or
may not
prove
to be a
relatively stable
developmental quality. Further,
it is
possible
that
the
relative stability
of the
various
sta-
tuses
will
vary
by the
topic area
in
which
identity
is
considered.
The
other question best approached
in
empirical fashion concerns
the
timing
of
identity
changes.
In his
discussion
of the
epigenetic principle, Erikson (1959) refers
not
only
to a
proper sequence
of
stage
de-
velopment
but to a
proper
rate.
Stage
5 in
his
theory
can be
presumed
to
cover much
or all of the
period
from
puberty through
the
college years.
He
does
not,
however, suggest
a
timetable
of the
ages
at
which particular
identity issues
are
most likely
to
arise,
be-
come
a
focus
of
development,
and
reach
eventual
resolution. Although wide individ-
ual
differences
in the
timing
of
identity
de-
velopment
may be
expected,
it
should
be of
interest
to
determine
if
there
are
character-
istic times
(or
developmental settings)
as-
sociated with
the
greatest probability
of
change.
Sex
Differences
in
Identity Development
Since Erikson (1963,
1968, 1975)
wrote
extensively
on the
differences
in
psycholog-
ical
functioning
between
the
sexes,
it
might
be
expected
that
he
would anticipate
sex
differences
in the
patterns
of
identity
for-
mation.
It is
questionable, however,
as to
whether
he
intended such
a
hypothesis.
Be-
fore
addressing that question,
it is
helpful
to
look
at the
differences
that
he
does
de-
scribe.
Working
from
his
experience with
the
play
constructions
of
preadolescent participants
in
a
long-term development study, Erikson
discussed differences between
the
sexes
in
the
real
and
symbolic
use of
space.
He
drew
an
association between
the
play construc-
tions
and
anatomically based genital modes.
Males, with external genital organs,
erect-
able
and
intrusive, developed play construc-
tion
characterized
by
height
and
downfall,
strong motion
and its
channelization
or ar-
rest. Females,
with
internal genital organs
having
vestibular access, constructed
static
interiors that were open, simply enclosed,
and
peaceful
or
intruded upon. However,
Erikson
did not
conclude that intrusion
and
enclosing
were
the
exclusive prerogatives
of
males
and
females, respectively. Rather,
these differences were seen
as
ones
of
"pre-
disposition
and
prediction."
He
wrote that
where
cultural
and
historical conditions
per-
mit,
each
sex may be led "to
make
use of,
to
share,
and at
times
to
imitate,
the
con-
figurations
most typical
of the
other
sex"
(Erikson,
1975,
p.
233).
The
emphasis Erikson
(1975)
placed
on
the
presence
of
anatomically based
sex
dif-
ferences
in
behavior reflects
an
effort
to
counter
the
assertions made
by
some that
all
sex
differences
are
learned.
He did not
pos-
tulate
different
developmental processes
for
males
and
females. Thus, while biological
predispositions
may be
expected
to
play
some role regarding
the
content
of
identity
choices, they would
not
necessarily
be in-
volved
in the
processes
by
which identity
elements
are
selected. Both sexes
may
undergo comparable experiences
of
crisis
and
commitment.
It
should
be
recognized
that
any sex
differences observed
in the
fre-
quency
of
presence
in the
different
identity
statuses
can be
interpreted
as a
function
of
cultural
influences,
not a
result
of the un-
folding
of
different
developmental capabili-
ties.
Antecedent Conditions
Relating
to
Identity
Development
The
choice
of a
developmental pathway
for
identity formation
may be
expected
to
be
influenced
by a
variety
of
interrelated
variables.
The
following
is a
summary
pre-

IDENTITY DEVELOPMENT
345
sentation
of
hypotheses
relating
antecedent
conditions
to the
course
of
identity
devel-
opment.
1.
The
greater
the
extent
of
identification
with
parents prior
to or
during adolescence,
the
greater
the
likelihood will
be of
forming
and
maintaining personally meaningful
commitments. With strong identification,
entrance into,
and
maintenance
of, the
fore-
closure
status
appears
most probable.
2.
Differences
in
parenting styles (Becker,
1964) will
be
reflected
in
differences
in
path-
ways
of
identity formation. Individuals with
permissive,
neglecting,
or
rejecting
parents
may
be
expected most frequently
to be
iden-
tity
diffuse
and to
have
difficulty
in
success-
fully
resolving identity crises should they
occur. Those
from
authoritarian families
may
take either
of two
quite divergent paths,
becoming foreclosed
on
parental choices
or
rebelling
and
going through
an
identity cri-
sis, perhaps
of
quite strenuous proportions.
The
former
alternative
should
be
more com-
mon
under circumstances where
an
adoles-
cent
is
able
to
earn
parental
respect,
whereas
the
latter should more frequently occur
where
parental approval cannot
be
gained.
Both
protective
and
indulgent parenting
styles
may be
expected
to
yield outcomes
characterized either
by
relative insecurity
(identity
diffusion)
or by
docile conformity
(foreclosure). Finally, democratic parenting
should
be
relatively conducive
to the
consid-
eration
of
identity
alternatives
and the
form-
ing
of
personal commitments (identity
achievement).
Presence
in the
foreclosure
status
may
also
be
expected where
any
early
decisions
are
functional
for the
adolescent
and
meet with parental support.
3.
The
greater,the
range
of
identity
al-
ternatives
to
which
an
individual
is
exposed
prior
to or
during adolescence,
the
greater
the
likelihood will
be of
undergoing
an
iden-
tity crisis. Thus, homogeneous communities
may
be
conducive
to the
forming
and
main-
taining
of
foreclosure commitments, whereas
more
heterogeneous
communities
may
serve
to
facilitate
the
entrance into
an
identity
crisis.
4.
The
greater
the
availability
of
model
figures
perceived
as
living successfully,
the
greater
the
probability
will
be
that
a
person
will
form
meaningful
commitments. This
should have
particular
relevance
for the
like-
lihood
of
successfully resolving
an
identity
crisis, since successful adult models
may
serve
as
examples
of how to
cope with dif-
ficulties
and
may
suggest identity content
alternatives that
can
be
profitably explored.
Similarly,
the
parents'
own
identity
statuses
may
influence
the
process
of
identity devel-
opment through
the
model value they have
for
the
adolescent.
5.
The
nature
of the
social expectations
pertaining
to
identity
choices
arising within
the
family,
the
schools,
and the
peer group
will
contribute
to the
particular
identity
de-
velopment
pathways employed. Where
one
is
exposed
to
social groups that seldom ques-
tion received authority,
an
identity crisis
would
appear less probable than among
groups
where questioning
is
more common
and
even encouraged.
6.
Consistent with
the
epigenetic princi-
ple,
it is
anticipated that
to the
extent that
preadolescent
personality provides
an
appro-
priate foundation
for
coping
with
identity
concerns
(i.e.,
there
are
sufficient
levels
of
basic trust, autonomy, initiative,
and
indus-
try),
the
more successfully identity devel-
opment
is
likely
to
proceed. Similarly, suc-
cess
in
such other aspects
of
development
as
formal
operations
and
moral
reasoning
may
also contribute
to
advances
in
identity for-
mation.
Research Evidence
Direction
and
Timing
of
Development
Development
prior
to and
during
the
high
school
years. Prior
to the
high school years,
there appears
to be
little interest
in
identity-
related questions.
Ciaccio
(1971) used
a
Thematic Apperception Test-type story-tell-
ing
projective instrument
to
assess
the
extent
of
concerns associated with each
of the first
five
stages
described
by
Erikson.
The
sam-
ples consisted
of
males
at
three
age
levels:
5,
8,
and
11
years.
The
results
of a
"unit
utterance" scoring procedure indicated that
the
11-year-old
boys were beginning
to ev-
idence identity concerns,
but the
frequency
of
such concerns
was
far
lower than
for
con-
cerns associated
with
the
earlier stage com-
ponents
of
industry
and
initiative. Similarly,

346
ALAN
S.
WATERMAN
in
two
studies using
the ego
identity inter-
view
with
11-
to
13-year-olds
(Meilman,
1979;
Archer,
Note
2), the
identity achieve-
ment
status
was
observed
in
less than
4% of
the
topic area scoring decisions.
The
results
of
cross-sectional studies
of
development
during
the
high school years
using
paper-and-pencil
instruments have
re-
vealed
only limited
differences.
Pomerantz
(1979)
found
twelfth-grade females
to
score
significantly
higher than eighth-grade
fe-
males
on the
Rasmussen
Ego
Identity
Scale,
but
the
difference
for
males, though
in the
expected direction,
was not
significant.
La
Voie
(1976)
studied high school sophomores,
juniors,
and
seniors
of
both sexes using Mar-
cia's
Ego
Identity-Incomplete Sentences
Blank
and
found
a
nonsignificant increase
with
increasing grade level. Similarly, How-
ard
(1960),
in a
comparison
of
sophomore
and
senior high school
girls,
found
a
nonsig-
nificant
increase
in
scores
on her own
iden-
tity questionnaire.
