See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/14764966

The moral development of medical students: A pilot study of the

possible influence of medical education

Article  in  Medical Education · January 1993

DOI: 10.1111/j.1365-2923.1993.tb00225.x · Source: PubMed

CITATIONS READS

122 851

4 authors, including:

Donnie J. Self Dawn E Schrader

Texas A&M University System Health Science Center Cornell University
78 PUBLICATIONS   1,580 CITATIONS    25 PUBLICATIONS   427 CITATIONS   

SEE PROFILE SEE PROFILE

Dewitt C Baldwin
Accreditation Council for Graduate Medical Education
137 PUBLICATIONS   5,722 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Resident work and sleep hoursurs View project

Professionalism and professional values in orthopedicd View project

All content following this page was uploaded by Dawn E Schrader on 02 May 2018.

The user has requested enhancement of the downloaded file.

Medical Education 1993, 27, 26-34
The moral development of medical students: a pilot study
of the possible influence of medical education
D. J. SELF, D. E. SCHRADERt, D. C. BALDWIN JRS & F. D. WOLINSKYS
Departments ofHumanities in Medicine, Philosophy, and Pediatrics, College of Medicine, Texas A & M
University, College Station, Texas, tDepartment of Education, Cornell University, Ithaca, N e w York,
$American Medical Association, Chicago, Illinois and §Department of Medicine, Indiana University School
of Medicine, Indianapolis, Indiana
Summary. Medicine endorses a code of ethics
and encourages a high moral character among
doctors. This study examines the influence of
medical education on the moral reasoning and
development of medical students. Kohlberg’s
Moral Judgment Interview was given to a sample
of 20 medical students (41.7% of students in that
class). The students were tested at the beginning
and at the end of their medical course to
determine whether their moral reasoning scores
had increased to the same extent as other people
who extend their formal education. It was found
that normally expected increases in moral
reasoning scores did not occur over the 4 years of
medical education for these students, suggesting
that their educational experience somehow
inhibited their moral reasoning ability rather
than facilitating it. With a range of moral
reasoning scores between 315 and 482, the
finding of a mean increase from first year to
fourth year of 18.5 points was not statistically
significant at the P 6 0.05 level. Statistical
analysis revealed no significant correlations at the
P < 0.05 level between the moral reasoning
scores and age, gender, Medical College
Admission Test scores, or grade point average
scores. Along with a brief description of Kohl-
berg’s cognitive moral development theory,
some interpretations and explanations are given
for the findings of the study.
Key words: *morals; *educ, med, undergrad;
Correspondence: Donnie J . Self PhD, Department
of Humanities in Medicine, Texas A&M University
Health Science Center, 164 Reynolds Medical Build-
ing, College Station, Texas 77843-1 114, USA.
26
ethics, med; attitude of health personnel;
students, med/psychol; attitude to health; think-
ing; pilot projects
Introduction
The Flexner Report (1910) led to major emphasis
of medical education on the acquisition of scien-
tific and technical knowledge, rather than on the
teaching of values and attitudes. Pellegrino,
however, has pointed out that medicine is, and
has always been, inherently a moral enterprise
(Pellegrino & Thomasma 1981). This is substan-
tiated by even a casual review ofthe historical and
contemporary codes of ethics of organized medi-
cine (Burns 1977;Reiser et al. 1977;Bulger 1989).
Most descriptions of the essential characteristics
or qualities of medical staff include the qualities
o f high moral character and following a code of
professional and ethical behaviour as important
considerations. Thus, the relationship of medical
education to the moral development of medical
students deserves careful consideration.
Instruments for gathering empirical data on
the development of moral reasoning in doctors
and medical students have not been available
until recently. As a result of the work of Kohl-
berg and his colleagues, however, several instru-
ments for assessing moral reasoning
development have been developed that are
appropriate for use with medical students (Kohl-
berg 1984; Rest 1979; Gibbs & Widaman 1982).
Relatively few moral reasoning studies have
been conducted with medical students. Most of
these have been cross-sectional, or ‘slice of time’,
in nature, representing one point in medical
 Moral development of medical students
education. The most extensive studies have been
done by Sheehan, Baldwin, Self and Dyer, some
of which have not been published (Sheehan et al.
1981; Baldwin et al. 1991; Self et al. 1992a8cb; A.
Dyer, personal communication). Benor et al.
(1984) have attempted to relate moral reasoning
to the medical school admissions process while
several other reports involve pre- and post-
testing medical students, and relating changes in
their moral development to the teaching of
medical ethics. The most extensive studies of this
nature have been undertaken by Self et al. (1989,
1992b)and by Galaz-Fontes etal. (1989). Thereis
also one report in the literature of a successful
intervention with undergraduate premedical
students in a medical ethics course (Goldman &
Arbuthnot 1979). To date, however, there have
been no published longitudinal studies on the
moral development of medical students in the
USA. As a result, very little is known about the
relationship of medical education to moral
reasoning and moral development.
Earlier studies have explored the changes in the
attitudes of students during medical education,
indicating an increase in cynicism and loss of
idealism (Eron 1955; Christie & Merton 1958).
More recent studies report an increase in detach-
ment and a decrease in empathy of medical
students during their professional education
(Zabarcnko & Zabarenko 1978). The relation-
ship between attitudes and moral reasoning,
however, is not well understood. There are
considerable data from the research of Kohlberg
on the general moral development of young
