CHAPTER 2

Moral Development[1]

American postsecondary education traditionally has had a clearly articulated role in the development of citizens who would both think and act morally.  The religiously affiliated liberal arts colleges of the early 1800s attempted to integrate curricular and noncurricular experiences in such a way that students graduated into the larger world both wiser and more sensitive to their moral and ethical responsibilities.  The rise of public research universities and the attendant fragmentation of knowledge that accompanied the development of academic disciplines undoubtedly functioned to shift the focus of the curriculum and the faculty away from the development of student character.  Yet, despite this evolutionary change in the structure of institutions, the tradition of liberal education and its concern with developing the whole person persist as an important theme in American postsecondary education.  It is widely expected that the college experience should contribute not only to cognitive growth, but also to the capacity to apply reason and intellect to interpersonal, social, political, and ethical questions (Ignelzi, 1990; Ikenberry, 1997; King, 1997; Mathiasen, 1998; Pascarella, 1997).  In this chapter, we attempt to synthesize the accumulated evidence on the impact of postsecondary education on both principled moral reasoning and moral behavior.

Change During College

Conclusions from How College Affects Students

The weight of evidence from a large number of studies that used different instruments and were conducted in different cultures clearly indicates that college is linked with statistically significant increases in the use of principled reasoning to judge moral issues.  Not only do upperclassmen tend to show higher levels of principled reasoning than freshmen or sophomores, but students also make statistically insignificant increases in the use of moral reasoning during college.  The relative magnitude of these gains is difficult to determine from the evidence presented.  However, the exact magnitude of the gain may not be as important as the fact that the major movement from conventional to postconventional or principled judgment during college is itself an important event in moral development.

Evidence from the 1990s

As with our previous synthesis, nearly all the evidence pertaining to change during college focuses on students' principled moral reasoning or judgment, as conceptualized by Lawrence Kohlberg (1981; 1984).  Kohlberg's theoretical framework is certainly not the only valid model of moral or ethical development, but it appears to be, by far, the dominant theoretical framework guiding inquiry into the impact of postsecondary education on such development.  His theory can be briefly summarized in the following way:

(D)evelopment is described as proceeding through six stages embedded within three levels.  The first two stages (Level I) are considered preconventional.  At this level, moral reasoning is highly egocentric in that it is based on the person's concerns for his or her own interests and for those of specific others the individual might care about.  At Level II (Stages 3 and 4), conventional moral reasoning takes over.  This reasoning is based on a concern with maintaining the social order.  Moral judgments are guided by obedience to rules and meeting the expectations of others, particularly those in positions of authority.  The orientation toward maintaining the system is replaced at Level III (Stages 5 and 6) by a post-conventional or principled perspective.  The basis of this kind of reasoning is a view of morality as a set of universal principles for making choices among alternative courses of action that would be held by any rational moral individual.  These are considered first principles in that they exist independently of and prior to societal codification.  Hence, a central emphasis in the post-conventional stage is on principles for choosing the most just arrangement for individuals within society (Pascarella, 1997, pp. 48-49).

Kohlberg's (1981; 1984) theoretical work has led to the development of a number of instruments designed to estimate an individual's level of principled moral reasoning or judgment. (Although the terms "principled moral reasoning" and "principled moral judgment" appear to be used interchangeably by scholars in the field, for purposes of consistency we will use "principled moral reasoning.")[2]  The two most visible instruments in the body of research are probably the Moral Judgment Interview (MJI) (Colby, Kohlberg, Gibbs, & Lieberman, 1983) and the Defining Issues Test (DIT) (Rest, 1979, 1986).  Each instrument presents a series of moral dilemmas and attempts to determine the extent to which an individual uses principled reasoning or prefers principled considerations in making a judgment or decision about each dilemma.  Clearly, there is a cognitive component to moral development as measured by both the MJI and the DIT.  For example, both Hill (1995) and McNeel and Granstrom (1995) present evidence to suggest that moral reasoning measured by either the MJI or the DIT has significant, positive correlations with standardized measures of critical thinking.[3]

In our previous synthesis, we concluded that the extent or level of principled moral reasoning was positively associated with level of formal postsecondary education, and that students generally made statistically significant gains in principled moral reasoning during college.  Unfortunately, because many studies failed to report complete statistical data, an estimate of the actual magnitude of the growth in principled reasoning made during college could not be made.  However, McNeel (1994a) conducted a valuable meta-analysis of 13 cross-sectional and 9 longitudinal studies of student moral development as measured by the Defining Issues Test.  The individual samples in his synthesis were from 12 different colleges and universities (7 liberal arts colleges, 3 universities, and 2 Bible colleges).  We took McNeel's raw data and conducted a reanalysis of his results, weighting each individual study by the size of its sample.  The results of our reanalysis differed in only trivial ways from those reported by McNeel.  For both cross-sectional studies (i.e., comparing different samples of freshmen and seniors at the same time) and longitudinal studies (i.e., comparing a sample of freshmen with themselves as seniors four years later), the average advantage of seniors over freshmen in principled moral reasoning was about .77 of a standard deviation (28 percentile points).  Moreover, the major change that took place during college was that students shifted from using moral reasoning that concedes to societal authority (conventional moral reasoning) to reasoning that is based on the application of universal moral principles (post-conventional moral reasoning).  We found little in the results of studies not included in McNeel's meta-analysis that would conflict with his findings of a strong association between principled moral reasoning and extent of exposure to postsecondary education (e.g., Cummings, Dyas, & Maddux, 2001; Duckett et al., 1992; Duckett & Ryden, 1994; Foster & La Force, 1999; Galotti et al., 1991; Gielen & Markoulis, 1994; Good & Cartwright, 1998; Hill, 1995; Jeffrey, 1993; Lind, 1997; Paradice & Dejoie, 1991).

Net Effects of College

Conclusions from How College Affects Students

The weight of evidence indicates that the college experience itself has a unique positive influence on increases in principled moral reasoning.  This influence appears to be substantially greater in magnitude than that due merely to maturation and cannot be attributed solely to initial differences in moral reasoning, intelligence, or social status between those who attend and those who do not attend college.  Of course, even with these important variables controlled statistically, one cannot be certain that all confounding influences have been taken into account.  Individual differences among students related to their likelihood of attending or not attending college may be the true causal influences underlying differential change in moral reasoning rather than the college experience itself.  This particular confounding influence is a potential threat to the internal validity of even the most carefully conducted longitudinal studies of the effect of college on moral reasoning.

It is also important to note that the evidence suggesting a net positive influence of college on the development of principled moral reasoning is consistent with more broadly based findings on the effects of college on values.  These findings, from an extensive series of national samples, suggest that college attendance is associated with a humanizing of values and attitudes concerning the rights and welfare of others.

Evidence from the 1990s

Principled Moral Reasoning

As we have seen for previous outcomes, the fact that students demonstrate substantial growth in principled moral reasoning during college does not necessarily mean that exposure to postsecondary education causes that growth.  A number of rival hypotheses (e.g., cohort differences between freshmen and seniors in cross-sectional studies; maturation or growing older in longitudinal studies) could account for the observed difference between freshmen and seniors.  However, consistent with the conclusions from our previous synthesis, there is evidence from the 1990s to suggest that exposure to postsecondary education has a statistically significant, net positive influence on principled moral reasoning.

For example, Wilson, Rest, Boldizar, and Deemer (1992) followed a sample of 102 high school graduates for nearly a decade to determine the effects of educational attainment on principled moral reasoning, as measured by the Defining Issues Test (DIT).  The DIT was administered at the beginning of the study and approximately 10 years later when the study participants were between 26 and 27.  With statistical controls for initial DIT score and a measure of dedication to career or education (e.g., motivation to do well in career or amount of time devoted to educational pursuits), level of postsecondary educational attainment had a strong, positive total effect on DIT scores 10 years later of .34.  (This can be interpreted as a one standard deviation increase in educational attainment leading to about a third of a standard deviation increase in principled moral reasoning, net of high school DIT score and level of dedication to career or education.)  The addition of controls for occupational status and the extent to which the individual was in a career that was personally meaningful and challenging reduced the net of effect of educational attainment on young adult moral reasoning, but only to about .28.  Thus, even in the presence of controls for precollege moral reasoning, dedication to career/education, occupational status, and sense of fulfilling work, a one standard deviation increase in postsecondary educational attainment led directly to between a quarter and a third of a standard deviation increase in moral reasoning at age 26-27.  There was, however, also a small indirect, positive effect of postsecondary educational attainment on young adult moral reasoning, transmitted through the positive effect of educational attainment on occupational status and fulfilling work.

Results quite consistent with those of Wilson, Rest, Boldizar, and Deemer (1992) have also been reported in cross-sectional studies by Rykiel (1995), with a single community college sample, and by Finger, Borduin, and Baumstark (1992), with a single four-year college sample.  With statistical controls for a standardized measure of verbal aptitude, Rykiel found that graduating second-year students had a statistically significant advantage over incoming first-year students of about .35 of a standard deviation (14 percentile points) in DIT principled moral reasoning scores.  (Since Rykiel's results are reported in terms of differences between statistically adjusted means and those of Wilson et al. [1992] are reported in terms of standardized regression coefficients, the results of the two studies are not directly comparable.) Finger, Borduin, and Baumstark found that of all of the predictors they considered (year in college, involvement in social activities, age, socioeconomic status, parental control, and parental warmth), year in college was by far the strongest predictor of principled moral reasoning--accounting for 13% of the variance in DIT scores.  Moreover, consistent with the conclusions of our previous synthesis, they also reported that age was not as strong a predictor of moral reasoning as level of postsecondary education completed (See also Cummings et al., 2001; Paradice & Dejoie, 1991).