Cross-sectional studies
involving
the ego
identity
interview were conducted
by
Meil-
man
(1979)
and
Archer
(Note
2). The
par-
ticipants
in
Meilman's study were college-
bound
males aged
12,
15,
and 18
years.2
Those
in
Archer's study were males
and fe-
males
from
Grades
6, 8, 10, and 12,
without
restriction
as to
future
plans.
In
both studies,
at the
youngest
age
level virtually
all
par-
ticipants
were
in the
foreclosure and/or
identity
diffusion
statuses.
The
foreclosure
status
was
more
frequent
in the
areas
of sex
role
attitudes (included only
in
Archer's
study),
avocational choice (included
only
in
Meilman's study),
and
religious beliefs (in-
cluded
in
both studies),
and the
identity dif-
fusion
status
was
most frequent
in the
area
of
political ideology (included
in
both stud-
ies).
In
comparison,
the
oldest high school
samples
evidenced more
instances
of the
identity
achievement status
in
both studies,
and
these increases
with
age
were distributed
across
all
content areas. There were corre-
sponding
decreases
in
instances
of the
iden-
tity
diffusion
status, also distributed across
the
various content areas.
The
frequencies
of
the
moratorium
and
foreclosure statuses
did
not
change substantially during
the pe-
riod considered here.
Hauser
(1971)
conducted
a
longitudinal
study
with
a
small sample
of
white
and
black
male
high school students
from
lower socio-
economic status backgrounds.
He
used
a
combination
of
interview
and
Q-sort proce-
dures over
a
3-year period.
He
found
the
white
adolescents
to
show
a
pattern
of
"pro-
gressive identity formation" characterized
by
frequent changes
in
self-concept during
the
early high school years
followed
by in-
creasing
consistency
and
stability
as the
per-
son
approached high school graduation.
In
contrast,
the
black adolescents showed
a
general
stability
in
their identity elements
over
the
entire study period,
a
pattern Hau-
ser
termed "identity foreclosure."
He
inter-
preted this lack
of
change
as
reflecting
a
problem
in
development
in
that important
developmental
issues
had
been dodged rather
than
resolved. Many
of the
stable identity
elements were
defined
in the
negative;
that
is, the
adolescent
knew what
he did not
want
to
become rather than what alternative
he
actively
wanted
to
pursue.
The
result
was a
rigid
and
impoverished
self-definition.3
Hau-
ser
speculated
that
these racial
differences
may
have resulted
from
greater discrimi-
nation
directed toward poor blacks than
to-
ward
whites
from
the
same background.
Development
during
the
college years.
It
is
during
the
college years
that
the
greatest
gains
in
identity formation appear
to
occur.
College
environments
provide
a
diversity
of
experiences that
can
serve both
to
trigger
consideration
of
identity issues
and to
sug-
gest alternative resolutions
for
identity con-
cerns.
The
results
of
numerous studies con-
firm
that,
in
general, senior
men and
women
have
a
stronger sense
of
personal identity
than
do
their
freshman
counterparts
and
that
the
identity commitments held
as se-
niors
are
more likely
to
have been arrived
at
through
the
successful resolution
of
iden-
tity
crises.
Cross-sectional studies using paper-and-
pencil
measures
of
identity yield
a
consistent
pattern
of findings,
with
high scores asso-
ciated with advancing
age
and/or years
in
2
Meilman (1979) also included
samples
of
males
aged
21 and 24
years.
3
In
Marcia's
(1966) classification system, such
in-
dividuals
would probably have been
termed
identity
diffuse.

IDENTITY DEVELOPMENT
347
college.
Such
results
have been
found
for the
Dignan
Ego
Identity Scale (Dignan, 1965;
Stark
&
Traxler, 1974; Thompson, 1963)
and
the
Stage
5
scale
of the
Inventory
of
Psychosocial Development (Constantinople,
1969; Whitbourne, Jelsma,
&
Waterman,
1982),
though
the
effects
were
not
always
statistically
significant.4
Longitudinal stud-
ies
using
the
Inventory
of
Psychosocial
De-
velopment provide further evidence
of
iden-
tity development during
the
college years
(Constantinople,
1969, 1970;
Whitbourne
et
al.,
1982).
Fry
(1974)
found
significant gains
on
the
Stage
5
components
for a
sample
of
college students
from
rural backgrounds
but
a
retrogressive
shift
for
those
from
urban
backgrounds.
There have been three longitudinal studies
tracing identity development among college
students using
the ego
identity interview pro-
cedure.
Adams
and
Fitch
(in
press) inter-
viewed
males
and
females
from
different
col-
lege
cohorts
at a
state
university
in
successive
years.
In the
other studies, changes
from
the
freshman
to the
senior
year were
assessed
among
male
students
at a
technological
in-
stitute (Waterman
&
Waterman, 1971;
Waterman, Geary,
&
Waterman,
1974)
and
at a
private liberal arts college (Waterman
&
Goldman, 1976).
In
each
of the
studies,
the
interview covered
the
areas
of
vocational
choice,
religious beliefs,
and
political
ideol-
ogy.
The
results
from
the
three schools were
quite
similar.
1.
College
clearly facilitates identity
de-
velopment
in the
area
of
vocational plans.
In
both studies covering
the
4-year period
from
the
freshman
to the
senior year, there
were significant increases
in the
frequency
of
students
in the
identity achievement sta-
tus and
decreases
in the
frequency
of
stu-
dents
in the
moratorium status.
In
addition,
there
was a
decrease
in the
frequency
of
stu-
dents
in the
identity
diffusion
status
at the
liberal
arts
college.
In the
1-year
study,
there
was
a
significant decrease
in the
frequency
of
identity
diffusions
and a
nonsignificant
increase
in the
frequency
of
identity
achievers.
2.
College experiences appear
to
under-
mine
traditional religious beliefs without
necessarily helping
the
students
to
establish
alternate belief systems.
At all
three colleges
there
were significant
decreases
in the
fre-
quency
of
students
in the
foreclosure status.
At
the
state
university there
was an
increase
in
the
frequency
of
students
in the
identity
achievement
status,
and at the
technological
school there
was an
increase
in the
frequency
of
identity
diffusions.
3.
Although
the
development
of
new,
clearer commitments
in the
area
of
political
ideology
was
observed
to
occur during
the
college years,
a
substantial proportion
of the
students demonstrated
little
interest
in the
topic. Significant increases
in the
frequency
of
the
identity achievement status were
ob-
served
over
1
year
at the
state
university
and
over
4
years
at the
technological institute.
At the
state
university
there
was
also
a
sig-
nificant
increase
in the
frequency
of the
moratorium
status. Significant decreases
were
found
for the
identity
diffusion
status
at the
state
university
and the
foreclosure
status
at the
technological institute.
No
sys-
tematic
changes were found
for the
students
at the
private liberal arts colleges.
In
both
4-year
studies,
more than half
of the
partic-
ipants
were approaching graduation without
clear beliefs
in
this area
and
without trying
to
form
any
commitments.
The
patterns
of
identity development
ob-
served
were relatively consistent across
the
three studies,
but
some
differences
were
ev-
ident.
The
differences
may
have been
due to
the
distinctive characteristics
of the
student
populations
at the
three
schools.
Differences
in
the
timing
of the
studies
may
also have
affected
the
results,
particularly
in the
area
of
political
ideology.
For
example,
the first
4-year study
was
conducted
at the
techno-
logical institute
from
1968
to
1972 during
the
height
of the
antiwar protest.
The
second
study,
conducted
at the
private liberal arts
college, took place
from
1970
to
1975, with
the
senior-year interviews conducted after
the
issue
of the war had
largely subsided
and
following
the
Watergate
scandal.
It
should
also
be
noted that
in all
three studies,
fcillow-
up
data
were only obtained
from
individuals
4Protinsky
(1975)
found
a
similar
difference
in a
comparison
of
junior
high
school students
and
college
students
for the
Stage
S
items
of the
Rasmussen
Ego
Identity
Scale.
However,
the
broad
age
range
of the
participants
in the
study
does
not
permit
a
determination
as to the
timing
of
development.

348
ALAN
S.
WATERMAN
who
continued
at
their respective schools
throughout
the
period
of the
research.
As
yet, nothing
is
known about
the
patterns
of
development
of
those
who
withdraw
from
college.
Development during
the
adult years.
To
date only limited research
has
been con-
ducted concerning identity among adults.
Waterman
and
Waterman
(1975)
compared
the
distribution
of the
identity status
of
col-
lege males
and
their fathers. They
found
that
the
fathers (men between
the
ages
of 40 and
65
years) were more frequently
in the
fore-
closure status
in the
three
topic areas cov-
ered
in the
interview (vocational choice,
re-
ligious
beliefs,
and
political ideology). Their
sons
were more likely
to be in the
morato-
rium
and
identity
diffusion
statuses.
A
very
high
percentage
of
foreclosures
was
also
found
in a
sample
of
married adult women
between
the
ages
of 23 and 50
years
(Water-
man,
Note
3). All of the
women were from
middle-class
communities, were
in
their ini-
tial marriage,
and had at
least
one
child aged
17
years
or
less
living
at
home.