adults of comparable age to typical medical
students (Colby & Kohlberg 1987a). These data
suggest a general increase in moral reasoning
skills with maturity. Other research reports,
notably by Rest (Rest et al. 1978; Rest 1986).
corroborate a steady incrcasc in moral reasoning
as a function of increasing age and level of
education. This report addrcsses the relationship
of moral reasoning to medical education.
The hypothcsis of this study is that the mcdical
education cxpericncc inhibits the normally
expected increase in moral reasoning of medical
students. Stated in the null form, the study
hypothesizes that there will be no significant
increase in the moral reasoning scores of mcdical
students from their first to fourth years. This is
based, in part, on the view that the rigid,
27
hierarchical, authoritarian structure of medical
education does not promote tolerance €or
differing values, support the conceptual explor-
ation of the fundamental values in medicine, or
encourage the cognitive conflict that have been
found to be important elements in moral
reasoning growth and development. Rather,
medical education seems to promote an environ-
ment focused on convergent thinking, getting
the ‘right’ answer, and maintenance of the rules
and regulations of the system, which according
to cognitive moral development theory encou-
rages a ‘conventional level’ moral ethos.
Methods
The study involved an examination of the change
in moral reasoning of 20 medical students who
were tested at the begining of their first year and
again at the end of their fourth year of medical
education. The sample represented 41.7% of the
students for that class. The students were
recruited as non-paid volunteers after appro-
priate informed consent. The project was
approved by the Institutional Review Board.
Although this was not a random sample of
medical students, thus creating a potential bias in
the data, the original sample represented nearly
halfofthe students for that cohort and there were
no statistically significant differences between
them and the rest oftheir class-mates with regard
to age, gender, grade point average scores, or
Medical College Admission Test scores. Thus,
they appear to be comparable to the other
medical students. The implied comparison is
based on reports in the literature which demon-
strate a correlation between an increase in moral
reasoning and an increase in age and educational
attainment as previously noted.
The instrument used for both the pre-test and
post-test mcasuremcnt of moral reasoning was
the Moral Judgment Intcrvicw (MJI) developed
by Kohlberg (1984). It consists of a 45-minutc,
scmistructurcd, oral, tape-recorded intcrvicw in
which subjects are asked to rcsolvc a scries of
three hypothetical moral dilemmas. Each dil-
emma is followed by a systematic set of opcn-
cndcd probc questions designed to enable the
subject to reveal the logic of his or her moral
reasoning.
28 D.]. Selfet al.
The dilemmas include the well-known Heinz
dilemma of a man considering whether or not to
steal a drug to save the life of his wife when all
legal means have been exhausted. This forces one
to think about the issues of life versus law. This
dilemma is followed by the judge’s dilemma of
what punishment to give Heinz after Heinz is
caught and found guilty of stealing the drug.
This forces one to think about the issues of
conscience versus punishment and the relation-
ship of legality to morality. The third dilemma is
one of a 14-year-old boy considering whether or
not to defy his father’s authority when his father
breaks a contractual agreement that his father had
made with him earlier. This forces one to think
about the issues of contract versus authority.
Following each dilemma a series of structured
probe questions are used to change the circum-
stances of each dilemma a little to see if one’s
view of morality and the logic of one’s moral
reasoning changes with the circumstances.
A transcript of the interview is scored, yielding
two numerical values. One score, the global
stage score, represents a category describing the
stage structure of a person’s reasoning in Kohl-
berg’s cognitive moral development theory
(Kohlberg 1976). The other score, the weighted
average score, is a continuous score that ranges
from a possible low of 100 to a maximum high of
500 and is correlated to the stages in cognitive
moral development theory. All interviews were
scored by the same scorer, thus eliminating the
possible problems of inter-rater reliability
between scorers. The scorer for this study had
been trained by Kohlberg at Harvard University
and had a reliability of 100% within a whole stage
and 75% within a third ofa stage when compared
to the training protocols in the scoring manual.
Kohlberg’s theory, which provides the theoreti-
cal basis for this study, has been described in
summary and in detail elsewhere (Kohlberg
1969, 1981). For purposes of review, the follow-
ing brief summary is abstracted from a study
reported in Academic Medicine (Self et al. 1989).
Theoretical overview
Based on 30 years of replicated research, Kohl-
berg’s theory provides three levels of moral
development known as preconventional mor-
ality, conventional morality, and postconven-
tional or principled morality. Each level contains
two stages. In the preconventional level, stage 1
is an authority-punishment stage in which what
is considered right is whatever the authority
figures say to do, and the reason for doing it is to
avoid punishment. Stage 2 is an egoistic instru-
mental exchange in which what is considered
right is whatever meets one’s own needs, but
with a sense of fairness in terms of equal
exchange between parties in agreement. For
example, a ‘what’s in it for me’ attitude or ‘I’ll
scratch your back, if you’ll scratch mine’
approach to morality pervades this stage.
In the conventional level, stage 3 morality is
concerned with mutual interpersonal expecta-
tions, peer relationships, and interpersonal con-
formity in which what is considered right is what
is expected by people close and important to you.
Stage 4 involves a concern for societal mainten-
ance and a conscience orientation in which one
fulfills one’s agreed-upon duties and contributes
to the welfare of the whole group, institution or