Inconsistencies in the way in which the evidence is reported in different investigations makes it difficult to estimate the magnitude of the net impact of college on growth in principled moral reasoning.  However, the weight of evidence suggests that at least a substantial part of the growth in principled moral reasoning that occurs during college is uniquely attributable to the college experience itself.  The significant positive association between extent of exposure to postsecondary education and level of principled moral reasoning persisted even in the presence of controls for potentially important confounding influences such as level of precollege moral reasoning, verbal ability, maturation, family socioeconomic status, and occupational level.  The reasons why postsecondary education may facilitate growth in principled moral reasoning are not completely clear.  However, Rest (1994) and Rest and Narvaez (1991) suggest that part of the explanation may be that college provides a relatively challenging and stimulating environment that leads students to overhaul and rethink the fundamental ways in which they form moral judgments.  College may do this in large measure because it encourages students to think about the larger social contexts of history, institutions, and broad intellectual and cultural trends--many of which involve moral and ethical issues. Consistent with such an explanation is evidence reported by Mason and Gibbs (1993) that academic perspective taking (i.e., exposure to broad perspectives concerning intellectual or social issues) is a strong predictor of advanced levels of moral reasoning among college students.

Despite rather consistent findings in support of a net positive impact of postsecondary education on growth in principled moral reasoning, it is possible that the evidence we reviewed is still confounded by other threats to internal validity, such as the interaction of selection and change.  For example, compared to their counterparts with less exposure to college, individuals with relatively high levels of postsecondary educational attainment might simply be more open to the influence of the college experience to begin with.  This difference in receptiveness to the impact of education on moral reasoning, more so than the college experience itself, might underlie the positive link between postsecondary educational attainment and level of principled moral reasoning.  Thus, as we cautioned in our previous synthesis, differences among individuals related to their likelihood of attending college persist as a threat to the internal validity of studies estimating the net impact of postsecondary education on principled moral reasoning.

Moral Reasoning and Moral Action

The primary instruments used to measure growth during college in principled moral reasoning, as well as the net impact of college on that growth (i.e., the Defining Issues Test and the Moral Judgment Interview) estimate the extent to which a person uses principled reasoning in making a judgment or decision about a moral dilemma.  These instruments are not, in and of themselves, assessments of an individual's actual moral action or behavior.  Nevertheless, although some inconsistencies exist between reasoning and behavior (e.g., Crow, Fok, Hartman, & Payne, 1991), we found clear evidence in our previous synthesis of a statistically significant and positive association between principled moral reasoning (as measured by the DIT and MJI) and principled moral behavior such as resistance to cheating, peer pressure, and unlawful or oppressive authority; "whistle blowing" on corruption; the keeping of contractual promises; political and social activism; nonaggression; and helping behavior (see Pascarella & Terenzini, 1991, pp. 363-364, for a review of these studies).  In research published during the 1990s, level of principled moral reasoning has been found to be positively linked to community involvement and a sense of civic responsibility (Rest, Thoma, & Edwards, 1997); ethical behavior in professions such as accounting (Arnold & Ponemon, 1991; Bernardi, 1991; Ponemon, 1992; Ponemon & Gabhart, 1990, 1993), dentistry (Bebeau, 1994), and medicine (Baldwin, Adamson, Sheehan, Self, & Oppenberg, 1996; Self & Baldwin, 1994); and the independent clinical performance of undergraduate nursing students (Duckett & Ryden, 1994).  There is also evidence suggesting level of principled reasoning is negatively associated with student self-reported academic misconduct (Cummings et al., 2001), and that college students involved in disciplinary matters exhibit lower levels of principled moral reasoning on the DIT than typical college students in normative research samples (Mullane, 1999).

Thus, it would appear that principled moral reasoning does not simply measure an individual's ability to make sophisticated responses to hypothetical moral dilemmas; it is also systematically associated with, and perhaps even a precursor to, principled moral action.  Growth in principled moral reasoning during college, therefore, should, at least indirectly, enhance the possibility of principled moral action.  As Rest and Narvaez (1991) caution, however, while principled moral reasoning may positively influence moral behavior, principled reasoning alone is not sufficient to determine moral behavior.  Rest (1994) further suggests that using principled reasoning to judge moral issues is only one of four components leading to moral action.  The other three components are moral sensitivity (being aware that a situation has a moral dimension to it and how one's action could affect others involved), moral motivation (prioritizing moral considerations relative to other situations), and moral character (the capacity to implement and persist in one's moral course of action).  Unfortunately, compared to principled moral reasoning, we uncovered little or no systematic evidence concerning the influence of postsecondary education on these other three enabling precursors of moral behavior.

Between-College Effects

Conclusions from How College Affects Students

Evidence concerning between-college effects on moral reasoning is nearly nonexistent.  Our own reanalysis of available data suggests that such differences may indeed exist.  However, differences in student recruitment and selection standards among institutions make any causal conclusion drawn from such analysis highly tentative.  Findings from one study are consistent with the notion that the student peer context may function to accentuate initial differences among colleges in terms of gains in principled moral reasoning.  However, this finding awaits replication.

Evidence from the 1990s

Principled Moral Reasoning

Credible evidence concerning the impact of different kinds of colleges on the development of principled moral reasoning in students was virtually nonexistent until the early 1990s, when McNeel (1994a) constructed an informative 12-institution data set that provides some tantalizing hints about between-college effects.  This data set is essentially the same as that we reanalyzed in the previous section on change during college.  However, McNeel also took the results from the 21 individual samples in his original meta-analysis and aggregated them according to institutional type (liberal arts colleges, Bible colleges, or universities) and study design (longitudinal or cross-sectional).  Once again we weighted the studies according to sample size and reanalyzed McNeel's raw data.  The results with the weighted samples differ only slightly from those originally reported by McNeel and suggest interesting differences between institutional types.  The largest freshman-to-senior gains or differences in principled moral reasoning were made at the private liberal arts colleges (average weighted senior advantage over freshmen across longitudinal and cross-sectional studies = .87 of a standard deviation, 31 percentile points).  Somewhat smaller, but still substantial gains were made at large, public universities (average weighted senior advantage over freshmen across studies = .62 of a standard deviation, 23 percentile points); and the smallest gains were made at the Bible colleges (average weighted senior advantage over freshmen across studies = .13 of a standard deviation, 6 percentile points).  Strikingly similar results are reported by Good and Cartwright (1998) with a smaller sample of three institutions--one of each type.

Such findings are, of course, limited by very small sample institutions and by potential rival explanations for the differences found between institutional types.  For example, liberal arts and Bible colleges might simply enroll populations of students characterized by major differences in their receptivity to those collegiate experiences that facilitate a shift from a conventional to postconventional perspective about moral issues.  At the same time, however, it is important to note that in the longitudinal investigations, the liberal arts colleges in McNeel's (1994a) sample started out with the highest entering freshman scores in principled moral reasoning and made the largest freshman-to-senior gains.  Conversely, students at the Bible colleges started out with the lowest entering principled reasoning scores and exhibited the smallest freshman-to-senior gains.  These two trends are just the opposite of what one would expect from regression artifacts (i.e., the artificial tendency for samples that have lower initial scores on a test to show greater gains on that test than samples that start out with initially higher scores on the same test).  As such, it argues for the possibility that the between-college differences found in principled moral reasoning gains represent actual institutional effects.  Traditional liberal arts education, combined with the unique social-psychological environments of small liberal arts colleges, may, in fact, be particularly conducive to the fostering of growth in principled moral reasoning.  Findings consistent with this conclusion are also reported by Ponemon and Glazer (1990).

It is also worth noting that, while McNeel's (1994a) data suggest that a conservative religious ideology may sometimes inhibit growth in principled moral reasoning (e.g., the Bible colleges), this is by no means an inevitability.  As McNeel points out, most of the liberal arts colleges in his sample were also religiously affiliated, with a strong commitment to developing students' personal Christian faith.  This was also the case with the single liberal arts college in the Good and Cartwright (1998) study.  Yet, what apparently counts in terms of their strong impact on students' growth in principled moral reasoning is the fact that the religious orientation of the liberal arts colleges was integrated within a genuine focus on liberal arts education.

Moral Behavior

A modest body of research has focused on the institutional characteristics that appear to inhibit academic dishonesty or academic cheating.  Nearly all of this research has been conducted by McCabe and his colleagues (McCabe & Trevino, 1993, 1996, 1997; McCabe, Trevino, & Butterfield, 1996, 1999).  Of particular relevance has been their concern with the contextual factors that influence student cheating behavior.  Analyzing a number of multi-institutional samples, they found that peer behavior formed a normative context for cheating.  A student was significantly more likely to admit (anonymously) to academic dishonesty if he or she observed another student cheating on an exam, and significantly less likely to admit to academic dishonesty if he or she perceived that close student acquaintances would disapprove of cheating behavior.  These associations remained statistically significant even in the presence of controls for such factors as age, sex, college grades, parental education, extracurricular involvement, involvement in intercollegiate athletics, and fraternity or sorority membership (McCabe & Trevino, 1993, 1996, 1997).  Consistent with these findings, McCabe and his colleagues also found that, irrespective of size and academic selectivity, institutions with honor codes or honor systems that are enforced by students also had significantly less student self-reported academic dishonesty than institutions without such honor codes or systems (McCabe & Trevino, 1993, 1997; McCabe et al., 1996).  They suggest three possibilities to account for this difference: 1) students at honor code schools are responsible for reporting academic dishonesty, not the faculty or administration; 2) academic integrity expectations are clearly defined to students at the start of their careers at honor code schools; and 3) students at honor code schools typically receive the privilege of unproctored examinations, which they wish to preserve.