The
greatest
concentrations
of
foreclosure decisions were
in
the
topic
areas
of
family/career priorities,
religious
beliefs,
and sex
role attitudes.
It
seems reasonable
to
suggest that
the
high
percentage
of
foreclosures among these adult
men
and
women were
a
function
of the so-
cietal conditions that were present when they
were
in the
stages
of
adolescence
and
youth
(e.g.,
the
depression, World
War II, or the
Korean War).
These
studies
may
tell
us
more about
the
impact
of
sociohistorical
events
than about ontogenetic development.
O'Connell
(1976)
conducted
a
retrospec-
tive
study
of
identity synthesis
in
married
women
with school-aged children.
The
women
were classified
as
traditional
(full-
time homemakers), neotraditional (having
resumed
a
career after
an
interruption
for
child bearing),
or
nontraditional (continu-
ously
committed
to
their careers).
On the
Sense
of
Identity Inventory, women
in all
samples indicated they
had
experienced
an
increasingly
strong sense
of
identity
as
they
moved
from
adolescence through
the
periods
when
they were
first
married,
had
their
first
child,
and
their children were
of
preschool
age,
to the
point where their children were
of
school age.
O'Connell
observed
that
after
marriage
"traditional
and
neotraditional
women's
sense
of
identity seems
to
undergo
a
moratorium
[i.e.,
a
hiatus] which
is not
terminated until
the
school children
stage
of
the
life cycle, while nontraditional women
develop their sense
of
identity
in a
more
straightforward
progression"
(p.
683).
Two
longitudinal studies have
traced
iden-
tity
development
from
the
college
years
into
adulthood. Whitbourne
and
Waterman
(1979),
using
the
Inventory
of
Psychosocial
Development, conducted
a
follow-up with
male
and
female college alumni
10
years
after
they
had
completed
the
instrument
as
undergraduates.
A
significant increase
was
found
for the
Stage
5
scale assessing identity
versus
identity
diffusion.
Marcia
(1976),
us-
ing
the ego
identity interview, followed
up
on
30 men who had
originally been inter-
viewed
6
years earlier while
in
college.
All
were
living
within
a
50-mile radius
of the
city
in
which
their
college
was
located.
The
results were reported
in
terms
of
changes
in
overall
identity status,
a
judgment arrived
at by
combining information
from
the vo-
cational, religious,
and
political content
areas
of
the
interview.
The
identity statuses, aside
from
the
moratorium status, were fairly sta-
ble
(70%).
The
individuals
in the
foreclosure
and
identity
diffusion
statuses were
found
to
be
more
stable
than were
those
in the
iden-
tity achievement status.
Six of the 7
identity
achievers
at the
time
of the first
interview
were
committed,
as
well,
in the
follow-up.
However,
3 of the men
were classified
as
foreclosures.
These must
be
considered
as
anomolous
changes, since once
a
person goes
through
an
identity crisis,
he or she can no
longer
fit the
technical
definition
of the
fore-
closure
status.5
Five
of the
7
moratoriums
in
college
had firm
commitments
at the
fol-
low-up,
and 2 had
become
diffuse.
Only
1
of
the
16
participants
who
were foreclosed
or
diffuse
at the
beginning
of the
study
had
become
an
identity achiever, while
2
others
were
going through
an
identity crisis
6
years
later. Thus
it
appears that
the
adult years
5
Waterman
et
al.
(1974)
analyzed instances
of an-
omolous
scoring
and
found
them
due to (a)
respondents'
failing
to
report some
significant
piece
of
information
in
one
interview
that
was
reported
in
another
interview
and
(b)
inconsistent
application
of the
scoring standards.

IDENTITY
DEVELOPMENT
349
are a
period
of
strengthening
of
identity
but
not
a
time when many
new
identity issues
are
raised
or
novel
possibilities
considered.6
Comparative
Stability
of the
Identity
Statuses
The
three longitudinal studies
of
identity
development during
the
college
years
have
also provided data
on the
relative stability
of
the
identity
statuses.
Comparisons
of
rel-
ative stability
can be
made
for a
1-year
pe-
riod
at the
state university (Adams
&
Fitch,
in
press)
and for a
4-year period
at the
tech-
nological institute (Waterman
&
Water-
man,
1971; Waterman
et
al,
1974)
and the
private
liberal arts college (Waterman
&
Goldman, 1976). Table
1
contains
the
per-
centage
of
instances
in
which
an
individual
was
in the
same identity status
at
both
the
beginning
and end of the
study period
in a
particular content area
of the
interview.7
As
expected,
the
moratorium status
was
the
least stable
of the
four
statuses
across
the
period
from
the
freshman
to the
senior
year
and was the
second
least
stable
status
in
the
shorter study.
In the two
4-year stud-
ies,
there
were
34
instances
of
identity
crises
in
particular interview areas
found
at the
start
of the
freshman year.
Not one of
these
crises
was
continuing during
the
senior year.
Of
these
crises
over
three
quarters
ended
in
the
formation
of
clearly
defined,
personally
meaningful
commitments.
At the
technolog-
ical institute
76% of the
moratoriums
had
become
identity
achievers,
and the
corre-
sponding
figure at the
private liberal arts
college
was
85%. Considering only those
in-
stances
in the
1-year
study
in
which
the
per-
son
had
emerged
from
the
moratorium sta-
tus,
in 85% of
cases
firm
commitments
had
been developed. Thus, there
is a
very high
probability
that identity crises experienced
in
college
will
be
successfully resolved.
The
stability
of the
other identity statuses
was
generally comparable.
It was
evident
that
in
instances where individuals
had firm
commitments
at the
start
of the
study,
ap-
proximately
one
third
to one
half were
no
longer
in the
same identity status
by the
time
of
the
follow-up.
This
finding for the
identity
achievement
status makes
it
clear that
the
successful
resolution
of an
identity
crisis
does
not
give
a
permanence
to the
commit-
ment
formed. Although
the
achievement sta-
tus is
slightly more
stable
than
the
foreclo-
sure status, both statuses
are
affected
by
factors leading
to the
initiation
of a
crisis
or
to
regression.
When
the
comparative
stability
data
are
analyzed separately
for the
different
content
areas
of the
interview, some differences
emerged.
The
identity achievement
and
fore-
closure
statuses
were more
stable
in the
area
of
vocational plans than
in the
other topic
areas. Conversely,
the
identity
diffusion
sta-
tus was
more stable
in the
areas
of
religious
beliefs
and
political ideology than
in the
area
of
vocational plans.
The
moratorium status
was
highly unstable
in all
interview
areas.
Sex
Differences
in
Identity Development
The
paper-and-pencil
measures
of the
identity construct indicate
few
differences
between
males
and
females.
Sex
differences
have
not
been
found
on the
identity items
of
the
Inventory
of
Psychosocial Develop-
ment
(Constantinople, 1969; Waterman
&
Whitbourne,
1981; Whitbourne
&
Water-
man,
1979);
the
Identity
Achievement
Sta-
tus
Scale
(Simmons, 1970);
the
Texas Psy-
chosocial
Identity
Scale
(Dunivant
&
Bieri,
Note
4); the
Objective Measure
of Ego
Iden-
tity
Status
(Adams,
Shea,
&
Fitch,
1979);
Rasmussen's
Ego
Identity
Scale
(Pomer-
antz,
1979);
or
Marcia's
Ego
Identity-In-
6
The
limited
number
of
instances
in
which
the
adult
men
(Waterman
&
Waterman, 1975)
and
adult
women
(Waterman, Note
3)
were
found
to be in the
identity
achievement
status
for the
most part
did
involve
crises
occurring
during
the
adult years. Because these fre-
quencies
were
small, however,
the findings can be
viewed
as
consistent
with
the
conclusion that
the
adult years
do not
constitute
a
period
of
extensive
redirection
of
identity
choices.
In
this regard,
the
phenomenon
of
midlife
crisis should
be
studied
from
an
identity status
perspective.
If
such crises
are or
become
frequent
within
the
society,
and if
such crises entail
the
reflective con-
sideration
of
identity
alternatives,
the
above
stated
con-
clusion
will
have
to be
modified.
7
The
relative stability percentages reported here
are
underestimates
of the
actual extent
of
change, since
in-
dividuals
who
went
through several changes during
the
study
period
and who
resumed
the
same status they were
in
initially
would
be
counted
as
stable.
The
content
of
the
identity elements
at the
second interview
may
have
been
the
same
or
different
from
their
initial
ideas.

350
ALAN
S.
WATERMAN
Table
1
Percent
Stability
of the Ego
Identity
Statuses
in
Longitudinal Research
Identity
status
Identity
achievement
Moratorium
Foreclosure
Identity
diffusion
1-year
stability
at
state
university"
68
50
61
42
4-year
stability
at
technological
instituteb
59
9
44
50
4-year
stability
at
private
liberal
arts
college'
54
0
46
67
"
Data
from
Adams
and
Fitch,
in
press.
bData
from
Waterman
and
Waterman, 1971,
and
Waterman, Geary,
and
Waterman, 1974.