society. Right is defined in terms of that which
maintains a smoothly running society and avoids
the breakdown of the system.
In the postconventional or principled level
moral reasoning, stage 5 emphasizes individual
rights such as life and liberty, but endorses a
social contract which protects all people’s rights
with a commitment that is freely entered upon to
serve the greatest good for the greatest number.
It is based upon a rational calculation of the best
welfare of all humankind. Lastly, stage 6 is based
on a commitment to universal ethical principles
ofjustice, equality, autonomy and respect for the
dignity of all human beings as individual per-
sons. Although laws and social agreements are
usually valid because they are based on these
principles, when they violate these principles,
one acts in accordance with the principles. What
is right is that which is required by a personal
commitment to these universal ethical principles
of justice, equality, autonomy, dignity and
respect of persons.
The validity of Kohlberg’s system has been
well established cross-culturally and under a
wide variety of socio-economic situations in
scores of studies in 26 cultures. The studies come
from both Asian and Western cultures, in both
the Northern and Southern hemispheres, and
include both longitudinal and cross-sectional
 Moral development of medical students
studies (Snarey 1985; Rest 1986). According to
the theory, people proceed through these stages
as they mature. The sequence is invariant,
although the rate and end-stage reached vary
with the individual. It is important to understand
that the theory is based upon the claim that only
the structure or justification the person uses
determines the stage score. Scores are not as-
signed based on the content or a particular set of
values or moral beliefs that the person holds.
What is being tested by the MJI is the person’s
structure of moral reasoning and not the person’s
particular set of moral beliefs or values nor the
person’s actual behaviour. For example, one
could be scored at stage 4 while holding either
conservative or liberal values. Indeed, whether
one holds conservative or liberal values is
immaterial to the structure of the reasons one uses
in supporting one’s values.
Kohlberg’s cognitive moral development
theory is not without its limitations and diffi-
culties. Some of the theoretical problems have
been reviewed by Reed (1987), Blum (1988) and
Gillon (1979). One problem not often noted is
the tendency in cognitive moral development
theory to equate moral reasoning and justice
reasoning. In the earlier description of the theory
above, it was noted that the higher stages in
Kohlberg’s theory rely on universal ethical prin-
ciples, especially the principle of justice. Ulti-
mately Kohlberg’s theory is a justice-based
theory with the principle of justice being the
highest form of morality, although Kohlberg
argues that the theory includes the components
of both justice and care. The correlative prin-
ciples of equality, autonomy and respect for the
dignity of all human beings are derived from the
concept of justice grounded in the moral phil-
osophies of Kant and Rawls (Kant 1956, 1959;
Rawls 1971). Kohlberg’s notion of the impar-
tiality of moral obligations to act toward all
human beings without special consideration to
friends and relatives is based on this concept of
justice. Thus what moral reasoning tests purport
to measure is a subject’s use of or appeal to
various forms of justice reasoning when
resolving moral dilemmas, with justice
reasoning being thought to be equivalent to
moral reasoning. This equivalency is a widely
held assumption throughout much of cognitive
moral development work.
29
Whether justice and moral reasoning are
equivalent or not, the approach of cognitive
moral development is that ethics is fundamen-
tally a principled approach to rational decision-
making about value issues. But not everyone
accepts this position. There are those who hold
that ethics is not and ought not to be a principle-
based, rule-governed, rational decision-making
endeavour. Gilligan and Noddings believe that
ethics is basically non-rational (as opposed to
irrational) and based on a commitment to caring
for others rather than on principles for rule-
governed, rational decision-making (Gilligan
1977, 1982; Gilligan & Attanucci 1988; Noddings
1984). Theirs is an ethic of care approach in
which morality is governed by a concern for and
responsiveness to others which espouses a con-
demnation of all exploitation, violence and
harm.
Results
The original sample consisted of 22 (45.8%) of
the 48 first-year students in the College of
Medicine. Complete moral reasoning data pre-
and post-medical education were collected on 20
of the 22 original subjects representing a 9.1%
sample loss. The analyses were restricted to those
20 subjects (41.7% of the students for that class)
for whom complete data were available. Demo-
graphic data were also obtained, including age,
gender, undergraduate grade point average
scores, and Medical College Admission Test
scores. As Table 1 indicates, the weighted aver-
age scores (WASs) ofthe medical students ranged
from 315 to 482 during the first year, and from
341 to 454 during the fourth year. The mean
increase from the first year to the fourth year of
the study of 185 WAS points was not statistically
significant at the P 6 0.05 level, thus supporting
the null hypothesis. From this, it can be inferred
that the medical education experience somehow
inhibited the normally expected growth in moral
reasoning development of these students.
O f additional interest is the finding that the
range of moral reasoning WASs narrowed
between the first and fourth years. During the
first year, the students’ WAS spread was 167,
which corresponds to approximately one and a
half moral development stages. During the
30 D . ] . Selfet al.
Table 1. Statistics on MJ1 moral reasoning data of medical students
1st year 4th year
Mean
Range Mean Range Mean change &value*
Weighted 315-482 378.2 341454 396.6 18.4 1.11
average score not significant
Global 3.0-5.0 3.9 3.5-4.5 4.0 0.1 1-40
stage score not significant
*t-value computed using pooled estimate of common variance