While the above findings suggest that the implementation of academic honor codes can foster a culture of academic integrity that functions to reduce cheating behavior, a degree of caution is appropriate in accepting the causal influences that underlie their results.  An equally plausible competing hypothesis is that schools with publicly proclaimed academic honor codes simply attract and enroll students for whom academic integrity is a higher personal priority when they enter college.

Within-College Effects

Conclusions from How College Affects Students

The evidence pertaining to the influence of different college experiences on principled moral reasoning is somewhat equivocal in terms of offering consistent, replicable findings.  This may not be too surprising in that developmental theory in no way guarantees that one particular type of life experience is the preeminent cause of development in principled moral reasoning.  A certain specific experience, then, might foster development if it happens to a receptive and reflective individual and if it is accompanied by other experiences in a cumulative and mutually reinforcing pattern.  The key role of college in fostering principled moral reasoning may therefore lie in providing a range of intellectual, cultural, and social experiences from which a range of different students might potentially benefit.

From this perspective there is modest support for what, according to Kohlberg's theories, are salient experiences in the fostering of growth in moral development.  College experiences in which an individual is exposed to divergent perspectives (e.g., living away from home, intellectual interactions with roommates) or is confronted with cognitive moral conflict (e.g., courses presenting issues from different perspectives) were reported by students as having a salient influence on their moral development.  Also consistent with Kohlberg's expectations are students' specification of the importance of interactions with upperclassmen in residential facilities (i.e., exposure to more advanced stages of moral reasoning) and assuming new personal responsibilities, such as social role taking.  These experiences form a major part of the intellectual and interpersonal opportunities that characterize residential colleges and universities.  However, it would appear to be the extent to which an individual takes advantage of these opportunities, particularly those having an intellectual or academic content, that is the key determinant of growth in moral reasoning during college.

Little consistent evidence exists to suggest that academic major or broad curricular categorizations are systematically associated with differences in the development of moral judgment.  Certain specific curricular or course interventions with college students, however, do appear to foster the increased use of principled reasoning in judging moral issues.  The most consistently effective interventions are those that emphasize moral dilemma discussion or personality development.  This conclusion seems to hold regardless of the methodological rigor of the study.  While these academic interventions would appear to enhance moral development, it is clear from the evidence reviewed that their effect, as well as the effect of any specific college experience, is smaller than the effect that can be reasonably attributed to four years of college.  One possible explanation for this finding is that, similar to its impact on cognitive development, the influence of college on principled moral reasoning is the result not so much of any single experience but rather of the cumulative impact of a set of mutually reinforcing experiences.  Another possibility is that the influence of different course interventions or specific college experiences on moral reasoning is conditional rather than general (i.e., the magnitude of the influence varies for different kinds of students).

Evidence from the 1990s

The decade of the 1990s saw a continuation of rigorous attempts to isolate and identify the within-college academic and nonacademic experiences that influenced growth in principled moral reasoning.  In addition, researchers also sought to estimate the within-college experiences linked to several forms of moral behavior.  We have organized our synthesis of this body of evidence within the following general categories: moral development interventions, service-learning, major field of study, extracurricular/peer involvement, intercollegiate athletic involvement, off-campus learning, off-campus employment, and interaction with faculty.

Moral Development Interventions

Didactic Courses

As indicated from the conclusions of our previous synthesis, two types of moral development interventions, those focusing on dilemma discussion and those emphasizing personality development, appeared to be particularly effective in fostering the use of principled moral reasoning.  As pointed out by Rest and Narvaez (1991), the role of the instructor in both of these types of interventions is more as a facilitator than as an information provider.  However, a subsequent quasi-experiment by Penn (1990) suggests that growth in the use of principled moral reasoning is enhanced even more when students are exposed to direct instruction in philosophical methods of ethical analysis as well as to dilemma discussion and personality development.  Classes that had all three elements, or all three elements plus instruction in formal logic, demonstrated one-semester growth in principled moral reasoning (as measured by the DIT) that averaged about .92 of a standard deviation (32 percentile points).  This average effect size was more than twice as large as those for dilemma discussion and personality development interventions (.41 and .36, respectively) noted in our previous synthesis.  A replication of Penn's didactic approach in a general education course by McNeel (1994b) yielded one-semester growth in principled moral reasoning of .65 of a standard deviation, still substantially larger than the typical effect sizes for dilemma discussion and personality development.  Such evidence suggests that growth in principled moral reasoning is not always achieved most efficiently through the process of self-discovering the appropriateness and power of such reasoning from discussion of controversial moral dilemmas with one's peers.  Rather, the impact of moral problem-solving with one's peers may be an even more powerful inducement to growth in principled thinking if one is first taught the basic component skills of moral reasoning (e.g., formal logic, role taking, justice operations) (McNeel, 1994b).

Ethics Courses

Another purposeful intervention designed to facilitate growth in principled moral reasoning is a course focusing on ethics.  This approach has been particularly visible in undergraduate business curricula (e.g., Armstrong, 1993; Ponemon, 1993; Shaub, 1994; St. Pierre, Nelson, & Gabbin, 1990).  The weight of evidence from these studies is somewhat equivocal.  Armstrong (1993) and Shaub (1994) report findings suggesting that business students exposed to courses or interventions having a purposeful focus on professional ethics and ethical issues show significantly greater growth in the use of principled moral reasoning (as measured by the DIT) than similar students not exposed to such courses or interventions.  However, both Ponemon (1993) and St. Pierre, Nelson, and Gabbin (1990) found that the inclusion of ethical content in courses, or exposure to a specific course in ethics during college, had a trivial impact on growth in the use of principled moral reasoning.  Similarly, Smith and Oakley (1996) reported that completion of a course in business ethics had no significant effect on students' attitudes toward ethical business behavior.  We suspect that the mixed findings for this body of evidence reflect to some extent the fact that the ethics interventions in the various studies differed substantially in their content, emphasis, and implementation.  On balance, across all studies the effect on principled moral reasoning of exposure to either ethics courses or interventions is probably positive, though quite modest in magnitude.  Beyond this rather tentative speculation, however, we hesitate to offer a firm conclusion.

Ethics in the Curriculum

There is also evidence from two independent, longitudinal studies to suggest that purposefully integrating ethical content into an undergraduate professional curriculum (i.e., nursing) may foster increased growth in principled moral reasoning (Duckett & Ryden, 1994; McNeel et al., 1997).  Duckett and Ryden estimated the impact of a nursing curriculum at a public university that built ethics content vertically into the curriculum and provided ethically focused learning experiences in existing courses across all levels.  Across four cohorts of students exposed to the curriculum, they found growth in principled moral reasoning during what appeared to be the last half of college to average about .41 of a standard deviation (16 percentile points).  The absence of a control group, however, made it difficult to determine just what part of that growth was attributable to the purposefully enhanced ethics curriculum.

A somewhat less ambiguous estimate of the unique impact of an ethics-oriented curriculum on principled reasoning is presented by McNeel, Schaffer, and Juarez (1997) in a study of undergraduate nursing students at a small liberal arts college.  The intervention purposefully integrated ethical content and ethical growth experiences in both the general education and nursing program curricula.  For example, the general education curriculum included several required courses with a major ethics component.  The nursing curriculum implemented such things as an ethics seminar, specific ethics content in senior courses, and small group meetings with a faculty member focusing on ethical issues and relationships.  Employing a variation on a time-series design, McNeel and colleagues compared the freshman-to-senior growth in principled moral reasoning of three cohorts of students prior to, and three cohorts of students after, the implementation of the ethics-oriented curriculum.  Both the preceding and subsequent cohorts entered college with about the same levels of principled reasoning on the DIT.  However, the three cohorts entering after the ethics-oriented curriculum was implemented demonstrated average freshman-to-senior growth in principled reasoning that was significantly larger in magnitude (1.54 of a standard deviation, 44 percentile points) than the average corresponding growth in principled reasoning shown by the three cohorts entering prior to the implementation of the curriculum (1.34 of a standard deviation, 41 percentile points).  Thus, principled reasoning growth was more pronounced among the cohorts who experienced the ethics enriched curriculum.[4]

Ability-Based Curricula

Although it is not experimental in design, and awaits replication, Mentkowski and Associates' (2000) assessment of the impact of Alverno College's Ability-Based Curriculum suggests that student progress through that curriculum may enhance the development of principled moral reasoning.  The Ability-Based Curriculum is an integrated set of general education and disciplinary courses constructed according to a common developmental framework.  It attempts to provide connected liberal arts learning experiences in courses that increase student competencies in eight broad areas (e.g., communication, analysis, problem solving, valuing and decision making, and the like), and at six increasingly complex developmental levels within each competency or ability category.  For example, within the valuing and decision making competency, developmental growth proceeds from identifying one's own values, to engaging in valuing in decision making in multiple contexts, to applying one's own values in an area of knowledge in a professional context.  Progress through the curriculum appeared to be measured by a cumulative record of ability evaluations that students complete.  As students completed courses, they were evaluated in terms of their developmental level within the eight ability or competency dimensions.  Mentkowski and her associates found that, even in the presence of statistical controls for age, entering intellectual development and critical thinking levels, and entering levels of principled moral reasoning (as measured by the DIT), the measure of progress through the developmentally-sequenced curriculum had a modest, but statistically significant and positive direct effect on women's levels of principled reasoning after two years of college.

