You are on page 1of 9

Developmental Psychology

1994, Vol. 30, No. 5,624-632

Copyright 1994 by the American Psychological Association, Inc


0012-1649/94/S3.00

Family Patterns of Moral Judgment During


Adolescence and Early Adulthood
Betsy Speicher
Family patterns of moral reasoning were compared in a cross-sectional sample from the Oakland
Growth Study (H. E. Jones, 1939a, 1939b) and a longitudinal sample from L. Kohlberg's (1958)
study of moral judgment development. The 221 offspringin the 2 samples( 121 male and 100 female)
ranged from 10 to 33 years old. Age, sex, cognitive stage, IQ, socioeconomic status, and education
were controlled in the data analyses. There were consistent family patterns of moral reasoning in the
2 samples when both sex and background variables were controlled. Developmental patterns indicated that, during adolescence, parent moral judgment was related to offspring moral reasoning but
was a stronger predictor of moral judgment among girls than boys. During young adulthood, fathers'
moral judgment and education were the strongest predictors of both sons' and daughters' moral
reasoning. However, education, not parent moral reasoning, limited the moral stage attained by
adult offspring.

This study of familial patterns of moral judgment compares


relationships between parent and offspring moral reasoning
from early adolescence to adulthood in a cross-sectional sample
from the Oakland Growth Study (Jones, 1939a, 1939b) and a
longitudinal sample from Colby and Kohlberg's (1987) 20-year
study of moral judgment development. Thefindingspresented
here are part of an expanding body of cognitive-developmental
research on the role of parents and family processes in the development of moral judgment.
In his classic "Stage and Sequence" paper, Kohlberg (1969)
described the cognitive-developmental approach to socialization and his six-stage model of moral judgment development.
Strongly influenced by Piaget's (1932/1965) theory of cognitive
development, Kohlberg focused on moral cognition and viewed
moral development as a process of increasing differentiation
and integration of perspectives of self and other in resolving
moral conflict. Accordingly, Kohlberg's approach to socialization emphasized the importance of social experiences that produce cognitive disequilibrium, challenge and reveal inadequacies in the child's reasoning, provide exposure to higher stage
reasoning, and encourage social perspective taking.
Like Piaget (1932/1965), Kohlberg (1969) downplayed the
role of the family in the development of moral reasoning and

focused on the peer group and schools. He argued that parent


moral reasoning should neither limit nor be related to the stage
attained by offspring. According to Kohlberg, neither family
participation nor identification with parents is critically necessary for moral development; the family is merely one of several
social institutions that promote moral development through the
creation of "role-taking opportunities" (Kohlberg, 1969, p.
399). As suggested by Powers (1988), this theoretical position
stemmed from an attempt to distinguish cognitive-developmental theory from psychoanalytic and social-learning approaches that focused on parents as the primary contributors to
a child's moral socialization.
Perhaps as a result of Kohlberg's (1969) theoretical bias
against families, there has been relatively little research exploring the contribution of the family to the development of moral
judgment. However, as cognitive-developmental researchers increasingly explore the role of family factors in the development
of children's moral reasoning, there is a growing body of empirical evidence that Kohlberg had underestimated the importance
of the family. The confluence of results suggests that, in both
cognitive and affective domains, reported and observed parental
behavior is related to their children's moral reasoning.
In the cognitive domain, studies comparing relationships between various discipline techniques and children's moral reasoning indicate that advanced moral reasoning is most consistently related to reported parental use of other-oriented inductive discipline, that is, cognitive techniques that emphasize
reasoning, encourage the child to take perspectives of others,
and rationally teach the child to understand the consequences
of actions (Buck, Walsh, & Rothman, 1981; Holstein, 1972;
Parikh, 1980; Shoffeitt, 1971). Results of observational studies
examining relationships between family interaction styles and
children's moral reasoning provide support for both the importance of family environmental factors and the cognitive-developmental view that moral development is promoted by role-taking opportunities, including parent encouragement of their
child's participation in family discussions (Holstein, 1969; Par-

Betsy Speicher, Harvard University (now at Clinical Developmental


Institute, Belmont, Massachusetts).
This article is a partial report of my doctoral dissertation, Harvard
Graduate School of Education, 1982. It was presented in April 1985 at
the biennial meeting of the Society for Research in Child Development,
Toronto, Ontario, Canada.
I am grateful to the Institute of Human Development, University of
California, Berkeley, for generously providing access to the data from
the Oakland Growth Study, and to the late Lawrence Kohlberg, who
donated data from his longitudinal study of moral judgment development.
Correspondence concerning this article should be addressed to Betsy
Speicher, 11 Trotting Horse Drive, Lexington, Massachusetts 02173.
624