fourth year, the WAS spread was 113 points, or
approximately one stage. Taken together, these
represent a theoretical difference between years 1
and 4 of approximately three-quarters of a glo-
bal stage - a significant reduction in moral
reasoning variance.
Table 2 demonstrates the correlation matrix of
demographic characteristics with changes in
moral reasoning WASs and Global Stage Scores.
There were no significant correlations at the P <
0.05 level between the change in moral reasoning
WASs or global stage scores and either age,
gender, Medical College Admission Test scores,
or grade point average scores. Therefore, it is
unlikely that any sampling biases related to these
factors attended the study outcome.
As an aggregate, these data provide little
information to explain the failure of medical
students to grow in moral reasoning during the
period. Nor do they offer possible directions for
educational changes which would enhance
medical students’ moral reasoning development
during these years. Therefore, data on individual
subjects were examined to provide insight into
trends of development which might prove useful
in developing programmes for medical students
or for changing the structure of medical edu-
cation.
Table 3 describes the medical students’ indi-
vidual demographic characteristics associated
with their moral reasoning scores. These data
may provide an explanation for the lack of
change in some individuals, the growth in
others, and the regression in still others.
Overall, the subjects scored in Kohlberg’s
conventional level of moral reasoning, although
five subjects (5, 7, 10, 16 and 18) scored in the
postconventional level on the pretest. Of those
students who were postconventional, three
regressed to the conventional level at year 4 of
medical education, although another three pro-
gressed to a postconventional level. There were
no significant gender differences between either
the conventional subjects or the postconven-
tional subjects, in that some subjects from both
genders regressed and some subjects from both
gender did not and all who regressed did so to the
same extent, namely, half a stage. Similarly,
some subjects from both genders progressed to
the postconventional level.