One reasonable conclusion from this finding is that Alverno's Ability-Based and developmentally oriented curriculum is particularly effective in promoting students' use of principled reasoning.  However, because of the absence of a control condition (e.g., a traditional and less integrated and developmentally-based liberal arts curriculum), such a conclusion involves a degree of risk.  As Mentkowski and Associates (2000) point out, while there are some clear logical links between the experiences fostered by the Ability-Based Curriculum and growth in principled moral reasoning, the design of their investigation makes it difficult to determine the magnitude of the unique advantage of an Ability-Based Curriculum over other less purposefully structured and integrated approaches to liberal arts education.


Other Interventions

To be sure, other studies have estimated the impact of purposeful interventions designed to foster growth in moral reasoning or moral development during college.  It is difficult, however, to fit these studies into a larger body of evidence.  As a group, they differ substantially from each other in terms of the focus and content of their intervention, and, as far as we could tell, are single studies that await replication.  Nevertheless, they are of some interest and deserve at least brief mention.  For example, Beller and Stoll (1992) report the results of a rigorously conducted true experiment that estimated the impact of a moral reasoning intervention designed for undergraduate student athletes.  Student athletes at a single institution were randomly assigned to two conditions: a "moral reasoning for sport" course and a control group that did not take the course.  The moral reasoning course focused on analyzing four values (honesty, responsibility, justice, and beneficence) and was integrated with readings on moral and ethical issues in sports (e.g., drug testing, gambling, rule violations, winning, and the like).  Compared to their counterparts in the control condition, student athletes exposed to the semester-long intervention demonstrated significantly larger gains on the Hahm-Beller Values Choice Inventory (a measure of principled moral judgment correlating .82 with the DIT), and larger, though only marginally significant, gains in principled reasoning as measured by the DIT.

Similarly, Mustapha and Seybert (1990) sought to determine if two different approaches to general education led to differences in students' principled moral reasoning.  The traditional (control) curriculum provided liberal arts studies through separate departmental courses.  The experimental curriculum consisted of an integrated six-course, six-semester sequence.  Each course combined the study of two or three subjects, was multidisciplinary, and was organized around the central idea of decision making.  Instructional methods sought to implement Kohlberg's theory that moral decision making is a cognitive developmental process learned through open socratic inquiry and active learning participation.  Students were confronted with situations requiring moral decision making throughout the curriculum.  With statistical controls for a standardized measure of intelligence, students in the experimental general education curriculum had a statistically significant advantage in principled moral reasoning over their counterparts in the traditional general education curriculum of about .50 of a standard deviation (19 percentile points).  Though potentially confounded by the fact that individuals could self-select themselves into the experimental curriculum, the results are nevertheless consistent with the conclusions of our previous synthesis that learning to make decisions about moral dilemmas (dilemma discussion) positively influences growth in principled moral reasoning.  Indeed, the Mustapha and Seybert findings suggest that this conclusion holds at the curricular as well as the course intervention level.[5]

Service-learning

A modest body of research has attempted to estimate the net impact of service-learning on various dimensions of moral development.  The results of this research suggest that involvement in service-learning has a somewhat inconsistent impact on growth in principled moral reasoning.  We uncovered two independent and rigorously conducted quasi-experiments that address this issue.  Both Boss (1994) and Cram (1998) randomly assigned two intact classes to different sections of one-semester undergraduate ethics courses, each of which included extensive discussion of moral dilemmas and ethical issues.  The experimental section in each study also included an active community service component that was integrated into the course content, whereas the control section had no service-learning component.  Boss found that the section with the service-learning component demonstrated significantly greater growth in principled moral reasoning (as measured by the DIT) than did the section without the service-learning component.  However, Cram reported that growth in DIT principled reasoning scores was almost identical between the experimental (service-learning) and control sections.  Since the two studies are quite similar in both methodological rigor and treatment implementation, it is difficult to choose one set of results as more valid than the other.[6]

Evidence pertaining to the impact of service-learning on aspects of moral/ethical development other than principled moral reasoning tends to be consistent with the findings of Boss (1994).  For example, in their comprehensive, multi-institutional evaluation of students in service-learning courses, Eyler and Giles (1999) found that, even with controls for age, gender, race, socioeconomic status, other service involvement, and dependent variable scores at the start of the semester, exposure to service-learning courses had a modest, but statistically significant, positive effect on the end-of-semester importance students placed on social justice.  Similar results are reported by Astin and Sax (1998) and Gray, Geschwind, Ondaatje, Robyn, Klein, Sax, Astin, and Astin (1996) for the net impact of service-learning or service involvement on sense of civic responsibility and the importance of service to the community.  Consistent with the cognitive impacts of service-learning, Eyler and Giles also report evidence suggesting that the net, positive effect of service-learning experiences on growth in the importance students place on social justice is even further enhanced when the service experience is integrated into the course content and provides opportunities for reflection through discussion and writing.  Evidence consistent with this aspect of Eyler and Giles' investigation is also reported by Myers-Lipton (1994).  Students in a program that closely integrated service and reflective learning showed greater two-year increases in civic responsibility and decreases in social prejudice than their counterparts involved in service without the reflective learning component.[7]

Major Field of Study

The small body of evidence pertaining to the net impact of major field of study on principled moral reasoning is inconsistent and provides little basis for an unambiguous conclusion.  McNeel and colleagues (1994a; 1996) summarize a series of longitudinal analyses from a single-college sample suggesting that major field of study is significantly linked to differences in freshman-to-senior growth in principled reasoning (as measured by the DIT).  Specifically, somewhat smaller principled reasoning growth appeared to accrue to students in certain vocational majors (e.g., education and business) than in other disciplines or fields of study (e.g., psychology, English, nursing, social work, humanities).  In contrast, however, Jeffrey's (1993) single-institution, cross-sectional investigation reported that differences between lower-division and senior students in principled moral reasoning were essentially the same for business and accounting majors as they were for liberal arts majors.  Similar inconsistent results have been reported by St. Pierre, Nelson, and Gabbin (1990) and Snodgrass and Behling (1996). Given the small body of evidence and the generally inconsistent findings, we believe any conclusion about the impact of major field of study on principled moral reasoning is premature.

Extracurricular/Peer Involvement

In their cogent discussion of the factors that influence growth in principled moral reasoning during college, Rest and Narvaez (1991) argue that the extracurricular milieu of college may play a potentially significant role in moral and ethical development.  There is a modicum of evidence to support their argument.  For example, Finger, Borduin, and Baumstark (1992) employed a regression approach to determine which among a range of factors was significantly linked to students' level of principled moral reasoning.  The factors included: year in college, age, socioeconomic status, relationships with parents, and frequency of informal social/extracurricular activities during college.  After year in college, level of social/extracurricular involvement during college was the next best predictor of principled reasoning, accounting for five percent of the variance in students' DIT scores.  Consistent results are reported by Lind (1997), who suggests that the role-taking opportunities that often accompany extracurricular involvement are an important catalyst in the development of principled moral reasoning.

In contrast with the evidence pertaining to the development of principled moral reasoning, the effects of extracurricular involvement on what might be considered ethical behavior are contradictory.  For example, Berger's (1998) analysis of eight religiously-affiliated colleges from the 1992-96 iteration of the Cooperative Institutional Research Program Data sought to determine the factors explaining students' community service involvement during college.  With statistical controls for such factors as previous community service, demographic characteristics, academic major, socioeconomic background, residence on campus, and high school grades, level of involvement in clubs and organizations had generally significant and positive effects on students' community service involvement.  Conversely, McCabe and Trevino (1997) found that level of extracurricular involvement during college was significantly and positively linked with self-reported academic dishonesty.  This positive relationship between extracurricular activities and cheating behavior persisted in the presence of statistical controls for such potentially confounding influences as age, gender, college grades, severity of institutional penalties for cheating, faculty support for academic integrity, perceived peer academic dishonesty, and perceived peer disapproval of cheating.

Although the Berger (1998) and McCabe and Trevino (1997) investigations consider different kinds of "moral behavior," with different regression specifications, it is still difficult to account for their diametrically opposed findings.  One possible explanation is that different kinds of extracurricular or social involvement can have different impacts on moral reasoning and moral behavior.  Consistent with this notion is recent evidence reported by Derryberry and Thoma (2000).  They found that students' principled moral reasoning levels were inversely related to the density of their friendship networks in college.  Highly dense networks are those in which each person is known by each other.  An example might be a fraternity or sorority.  Low-density friendship networks are those characterized by multiple independent friendships with a diversity of different kinds of individuals who don't typically interact with one another (e.g., friends from a residence facility, friends from one's major, friends from on-campus work, and the like).  Low-density friendship networks provide a more diverse social environment during college than high-density networks; and, therefore, greater exposure to different ideas, values, and experiences.