'
Data
from
Waterman
and
Goldman,
1976.
complete
Sentences
Blank
(La
Voie,
1976).
Dunivant
and
Bieri
(Note
4) did find
females
to
score higher than males
on
Gruen's
(1960)
Ego
Identity
Scale
and
Block's
(1961)
Role
Consistency Index.
At
least
10
studies
have been
conducted
that
can be
used
for
comparisons
of the
dis-
tribution
of the
various identity statuses
be-
tween
males
and
females.
Two
involved sam-
ples
of
high school upperclassmen
or
recent
high school graduates (Waterman
&
Archer,
1979; Grotevant,
Note
5); six
involved col-
lege samples (Adams
&
Fitch,
in
press;
Hodgson
&
Fisher, 1979; Orlofsky, 1978;
Poppen, 1974; Waterman
&
Nevid, 1977;
Rothman,
Note
6); one
involved
a
sample
of
17- to
18-year-old Danish students
(Matteson, 1977);
and one
involved adults
(Tesch,
1980). Some
differences
emerged
in
particular content areas.
Vocational
choice.
No sex
differences
were observed
in
either high school sample
or in five of the six
college samples.
In one
college sample, males were more frequently
in
the
identity achievement status,
and fe-
males were more frequently foreclosed
or
diffuse
(Hodgson
&
Fisher,
1979).
Also,
the
Danish
males showed more exploration
of
occupational alternatives than
did the fe-
males
(Matteson,
1977).
No
difference
was
found
for the
adult sample.
Religious
beliefs.
Among
the
high school
students, Waterman
and
Archer (1979)
found
that
males were disproportionately
present
in the
foreclosure category
and fe-
males were overrepresented
in the
identity
diffusion
status. Grotevant
(Note
5)
found
no
difference
in
this
area.
At the
college
level,
two
studies indicated tendencies
for
males
to
more
likely
be
identity achievers
(Adams
&
Fitch,
in
press; Hodgson
&
Fisher, 1979);
one
showed
a
significant
ef-
fect
for
females
to be
more frequently
in the
achiever status (Orlofsky, 1978);
and
three
revealed
no
differences
(Poppen,
1974;
Waterman
&
Nevid, 1977; Rothman,
Note
6). The
adult sample
did not
yield
a
signif-
icant
difference.
Political ideology. Prior
to the
college
years, Grotevant
(Note
5)
found
males
to
be
more frequently
in the
identity achieve-
ment
status
and
females more
often
in the
moratorium status.
In
contrast, Waterman
and
Archer
(1979)
found
males
to be
over-
represented
in the
foreclosure
category
and
females
to be
more frequently
diffuse.
Sig-
nificant
differences
were also observed
in two
of
the
college studies, with males more
often
in
the
identity achievement status
and fe-
males overrepresented
in the
foreclosure sta-
tus
(Adams
&
Fitch,
in
press; Hodgson
&
Fisher, 1979).
No
differences
were
found
in
the
other
four
college
studies,
nor
were
any
found
in the
study
of
adults.
Attitudes
toward
premarital
intercourse.
In
three
of the five
college
studies
that
in-
cluded this topic area, females were
found
to be
more
frequently reported
as
identity
achievers than were males, whereas
the
males were more frequently foreclosed
or
diffuse
(Orlofsky, 1978; Poppen, 1974;
Waterman
&
Nevid, 1977). Rothman
(Note
6)
found
the
identity achievement status
to
be the
most frequent
for
both males
and fe-
males, with males more likely
to be in the
moratorium status
and
females more likely
to be
foreclosed.
No
differences
were
found
by
Hodgson
and
Fisher
(1979).
Matteson

IDENTITY DEVELOPMENT
351
(1977)
reported
that
among Danish stu-
dents, females were more likely than males
to
undergo
exploration
of
sexual
identity
alternatives.
Sex
role attitudes.
At the
high
school
level,
no sex
differences
were observed
in
either
study.
Only
one of the
college
studies
included
the
topic
of sex
roles,
and as in the
area
of
sexual
attitudes,
females were more
likely
to be in the
identity achievement
and
moratorium
statuses, whereas
the
males
were
more likely
to be
foreclosed
or
diffuse
(Hodgson
&
Fisher,
1979). Tesch (1980)
found
a
trend toward
a sex
difference
in
this
area
for
adults, with females somewhat more
likely
to be in the
moratorium
status
and
males overrepresented
in the
foreclosure cat-
egory.
Psychological
significance
of the
identity
statuses
for
males
and
females.
The
results
of
these analyses
for sex
differences suggest
that
in
most interview topic
areas,
males
and
females
undergo similar patterns
of
identity
development. Only
in the
area
of
attitudes
toward
sexual behavior
was
there general,
though
not
complete,
consistency
among
the
studies, indicating
the use of
different
pro-
cesses.
(Not enough
research
on sex
role
attitudes
has
been conducted
to
reach
a
con-
clusion
regarding
this
topic.)
It is
clear
that
males
and
females
are
more similar than
different
in
their
use of
developmental pro-
cesses,
but
there
is
reason
to
believe that
the
identity
statuses
may
nevertheless
have dif-
ferent
psychological implications
for the
sexes
(Marcia,
1980).
For
both
sexes,
the
identity achievement status
is
associated
with
relatively good adaptive
capacities,
whereas
the
identity
diffusion
status
is
often
associated with
difficulties
in
coping.
For the
moratorium
and
foreclosure statuses, how-
ever,
sex
differences emerge. Among males,
individuals
in the
moratorium
status
look
generally
similar
in
behavioral
traits
to
iden-
tity achievers,
and
both stand
in
contrast
with
males
in the
foreclosure
and
identity
diffusion
statuses.
Among females
the
pat-
terns
are
more complex, with women
in the
foreclosure
and
identity
achievement sta-
tuses looking most similar
on
some qualities,
while
on
others
it is the
identity
achievement
and
moratorium women
who
appear most
similar.
Marcia
(1980)
suggests
that
"the
relative
adaptiveness
of the
Moratorium
and
Foreclosure status
for
women
is a
function
both
of the
dependent
variables
used
in a
particular study
and of the
existing cultural
supports
for
women's
explorations
of
alter-
natives"
(p.
175).
Antecedent
Conditions Relating
to
Identity
Development
A
large number
of
hypotheses were
of-
fered
regarding variables that
may
contrib-
ute to
determining
the
course
of
identity
development,
but
only
a few
have been
in-
vestigated.
The
search
for
antecedent
influ-
ences
has
focused
on
family variables,
de-
velopmental
characteristics believed
to be
hierarchically related
to
identity,
and
per-
sonal activities
and
interests.
Family
variables. Considerable atten-
tion
has
been directed toward
identifying
the
nature
of
differences
in
family
relationships
and
parenting styles characteristic
of
indi-
viduals
in the
various
identity
statuses.
Since
Marcia
(1980)
has
reviewed this literature
in
some
detail,
only
a
brief summary
of the
findings
will
be
presented here.
As
would
be
expected,
foreclosures have
been
found
to
have
the
closest relationships
with
their
parents.
Both sons
and
daughters
evaluate their parents most favorably
and
describe
their
families
as
child
centered.
The
fathers
of
foreclosures
are
seen
as
relatively
possessive
and
intrusive toward
their
sons
while
being more supportive
and
encourag-
ing of
their daughters.
In
turn, foreclosure
sons
are the
most willing
to
involve their
families
in
making important
life
decisions.
In
contrast, identity
diffusions
of
both sexes
reported
the
most distance
from
their fam-
ilies.
The
parents
of
identity
diffusions
were
seen
as
indifferent,
inactive, detached,
not
understanding,
and
rejecting. Both males
and
females
in the
moratorium
and
identity
achievement statuses were relatively critical
of
their parents
and
were likely
to
report
themselves
to be in
conflict with
their
fam-
ilies. Sons
in
these statuses were
not
likely
to
turn
to
their
families when making
im-
portant
life
decisions. This tension within
the
family
appears
related
to
ambivalence
on the
part
of
both parents
and
offspring
over
the
latter's
attempts
at
individuation.
With
the

352
ALAN
S.
WATERMAN
successful
resolution
of an
identity
crisis,
better
family
relationships
may be
estab-
lished. Finally, there
is
evidence that,
in
gen-
eral,
the
nature
of the
relationship with
the
parent
of the
same
sex is
more strongly
re-
lated
to
identity
status
than
is the
relation-
ship
with
the
opposite-sex parent.
It is an
attractive hypothesis that parental
behavior
contributes
to
identity formation,
and
the
pattern
of
research results obtained
provides readily
interpretable
links between
family
functioning
and the
various identity
statuses. However,
for
both methodological
and
conceptual reasons,
it
cannot
be
con-
cluded that
the
family
variables studied
ac-
tually
do
influence
the
process
of
identity
development.