625

FAMILY PATTERNS OF MORAL JUDGMENT

ikh, 1980), parent use of reasoning (Buck etal, 1981), and parent discussion styles that elicit and re-present the reasoning of
others (Walker & Taylor, 1991).
In the affective domain, the role of positive family relationships was first suggested by studies exploring connections between children's moral reasoning and self-report measures of
parent affection and involvement (Fodor, 1973; Hart, 1988;
Hoffman & Saltzstein, 1967; Holstein, 1969; Parikh, 1980;
Shoffeitt, 1971; Speicher, 1992). These studies consistently
found relationships between children's moral reasoning and
parent warmth or involvement. In Kohlberg's (1958) own longitudinal sample, Hart (1988) found that reported paternal
affection and involvement, measured during adolescence, predicted sons' moral judgment during childhood, adolescence,
and adulthood. Results of observational studies also have consistently found an association between positive parent-child interaction, for example, parental warmth (Buck et al., 1981), low
maternal hostility (Jurkovic & Prentice, 1974), maternal affective support (Powers, 1982/1983), and high parent supportive
interactions (Walker & Taylor, 1991) and moral judgment maturity.
The final aspect of the family social environment studied by
cognitive-developmental researchers is the parents' own moral
reasoning. As noted by Powers (1988), parent moral reasoning
is seen as a potentially important aspect of the family social
environment for two reasons. First, experimental research has
shown that moral reasoning is stimulated by exposure to optimal ranges of higher stage thinking (Berkowitz, Gibbs, &
Broughton, 1980; Rest, Turiel, &Kohlberg, 1969;Turiel, 1966;
Walker, 1982). Therefore, it seems reasonable to examine
whether there is a relationship between parent and child moral
judgment because parents whose own moral judgment development is more advanced than that of their children should be
better able to stimulate moral reasoning by exposing children
to more complex moral arguments. Second, parent moral judgment may be related to child-rearing philosophy and behavior.
Research results indicate that higher stage parents are significantly more likely than lower stage parents to report the use
of cognitive methods of moral transmission and democratic
family decision making (Speicher-Dubin, 1982); to encourage
their children's participation in moral discussions (Holstein,
1969; Parikh, 1975); and to be supportive, sharing, and challenging in family discussions (Powers, 1982/1983). It seems
likely that both differential parent behavior and moral reasoning contribute to hypothesized relationships between parent
and offspring moral judgment at various points in a child's development.
Although previous studies of family moral judgment patterns
found some significant relationships, results were inconsistent
and varied depending on the age of the offspring and the sex of
the parents and children (Buck et al., 1981; Dunton, 1988/
1989; Haan, Langer, & Kohlberg, 1976; Holstein, 1969, 1975;
Parikh, 1980; Powers, 1982/1983; Shoffeitt, 1971; Walker &
Taylor, 1991). Inconsistencies across studies outweighed consistent patterns, and there were serious methodological weaknesses
in many of the studies, including the use of unreliable and unstandardized versions of Kohlberg's (1958) scoring system, data
analyses that unnecessarily reduced variability of scores, and
inconsistent use of control variables such as age, IQ, socioeco-

nomic status (SES), sex, and education in the research designs.


Even among the studies that used the final, published Standard
Issue moral judgment scoring system (Colby et al., 1987), consistent patterns did not emerge (Dunton; 1988/1989; Powers,
1982/1983; Walker & Taylor, 1991).
The present study is a reanalysis of cross-sectional parentoffspring moral judgment data from the Oakland Growth Study
(OGS), previously reported by Haan et al. (1976), and an analysis of parent-offspring moral judgment data from Kohlberg's
(1958) longitudinal study. Research methodology has been improved in three ways. First, all of the moral judgment data were
scored according to the final standardized and published Standard Form Scoring system (Colby et al., 1987). Second, potentially confounding variables such as age, sex, cognitive stage, IQ,
SES, and education were controlled in the data analyses. Third,
comparisons of longitudinal and cross-sectional family moral
judgment relationships permitted an exploration of consistent
patterns of results.
The purpose of this study was to describe family patterns of
moral reasoning across age and sex groups and to explore crosssectional and longitudinal consistencies in the patterns of results. Although a correlational approach does not reveal how
parents actually reason with their children (Walker & Taylor,
1991), a description of developmental patterns can suggest directions for more explanatory research.

Method
Subjects
The subjects from the OGS, a longitudinal study begun by Jones
(1939a, 1939b), were all lower- to upper-middle-class White natives of
California and included 82 mothers and 75 fathers from the original
longitudinal sample, and 100 male and 100 female offspring. There
were 98 adolescent offspring (ages 10 to 18 years) and 102 early adult
offspring (ages 19 to 31 years). All of the subjects were interviewed with
the Kohlberg Moral Judgment Interview (Colby et al., 1987) by researchers at the Institute of Human Development, University of California, Berkeley, during a follow-up wave in 1969/1970. A small subset was
reinterviewed in 1975. Although the OGS parent cohort was followed
longitudinally, the present sample is cross-sectional because the offspring ranged in age from 10 to 31 years and most of the subjects were
interviewed only during the 1969/1970 follow-up wave.
A comparison of the educational and occupational attainment of
male and female parents shows fairly wide disparities between the two
groups. A majority (60%) of the male parents were college graduates,
whereas an overwhelming majority (83%) of the female parents had not
graduated from college. Similarly, almost half of the male parents were
professionally employed; a subtantial majority (79%) of the female parents were either clerical and sales personnel or were unemployed.
The subjects from Kohlberg's (1958) longitudinal study of moral
judgment development included 21 males whose parents (19 mothers
and 18 fathers) completed moral judgment interviews in 1959. The original Kohlberg sample consisted of three age groups: 10-, 13-, and 16year-olds. Although moral judgment interviews were administered six
times at 3-year intervals since 1955, the number of interviews per subject varied.
Socioeconomically, the present Kohlberg sample included 8 subjects
with lower-middle-class backgrounds and 13 subjects with upper-middle-class backgrounds. Educational and occupational attainment of the
female parents was not available. All of the subjects were White residents of the suburbs of Chicago.