Derryberry and Thoma (2000) point out that their findings could reflect the fact that individuals with high levels of principled reasoning may simply be more open to diverse (low-density) friendship networks and tend to enter into them more frequently.  However, other investigations with more extensive controls for confounding influences tend to suggest a causal link between diverse friendship networks during college and both principled moral reasoning and principled behavior.  For example, Kilgannon and Erwin (1992) sought to determine the impact of fraternity or sorority membership (compared to not being Greek-affiliated) on growth in principled moral reasoning (as measured by the DIT) during the first two years of college.  According to Derryberry and Thoma, Greek-affiliated students would be more likely than those not joining fraternities and sororities to have high-density friendship networks that tend to insulate individuals from exposure to the kinds of divergent sociopolitical and cultural perspectives that have been shown to foster growth in principled moral reasoning.  One might therefore expect Greek-affiliated students to demonstrate less growth in principled reasoning than their counterparts who do not join fraternities or sororities.  This is generally what Kilgannon and Erwin found.  In the presence of statistical controls for precollege level of principled moral reasoning, women who joined sororities demonstrated significantly lower principled reasoning after two years than did women who were not sorority members.  The corresponding comparison for men was in the same direction, though not statistically significant.

Evidence suggesting that fraternity or sorority membership inhibits growth in principled moral reasoning is not an isolated finding.  There is also evidence indicating a negative relationship between Greek affiliation and moral or ethical behavior.  For example, it appears that fraternity and sorority members are more likely to admit to academic dishonesty during college than their counterparts who are not Greek-affiliated (Kirkvliet, 1994; McCabe & Bowers, 1996).  Of course this could, to some extent, reflect the possibility that individuals who choose to join fraternities and sororities have a stronger disposition toward academic dishonesty when they enter college.  Nevertheless, when McCabe and Trevino (1997) controlled for important individual and contextual influences such as age, sex, college grades, parental education, extracurricular and athletic involvement, and the peer environment with regard to cheating at the institution attended, fraternity and sorority members still reported being involved in a significantly higher level of academic dishonesty than their non-Greek-affiliated counterparts.

From a different perspective, there is also evidence suggesting that low-density, diverse friendship networks may have a positive impact on what might be considered other ethically-oriented behaviors.  In Berger's (1998) eight-institution analysis of the factors influencing community service involvement, both interaction with different racial/ethnic groups and involvement in multicultural activities tended to significantly enhance the likelihood of participating in community service during college.  These positive effects remained statistically significant even in the presence of statistical controls for salient confounding influences, such as prior community service involvement, campus residence, academic major, extracurricular involvement, race, sex, socioeconomic status, humanistic values at entry to college, and high school academic achievement.  With the same set of controls in effect, being a member of a fraternity or sorority had a significant negative impact on community service involvement in one of the three analyses conducted.  However, there is evidence that conflicts with Berger’s latter finding. Pierson (2002) found that Greek affiliation actually increased the likelihood of volunteer work in college, even with controls for secondary school volunteerism, individual background traits, and other academic and non-academic experiences during college.

Intercollegiate Athletic Involvement

A small body of research has focused, either directly or indirectly, on the relationship between intercollegiate athletic participation during college and moral development.  This literature provides little convincing evidence that involvement in intercollegiate athletics has a causal influence on either principled moral reasoning or moral behavior.  Baldizan and Frey (1995) found only small, chance differences between male and female intercollegiate athletes and male and female nonathletes in DIT principled reasoning scores.  Similarly, Beller, Stoll, Burwell, and Cole (1996) found no significant difference in the use of principled reasoning between intercollegiate athletes participating in individual sports and nonathletes.  Intercollegiate athletes participating in team sports had significantly lower principled reasoning scores than either individual sport athletes or nonathletes.  However, this may simply reflect a recruitment rather than a socialization effect, in that athletes participating in team sports might have started college with lower levels of principled moral reasoning to begin with.  The design of the Beller et al. study makes it difficult to determine which of the two effects is more probable.

The same inconclusiveness holds for the relationship between intercollegiate athletic participation and moral/ethical behavior.  Nixon (1997) found that, irrespective of sex, intercollegiate athletes are no more likely to demonstrate aggressive behavior (injuring another person) outside of their sport than nonathletes.  Similarly, the results of McCabe and Trevino's (1997) multi-institutional study of academic dishonesty indicate that participation in intercollegiate athletics has only a small and nonsignificant effect on cheating behavior when important confounding influences are taken into account.

Off-Campus Learning Experiences

Consistent with Derryberry and Thoma's (2000) contention that interaction with diverse peers and divergent intellectual and social perspectives tends to enhance students’ growth in principled moral reasoning, there is single-study evidence to suggest that participation in off-campus learning experiences positively influences such growth.  McNeel (1994a) compared the freshman-to-senior growth in principled moral reasoning of students who participated in off-campus learning (e.g., study in foreign countries, study in Washington, D.C., or other unspecified off-campus experiences) with students not participating in such programs.  Those students participating in off-campus learning experiences demonstrated gains in the use of principled moral reasoning during college that were, on average, about 1.7 times as large as the corresponding gains shown by their counterparts who did not participate in off-campus experiences.  As McNeel points out, the causal inferences from such a finding are tenuous at best.  For example, those who choose to participate in off-campus learning experiences may simply be on a particularly strong growth trajectory during college because they are especially receptive to the developmental impacts of liberal arts education.  At the same time, it is worth noting that those participating and those not participating in off-campus learning experiences started college with only trivial and chance differences in their levels of principled moral reasoning.  Consequently, it is unlikely that the results reported by McNeel are simply a reflection of regression artifacts.

Off-Campus Employment

The small body of evidence we uncovered suggests that off-campus employment during college has a negative influence on both principled moral reasoning and moral/ethical behavior.  Rykiel's (1995) study of community college students found that, net of verbal ability, individuals working off-campus 25 hours or more per week had significantly lower DIT principled reasoning scores than did their counterparts employed off-campus for less than 20 hours per week.  Similarly, Berger (1998) reported that hours of off-campus work per week tended to negatively influence community service involvement during college, even when confounding influences such as student demographic characteristics, prior community service involvement, place of residence, and academic major were taken into account.  It may well be that extensive involvement in off-campus work simply diminishes the time one has available to either commit oneself to community service, or to be involved in the social and academic experiences of college that contribute to growth in principled reasoning.  At the same time, however, both the Rykiel and Berger studies tell us only that the more hours one works off-campus per week the lower his or her level of moral reasoning or probability of participating in what might be regarded as moral/ethical behavior.  The studies do not inform us with regard to differences between students who work off-campus during college and those who do not, the impact of off-campus work related to one's academic major or intended career, or how on-campus work might influence dimensions of moral or ethical development.

Interaction With Faculty

One of the conclusions from our previous synthesis was that principled moral reasoning was enhanced by exposure to and interaction with individuals at more advanced stages of principled reasoning.  While this interaction might be most likely with upper-class peers, it could also occur through interactions with faculty.  Although based on a single-sample finding that awaits replication, there is evidence to support the contention that out-of-class contact with faculty members in a small, liberal arts college is linked to growth in principled moral reasoning. McNeel (1994a) found that students who reported at least some out-of-class contact with faculty during college had freshman-to-senior gains in DIT principled reasoning scores that were about three times as large as their counterparts who reported no such contact during college.  Furthermore, the two groups differed in only trivial, chance ways in level of principled reasoning when they entered college; so it is unlikely that the finding simply reflects regression artifacts.  McNeel speculates that informal, out-of-class interactions with faculty may have a strong impact on students' principled reasoning because such interactions provide a setting in which faculty maturity in reasoning about moral issues with which students may be struggling can be modeled for students in a personal and nonthreatening way.  Because of the nonexperimental design of McNeel's study, however, the causal direction of his results is somewhat ambiguous.  A reasonable alternative explanation for his finding is that students who are the most receptive to the impacts of liberal arts education, and who are experiencing the largest changes in how they reason about moral issues during college, may simply be the most strongly oriented toward interaction with faculty outside of class.

A related investigation by Nevins and McNeel (1992) at the same small liberal arts college suggests that it may be possible to develop faculty teaching workshops that enhance a professor's influence on students' growth in principled reasoning.  In each of three separate years, a year-long faculty development program was implemented that sought to introduce faculty to concepts of developmentally-based instruction that could be used in their classes.  The faculty who chose to participate studied a range of different models of how students develop during college (e.g., Erikson, Gilligan, King and Kitchener, Kolb, Perry).  In addition, general concepts such as maturational levels, developmental transformations, individual differences (including learning style) and a balance between challenge and support were presented and discussed.  Each faculty participant chose a target course to revise in light of what he or she had learned, and was given release time to carry out the revision.  In order to determine the impact of the faculty-development program, Nevins and McNeel divided students into groups according to the amount of contact they had with the workshop-trained faculty, and then compared their growth in DIT principled reasoning during the first two years of college.  Students with moderate or high levels of exposure to the workshop-trained faculty in the professors' target courses demonstrated growth in principled moral reasoning during the first two years of college that, on average, was about 2.9 times as large as the corresponding growth shown by students with little or no such exposure.  A similar trend in DIT scores was shown for students with different levels of exposure to the workshop faculty in their nontarget courses.  As Nevins and McNeel candidly admit, there are plausible alternative explanations for their findings, not the least of which is that the faculty-development program may have attracted the most effective teachers.  Despite this limitation, however, the faculty-development workshop they describe and assess represents a potentially significant, if yet unreplicated, intervention for enhancing the impact of faculty and their courses on student moral development.

Conditional Effects of College

Conclusions from How College Affects Students

Almost no research has systematically looked for conditional effects.  The little evidence that does exist suggests that an instructional intervention that stresses exposure to intensive moral arguments and discussion has more positive effects on principled moral judgment for subjects at higher levels of cognitive development (formal reasoners) than for subjects at lower levels (concrete reasoners).  This conditional effect underscores the notion that moral development does not occur in isolation from other areas of student development during college but rather is a part of a network of mutually supporting changes.