In
most studies,
the
research-
ers
have relied solely
on
verbal report mea-
sures obtained
from
adolescent
or
youth
re-
spondents. These measures
are
subject
to
errors
of
memory, defensive distortion,
and
conscious
impression management.
The
same
problems exist
in
those studies where
infor-
mation
was
obtained
from
the
parents. Even
if
the
accuracy
of the
reports were assumed,
it
would still
not be
possible
to
reach
a
con-
clusion
concerning
any
causal contribution
of
family
variables.
The
type
of
behavior
shown
by a son or
daughter
may
well
have
elicited particular responses
from
the
par-
ents, which could account
for the
observed
relationships.
It is far
easier
to
provide
a
sup-
portive,
child-centered environment when
children
identify
strongly with
the
parents
and
follow family
traditions
without ques-
tioning.
In
contrast,
the
ambivalence
and
conflict
found
in the
family
backgrounds,of
moratoriums
and
identity achievers
may be
a
response
to the
sometimes extreme
shifts
in
behavior engaged
in by a son or
daughter
going
through
a
stressful,
and
stressing,
identity
search. Similarly,
it is
possible that
the
lack
of
involvement shown
by the
parents
of
identity
diffusions
may be at
least par-
tially
the
result
of a
child's
giving
few
strongly
directed behavioral stimuli
to
which
the
parents
can
respond.
It is
probably
not
possible
to
entirely unravel
the
complex
causal
effects
present
in any
relationship
of
long
duration. Long-term, longitudinal, pre-
dictive
studies appear necessary
if any
mean-
ingful
evaluation
is to be
made
of the
like-
lihood
that
parenting
behaviors
contribute
to the
process
by
which identity
is
formed.
Along
another line, Waterman
and
Waterman (1975) hypothesized that parents
may
serve
as
role models
for the
type
of de-
cision-making
processes
involved
in
identity
formation.
However, they
found
no
relation-
ship between
the
identity statuses shown
by
a
sample
of
fathers
and
their college-attend-
ing
sons.
Variables
relating
to
family
stability
have
also been investigated
as
possible contribu-
tors
to
identity development with conflicting
results
to
date. Oshman
and
Manosevitz
(1976)
found
that
a
sample
of
college males
who
experienced father absence
from
an av-
erage
age of
about
10
years
had
significantly
lower
full
scale scores
on the
Rasmussen
Ego
Identity Scale than
did
samples
from
intact
families
and
families
in
which
the
mothers
had
remarried.
The
directions
of
differences
were
the
same
on all Ego
Identity
Scale
stage
scales,
but the
effect
for the
Stage
5
(identity
vs.
identity
diffusion)
items
was not
signifi-
cant. Jordan
(1970)
found
that college males
in
the
identity
diffusion
status were more
likely
to
have come
from
broken homes than
were
those
in the
other statuses.
In
contrast,
St.
Clair
and Day
(1979)
reported
that
in
a
sample
of
high school females,
two
thirds
of
the
students
in the
identity achiever status
came
from
homes disrupted
by
divorce
or
the
death
of one
parent, whereas less than
one fifth of the
students
in
each
of the
other
statuses
were
from
broken homes.
Also,
Grossman, Shea,
and
Adams (1980)
found
that male college students
with
a
history
of
parental divorce scored higher
on
Marcia's
Ego
Identity-Incomplete
Sentences Blank
than
did
males
from
intact
families
or fe-
males
from
either divorced
or
intact families.
Developmental
characteristics.
One
test
of
the
epigenetic
principle—that
the
suc-
cessful
resolution
of the
crisis associated
with each
of the
earlier
stages
of
psychoso-
cial
development provides
the
foundation
for
the
successful development
of
identity—is
furnished
by the
pattern
of
intercorrelations
among
the
stage scales
of the
Inventory
of
Psychosocial Development (Constantinople,
1969). This instrument
is
designed
to
assess
the
personality components associated
with

IDENTITY
DEVELOPMENT
353
each
of the first six
stages
in
Erikson's
the-
ory.
As
expected,
the
scores
on the
Stage
5
scale
for
identity versus identity
diffusion
have been shown
to
yield significant corre-
lations
with each
of the
earlier
stage
scales
in
samples
of
undergraduates drawn
from
five
colleges
and a
sample
of
adult college
alumni
(La
Voie
&
Adams,
in
press; Water-
man,
Buebel,
&
Waterman, 1970;
Water-
man
&
Whitbourne, 1981). Marcia
and
Miller
(Note
7),
working with
a
sample
of
adult women, compared
the
identity sta-
tuses,
as
assessed
by the ego
identity
inter-
view,
with
regard
to
scores
from
the
Inven-
tory
of
Psychosocial Development. Women
in
the
identity achievement status
had the
highest
scores
on
each
of the
early
stage
scales,
whereas those
in the
identity
diffusion
status consistently
had the
lowest scores.
Rothman
(1978)
conducted
a
similar study
with
a
college
student
sample using
the
Ras-
mussen
Ego
Identity
Scale
instead
of the
Inventory
of
Psychosocial Development.
Again
the
identity
diffusion
status
was as-
sociated
with
the
lowest scores
on
each
of
the
early
stage
scales,
but no
clear
pattern
existed among
the
other statuses.
The
stage
components
of
autonomy
and
industry most
clearly
discriminated
among
the
statuses.
Ego
identity
has
also been presumed
to
rest upon
a
developmental
foundation
pro-
vided
by
other aspects
of ego
functioning
than
those
focused
on by
Erikson. Efforts
have been made
to
relate
progress
in
identity
formation
to
Loevinger's
ego
stages, Kohl-
berg's stages
of
moral reasoning,
and
Pia-
get's
stages
of
cognitive functioning.
In
each
instance
there
is a
possibility
that
a
certain
level
of
stage
functioning
may
constitute
a
necessary
but not
sufficient
condition
for
adopting
a
particular
path
of
identity
for-
mation.
A
positive relationship between identity
and
Loevinger's
ego
stages
was
obtained
by
Adams
and
Shea
(1979)
in a
sample
of
male
and
female
college
students.
They
reported
that while
20% of
individuals
in the
identity
diffusion
status were
in the A
stage
of
self-
protective functioning (the lowest stage
ob-
served
in the
sample),
less
than
4% of
those
in
each
of the
other statuses were
at
that
level.
In
contrast, only
5% of
identity dif-
fusions
were
at or
above
the
1-4
stage
of
conscientious
functioning;
the
corresponding
percentages
for the
other status were
19%
for
foreclosures,
29% for
moratoriums,
and
37%
for
identity achievers.
The
only
two in-
dividuals
at the
1-6
stage
of
integrated
func-
tioning
were
in the
identity achievement sta-
tus. These data suggest
that
the
attainment
of
Stage
1-4
could constitute
a
necessary
but
not
sufficient
condition
for
movement
out of
the
identity
diffusion
status. (The
few ex-
ceptions
found
in the
study
may
have been
a
function
of
measurement
error
on
either
variable.)
The
possibility also exists that
Stage
1-6
is a
necessary condition
for be-
coming
an
identity achiever,
but the ex-
tremely small sample involved renders such
a
relationship highly speculative.
Adams
and
Fitch
(1981)
conducted
a 1-
year
follow-up
with
the
sample
from
the
pre-
vious study. They found
considerable
change
in
both identity status
and ego
stage.
Cross-
lagged correlations
using
continuous mea-
sures
on
both variables yielded evidence
of
contemporaneous change
but not
sequen-
tially.
(It
should
be
noted
that
the
contin-
uous
measure
of
identity
did not
preserve
the
uniqueness
of the
different
identity
statuses.)
A
positive relationship between identity
status
and
level
of
moral reasoning
was ob-
tained
in
three studies
of
college samples
using
Kohlberg's
interview procedure
or a
modification
of it
(Podd,
1972; Poppen,
1974; Rowe
&
Marcia,
1980).
The
identity
achievement
status
was
most strongly asso-
ciated
with
postconventional moral reason-
ing, whereas
the
foreclosure
and
identity
diffusion
statuses
were
overrepresented
at
the
preconventional
and
conventional levels.
Cauble
(1976)
did not find a
difference
be-
tween
the
combined identity achievement
and
moratorium
statuses
and the
combined
foreclosure
and
identity
diffusion
statuses
using
Rest's
Defining
Issues Test
as the
mea-
sure
of
moral reasoning.
The
study
by
Podd
provided
data
in a
form
most
suitable
for
determining
whether
the
postconventional
level
of
reasoning serves
as a
necessary
but
not
sufficient
condition
for
entering
the
mor-
atorium
and/or
identity achievement sta-
tuses.
He
found
51%
of the
participants
in
the
identity achievement status
to be
func-

354
ALAN
S.
WATERMAN
tioning
at the
postconventional level.
The
corresponding
percentages
for the
other sta-
tuses were
31% for
moratorium,
12% for
foreclosure,
and 9% for
identity
diffusion.
Thus, although
a
relationship
between
the
two
developmental schemes
is
apparent,
postconventional reasoning
does
not
serve
as
a
necessary condition
for
entering
or
suc-
cessfully
resolving
an
identity crisis.