626

BETSY SPEICHER

Measures
Moral judgment. Moral judgment was measured from responses to
the semistructured Kohlberg Moral Judgment Interview, which consists
of hypothetical dilemmas and a series of probing questions designed
to elicit the structure of the individual's moral reasoning. All moral
judgment interviews were scored blind according to the Standard Issue
Scoring System (Colby et al., 1987), which is both a reliable and valid
psychometric measure. Colby and Kohlberg (1987) presented strong evidence of the reliability (test-retest, interrater, and alternate form) and
construct validity of the Standard Issue Scoring System.
The Colby et al. (1987) manual contains stage descriptions and prototypic responses for the five moral judgment stages. At Stage 1, which
is normative during middle childhood, morality is equated with avoidance of punishment and obedience to rules and authority for its own
sake. Older children and early adolescents typically reason at Stage 2
and view morality from a concrete individualistic perspective in which
right action is relativistic and involves serving one individual's interests,
or making concrete exchanges. The primary concerns at Stage 3, which
usually develops during adolescence, are maintaining interpersonal relationships, following the golden rule, living up to the expectations of
others, being a good person with good motives, and so on. At Stage 4,
the individual takes a broad societal, rather than individual or interpersonal, perspective in making moral judgments and assesses the effects of
actions on the social system. Finally, at Stage 5 the individual takes a
prior-to-society perspective and judges the morality of actions in terms
of upholding universal human rights and the social contract, and maximizing social utility and welfare.
Most of the Kohlberg sample interviews were scored by the same
three experienced raters whose reliability was reported by Colby and
Kohlberg (1987); interviews of 3 subjects were cases used in the construction of the Colby et al. (1987) manual. All other moral judgment
interviews were scored by the investigator, one of the three raters who
coded the Kohlberg longitudinal data.
Interrater reliability for the OGS moral judgment data was obtained
from a subset of 30 parent and offspring interviews that were independently scored according to the Colby et al. (1987) manual. Standard
issue scoring of Kohlberg Moral Judgment Interviews yields two types
of scores. One is a nine-integer stage score ranging from Stages 1 to 5 and
including 4 transitional stages (e.g., Stage 3/4). The other is a weighted
average score (WAS), which is a continuous score, ranging from 100 to
500, that is derived from a weighted average of all scorable responses in
an interview multiplied by 100. Eighty-seven percent of the nine-integer
stage scores were within one third of a stage. The Pearson product moment correlation, based on continuous WAS, was .92 and was comparable with correlations reported by Colby and Kohlberg (1987).
Background variables. In the OGS sample, intelligence was assessed by the individually administered Wechsler-Bellveau Intelligence
Scales, and in the Kohlberg sample, intelligence was measured from
group IQ tests (e.g., the Otis Group Intelligence Scale and the Thurstone
Primary Mental Abilities Test). Social class of OGS male and female
parents was measured by the Hollingshead Index of Social Position
(Hollingshead & Redlich, 1958). In the Kohlberg sample, social class
was a dichotomous measure (upper-middle-class and lower-middleclass) based on the educational and occupational level of the male parent. Job status was based on the 7-point Hollingshead scale. For Kohlberg subjects, education was measured in years, and for OGS subjects,
education was measured by the Hollingshead Education Scale. In the
OGS sample, cognitive stage was measured from three Piagetian tests
adapted from Inhelder and Piaget (1958) and scored by Institute of Human Development researchers by a scoring system described by Kuhn,
Kohlberg, Langer, and Haan (1977).

Design
Parent-offspring moral judgment data from the cross-sectional OGS
study and from Kohlberg's longitudinal study were analyzed separately,

and the patterns of relationships were compared during adolescence


(ages lOto 18) and early adulthood (ages 19 to 33). These two age ranges
were selected for two reasons. First, the number of subjects in the two
groups was approximately equal and the sample size in each of the two
age groups was sufficiently large to permit control of background variables. Second, because normative middle-class adolescents typically begin the maturational process of establishing physical independence
from parents during the college years, the inclusion of 19-year-olds in
the young adult group was reasonable from a developmental perspective.
Parent-offspring moral judgment correlations were analyzed separately for each sex and age group. That is, for each of the two age groups,
there was an analysis of father-son, mother-son, father-daughter, and
mother-daughter moral judgment relationships. When background
variables (age, parent SES, offspring education, IQ, and cognitive stage)
were significantly related to offspring moral judgment in a particular
age or sex group, these variables were controlled in analyses of relationships between parent and offspring moral judgment.

Results
Relationships Between the Background Variables and
Offspring Moral Judgment
Correlations between background variables and offspring
moral reasoning can be found in Tables 1 and 2. With the exception of age, which was strongly related to moral reasoning
during late childhood and adolescence but unrelated to moral
reasoning during adulthood, correlations between background
variables and moral judgment in the OGS sample generally increased between adolescence and adulthood. For example,
whereas none of the correlations between cognitive stage and
moral judgment was significant between ages 10 and 18, there
were significant relationships between these two variables
among both males and females during adulthood. Similarly,
correlations among intelligence, parent SES, and moral judgment were lowest in the adolescent OGS age group and highest
in the adult group. Increasing correlations between intelligence
and moral judgment in the male Kohlberg sample is consistent
with patterns seen in the OGS sample. The only inconsistency
across samples was a developmental pattern of decreasing correlations between fathers' SES and moral judgment among
Kohlberg sample sons.