Evidence from the 1990s

Once again we uncovered relatively little in the way of research that systematically investigated the conditional effects of postsecondary education on moral development.  However, consistent with the conclusions of our previous synthesis, we found evidence indicating that a certain level of formal reasoning may be required for students to realize maximum benefits from instructional interventions designed to increase principled reasoning.  In an eight-week interpersonal skills training class, Santilli and Hudson (1992) found that only those students with relatively high (consolidated) formal reasoning scores showed gains in DIT principled moral reasoning.  Students with relatively low (early basic) formal reasoning levels demonstrated only chance gains in principled reasoning.

There is also evidence to suggest that the impact on principled moral reasoning of hours per week employed off-campus varied in magnitude for different aged students.  Recall from Rykiel's (1995) longitudinal study of community college students that, net of tested verbal ability, students working less than 20 hours per week had higher DIT principled reasoning scores than their counterparts employed off-campus for 25 or more hours per week.  The positive effect on principled reasoning of working less than 20 hours per week was larger for older (nontraditional aged) students than it was for younger (traditional aged) students.  However, we uncovered no independent study replicating this finding, so it is difficult to interpret its meaning within the context of other evidence.

Long-Term Effects of College

Conclusions from How College Affects Students

Evidence from ten- and twenty-year longitudinal studies is clear in identifying the positive, long-term influence of college on principled moral reasoning.  The estimated advantages that accrue to those who attend college (versus those who do not) do not diminish over time but tend to increase.  In large measure, this may be due to the tendency for college to channel individuals into posteducation occupations and lifestyles characterized by a level of continuing intellectual stimulation and challenge that either maintains or further enhances principled moral judgment.  Conversely, having only a high school diploma may tend to channel one into occupational or lifestyle environments characterized by a relatively low level of intellectual stimulation.  Thus, level of principled moral judgment may actually regress over time.

Evidence from the 1990s

We did not find the kind of longitudinal evidence in the research of the 1990s that permitted long-term comparison of trends in principled moral reasoning growth among individuals completing different amounts of postsecondary education.  However, the scant evidence we did uncover is generally consistent with the conclusions of our previous synthesis in suggesting that level of principled reasoning tends to persist or modestly increase for college graduates over the years immediately following graduation.

The most useful investigation in this regard is Mentkowski and Associates' (2000) longitudinal assessment of a sample of women attending Alverno College, a small single-sex liberal arts college.  Mentkowski and her colleagues assessed the sample with the Defining Issues Test (DIT) at four different times: when they entered college; after the first two years of college; at graduation from college; and about five years subsequent to graduation.  [The results of the first three assessments were incorporated in the results of our earlier synthesis (Pascarella & Terenzini, 1991) or our reanalysis of McNeel's (1994a) meta-analysis.]  They found that during the five years after graduating from college, women's growth in principled moral reasoning on the DIT tended to plateau (although it showed a very small growth trend).  The DIT mean score of the alumna was only slightly higher than it was when they were graduating seniors; suggesting that the rather dramatic principled reasoning growth that occurred during college (about .80 of a standard deviation in the Alverno sample) was maintained during the five years subsequent to graduation.  A smaller sample of women completed the Moral Judgment Interview (MJI) during the same four assessments, and actually showed a modest increase in principled reasoning between graduation and the five-year follow-up (about .18 of a standard deviation).  Such evidence reinforces the conclusion that the developmental trends in principled moral reasoning that occur during college do not retrogress after college.  As suggested in our previous synthesis, this is probably attributable in no small way to the particular occupations and life-style choices that are characteristic of, or available to, college graduates.

Results generally consistent with those of Mentkowski and Associates (2000) are reported by Astin, Sax, and Avalos (1999) in their longitudinal study of factors influencing community service involvement among alumni nine years after enrolling in college.  With controls for service involvement in high school, as well as other relevant confounding influences, engagement in volunteer community service during college had a significant positive effect on hours spent in voluntary community service activities as alumni.  Thus, as with increases in the use of principled reasoning that occur during college and persist into the early post-graduation years, engaging in what might be considered ethically-oriented behavior during college tends to continue when one leaves college.

Summary

Change During College

As with our previous synthesis, nearly all the evidence pertaining to change during college concerns students' use of principled reasoning in judging moral issues.  Also consistent with our previous synthesis, the evidence is consistent in suggesting that level of principled moral reasoning is positively associated with level of postsecondary educational attainment, and that students generally make statistically significant gains in principled reasoning during college. The major contribution of the research of the 1990s was that it permitted us to compute an effect size.  We estimate that the average advantage of seniors over freshmen in principled moral reasoning is about .77 of a standard deviation (28 percentile points).  It is also the case that the major change that takes place during college is a shift from using moral reasoning that concedes to societal authority (conventional moral reasoning) to reasoning that is based on the application of universal moral principles (post-conventional moral reasoning).


Net Effects of College

Inconsistencies in the way in which the evidence is reported in different studies makes it difficult to estimate the magnitude of the net impact of college on growth in principled moral reasoning.  The weight of evidence, however, suggests that at least a substantial part of the growth in principled reasoning that occurs during college is uniquely attributable to the college experience itself.  The reasons why postsecondary education may be particularly effective in facilitating growth in principled moral reasoning are not completely clear.  However, at least part of the explanation may be that college provides a relatively challenging and stimulating environment that leads students to overhaul and reconsider the ways in which they formulate moral judgments.  Despite generally consistent findings in support of a net positive impact of postsecondary education on growth in principled reasoning, it is still possible that the body of evidence is confounded by threats to internal validity such as the interaction of selection and change (i.e., differences among individuals related to their likelihood of attending college which also predict growth in principled reasoning).

Consistent with our previous synthesis, we found extensive evidence of a positive relationship between level of principled moral reasoning and the likelihood of principled behavior.  Thus, growth in principled reasoning during college should, at least indirectly, increase the probability of principled action.  However, principled reasoning, in and of itself, is probably insufficient to determine principled behavior because it is only one of several individual influences on moral action.  Others are such things as moral sensitivity, moral motivation, and moral character.  Compared to principled moral reasoning, we know relatively little about the impact of college on these other precursors of moral behavior.


Between-College Effects

Despite being based on a small sample of institutions, and the presence of some competing explanations for the results, we uncovered evidence suggesting discernible between-college effects on student growth in principled moral reasoning.  Specifically, the largest growth appeared to occur at small, private liberal arts colleges, somewhat more modest growth occurred at public universities, and the least growth happened at Bible colleges.  The findings suggest that traditional liberal arts education, combined with the unique social psychological environments of small liberal arts colleges, may be particularly conducive to fostering growth in principled moral reasoning.

In addition to principled reasoning, we uncovered evidence suggesting between-college effects on principled moral behavior.  Irrespective of size and academic selectivity, institutions with honor codes or honor systems that are enforced by students tend to have lower levels of student self-reported academic dishonesty than institutions without such honor codes or systems.  One interpretation of this finding is that the presence of such honor codes or systems helps foster a culture of academic integrity that functions to reduce cheating behavior.  However, it is also possible that schools with visible academic honor codes tend to attract and enroll students for whom academic integrity is a higher personal priority when they enter college.

Within-College Effects

The relatively extensive literature on within-college effects on moral development suggests the following general conclusions:

1.   In our previous synthesis, we found meta-analytic evidence indicating that moral development interventions focusing on moral dilemma discussion and personality development were particularly effective in fostering the use of principled moral reasoning.  The role of the instructor in these types of interventions is more as a facilitator than information provider.  However, quasi-experimental evidence from the 1990s suggests that growth in principled moral reasoning is even further enhanced when students are exposed to direct instruction in philosophical methods of ethical analysis, as well as to dilemma discussion and personality development.

2.   Evidence regarding the effect of exposure to ethics courses or ethics interventions on principled moral reasoning is somewhat mixed.  On balance, we conclude that there is probably a very modest positive effect.

3.   Evidence from two independently conducted longitudinal studies suggests that purposefully integrating ethical content into an undergraduate professional curriculum in nursing may foster increased growth in principled moral reasoning.

4.   While not experimental in design, and based on a single sample, there is evidence suggesting that progress through Alverno College's liberal arts Ability-Based Curriculum may enhance the development of principled moral reasoning.  However, the absence of a comparison or control curriculum makes it difficult to determine the net advantage of the Ability-Based Curriculum over other less purposefully structured and integrated approaches to liberal arts education.

5.   Single-sample, quasi-experimental evidence suggests that principled moral reasoning may also be significantly enhanced by a general education curriculum that integrates moral and ethical decision-making throughout a multidisciplinary six-course sequence.  Though potentially limited by student self-selection into the experimental curriculum, the results are consistent with our 1991 conclusion that learning to make decisions about moral dilemmas (dilemma discussion) positively influences growth in principled reasoning.

6.   Quasi-experimental evidence pertaining to the impact of service-learning on principled moral reasoning is mixed.  However, additional evidence suggests that service-learning or service involvement positively influences such outcomes of college as:  the importance students place on social justice, sense of civic responsibility, and importance of service to the community.  It would also appear that the impact of service experiences on these and similar outcomes is enhanced when the service experience is integrated into course content and provides opportunities for reflection through discussion and writing.

7.   We found little evidence across studies to indicate that major field of study has a consistent impact on growth in principled moral reasoning.