It has
also been hypothesized that
formal
operational
thought
is
related
to the
likeli-
hood
of
undergoing
and
resolving
an
identity
crisis (Inhelder
&
Piaget,
1958; Kohlberg
&
Gilligan,
1971).
However,
the
research
results
obtained
to
date
have been
largely
negative.
Neither Berzonsky, Weiner,
and
Raphael
(1975)
nor
Cauble
(1976)
have
ob-
tained support
for the
hypothesis. Rowe
and
Marcia
(1980)
obtained supportive
results,
but
their sample size
was too
small
for
mean-
ingful
statistical
analysis.
Marcia
(1980)
suggested that methodological problems
in
the
earlier
research
may
account
for the
neg-
ative
results.
Since
virtually
all of the
research
on de-
velopmental
characteristics
has
involved
as-
sessment
at
only
a
single point
in
time,
it is
not
possible
to
conclude that these variables
actually
constitute
antecedents
of
progress
in
identity formation.
The
associations
ob-
served
could
be
operating
in
either direction,
or
the
change
may be
occurring simulta-
neously
rather than sequentially. Again, lon-
gitudinal
research
is
needed
to
establish
whether
any of the
stage constructs
are
pre-
dictive
antecedents
of
identity development.
Interests
and
activities.
A
large number
of
correlates
of the
identity statuses have
been
identified,
any of
which might contrib-
ute
to the
process
of
identity development;
however,
in
only
a
couple
of
instances
is
there reason
to
believe that
a
sequential
ar-
rangement
may
exist.
In the
longitudinal
studies conducted
at the
technological insti-
tute
and the
private liberal arts college
(Waterman
&
Waterman,
1971;
Waterman
et
al,
1974; Waterman
&
Goldman, 1976),
the
College Student Questionnaire-Part
1
(Peterson, 1965)
was
given
to all
partici-
pants
at the
beginning
of
their freshman
year.
The
Cultural Sophistication scale,
which
taps
an
interest
in
such areas
as
art,
music, literature,
and
foreign
films, was
found
to be
predictive
of
identity formation.
At
both
schools, freshmen
who
were
identity
achievers
scored
higher
on the
scale than
did
freshmen
who
were
not
identity
achievers.
More importantly, students
who
became
identity achievers during
their
college
years
had
more cultural interests
as
freshmen than
did
students
who did not
enter
the
status.
Cultural interests
may
contribute either
to
the
creation
of
identity
crises
and/or
their
resolution.
The
exposure
to new
ideas through
cultural media
may
serve
to
challenge
the
views
with which
a
person
was
raised
and
suggest more promising
identity
alternatives
to
which
commitments
can be
formed.
A
possibly
related
variable shown
to be
associated
with
identity development
is ex-
pressive writing activity. Among samples
of
both
high school
and
college students, males
and
females
who
wrote poetry were
far
more
likely
to be in the
identity achievement status
than were students
who had
never written
poetry
(Waterman
&
Archer, 1979; Water-
man, Kohutis,
&
Pulone,
1977).
No
differ-
ences were
found
between students
who had
kept
a
personal journal
or
diary
and
those
who
had
not. Since expressive writing
was
found
to
have
started
relatively early
in ad-
olescence,
the
poetry writing activities were
almost certainly antecedent
to the
achieving
of
a
stable sense
of
identity.
It had
been originally expected that poetry
writing
and
journal keeping would yield sim-
ilar results.
The
differences
may be ex-
plained
in
terms
of the
ways
in
which
the
students approached
the two
types
of
expres-
sion.
Both groups
saw
their writings
as
help-
ing
to
increase their level
of
self-understand-
ing,
but
there were
differences
in the
themes
written about
and in the
perceptions
of the
functions
the
writing served (Waterman
&
Archer, 1979). Journal keeping appeared
to
be
used primarily
for
making
a
record
of
current activities
and for
describing rela-
tionships
with
friends
and
parents.
In
con-
trast, poetry writing
was
seen
as a
vehicle
for
the
creative expression
of
more abstract,
emotional concerns.
Further,
poetry writers
were
much more likely
to
want
to
share their
writings
than were journal keepers. Thus,
poetry
writing
may be a
technique that
is

IDENTITY
DEVELOPMENT
355
instrumental
in
working through
an
identity
crisis
by
aiding
in the
exploration
of
possible
identity alternatives
and by
serving
as a
means
for
gaining feedback from
others.
Rather than there being
a
direct,
facili-
tative connection between cultural interests,
poetry writing,
and
identity formation,
it is
possible that
all
three
may
relate
to
some
common
underlying psychological quality
involving
curiosity
and
exploration.
The im-
petus
to
explore
the
realm
of
ideas
and
feel-
ings
through cultural media
and
personal
emotional
states
through poetry writing
may
also
be
involved
in the
tendency
to
consider
a
variety
of
identity alternatives.
One
other variable that
may
serve
as an
antecedent
of
identity formation
is the in-
dividual's work history. Munro
and
Adams
(1977)
reported
a
comparison
of the
distri-
bution
of the
identity statuses between sam-
ples
of
college
students
and
working youth
of
equivalent age. They
found
the
working
sample
to be
more
frequently
in the
identity
achievement
status
and the
college
sample
to be
more
often
in the
identity
diffusion
status.
The
groups
differed
significantly with
respect
to
identity
status
in the
areas
of re-
ligious
beliefs
and
political ideology
but not
with
respect
to
vocational
choice.
Munro
and
Adams
(1977)
suggested that
full-time
em-
ployment
"might stimulate rapid movement
toward identity formation, while
college
at-
tendance
might
be
seen
as an
extended mor-
atorium
period"
(p.
523).
Concluding
Comments
The
basic
hypothesis embodied
in
Erik-
son's theory
of
identity
development—that
movement
from
adolescence
to
adulthood
involves
changes
in
identity that
can be
characterized
as
progressive developmental
shifts—fares
very well
in
empirical
studies.
Substantial
evidence exists
for
both
an in-
creased probability that consideration will
be
given
to
identity alternatives
and an in-
creased probability that personally mean-
ingful
commitments
will
be
formed. With
respect
to the
timing
of
development,
it ap-
pears that
only
limited changes occur prior
to or
during
the
high school years.
The
most
extensive advances
in
identity formation
oc-
cur
during
the
time
spent
in
college.
(Not
enough
research
has
been conducted with
noncollege samples
of
equivalent
age to de-
termine
the
relative
contribution
of
ontoge-
netic
and
situational variables
to
such
changes.)
The
period immediately
after
col-
lege
appears
most often
to
involve
a
consol-
idation
of the
sense
of
identity rather than
the
continued
exploration
of
identity alter-
natives.
Of
the
topic areas that
may
become
the
focus
of
identity concern,
the
greatest doc-
umentation
of
progressive developmental
shifts
exists
for
vocational choice. Identity
questions concerning sexual expression have
not
been studied longitudinally,
but the
high
probability
of the
identity achievement sta-
tus,
at
least
among women,
indicate
similar
progress
in
this area. Less consistency
has
been
observed
for the
patterns
of
identity
change
in the
areas
of
religious beliefs
and
political ideology, perhaps
as a
function
of
differences
in the
background
characteristics
of
the
participants
in
various
studies
or
changing
sociohistorical conditions.
Comparisons
of the
patterns
of
identity
formation
shown
by
males
and
females yield
far
more evidence
of
similarities than dif-
ferences. With
respect
to the
processes
of
development,
the
sexes show generally sim-
ilar
probabilities
of
consideration
of
identity
alternatives
and
establishment
of
commit-
ments. Only
in the
area
of
attitudes toward
premarital sexual activity
are
differences
evident.
Here,
females
appear
more likely
to go
through
a
period
of
moratorium
and
achieve identity commitments; males
are
more likely
to
maintain
the
early commit-
ments
on
which they
had
foreclosed.
The
similarities between
the
sexes
in
develop-
mental processes
do not
contradict
the
sub-
stantial
evidence
from
other research
that
males
and
females
differ
in the
content
of
their identity choices (Frieze, Parsons, John-
son, Ruble,
&
Zellman, 1978).
To
date
the
research
has
yielded only lim-
ited information with regard
to the
anteced-
ents
of
developmental change. Although
the
findings
concerning
the
relationship
between
parental behavior
and the
identity statuses
of
sons
and
daughters
are
consistent with
theoretical
expectations,
there
are a
number

356
ALAN
S.
WATERMAN
of
alternative explanations available
that
do
not
involve
a
causal influence
of the
parental
variables. Similarly, extensive
data
that
are
consistent with
a
hierarchical arrangement
of
Erikson's
psychosocial
stages
exist,
but
the
methodological limitations
of the re-
search
do not
permit
a
determination
of the
direction
of
effect.
Only
a
couple
of
person-
ality
or
activity
variables
have been identi-
fied
as
predictive
of the
direction
of
future
identity
formation.
What
is
needed
at the
present time
is re-
search
that
traces
identity development over
a
relatively
broad
range
of
ages.