Relationships Between Parent and Offspring


Moral Reasoning
Zero-order and partial correlations between parent and offspring moral reasoning in the OGS cross-sectional sample and
Kohlberg longitudinal sample during adolescence and early
adulthood appear in Tables 3, 4, 5, and 6. In the OGS sample,
results were highly consistent across sex and age groups. During
both adolescence and young adulthood, there were moderate,
significant correlations between parent and offspring moral reasoning both when sex groups were combined and when correlations for boys and girls were analyzed separately. In the
Kohlberg sample, there were also positive associations between
parent and son moral judgment, although the only significant
findings were correlations between father and son moral judgment among the young adults.
Significant relationships between parent and offspring moral

627

FAMILY PATTERNS OF MORAL JUDGMENT

Table 1
Correlations Between Background Variables and Offspring Moral
Judgment: Oakland Growth Study Sample
19-31-year-olds

10-18-year-olds
Sons

Background variable

.79***
(52)

Age

Cognitive stage3
(correlational problem)
Intelligence

.12

(32)
.15

Mothers' SES

(49)
-.03
(47)

Fathers' SES

.15

(45)

Daughters

Children

.52***
(46)
-.07
(24)
.29*
(44)
-.02
(45)
.16
(43)

.66***
(98)
.09
(56)
.20*
(93)
-.03
(92)
.18
(88)

Education

Sons

Daughters

Children

.42*
(28)
.43**
(42)
.45**
(46)
.30*
(46)
.58***
(44)

.41*
(27)
.38**
(46)
.20
(53)

.44***
(55)
.41***
(88)
32***
(99)
.25*
(88)
.55**.
(92)

.22

(42)
.51***
(48)

Note. Sample sizes are in parentheses. SES = socioeconomic status.


a
Scores on the Piagetian correlational problem were more consistently correlated with moral reasoning
than scores on the combinatorial and pendulum problems, and they were used as the measure of cognitive
stage in analyses that used cognitive stage as a control variable.
* p < . 0 5 . **p<.0l. ***p<.001.

reasoning were further analyzed in a series of partial correlations and multiple regressions, with the effects of background
variables that were significantly related to offspring moral reasoning in a given sex and age group controlled. In the Kohlberg
young adult sample, these analyses entailed partial correlations
within 3-year age groups (see Table 6). This approach avoided
inclusion of repeated measures from the same longitudinal subjects in a single analysis. In the larger OGS cross-sectional sample, background variables that were significantly related to offspring moral judgment were controlled by entering these variablesfirstin multiple regressions equations.
With background variables and sex controlled, further consistent patterns in the two samples emerged. For example, when
sex groups in the adolescent OGS sample were combined, both
mothers', F(3,82) = 3.06,p < .05, and fathers', F(3,82) = 3.24/>
< .05, moral judgment significantly predicted adolescent moral
reasoning after age and IQ were controlled. However, when fam-

ily moral judgment relationships were analyzed within each sex


group, gender differences were revealed. Whereas both mothers', F(3, 37) = 3.82, p < .025, and fathers', F(3, 37) = 3.38, p <
.025, moral judgment significantly predicted daughters' moral
judgment after age and IQ were controlled, neither parents'
moral judgment was a significant predictor of sons' moral judgment after age was controlled (see Tables 3 and 4). These latter
findings among adolescent OGS boys are consistent with the
nonsignificant parent-son correlations in the Kohlberg sample
(see Table 5).
During early adulthood, thefindingsin the two samples also
showed similar patterns after background variables were controlled. As shown in Table 5, there were positive relationships
between mother and son moral judgment in the Kohlberg sample, but the correlations were not significant. Likewise, after
background variables in the OGS sample were controlled,
mothers' moral reasoning did not significantly predict offspring

Table 2
Correlations Between Background Variables and Sons' Moral Judgment: Kohlberg Sample
Age group (in years)
Background
variable
Intelligence
Father's job status
Fathers' SES
Education

10

.00
(10)
.58
(10)
.63*
(10)

13-14

16-18

20-22

24-26

.39

.08

.15

.46

(19)

(17)

(15)

(13)

28-30
.74**
(14)

.35

.25

.50

.13

.05

(19)

(17)

(15)

(13)

(14)
-.01
(14)
.56*
(12)

.29

.25

.47

.20

(19)

(17)

(15)

(13)
.57*
(10)

.41

(12)
Note. Sample sizes are in parentheses. SES = socioeconomic status.
*p<.05.
**p<.01.

32-33
.65
(8)
.60
(8)
.22
(8)
.59
(6)

628

BETSY SPEICHER

Table 3
Zero-Order and Partial Correlations Between Mother and Offspring
Moral Judgment: Oakland Growth Study Sample
Son

Daughter

Child

Age

Zero-order r

Partial r

Zero-order r

Partial r

Zero-order r

Partial r

10-18 years

.26*
(51)
.28*
(47)

.15"
(48)
.16
(25)
.14f
(25)

.40**
(45)
.31*
(51)

.31*"
(41)
.10"
(39)

.32***
(96)
.28**
(98)

.19*b
(86)
.04e
(44)

19-31 years

Note. Sample sizes are in parentheses.


* Age controlled. "Age and IQ controlled. 'Education, IQ, cognitive stage, and fathers' socioeconomic
status (SES) controlled. "Education and IQ controlled. "Education, IQ, cognitive stage, fathers' SES, and
mothers' SES controlled, tducation, IQ, cognitive stage, and mothers' SES controlled.
*p<.05. **p<.01. ***/;<.001.

moral judgment in either sex group or when sex groups were


combined (see Table 3). Mothers' moral judgment did not significantly increase the proportion of explained variance in offspring moral judgment after education, IQ, cognitive stage, and
fathers' SES were controlled in the adult male group, F(5, 19) =
.392, or after education and IQ were controlled in the adult female group, F(3, 35) = .585.
Relationships between father and son moral judgment in the
two young adult samples were also strikingly similar. In the
Kohlberg sample, father-son correlations were significant in the
combined age-19-to-33 adult Kohlberg group (in which there
were longitudinal repeated measures at 3-year intervals for most
subjects), and in each of four 3-year age groups (without repeated measures). Zero-order correlations within the 3-year age
groups ranged from .59 to .96. As Table 6 indicates, partial correlations between father and son moral judgment in the adult
Kohlberg age groups were in the moderate (.42) to high (.95)
range even after the effects of background variables were controlled.
The strength of fathers' moral judgment as a predictor of the
adult moral reasoning of the Kohlberg subjects can be further
demonstrated by the proportion of variance in sons' moral judg-