8.   Although we found some evidence to support the contention that the role-taking opportunities that accompany extracurricular/peer involvement in college have positive impacts on growth in principled reasoning, evidence with regard to their impact on actual moral behavior is mixed.  Different kinds of social or extracurricular involvement may well have different impacts on moral development.  In this regard, there is clear evidence supporting the hypothesis that involvement in low-density peer networks, which expose one to a diverse intellectual and social environment during college, fosters greater growth in moral development than involvement in high-density, relatively homogeneous peer networks, such as fraternities or sororities.  Indeed, the evidence we uncovered suggests that Greek affiliation may in fact inhibit growth in principled reasoning and increase the likelihood of academic dishonesty during college.

9.   We found little convincing evidence to suggest that intercollegiate athletic participation had more than a trivial, chance impact on either principled moral reasoning or academic dishonesty.

10.   Consistent with the contention that interaction with diverse peers and divergent intellectual and social perspectives tends to enhance growth in principled moral reasoning, we found single-study evidence that students participating in off-campus learning programs (e.g., study in foreign countries) showed greater growth in principled reasoning during college than their counterparts who did not participate in off-campus learning experiences.

11.   A small body of evidence suggests that off-campus employment during college has a negative influence on both principled moral reasoning and involvement in community service.  This may be because extensive off-campus work simply diminishes the time one has available to either commit oneself to community service or to be involved in the social and academic experiences in college that contribute to growth in principled reasoning.

12.   Although based on a single-sample, there is evidence to support the contention that out-of-class contact with faculty members has a positive influence on growth in principled moral reasoning.  A related finding suggests that it is possible to develop year-long faculty teaching workshops founded on introducing the concepts of developmentally-based instruction that enhance a professor's influence on students' growth in principled reasoning.

Conditional Effects of College

Consistent with the conclusions of our previous synthesis, we found evidence supporting the contention that a certain level of formal reasoning may be required for students to realize significant benefits from instructional interventions designed to increase principled moral reasoning.

Long-Term Effects of College

Consistent with the conclusions of our previous synthesis, we found evidence suggesting that increases in the use of principled moral reasoning that occur during college tend to persist into the years immediately following graduation.  Similarly, involvement in voluntary community service during college tends to continue when one leaves college.

 

Adams, M., & Zhou-McGovern, Y. (1994). The sociomoral development of undergraduates in a "social diversity" course:  Developmental theory, research, and instructional applications. Paper presented at the annual meeting of the American Educational Research Association, New Orleans.

Armstrong, M. (1993). Ethics and professionalism in accounting education. Journal of Accounting Education, 11, 77-92.

Arnold, D., & Ponemon, L. (1991). Internal auditors' perceptions of whistle-blowing and the influence of moral reasoning:  An experiment. Auditing:  A Journal of Practice and Theory, 10(2), 1-15.

Astin, A., & Sax, L. (1998). How undergraduates are affected by service participation. Journal of College Student Development, 39, 251-263.

Astin, A., Sax, L., & Avalos, J. (1999). Long-term effects of volunteerism during the undergraduate years. Review of Higher Education, 22, 187-202.

Baldizan, L., & Frey, J. (1995). Athletics and moral development:  Regulatory and ethical issues. College Student Affairs Journal, 15, 33-43.

Baldwin, D., Adamson, T., Sheehan, J., Self, D., & Oppenberg, A. (1996). Moral reasoning and malpractice: A pilot study of orthopedic surgeons. American Journal of Orthopedics, 25, 481-484.

Batchelder, T., & Root, S. (1994). Effects of an undergraduate program to integrate academic learning and service:  Cognitive, prosocial cognitive, and identity outcomes. Journal of Adolescence, 17, 341-355.

Bebeau, M. (1994). Influencing the moral dimensions of dental practice. In J. Rest & D. Narvaez (Eds.), Moral development in the professions:  Psychology and applied ethics (pp. 113-139). Hillsdale, NJ: Lawrence Erlbaum.

Beller, J., & Stoll, S. (1992). A moral reasoning intervention program for student athletes. Athletic Academic Journal(Spring), 43-57.

Beller, J., Stoll, S., Burwell, B., & Cole, J. (1996). The relationship of competition and a Christian liberal arts education on moral reasoning of college student athletes. Research on Christian Higher Education, 3, 99-114.

Berger, J. (1998). Organizational behavior and student outcomes:  A new perspective on college impact. Unpublished manuscript, University of New Orleans, New Orleans.

Bernardi, R. (1991). Fraud detection:  An experiment testing differences in perceived client integrity and competence, individual auditor cognitive style and experience, and accounting firms. Unpublished doctoral dissertation, Union College, Schenectady, NY.

Boss, J. (1994). The effect of community service work on the moral development of college ethics students. Journal of Moral Education, 23, 183-198.

Colby, A., Kohlberg, L., Gibbs, J., & Lieberman, M. (1983). A longitudinal study of moral judgment. Society for Research on Child Development Monograph, 48(1-2, Serial No. 200).

Cram, S. (1998). The impact of service-learning on moral development and self-esteem of community college ethics students. Unpublished doctoral dissertation, University of Iowa, Iowa City.

Crandall, C., Tsang, J., Goldman, S., & Pennington, J. (1999). Newsworthy moral dilmmas:  Justice, caring, and gender. Sex Roles, 40, 187-209.

Crow, S., Fok, L., Hartman, S., & Payne, D. (1991). Gender and values:  What is the impact on decision making? Sex Roles, 25, 255-268.

Cummings, R., Dyas, L., & Maddux, C. (2001). Principled moral reasoning and behavior of preservice teacher education students. American Educational Research Journal, 38, 143-158.

Derryberry, W., & Thoma, S. (2000). The friendship effect: Its role in the development of moral thinking in students. About Campus, 5(2), 13-18.

Duckett, L., Rowan-Boyer, M., Ryden, M., Crisham, P., Savik, K., & Rest, J. (1992). Challenging misperceptions about nurses' moral reasoning. Nursing Research, 41, 324-331.

Duckett, L., & Ryden, M. (1994). Education for ethical nursing practice. In J. Rest & D. Narvaez (Eds.), Moral development in the professions:  Psychology and applied ethics (pp. 48-66). Hillsdale, NJ: Lawrence Erlbaum.

Eyler, J., & Giles, D. (1999). Where's the learning in service learning? San Francisco: Jossey-Bass.

Finger, W., Borduin, C., & Baumstark, K. (1992). Correlates of moral judgment development in college students. Journal of Genetic Psychology, 153, 221-223.

Flanagan, O. (1982). Virtue, sex, and gender. Ethics, 92, 499-512.

Foster, J., & La Force, B. (1999). A longitudinal study of moral, religious, and identity development in a Christian liberal arts environment. Journal of Psychology and Theology, 27, 52-68.

Galotti, K., Kozberg, S., & Farmer, M. (1991). Gender and developmental differences in adolescents' conceptions of moral reasoning. Journal of Youth and Adolescence, 20, 13-30.

Gielen, U., & Markoulis, D. (1994). Preference for principled moral reasoning:  A developmental and cross-cultural perspective. In L. Adler & U. Gielen (Eds.), Cross-cultural topics in psychology (pp. 73-87). Westport, CT: Greenwood.

Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University Press.

Gilligan, C., & Attanucci, J. (1988). Two moral orientations:  Gender differences and similarities. Merrill-Palmer Quarterly, 43, 223-237.

Good, J., & Cartwright, C. (1998). Development of moral judgement among undergraduate university students. College Student Journal, 32, 270-276.

Gorman, M., Duffy, J., & Heffernan, M. (1994). Service experience and the moral development of college students. Religious Education, 89, 422-431.

Gray, M., Geschwind, S., Ondaatje, E., Robyn, A., Klein, S., Sax, L., et al. (1996). Evaluation of learn and serve America, higher education:  First year report (Vol. I). Washington, D.C.: RAND Institute on Education and Training.

Hill, K. (1995). Critical thinking and its relation to academic, personal, and moral development in the college years. Dissertation Abstracts International, 56, 4603B.

Ignelzi, M. (1990). Ethical education in a college environment: The just community approach. NASPA Journal, 27, 191-198.

Ikenberry, S. (1997). Values, character, leadership: Reexamining our mission. Educational Record, 78(3,4), 7-9.

Jeffrey, C. (1993). Ethical development of accounting students, business students, and liberal arts students. Issues in Accounting Education, 8, 26-40.

Kilgannon, S., & Erwin, T. (1992). Longitudinal study about the identity and moral development of Greek students. Journal of College Student Development, 33, 253-259.

King, P. (1997). Character and civic education: What does it take? Educational Record, 78(3,4), 87-93.

King, P., & Mayhew, M. (2002). Moral judgment development in higher education: Insights from the Defining Issues Test. Unpublished unpublished manuscript, University of Michigan, Ann Arbor, MI.

Kirkvliet, J. (1994). Cheating by economics' students:  A comparison of survey results. Journal of Economic Education, 25, 121-133.

Knox, P., Fagley, N., & Miller, P. (1998). Moral orientation in a sample of African American college student. Paper presented at the annual meeting of the American Educational Research Association, San Diego, CA.

Kohlberg, L. (1981). The meaning and measurement of moral development. Worcester, MA: Clark University Press.

Kohlberg, L. (1984). Essays on moral development.  The psychology of moral development (Vol. 2). San Francisco: Harper and Row.

Lind, G. (1997). Educational environments which promote self-sustaining moral development. Konstanz, Germany: University of Konstanz.

Mason, M., & Gibbs, J. (1993). Social perspective taking and moral judgment among college students. Journal of Adolescent Research, 8, 109-123.