In
partic-
ular, more information
is
needed about both
the
roots
of
identity
in the
years before
and
during high school
and the
nature
of
identity
changes during
the
adult years.
In
addition,
developmental research
is
particularly needed
on
samples
that
do not
attend college.
The use of
cohort-sequential methodology
(Schaie,
1965, 1970) would
aid in the
iden-
tification
of the
generality
of
developmental
patterns across cohorts.
It is
also recom-
mended
that
longitudinal studies
on
identity
development include
the
assessment
of
pos-
sible
predictor variables
of
identity
change.
Reference
Notes
1.
Marcia,
J. E.
Studies
in ego
identity. Unpublished
manuscript,
1976. (Available
from
J. E.
Marcia,
Department
of
Psychology,
Simon
Eraser
University,
Burnaby,
British Columbia, Canada
V5A
1S6.)
2.
Archer,
S. L. Ego
identity
development
among early
and
mid-adolescents. Paper presented
at the
meeting
of
the
Eastern Psychological Association,
New
York,
April
1981.
3.
Waterman,
A. S. Ego
identity
status
among mar-
ried,
adult women. Paper presented
at the
meeting
of
the
Eastern Psychological Association, Hartford,
Connecticut, April 1980.
4.
Dunivant,
N.,
Jr.,
&
Bieri,
J. A
multimethod
com-
parison
of
various measures
of
Erikson's
concept
of
ego
identity. Paper presented
at the
meeting
of the
Southwestern
Psychological
Association,
Dallas,
Texas, April 1973.
5.
Grotevant,
H. D.
Personal
communication, April
7,
1981.
6.
Rothman,
K. M.
Personal communication, Feb-
ruary
14,
1980.
7.
Marcia,
J.
E.,
&
Milter,
E. C. Ego
identity
in
mature
women.
Paper presented
at the
meeting
of the
East-
ern
Psychological Association, Hartford, Connecti-
cut, April
1980.
References
Adams,
G. R., &
Fitch,
S. A. Ego
stage
and
identity
status development:
A
cross-lag analysis. Journal
of
'Adolescence,
1981,4,
163-171.
Adams,
G.
R.,
&
Fitch,
S. A. Ego
stage
and
identity
status development:
A
cross-sequential analysis. Jour-
nal of
Personality
and
Social Psychology,
in
press.
Adams,
G.
R.,
&
Shea,
J. A. The
relationship between
identity
status,
locus
of
control,
and ego
development.
Journal
of
Youth
and
Adolescence, 1979,
8,
81-89.
Adams,
G.
R.,
&
Shea,
J.,
&
Fitch,
S. A.
Toward
the
development
of an
objective assessment
of
ego-iden-
tity
status. Journal
of
Youth
and
Adolescence,
1979,
8,
223-237.
Becker,
W. C.
Consequences
of
different
kinds
of pa-
rental
discipline.
In M. L.
Hoffman
& L. W.
Hoffman
(Eds.),
Review
of
child development research.
New
York:
Russell
Sage Foundation, 1964.
Berzonsky,
W.
M.,
Weiner,
A.
S.,
&
Raphael,
D. In-
terdependence
of
formal
reasoning. Developmental
Psychology,
1975,
//,
258.
Block,
J. Ego
identity, role
variability
and
adjustment.
Journal
of
Consulting Psychology, 1961,
25,
392-
397.
Bourne,
E. The
state
of
research
on ego
identity:
A
review
and
appraisal. Part
I.
Journal
of
Youth
and
Adolescence, 1978,
7,
223-251.
(a)
Bourne,
E. The
state
of
research
on ego
identity:
A
review
and
appraisal. Part
II.
Journal
of
Youth
and
Adolescence, 1978,
7,
371-392.
(b)
Cauble,
M. A.
Formal operations,
ego
identity,
and
prin-
cipled
morality:
Are
they
related?
Developmental
Psychology,
1976,
12,
363-364.
Ciaccio,
N. V. A
test
of
Erikson's theory
of ego
epi-
genesis. Developmental Psychology,
1971,
4,
306-
311.
Constantinople,
A. An
Eriksonian measure
of
personal
development
in
college students.
Developmental
Psy-
chology,
1969,
/,
357-372.
Constantinople,
A.
Some correlates
of
average
level
of
happiness
among college students. Developmental
Psychology,
1970,
2,
447.
Dignan,
M. H. Ego
identity
and
maternal identification.
Journal
of
Personality
and
Social Psychology, 1965,
/,
476-483.
Erikson,
E. H.
Young
man
Luther.
New
York: Norton,
1958.
Erikson,
E. H.
Identity
and
life
cycle: Selected papers.
Psychological
Issues, 1959,
/(I),
1-171.
Erikson,
E. H.
Childhood
and
society (2nd ed.).
New
York:
Norton
1963.
Erikson,
E. H.
Identity: Youth
and
crisis.
New
York:
Norton, 1968.
Erikson,
E. H.
Life
history
and the
historical moment.
New
York: Norton, 1975.
Frieze,
I.
H.,
Parsons,
J.
E.,
Johnson,
P.
B.,
Ruble,
D.
N.,
&
Zellman,
G. L.
Women
and sex
roles:
A
social psychological perspective.
New
York:
Norton,
1978.
i
Fry,
P. S.
Developmental changes
in
identity status
of
university
students
from
rural
and
urban back-

IDENTITY DEVELOPMENT
357
grounds.
Journal
of
College Student Personnel, 1974,
15,
183-190.
Grossman,
S. M.,
Shea,
J.
A.,
&
Adams,
G. R.
Effects
of
parental divorce during early childhood
on ego
development
and
identity formation
of
college stu-
dents. Journal
of
Divorce, 1980,
3,
263-272.
Gruen,
W,
Rejection
of
false
information
about oneself
as an
indication
of ego
identity. Journal
of
Consulting
Psychology,
1960,
24,
231-233.
Hauser,
S. T.
Black
and
white identity formation.
New
York:
Wiley-Interscience,
1971.
Hodgson,
J.
W.,
&
Fischer,
J. L. Sex
differences
in
identity
and
intimacy development
in
'college
youth.
Journal
of
Youth
and
Adolescence, 1979,
8,
37-50.
Howard,
L. P.
Identity conflicts
in
adolescent girls.
Smith
College
Studies
in
Social Work, 1960,
31,1-
21.
Inhelder,
B.,
&
Piaget,
J. The
growth
of
logical thinking
from childhood
to
adolescence.
New
York: Basic
Books, 1958.
Jordan,
D.
Parental antecedents
of ego
identity for-
mation. Unpublished master's thesis,
State
University
of
New
York
at
Buffalo,
1970.
Kohlberg,
L.,
&
Gilligan,
C. The
adolescent
as a
phi-
losopher:
The
discovery
of
self
in a
postconventional
world.
Daedalus, 1971, 100,
1051-1086.
La
Voie,
J. C. Ego
identity formation
in
middle
ado-
lescence. Journal
of
Youth
and
Adolescence, 1976,
5,
371-385.
La
Voie,
J.
C.,
&
Adams,
G. R.
Erikson developmental
stage
resolution
and
attachment
behavior
in
young
adulthood. Adolescence,
in
press.
Marcia,
J. E.
Development
and
validation
of
ego-iden-
tity
status.
Journal
of
Personality
and
Social Psy-
chology,
1966,
3,
551-558.
Marcia,
J. E.
Identity
six
years after:
A
follow-up
study.
Journal
of
Youth
and
Adolescence,
1976,5,145-160.
Marcia,
J. E.
Identity
in
adolescence.
In J.
Adelson
(Ed.),
Handbook
of
adolescent psychology.
New
York: Wiley, 1980,
Matteson,
D. R.
Exploration
and
commitment:
Sex
dif-
ferences
and
methodological problems
in the use of
identity
status
categories.
Journal
of
Youth
and Ad-
olescence,
1977,
6,
353-374.
.
Meilman,
P. W.
Cross-sectional
age
changes
in ego
identity
status
during
adolescence.
Developmental
Psychology,
1979,
IS,
230-231.
Munro,
G.,
&
Adams,
G. R.
Ego-identity formation
in
college
students
and
working youth. Developmental
Psychology, 1977,
13,
523-524.
O'Connell,
A. N. The
relationship between
life
style
and
identity synthesis
and
resynthesis
in
traditional,
neo-
traditional,
and
non-traditional women. Journal
of
Personality, 1976,
44,
675-688.
Orlofsky,
J. L.
Identity formation, nAchievement,
and
fear
of
success
in
college
men and
women. Journal
of
Youth
and
Adolescence,
1978,
7,
49-62.
Oshman,
H. P., &
Manosevitz,
M.
Father
absence:
Effects
of
stepfathers upon psychosocial development
in
males. Developmental Psychology,
1976,12,
479-
480.
Peterson,
R, E.
Technical manual: College
Student
Questionnaires. Princeton, N.J.: Educational
Testing
Service, 1965.
Podd,
M. H. Ego
identity status
and
morality:
The re-
lationship
between
two
developmental constructs.