ment explained by fathers' moral judgment after other variables


were controlled. Among subjects in their early 20s, fathers'
moral judgment explained 28% of the variance in sons' moral
reasoning after fathers' job and education were controlled, and
40% of the variance after fathers' SES and education were controlled. After controlling for IQ and education, fathers' moral
judgment explained 48% of the variance among subjects in their
late 20s.
As in the Kohlberg sample, fathers' moral judgment continued to be a strong predictor of adult sons' moral judgment after
background variables were controlled in the OGS adult male
sample. Fathers' moral judgment remained a significant predictor of sons' moral judgment after IQ, education, cognitive stage,
and fathers' SES were controlled, F(5, 19) = 3.52, p < .025, and
increased the amount of explained variance from 56% to 63%.
Fathers' moral judgment was also significantly related to
moral reasoning of adult daughters. Although education was
consistently a strong predictor of moral reasoning maturity during adulthood, fathers' moral judgment significantly increased
the explained moral judgment variance in female offspring after
education and IQ were controlled, F(3, 27) = 4.21, p < .025,
and in the combined male and female sample, after education,

Table 4
Zero-Order and Partial Correlations Between Father and Offspring
Moral Judgment: Oakland Growth Study Sample
Daughter

Son

Child

Age

Zero-order r

Partial r

Zero-order r

Partial r

Zero-order r

Partial r

10-18 years

.28*
(48)
.34**
(48)

.12"
(48)
.40* '"
(25)

.28*
(43)
.38**
(43)

.29* "
(41)
.37*'
(31) '

.28**
(91)
.37***
(91)

.20*"
(86)
.29** f
(44)

19-31 years

Note. Sample sizes are in parentheses.


* Age controlled. "Age and IQ controlled. 'Education, IQ, cognitive stage, and fathers' socioeconomic
status (SES) controlled. "Education, IQ, cognitive stage, and mothers' SES controlled. "Education and
IQ controlled. f Education, IQ, cognitive stage, fathers' SES, and mothers' SES controlled.
*/><.05. **p<.01. /><.001.

629

FAMILY PATTERNS OF MORAL JUDGMENT

Table 5
Zero-Order Correlations Between Father and Son
Moral Judgment: Kohlberg Sample
Zero-order r

Age

Mothers' moral
judgment

Fathers' moral
judgment

.06
(10)
-.19
(18)
-.24
(16)
.20
(15)
.16
(13)
-.37
(13)

.39
(9)
.45
(16)
.31
(14)
.70**
(13)
.66*
(12)
.59*
(11)
.96**
(6)
.20
(40)
.50***
(44)

10 years
13-14 years
16-18 years
20-22 years
24-26 years
28-30 years
32-33 years
10-18 years
19-33 years

.38
(7)
.18
(45)
.15
(47)

Note. Sample sizes are in parentheses.


*/><.05. **p<.01. **/><.001.

cognitive stage, fathers' SES, mothers' SES, and IQ were controlled, F(6, 37) = 3.41, p< .01.
Differences in the relative contribution of mothers' versus fathers' moral judgment as predictors of offspring moral judgment during adulthood may, in part, be related to wide disparities in educational and occupational attainment and in distributions of moral judgment stages among male and female
parents. Only 2% of the OGS mothers and none of the mothers
in the Kohlberg sample were reasoning at Stage 4 or Stage 4/5

Table 6
Partial Correlations Between Father and Son
Moral Judgment: Kohlberg Sample
Age group (in years)

Background variable
IQ
Education
Fathers* job
Fathers' SES
IQ and education
Fathers' job and education
Fathers' SES and
education

20-22
(n=13)
.62*
.59*
.79**
.53

24-26
(n=12)

28-30
(= 11)

.60*
.50

.52
.42

.50

.69*

32-33
(n = 6)
.95**
.95*
.94

.63*

Note. Because of the small number of subjects in each of the adult


age groups and the extremely high correlations coefficients required to
achieve statistical significance, all background variables that had a
greater than .40 correlation with sons' moral judgment were included
in the partial correlations. SES = socioeconomic status.
* p < . 0 5 . **p<.01.

compared with 35% of the OGS fathers and 17% of the Kohlberg fathers. In both samples there were also significant differences in WAS and in variability of scores. Among OGS fathers,
the mean WAS was 365 compared with 325 among mothers,
4374) = 9.02, p < .05, two-tailed, and there were significant
differences in variance, F( 181, 193) = 1.11, p < .05, two-tailed.
The mean WAS of Kohlberg sample fathers and mothers were
344 and 321, respectively, t(35) = 2.10, p< .05, two-tailed, and
there were also significant differences in variability, F( 17, 18) =
2.94, p < .05, two-tailed. These sex differences must be taken
into account when interpreting family patterns of moral reasoning.
Limitation on Development
In support of Kohlberg's (1969) theoretical expectations, results from both samples clearly indicate that parent moral judgment did not limit the adult stage attained by offspring. Among
OGS adult offspring, 30% exceeded their mothers' moral stage;
24% exceeded their fathers' stage; and 22% exceeded the moral
stage of both parents. In the Kohlberg sample, after age 22, 53%
of the adult children had exceeded the moral stage of both parents. Graduation from college, however, was a critical factor in
determining which subjects developed postconventional (Stage
5) reasoning and which subjects achieved more advanced moral
reasoning than their parents. All of the subjects who exceeded
the moral stage of both parents had graduated from college.
Ninety percent of the principled subjects had conventional
(Stages 3 and 4) parents, but all had graduated from college.
Discussion
Across both the Kohlberg longitudinal sample and the OGS
cross-sectional sample, there were numerous positive associations between parent and offspring moral judgment. Consistent
family patterns of moral reasoning were found in the two samples when both sex and background variables were controlled.
Developmental patterns indicated that, during adolescence,
parent moral judgment was related to offspring moral reasoning
but was a stronger predictor among girls than among boys;
among young adults, fathers' moral judgment and education
were the strongest predictors of both sons' and daughters' moral
reasoning; but education, not parent moral reasoning, limited
the moral stage attained by offspring.
The finding that there were very similar patterns of family
moral reasoning in the two samples after both sex and background variables were controlled underscores the importance
of controlling for potentially confounding variables and helps
to explain inconsistencies in much of the previous research. For
example, Walker and Taylor (1991) found positive but nonsignificant trends in parent-child moral judgment correlations in
a sample of 6- to 16-year-olds, but they did not control for age.
Given the presentfindingthat age was the strongest predictor of
adolescent moral reasoning (see Table 1), it is possible that
Walker and Taylor's results were confounded by age. During
early adulthood, age was unrelated to moral reasoning; the
strongest predictors were education and fathers' moral reasoning. It is noteworthy that in both the Kohlberg longitudinal sample and the OGS cross-sectional sample, fathers' moral reason-