Table 2. Correlation of demographic characteristics with changes in global moral

reasoning stage scores and weighted average scores
Change in weighted Change in global
average score stage score
Correlation Correlation
coefficient P-value coefficient P-value

Age -0.217 0.179’ -0.009 0.485

Gender* -0.125 0.300 0.055 0.408
MCAT -0.303 0.097 -0.298 0.101
GPA -0.094 0348 0.007 0.488
*Men coded as 1. women coded as 2.
 Moral development of medical students 31

Table 3. Demographic data and moral reasoning scores of medical students

Weighted Weighted Global Global Change
average average stage stage in
Identity Age score score Change score score global
number (4th year) Gender MCAT GPA (1st year) (4th year) in WAS (1st year) (4th year) stage
01 24 Male 53 2.15 326 375 +49 314 314 0
02 35 Male 62 3.35 315 360 +45 3 314 +0*5
03 28 Male 57 3.20 367 444 +77 314 415 +1.0
04 25 Female 52 2.79 348 401 +53 314 4 +05
05 26 Female 55 3.32 438 447 +9 415 415 0
06 34 Female 44 2.85 360 387 +27 314 4 +0.5
07 26 Male 53 2.61 443 454 +11 415 415 0
08 25 Female 63 3.25 364 427 +63 314 415 +1.0
09 25 Female 47 2.93 350 369 +19 314 314 0
10 31 Female 58 3.55 482 430 -52 5 415 -0.5
11 25 Female 56 3.75 353 37 1 +18 314 4 +0.5
12 24 Female 52 3.72 38 1 403 +22 4 4 0
13 26 Male 56 3.45 342 341 -1 314 314 0
14 24 Male 73 3.17 400 417 +17 4 4 0
15 24 Female 60 3.89 379 403 +24 4 4 0
16 34 Female 66 2.76 405 361 -44 415 4 -0.5
17 28 Female 51 2.75 330 348 +18 314 4 +0.5
18 26 Male 73 3.24 435 404 -31 415 4 -0.5
19 32 Female 55 3.13 404 429 + 25 4 415 +0.5
20 34 Female 51 2.82 341 361 +20 314 314 0