Developmental
Psychology, 1972,
6,
497-507.
Pomerantz,
S. C. Sex
differences
in the
relative impor-
tance
of
self-esteem, physical self-satisfaction
and
identity
in
predicting adolescent satisfaction. Journal
of
Youth
and
Adolescence, 1979,
S,
51-61.
Poppen,
P. J. The
development
of sex
differences
in
moral judgment
for
college males
and
females.
Un-
published
doctoral dissertation, Cornell University,
1974.
Protinsky,
H. O., Jr.
Eriksonian
ego
identity
in
adoles-
cents. Adolescence, 1975,
10,
428-432.
Rasmussen,
J, E. An
experimental approach
to the
con-
cept
of ego
identity
as
related
to
character disorder.
Dissertation Abstracts, 1961,
22,
1711-1712. (Uni-
versity
Microfilms
No.
61-3723)
Rothman,
K. M.
Multivariate analysis
of the
relation-
ship
of
psychosocial crisis variables
to ego
identity
status.
Journal
of
Youth
and
Adolescence, 1978,
7,
93-105.
Rowe,
I.,
&
Marcia,
J. E. Ego
identity
status,
formal
operations,
and
moral development. Journal
of
Youth
and
Adolescence, 1980,
9,
87-99.
Schaie,
K. W. A
general model
for the
study
of
devel-
opmental
problems.
Psychological
Bulletin,
1965,64,
92-107.
Schaie,
K. W. A
reinterpretation
of age
related
changes
in
cognitive structure
and
functioning.
In L. R.
Goulet
&
P. B.
Baltes
(Eds.),
Life-span developmental psy-
chology:
Research
and
theory.
New
York: Academic
Press,
1970.
Simmons,
D. D.
Development
of an
objective measure
of
identity,
achievement
status.
Journal
ofProjective
Techniques
and
Personality Assessment, 1970,
34,
241-244.
Stark,
P.
A.,
&
Traxler,
A. J.
Empirical validation
of
Erikson's
theory
of
identity
crises
in
late
adolescence.
Journal
of
Psychology, 1974,
86,
25-33.
St.
Clair,
S.,
&
Day,
H. D. Ego
identity
status
and
values
among high school females. Journal
of
Youth
and
Adolescence, 1979,
8,
317-326.
Tesch,
S. A.
Intimacy
and
identity development
in
early
adulthood.
Unpublished
doctoral
dissertation,
Uni-
versity
of
Rochester, 1980.
Thompson,
M. St. G.
Modifications
in
identity:
A
study
of
the
socialization process during
a
sister formation
program. Unpublished
doctoral
dissertation,
Univer-
sity
of
Chicago,
1963.
Waterman,
A.
S.,
&
Archer,
S. Ego
identity
status
and
expressive writing among high school
and
college
stu-
dents.
Journal
of
Youth
and
Adolescence, 1979,
8,
327-341.
Waterman,
A.
S.,
Geary,
P.
S.,
&
Waterman,
C. K. A
longitudinal
study
of
changes
in ego
identity
status
from
the
freshman
to the
senior
year
at
college.
De-
velopmental
Psychology, 1974,
10,
387-392.
Waterman,
A.
S.,
&
Goldman,
J. A. A
longitudinal
study
of ego
identity development
at a
liberal arts

358
ALAN
S.
WATERMAN
college. Journal
of
Youth
and
Adolescence, 1976,
5,
Waterman,
C.
K.,
&
Nevid,
J. S. Sex
differences
in the
361
-369. resolution
of the
identity crisis. Journal
of
Youth
and
Waterman,
A.
S.,
Kohutis,
E., &
Pulone,
J. The
role Adolescence, 1977,
6,
337-342.
of
expressive writing
in ego
identity formation.
De-
Waterman,
C.
K.,
&
Waterman,
A. S.
Fathers
and
sons:
velopmental
Psychology,
1977,
13,
286-287.
A
study
of ego
identity
across
two
generations.
Jour-
Waterman,
A.
S.,
&
Waterman,
C. K. A
longitudinal
na\
Of
Youth
and
Adolescence, 1975,
4,
331-338.
study
of
changes
in
ego
identity status
during
the
whjtbou
s
K
Je,
B
M
&
Waterman)
A.
S.
reshman
year
at
college.
Developmental
Psychology,
^
Eriksonian
measure'
of
per;onaiity
development
... A
c<
o
««.•
i.
C-
v
TU
i
.
in
college students:
A
reexamination
of
Constant!-
Waterman,
A.
S
&
Whitbourne,
S. K.
The
Inventory
^
f
d
; ,
lication.
Developmental
of
Psychosocial
Development:
A
review
and
evalua-
Psvcnoiof,y
1982
;«
369-371
tion.
JSAS
Catalog
of
Selected Documents
in
Psy-
^ycnoiogy,
wvi,
is,
iw
ill.
chology,
1981,
//,
5.
(Ms.
No.
2179) Whitbourne,
S.
K.,
&
Waterman,
A. S.
Psychosocial
Waterman,
C.
K.,
Buebel,
M.
E.,
&
Waterman,
A. S.
development during
the
adult
years:
Age and
cohort
The
relationship between resolution
of the
identity comparisons. Developmental Psychology, 1979,
IS,
crisis
and
outcomes
of
previous psychosocial crises.
373-378.
Proceedings
of the
78th
Annual
Convention
of the
American Psychological Association,
1970,
5,
467-
468. (Summary)
Received
June
4,
1981
•
Manuscripts Accepted
for
Publication
Stages
of
Drug
Use in
Adolescence: Personality, Peer,
and
Family Correlates. Judith
S.
Brook (Mount
Sinai Medical Center,
One
Gustave
L.
Levy Place, Annenberg 22-74,
New
York,
New
York 10029),
Martin Whiteman,
and Ann
Scovell Gordon.
Effects
of
Maternal
Age on
Parenting Role. Arlene
S.
Ragozin (Department
of
Psychiatry
and
Behavioral
Sciences, Child Development
and
Mental Retardation Center, University
of
Washington,
Seattle,
Washington
98295), Robert
B.
Basham, Keith
A.
Crnic, Mark
T.
Greenberg,
and
Nancy
M.
Robinson.
Left-Right Memory
in
Two-Year-Old
Children:
A New
Look
at
Search Tasks.
Lila
Ghent Braine
(Department
of
Psychology-Hollingworth Laboratories, Barnard College, Columbia University,
New
York,
New
York
10027)
and
Rebecca
A.
Eder.
Characteristics
of
Hiding Places
and the
Transition
to
Stage
IV
Performance. Carl
J.
Dunst, Penelope
H.
Brooks
(George
Peabody
College
for
Teachers,
Vanderbilt University, Nashville,
Tennessee
37203),
and
Pamela
A.
Doxsey.
Clinical
Examination
of
Visual
and
Social Responses
in
Infant
Rhesus Monkeys. Morton
J.
Mendelson
(Department
of
Psychology,
Stewart
Biological
Sciences
Building,
1205
Docteur
Penfield
Avenue,
Montreal,
PQ,
Canada
H3A
1B1).
A
Developmental Study
of
Recall From Natural Categories. Joseph
M.
Fitzgerald (Department
of
Psychology, Wayne
State
University,
Detroit,
Michigan
48202).
Reasoning,
Prohibitions,
and
Motivations
for
Compliance. Leon Kuczynski (Laboratory
of
Develop-
mental
Psychology, Building
15K,
9000 Rockville Pike, Bethesda, Maryland 20205).
A
Naturalistic Study
of
Children's
Social
Understanding
as
Expressed
in
Reflective Interviews, Group
Discussions,
and
Group Task Negotiations. Robert
L.
Selman (Judge Baker Guidance
Center,
Boston,
Massachusetts
02115),
Mira Zamansky Schorin, Carolyn
R.
Stone,
and
Erin Phelps.
Sex
Roles
as
Social Conventions:
The
Development
of
Children's Conceptions
of
Sex-Role Stereotypes.
D.
Bruce Carter (Syracuse University, Department
of
Psychology,
331
Huntington Hall,
150
Marshall
Street,
Syracuse,
New
York
13210)
and
Charlotte
J.
Patterson.
Network
Representation
of a
Child's Dinosaur Knowledge. Michelene
T. H. Chi
(Learning Research
and
Development Center,
University
of
Pittsburgh, LRDC
Bldg.,
3939
O'Hara
Street, Pittsburgh,
Pennsylvania
15260)
and
Kandi
Daimon Koeske.
Temporal Inferences
by
Young Children. Daniel
M.
Fajardo (The Commonwealth
of
Massachusetts,
University
of
Massachusetts, Department
of
Psychology, Amherst, Massachusetts
01003)
and
Benson
Schaeffer.
The
Role
of
Manipulation
in
Infants' Detection
of
Invariant
Properties
of
Objects. Holly
A.
Ruff
(Department
of
Pediatrics, Albert Einstein
College
of
Medicine,
1300
Morris Park Avenue, Bronx,
New
York 10561).
(Continued
on
page 430)