630

BETSY SPEICHER

ing remained significantly related to their adult children's reasoning after education and other background variables that
were significantly correlated with moral judgment were controlled.
Although consistent family moral judgment patterns found
in this study cannot explain the mechanisms of development,
the findings can be interpreted theoretically, and in conjunction
with other empirical evidence, to generate hypotheses for future
research. For example, sex differences in the relative strength of
parental moral judgment in predicting offspring moral reasoning may be related to stage disparity between parents and children. In support of Kohlberg's theoretical claim that cognitive
conflict is the mechanism that promotes moral judgment development, Walker and Taylor (1991) found that a relatively large,
one-stage difference between parent and child moral reasoning
was predictive of development over a 2-year period, and that
parents provide maximal exposure to higher stage reasoning by
accommodating their moral reasoning to that of their child during moral discussions. Given these results, it seems likely that
the relative strength of mothers' moral judgment as a predictor
of child moral judgment during adolescence and of fathers'
moral judgment as a predictor of offspring moral stage in adulthood may be related to sex differences in parent moral judgment. Mothers in both samples were significantly lower in stage
of moral reasoning, and their moral judgment evidenced significantly less variability than their male counterparts. During
adolescence, when subjects in both samples were in a transition
between Stages 2 and 3, the Stage 3 and Stage 3/4 reasoning of
their mothers may have been optimally discrepant to stimulate
development. In early adulthood, when Stages 4 and 5 reasoning
was developing, fathers' moral judgment was a strong predictor
and may have been optimally discrepant.
Sex differences in parent-child moral judgment relationships
during adolescence are unexplainable from a cognitive-developmental perspective. With age controlled, neither parents'
moral judgment was significantly related to sons' moral judgment in the two adolescent samples. Daughters' adolescent patterns were quite different: There were significant relationships
between the moral reasoning of both parents and their daughters. These results may reflect differential sex role socialization,
as Conger's (1977) assertion that boys have traditionally received more cultural support for independence and assertiveness whereas girls are expected to be dependent and compliant has strong empirical support (Block, 1983; Saegert & Hart,
1976, cited in Ruble, 1988; Zern, 1984). The OGS adolescent
girls may have been more dependent on parents and amenable
to parental influence.
The finding that family patterns of moral reasoning change
with development supports a stage-environment fit hypothesis
(Eccles et al., 1993) and suggests that environmental factors that
stimulate development may change as children mature. As previously noted, there is increasing empirical evidence of a connection between positive intrafamilial relationships and the development of moral reasoning during adolescence. In a self-report study of a subset of OGS adolescents, Speicher (1992)
found that adolescent moral judgment was most consistently
related to adolescent perceptions of affectively positive family
relationships, family communication, and parent understanding and support. Contrary to the expectations of cognitive-de-

velopmental theory, moral reasoning during adolescence was


unrelated to reports of democratic, child-participatory family
structures and decision making. Consistent with these results,
the observational studies of Walker and Taylor (1991) and Powers (1982/1983) both indicated that moral judgment development was negatively associated with affective conflict but positively associated with interactions that were affectively supportive and that encouraged perspective taking.
Although affective and interpersonal factors in the family environment may have an important role in the consolidation of
Stage 3 reasoning during adolescence, the finding that graduation from college was a critical factor in the adult stage of moral
judgment, and that education was one of the strongest predictors of moral judgment among adult offspring, suggests that
cognitive and intellectual stimulation may optimally promote
development during early adulthood. A strong relationship between higher education and advanced moral judgment also has
been reported by Colby and Kohlberg (1987), Rest and Narvaez
(1991), Rest and Thoma (1985), Walker (1986), and White
(1988). Normative developmental processes during the young
adult and college years may also underlie both the strong relationship between education and moral judgment and between
father and offspring moral judgment. The early adult years are
of critical importance in the development of both ideological
identity (Douvan & Adelson, 1966; Marcia, 1980) and values
(Perry, 1970); and the development of moral reasoning and
identity appear to be parallel processes (Marcia, 1980). The
challenge to previously held beliefs, values, and epistemology
inherent in the college experience may stimulate the young
adult's interest in examining and challenging parental beliefs,
resulting in intrafamilial moral dialogue and increased parental, and particularly paternal, contribution to identity and
moral judgment development. This interpretation is supported
by Rest and Narvaez's (1991) conclusion that "one of the influences of the college experience is that it provides general intellectual stimulation that causes students to overhaul and rethink the basic ways in which they make moral judgments" (p.
239).
The father-son relationship appears to be important both in
male identity development (Marcia, 1980) and in male moral
judgment development, as suggested by the present results and
by findings reported by Hart (1988). Hart found that in Kohlberg's longitudinal sample, identification with mothers and maternal affection-involvement were unrelated to sons' moral
judgment at any age. However, identification with fathers and
paternal affection and involvement, measured during adolescence, significantly predicted sons' moral judgment from childhood to adulthood.
Interpretation of the stronger paternal role in the moral judgment development of adult offspring is problematic from both
a cognitive-developmental and traditional socialization perspective. Kohlberg predicted weak parent-offspring moral
judgment correlations during adulthood, whereas most socialization paradigms emphasize the role of the mother as the primary socialization agent. Hart (1988) argued that the most parsimonious interpretation to explain his results was the developmental process of identification, which also may have
contributed to the strong father-son correlations in the present
study. (Although a process of same-sex identification cannot ex-