Apart from the regression of the postconven-
tional subjects, the general trend in the data was
that there was no significant change - either in
growth or regression - in the student’s moral
reasoning during their 4 years of medical edu-
cation. Forty-five per cent (n = 9) of the subjects
showed no change in their global stage score. In
fact, although there were no statistically signifi-
cant changes in WASs as a group, 35% ( n = 7) of
the individuals demonstrated substantive change
in their WASs (generally defined in the literature
as greater than 33 WAS points). The mean
change in WASs was 18.5 points, indicating that
there was a positive trend and suggesting some
changes were occurring in medical students’
moral reasoning -yet these changes may not be
toward positive development. Specifically, only
10% (n = 2) of the subjects decreased substan-
tially in WAS points (33 points or more), while
15% (n = 3) decreased by halfa global stage score
between year 1 and year 4.
In examining gender differences in WASs, two
men and two women regressed. The regression
was more marked in women (x = -48 points)
than in males (x = -16 points). In the three cases
where the regression was greater than 33 points,
the scores were among the highest in the sample.
In examining regression in global scores, two
women and one man regressed. The regression
in each case was half a stage. As with the WASs,
those who regressed scored among the highest in
the sample.
Discussion
Although this was not a random sample of
medical students, the participants were 41.7% of
the class, and there was no reason to think that
they were not comparable to the other medical
students since there were no statistically signifi-
cant differences at the P S 0.05 level between the
two groups with regard to age, gender, grade
point average scores, or Medical College
Admission Test scores. However, although this
small sample size makes it difficult to make broad
generalizations, the results obtained here suggest
32 0.1.Selfet al.
trends to be expected in a larger study. Clearly,
additional research needs to be done to investi-
gate, characterize and understand more fully the
relationship between medical education and
moral development.
Although it is possible that the student selec-
tion process at the school where the study was
conducted inadvertently selects students less sus-
ceptible to moral growth, this seems intuitively
unlikely and is empirically unsubstantiated.
Similarly, while the curricula and teaching staff
of other medical colleges might exert a more
positive influence on the moral development of
their students, there is no reason to believe this is
the case while there are plausible reasons to
believe this is not the case. The curricula of all
medical colleges are highly regulated by accredit-
ing bodies such that generally all medical
students take similar subjects in about the same
sequence, and frequently use the same textbooks
or related teaching materials. So while this pilot
study represents only a small sample from one
school, the results suggest that concern is war-
ranted, and that more research is needed to
explore the relationship of moral development
and medical education.
The results here indicate that the moral
reasoning of medical students does not signifi-
cantly increase over the 4 years of medical school
and that what change there is in moral reasoning
during the medical education years is not signifi-
cantly related to grade point average scores or
Medical College Admission Test scores. The
trends in the data demonstrate that the range of
moral judgement scores of the students decrease
after 4 years of medical education, indicating a
strong socializing factor of the medical experi-
ence. The fact that the majority of the postcon-
ventional subjects regressed to conventional
reasoning and the fact that those subjects scoring
in the lower conventional reasoning range
moved to the higher conventional reasoning
range indicate a movement towards homo-
geneity in moral thinking that may be due to the
medical education experience. The fact that other
subjects in the USA with similar age and edu-
cational level do not experience such a narrow-
ing-range effect, implicates qualities of the
medical educational system. Of course, with
only two points of measurement at the beginning
and end of medical education there is no way of
knowing what happened to the moral reasoning
of these students in between the measurements.
For example, it could be that there is a steady
increase in moral reasoning during the first 2
years followed by a steady decrease in moral
reasoning during the last 2 years of the medical
course or vice versa. These questions can only be
answered with a longitudinal study measuring
moral reasoning at each of the 4 years of the
medical course. Such a study is currently under
way.
The findings regarding gender differences fail
to support the argument by Gilligan (1982) that
Kohlberg’s moral reasoning theory is gender
biased. Both men and women scored similarly
on the pre- and post-data, and both a man and a
woman were scored a t the postconventional level
of moral reasoning in this sample (stage 45-5.0).
What is of interest is not that there are no gender
differences in the WAS or global stage scores, or
in the number of subjects of each gender who
change stage, but that changes, when they occur,
are somewhat more likely to be more substantial
for women than for men. The data do not
provide adequate information to explain this
finding readily, except that those who regressed
scored among the highest in the sample. This
raises some interesting questions about how
medical education or the moral atmosphere of
medical education may be differentially per-
ceived by men and women who are at postcon-
ventional levels of moral reasoning.
Despite the recent criticisms by Gilligan and
Noddings, and in light of Kohlberg’s acknow-
legement that a care ethic has indeed widened the
moral domain, the results of this pilot study
maintain their importance. Additional studies
need to examine whether or not justice is inhibi-
ted by the development of care in medical
students.
Even if morality is understood as a rule-gover-
ned, rational decision-making endeavour, Kohl-
berg’s theory about moral reasoning is still
limited in that it is only one component of the
complex issue of morality. Rest has pointed out
at least four components ofmorality (Rest 1986).
These include recognition or interpretation of an
issue as a moral issue; judgement about what is
right; priority of moral values over other perso-
nal values; and perseverance or ego strength to
implement one’s moral intentions. Kohlberg’s
 Moral development
theory only deals with the second component of
moral judgement. So even if Kohlberg’s theory
were an accurate and complete understanding of
moral reasoning, there still are other important
aspects of morality that have to be taken into
consideration and need to be studied.
However, Gillon (1979) challenges Kohlberg’s
theory not on its psychological foundations of
adequacy but on its philosophical foundations.
Basically he points out the need for a coherent
defence of why higher stages are better than
lower stages, i.e. why is Kantian (stage 6)
morality with reliance on the principles ofjustice
and autonomy superior to or better than the
egocentric hedonism ofstage 1 or the enlightened
self-interest of stage 2? This has been the centre
of debate in moral philosophy since before the
time of Socrates and has not been settled by
Kohlberg or his followers.
Nevertheless, since the medical profession
appears to accept the importance of moral
character and the pursuit of a high moral code of
ethics for its members, then closer attention
needs to be given to the structure of medical
education and the influence it has on the moral
growth and development of the students enter-
ing the profession. Well-defined, clearly estab-
lished techniques are available for improving
students’ moral reasoning. Early studies by Blatt
& Kohlberg (1975) and subsequently replicated
by many others (Blasi 1980) have shown that
moral reasoning stage change can be stimulated
by structured moral dilemma discussions which
create cognitive conflict by pitting arguments at
one stage of reasoning against arguments at a
different stage of reasoning. Recent studies have
reported significant increases in the moral
reasoning of students in medicine by the formal
teaching of medical ethics in the curriculum (Self
et al. 1989; Self et al. 1992a,b). These studies
indicated that small-group, problem-oriented,
case study moral dilemma discussions were the
most effective in this regard. The work of Rest
(1988) has shown that contrary to popular belief
it is not developmentally too late for moral
reasoning growth to occur in young adults.
Thus, it appears possible that the moral
reasoning and moral development of medical
students could be enhanced by improvements in
the structure of medical education.
of medical students 33
Conclusion
This study describes a method appropriate for
assessing the influence of medical education on
the moral reasoning and development of medical
students. It suggests that the moral growth and
development of medical students appears to be
inhibited during their education. These results
need to be replicated by others in different
settings, both within and outside the medical
profession.
Since the sample size is small and non-random,
it is difficult to draw definite conclusions or
broad generalizations from this study. Rather
these results should be considered a pilot project
suggesting trends to be further explored in larger
studies. Additional research needs to be done to
investigate, characterize and understand more
fully the relationship between medical education
and moral development. In particular, longitud-
inal studies need to be implemented to determine
if moral development resumes subsequent to
medical education.
If these findings are found to be an accurate
reflection of the state of medical education, then
serious attention needs to be given to how to alter
the structure of medical education so as to have it
promote the moral development espoused as
important to the medical profession.
Acknowledgment
This work was supported in part by a grant from
the American Medical Association’s Division of
Medical Education Research and Information.
References
Baldwin Jr D.C., Daugherty S.F. & Self D.J. (1991)
Changes in moral reasoning during medical school.
Academic Medicine (September Suppl.) 66, 1-3.
Benor D.E., Notzer N., Sheehan T.J. & Norman G.R.
(1984) Moral reasoning as a criterion for admission
to medical school. Medical Education 18, 423-8.
Blasi A. (1980) Bridging moral cognition and moral
action: a critical review of the literature. Psychologi-
cal Bulletin 88, 1-45.