FAMILY PATTERNS OF MORAL JUDGMENT

plain the strong father-daughter correlations, this cohort of female offspring may have identified with parental competence
rather than sex.)
Recognizing that this interpretation is contrary to the expectations of cognitive-developmental theory, Hart (1988) and
Gibbs and Schnell (1985) argued that different theoretical accounts, along with an empirically based understanding of normative developmental processes, may meaningfully complement each other in understanding the process of moral development. Kohlberg's (1969) constructivist approach, in this view,
is not undermined by results that support conceptions of developmental processes proposed by other theoretical paradigms.
In fact, the integration of various theoretical approaches, along
with knowledge gleaned from developmental research, may result in more powerful explanations of underlying developmental
processes than interpretations based on a single theoretical perspective.
References
Berkowitz, M. W., Gibbs, J. C , & Broughton, J. M. (1980). The relation
of moral judgment stage disparity to developmental effects of peer
dialogues. Merrill-Palmer Quarterly, 26, 341-357.
Block, J. H. (1983). Differential premises arising from differential socialization of the sexes: Some conjectures. Child Development, 54,
1335-1354.
Buck, L. Z., Walsh, W. F., & Rothman, G. (1981). Relationship between
parental moral judgment and socialization. Youth and Society, 13,
91-116.
Colby, A., & Kohlberg, L. (1987). The measurement of moral judgment
(Vol. 1). New York: Cambridge University Press.
Colby, A., Kohlberg, L., Speicher, B., Hewer, A., Gibbs, J., Candee, D.,
Gibbs, J., & Power, C. (1987). The measurement of moral judgment
(Vol. 2). New York: Cambridge University Press.
Conger, J. J. (1977). Adolescence and youth. New York: Harper & Row.
Douvan, E., & Adelson, J. (1966). The adolescent experience. New York:
Wiley.
Dunton, K. J. (1989). Parental practices associated with their children's
moral reasoning development (Doctoral dissertation, Stanford University, 1988). Dissertation Abstracts International, 49, 33O6A. (University Microfilms No. 88-26,133)
Eccles, J. S., Midgley, C , Wigneld, A., Buchman, C. M., Reuman, D.,
Flanagan, C , & Mac Iver, D. (1993). Development during adolescence: The impact of stage-environment fit on young adolescents'
experiences in schools and in families. American Psychologist, 48,
90-101.
Fodor, E. M. (1973). Moral development and parent behavior antecedents in adolescent psychopaths. Journal of Genetic Psychology, 122,
37-43.
Gibbs, J., & Schnell, S. (1985). Moral development "versus" socialization: A critique. American Psychologist, 40, 1071-1080.
Haan, N., Langer, J., & Kohlberg, L. (1976). Family patterns of moral
reasoning. Child Development, 47, 1204-1206.
Hart, D. (1988). A longitudinal study of adolescents' socialization and
identification as predictors of adult moral judgment development.
Merrill-Palmer Quarterly, 34, 245-260.
Hoffman, M. L., & Saltzstein, H. D. (1967). Parent discipline and the
child's moral development. Journal of Personality and Social Psychology, 5, 45-57.
Hollingshead, A. B., & Redlich, F. C. (1958). Social class and mental
illness. New York: Wiley.
Holstein, C. B. (1969). Parental consensus and interaction in relation to