Blatt M. & Kohlberg L. (1975) Theeffects ofclassroom
moral discussion upon ‘children’s level of moral
judgment. Journal ofMoral Education 4, 129-61.
Blum L. (1988)Gilligan and Kohlberg: implications for
moral theory. Ethics 98, 472-91.
Bulger R.J. (ed.) (1989) In Search ofthe Modern Hippo-
crates. University of Iowa Press, Iowa City, Iowa.
 34 D.]. Selfet al.
Burns C.R. (ed.) (1977) Legacies in Ethics and Medicine.
Science History Publications, New York.
Christie R. & Merton, R. (1958) Procedures for the
sociological study of the values climate of medical
schools. Journal ofMedical Education 33, 125-53.
Colby A. & Kohlberg L. (1987a) The Measurement of
MoralJudgement: Vol. I , Theoretical Foundations and
Research Validation. Cambridge University Press,
New York.
Colby A. & Kohlberg K. (1987b) The Measurement of
Moral Judgment: vol. 11, Standard Issue Scoring
Manual. Cambridge University Press, New York.
Eron L. (1955) The effects of medical education on
medical students. Journal of Medical Education 10,
55-6.
Flexner A. (1910) Medical Education in the United States
and Canada: A Report to the Carnegie Foundationfor
the Advancement of Teaching. Bulletin No. 4,
Updyke, Boston, Massachusetts.
Galaz-Fontes J.F., Pacheco-Sanchez M.E., Sierra-
Morales I., Commons M.L., Gutheil T.G. &
Hauser M.J. (1989) Medical school training and
moral reasoning in Mexico. Paper presented at the
Fourth Adult Development Symposium meeting,
Cambridge, MA.
Gibbs J.C. & Widaman K.F. (1982) Social Intelligence:
Measuring the Development of Sociomoral Rejection.
Prentice-Hall, Englewood Cliffs, New Jersey.
Gilligan C. (1977) In a different voice: women’s
conceptions of self and of morality. Harvard Edu-
cational Review 47, 481-517.
Gilligan C. & Attanucci J. (1988) T w o moral ori-
entations: gender differences and similarities. Mer-
rill-Palmer Quarterly 33, 223-37.
Gilligan C. (1982) In a Different Voice: Psychological
Theory and Women’s Development. Harvard Uni-
versity Press, Cambridge, Massachusetts.
Gillon R. (1979) Commentary.Journal ofMedical Ethics
5, 18&1.
Goldman S.A. & Arbuthnot J. (1979) Teaching
medical ethics: the cognitive-developmental
approach. Journal ofMedical Ethics 5, 170-80.
Kant I. (1956) Critique of Practical Reason. Trans. L.
Beck. Bobbs-Merrill, Indianapolis. (Originally
published 1788.)
Kant 1. (1959) Foundations ofthe Metaphysics ofMorals.
Trans L. Beck. Bobbs-Merrill, Indianapolis.
(Originally published 1785.)
Kohlberg L. (1969) Stage and sequence: the cognitive-
developmental approach to socialization. In: Hand-
book of Socialization Theory and Research (ed. by D.
Goslin), pp. 347-480. Rand McNally, Chicago,
Illinois.
Kohlberg L. (1976) Moral stages and moralization: the
cognitive developmental approach. In: Moral
Development and Behavior: Theory, Research and
Social Issues (ed. by T. Lickona), pp. 31-53. Holt,
Rinehart and Winston, New York.
Kohlberg L. (1981)Essays onMoral Development: Vol. I ,
View publication stats
The Philosophy of Moral Development. Harper and
Row, Publishers, San Francisco, California.
Kohlberg L. (1984) Essays on Moral Development: Vol.
11, The Psychology ojMoral Development. Harper and
Row, New York.
Noddings N. (1984) Caring: A Feminine Approach to
Ethics and Moral Education. University of California
Press, Los Angeles, California.
Pellegrino E.D. & Thomasma D.C. (1981) A Philo-
sophical Basis ofMedical Practice. Oxford University
Press, New York.
Rawls J. (1971) A Theory ofJustice. Harvard University
Press, Cambridge, Massachusetts.
Reed T.M. (1987) Developmental moral theory. Ethics
97,441-56.
Reiser S.J., Dyck A.J. & Curran W.J. (eds) (1977)Ethics
in Medicine: Historical Perspectives and Contemporary
Concerns. The Massachusetts Institute of Tech-
nology, Cambridge, Massachusetts.
Rest J.R., Davidson M. & Robbins S. (1978) Age
trends in judging moral issues: a review of cross-
sectional, longitudinal, and sequential studies of the
defining issues test. Child Development 49, 263-79.
Rest J.R. (1979) Development in Judging Moral Issues.
University of Minnesota Press, Minneapolis, Min-
nesota.
Rest J.R. (1986) Moral Development: Advances in
Research and Theory. Praeger, New York.
Rest J.R. (1988) Can ethics be taught in professional
schools? The psychological research. Ethics: Easier
Said Than Done 1 , 2 2 4 .
Self D.J. Baldwin Jr D.C. & Hurrey M.O. (1992a)
Description and evaluation of the use of film
discussion for teaching medical humanities. Acade-
mic Medicine (in press).
Self D.J., Baldwin Jr D.C. & Wolinsky F.D. (1992b)
Evaluation of teaching medical ethics by an assess-
ment of moral reasoning. Medical Education 26,
178-84.
Self D.J., Wolinsky F.D. & Baldwin Jr D.C. (1989)
The effect of teaching medical ethics on medical
students’ moral reasoning. Academic Medicine 64,
7559.
Sheehan T.J., Husted S.D. & Candee D. (1981) The
development of moral judgment over three years in
a group of medical students. Paper presented at the
American Educational Research Association meet-
ing, Los Angeles, California.
Snarey J.R. (1985) Cross-cultural university of social-
moral development: a critical review of Kohlber-
gian research. Psychological Bulletin 97, 202-32.
Zabrenko R.N.& Zabarenko L.M. (1978) The Doctor
Tree: Developmental Stages in the Growth .f Physi-
cians. University of Pittsburgh Press, Pittsburgh,
Pennsylvania.
Received 14 April 1992; editorial comments to authors
2J u l y 1992; acceptedforpublication 29 J u l y 1992