631

the child's moral judgment. Unpublished doctoral dissertation, University of California, Berkeley.
Holstein, C. B. (1972). The relation between children's moral judgment
level to that of their parents and to communication patterns in the
family. In R. C. Smart & M. S. Smart (Eds.), Readings in child development and relationships (pp. 484-494). New York: Macmillan.
Holstein, C. B. (1975, March). A longitudinal study of family influence
on adolescent moral judgment development. Paper presented at the
meeting of the Society for Research in Child Development, Denver,
CO.
Inhelder, B., & Piaget, J. (1958). The growth of logical thinking from
childhood to adolescence. New York: Basic Books.
Jones, H. E. (1939a). The adolescent growth study: I. Principles and
methods. Journal of Consulting Psychology, 3, 157-159.
Jones, H. E. (1939b). The adolescent growth study: II. Procedures. Journal of Consulting Psychology, 3, 177-180.
Jurkovic, G. J., & Prentice, N. M. (1974). Dimensions of moral interaction and moral judgment in delinquent and nondelinquent families.
Journal of Consulting and Clinical Psychology, 42, 256-262.
Kohlberg, L. (1958). The development of modes of moral thinking and
choice in the years 10 to 16. Unpublished doctoral dissertation, University of Chicago.
Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental
approach to socialization. In D. A. Goslin (Ed.), Handbook of socialization theory and research (pp. 347-480). Chicago: Rand McNally.
Kuhn, D., Kohlberg, L., Langer, J., & Haan, N. (1977). The development of formal operations in logical and moral judgment. Genetic
Psychology Monographs, 95, 97-188.
Marcia, J. E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handbook of adolescent psychology (pp. 159-187). New York: Wiley.
Parikh, B. (1975). Moral judgment and its relation to family environmental factors in Indian and American upper middle class families.
Unpublished doctoral dissertation, Boston University.
Parikh, B. (1980). Development of moral judgment and its relation to
family environmental factors in Indian and American families. Child
Development, 51, 1030-1039.
Perry, W. G., Jr. (1970). Intellectual and ethical development in the college years. New York: Holt, Rinehart & Winston.
Piaget, J. (1965). The moraljudgment of the child. New York: Free Press.
(Original work published 1932)
Powers, S. I. (1983). Family interaction and parental moral judgment as
a context for adolescent moral development: A study of patient and
non-patient adolescents (Doctoral dissertation, Harvard University,
1982). Dissertation Abstracts International, 43, 3753B. (University
Microfilms No. 83-08,501)
Powers, S. I. (1988). Moral judgment development within the family.
Journal of Moral Education, 17, 209-219.
Rest, J. R., & Narvaez, D. (1991). The college experience and moral
development. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook
of moral behavior and development: Vol. 2. Research (pp. 229-245).
Hillsdale, NJ: Erlbaum.
Rest, J. R., &Thoma, S. J. (1985). Relation of moral judgment development to formal education. Developmental Psychology, 21, 709-714.
Rest, J. R., Turiel, E., & Kohlberg, L. (1969). Relations between level
of moral judgment and preference and comprehension of the moral
judgment of others. Journal of Personality, 37, 225-252.
Ruble, D. N. (1988). Sex role development. In M. H. Bornstein & M. E.
Lamb (Eds.), Developmental psychology: An advanced textbook (pp.
411-460). Hillsdale, NJ: Erlbaum.
Shoffeitt, P. G. (1971). The moral development of children as a function
of parental moral judgments and child-rearing practices (Doctoral
dissertation, George Peabody College for Teachers, 1971). Dissertation Abstracts International, 32, 2427B-2428B. (University Microfilms No. 71-26,218)

632

BETSY SPEICHER

Speicher, B. (1992). Adolescent moral judgment and perceptions of


family interaction. Journal of Family Psychology, 6, 128-138.
Speicher-Dubin, B. (1982). Relationships between parent moral judgment, child moral judgment and family interaction: A correlational
study. Unpublished doctoral dissertation, Harvard University.
Turiel, E. (1966). An experimental test of the sequentiality of development stages in the child's moral judgments. Journal of Personality
and Social Psychology, 3, 611 -618.
Walker, L. J. (1982). The sequentiality of Kohlberg's stages of moral
development. Child Development, 53, 1330-1336.
Walker, L. J. (1986). Experiential and cognitive sources of moral development in adulthood. Human Development, 29, 113-124.

Walker, L. J., & Taylor, J. H. (1991). Family interactions and the development of moral reasoning. Child Development, 62, 264-283.
White, C. B. (1988). Age, education, and sex effects on adult moral reasoning. International Journal of Aging and Human Development, 27,
271-281.
Zern, D. S. (1984). Relationships among selected child-rearing variables
in a cross-cultural sample of 110 societies. Developmental Psychology,
20, 683-690.

Received February 26, 1992


Revision received February 17, 1994
Accepted February 17, 1994

Call for Nominations


The Publications and Communications Board has opened nominations for the editorships of
the Journal of Consulting and Clinical Psychology, the Journal ofEducational Psychology,
the Interpersonal Relations and Group Processes section of the Journal of Personality and
Social Psychology, Neuropsychology, and Psychological Bulletin for the years 1997-2002.
Larry E. Beutler, PhD; Joel R. Levin, PhD; Norman Miller, PhD; Nelson Butters, PhD; and
Robert J. Sternberg, PhD, respectively, are the incumbent editors. Candidates must be
members of APA and should be available to start receiving manuscripts in early 1996 to
prepare for issues published in 1997. Please note that the P&C Board encourages
participation by members of underrepresented groups in the publication process and would
particularly welcome such nominees. To nominate candidates, prepare a statement of one
page or less in support of each candidate.
For the Journal of Consulting and Clinical Psychology, submit nominations to
Hans H. Strupp, PhD, Department of Psychology, Wilson Hall, Vanderbilt University, Nashville, TN 37240, to FAX number (615) 343-8449, or to
STRUPPHH@CTRVAX.VANDERBILT.EDU.
For the Journal of Educational Psychology, submit nominations to Carl E.
Thoresen, PhD, School of Education, Stanford University, Stanford, C A 94305-3096,
to FAX number (414) 725-7412, or to CTHOR@LELAND.STANFORD.EDU.
For the Interpersonal Relations and Group Processes section of the Journal of
Personality and Social Psychology, submit nominations to Judith P. Worell, PhD,
Department of Educational and Counseling Psychology, 235 Dickey Hall, University of Kentucky, Lexington, KY 40506-0017, to FAX number (606) 257-5662, or
to CPDJUDYW@UKCC.UKY.EDU.
For Neuropsychology, submit nominations to Martha A. Storandt, PhD, Psychology Department, Box 1125, Washington University, 1 Brookings Drive, St. Louis,
MO 63130, or call (314) 935-6508.
For Psychological Bulletin, submit nominations to Richard M. Suinn, PhD,
Department of Psychology, Colorado State University, Fort Collins, CO 805230001,ortoRICHARD_SUINN.PSYCH@CNSMAIL.MSO.COLOSTATE.EDU.
First review of nominations will begin December 15, 1994.