Mathiasen, R. (1998). Moral education of college students: Faculty and staff perspectives. College Student Journal, 32, 374-377.

McCabe, D., & Bowers, W. (1996). The relationship between student cheating and college fraternity or sorority membership. NASPA Journal, 33, 280-291.

McCabe, D., & Trevino, L. (1993). Academic dishonesty:  Honor codes and other contextual influences. Journal of Higher Education, 64, 522-538.

McCabe, D., & Trevino, L. (1996). What we know about cheating in college. Change, 28, 29-33.

McCabe, D., & Trevino, L. (1997). Individual and contextual influences on academic dishonesty:  A multicampus investigation. Research in Higher Education, 38, 379-396.

McCabe, D., Trevino, L., & Butterfield, K. (1996). The influence of collegiate and corporate codes of conduct on ethics-related behavior in the workplace. Business Ethics Quarterly, 4, 461-476.

McCabe, D., Trevino, L., & Butterfield, K. (1999). Academic integrity in honor code and non-honor code environments: A qualitative investigation. Journal of Higher Education, 70, 211-234.

McNeel, S. (1994a). College teaching and student moral development. In J. Rest & D. Narvaez (Eds.), Moral development in the professions:  Psychology and applied ethics (pp. 26-47). Hillsdale, NJ: Lawrence Erlbaum.

McNeel, S. (1994b). Integrating psychology and philosophy in the teaching of ethics:  A replication of Penn's direct approach. Unpublished manuscript, Bethel College, St. Paul, MN.

McNeel, S., Abou-Zeid, B., Essenburg, T., Smith, R., Danforth, D., & Weaver, R. (1996). Business students:  Are they really different? Research on Christian Higher Education, 3, 33-56.

McNeel, S., & Granstrom, S. (1995). Principled reasoning, critical thinking, and spirituality. Unpublished manuscript, Bethel College, St. Paul, MN.

McNeel, S., Schaffer, M., & Juarez, M. (1997). Growth in moral judgment among baccalaureate nursing students. Research on Christian Higher Education, 4, 29-45.

Mentkowski, M., & Associates. (2000). Learning that lasts: Interpreting learning, development, and performance in college and beyond. San Francisco: Jossey-Bass.

Mullane, S. (1999). Fairness, educational value, and moral development in the student disciplinary process. NASPA Journal, 36, 86-95.

Mustapha, S., & Seybert, J. (1990). Moral reasoning in college students:  Effects of two general education curricula. Educational Research Quarterly, 14(4), 32-40.

Myers-Lipton, S. (1994). The effects of service-learning on college students' attitudes toward civic responsibility, international understanding, and racial prejudice. Unpublished doctoral dissertation, University of Colorado, Boulder.

Nevins, K., & McNeel, S. (1992). Facilitating student moral development through faculty development. Moral Education Forum, 17(4), 12-18.

Nixon, H., II. (1997). Gender, sport, and aggressive behavior outside sport. Journal of Sport & Social Issues, 21, 379-391.

Paradice, D., & Dejoie, R. (1991). The ethical decision-making process of information systems workers. Journal of Business Ethics, 10, 1-21.

Pascarella, E. (1997). College's influence on principled moral reasoning. Educational Record, 78(3,4), 47-55.

Pascarella, E., & Terenzini, P. (1991). How college affects students: Findings and insights from twenty years of research. San Francisco: Jossey-Bass.

Penn, W. (1990). Teaching ethics -- A direct approach. Journal of Moral Education, 19, 124-138.

Pierson, C. (2002). Volunteerism in college: Impacts on cognitive outcomes, learning orientations, and educational aspirations. Unpublished unpublished doctoral dissertation, University of Iowa, Iowa City, IA.

Ponemon, L. (1992). Auditor underreporting of time and moral reasoning:  An experimental-lab study. Contemporary Accounting Research, 9, 171-189.

Ponemon, L. (1993a). Can ethics be taught in accounting? Journal of Accounting Education, 11, 185-209.

Ponemon, L., & Gabhart, D. (1990). Auditor independence judgments:  A cognitive development model and experimental evidence. Contemporary Accounting Research, 7, 227-251.

Ponemon, L., & Gabhart, D. (1993). Ethical reasoning in accounting and auditing. Vancouver, Canada: Canadian General Accountants' Research Foundation.

Ponemon, L., & Glazer, A. (1990). Accounting education and ethical development:  The influence of liberal learning on students and alumni in accounting practice. Issues in Accounting Education, 5, 195-208.

Rest, J. (1979). Development in juding moral issues. Minneapolis: University of Minnesota Press.

Rest, J. (1986c). Moral development:  Advances in research and theory. New York: Praeger.

Rest, J. (1994). Background:  Theory and research. In J. Rest & D. Narvaez (Eds.), Moral development in the professions:  Psychology and applied ethics (pp. 1-25). Hillsdale, NJ: Lawrence Erlbaum.

Rest, J., & Narvaez, D. (1991). The college experience and moral development. In W. Kurtines & J. Gewirtz (Eds.), Handbook of moral behavior and development (pp. 229-245). Hillsdale, New Jersey.

Rest, J., Narvaez, D., Bebeau, M., & Thoma, S. (1999). Postconventional moral thinking: A neo-Kohlbergian approach. Mahwah, NJ: Lawrence Erlbaum Associates, Publishers.

Rest, J., Thoma, S., & Edwards, L. (1997). Devising and validating a measure of moral judgment:  Stage preference and stage consisting approaches. Journal of Educational Psychology, 89, 5-28.

Rykiel, J. (1995). The community college experience: Is there an effect on critical thinking and moral reasoning? Dissertation Abstracts International, 56, 3824A.

Santilli, N., & Hudson, L. (1992). Enhancing moral growth:  Is communication the key? Adolescence, 27, 145-160.

Self, D., & Baldwin, D. (1994). Moral reasoning in medicine. In J. Rest & D. Narvaez (Eds.), Moral development in the professions:  Psychology and applied ethics (pp. 140-153). Hillsdale, NJ: Lawrence Erlbaum.

Shaub, M. (1994). An analysis of the association of traditional demographic variables with the moral reasoning of auditing students and auditors. Journal of Accounting Education, 12, 1-26.

Smith, P., & Oakley, E. (1996). The value of ethics education in business school curriculum. College Student Journal, 30, 274-283.

Snodgrass, J., & Behling, R. (1996). Differences in moral reasoning between college and university business majors and non-business majors. Business and Professional Ethics Journal, 15, 79-84.

St. Pierre, K., Nelson, E., & Gabbin, A. (1990). A study of the ethical development of accounting majors in relation to other business and nonbusiness disciplines. Accounting Educator's Journal, 3, 23-35.

Tennant, S. (1991). Personal and moral development:  A developmental curriculum intervention for liberal arts freshmen. Dissertation Abstracts International, 51, 2657A.

Walker, L. (1984). Sex differences in the development of moral reasoning:  A critical review. Child Development, 57, 522-527.

Walker, L. (1991). Sex differences in moral reasoning. In W. Kurtines & J. Gewirtz (Eds.), Handbook of moral behavior and development:  Volume 2:  Research (pp. 333-364). Hillsdale, NJ: Lawrence Erlbaum Associates.

Wilson, K., Rest, J., Boldizar, J., & Deemer, D. (1992). Moral judgment development:  The effects of education and occupation. Social Justice Research, 5, 31-48.

 



[1] Parts of this chapter draw heavily on an earlier article: Pascarella, E. (1997). College’s influence on principled moral reasoning. Educational Record, 78 (Summer/Fall), 47-55, and on an excellent review by King and Mayhew (2002).

[2] Recently the “principled” part of principled moral reasoning has come under criticism by moral philosophers who object to Principalism (Rest, Narvaez, Bebeau, & Thoma, 1999).  In response to this, some scholars have replaced principled moral reasoning with postconventional moral reasoning (Rest et al., 1999). While we recognize this concern, most of the research literature we reviewed still uses the term principled moral reasoning or judgment.  Because of this, and for purposes of continuity with our previous synthesis, we continue to employ the term principled moral reasoning in this chapter.

[3] However, despite some theoretical arguments to the contrary (Gilligan, 1982; Gilligan & Attanucci, 1988), there is little empirical support for the notion that principled moral reasoning is significantly related to gender (Baldizan & Frey, 1995; Crandall, Tsang, Goldman, & Pennington, 1999; Flanagan, 1982; Galotti, Kozberg, & Farmer, 1991; Knox, Fagley, & Miller, 1998; McNeel, Schaffer, & Juarez, 1997; Walker, 1984, 1991).

[4] There was also a tendency for students who transferred into the program their junior year to show significantly greater growth in DIT principled reasoning test scores during the last two years of college than students already in the program.  However, because the former group started out with nearly a one standard deviation deficit in principled reasoning at the start of their junior year, relative to the latter group, it is likely that this comparison may be confounded by regression artifacts.

[5] Other quasi-experimental research has been less successful in finding significant effects of moral development interventions on principled moral reasoning.  See, for example, studies estimating the impact of a social diversity course (Adams & Zhou-McGovern, 1994), an interpersonal communications course (Santilli & Hudson, 1992), and a psychosocial/developmental course for freshmen (Tennant, 1991).

[6] Results reported by Gorman, Duffy, and Heffernan (1994) tend to support those of Boss (1994). However, the study design is not as rigorous as that of either Boss (1994) or Cram (1998).

[7] Batchelder and Root (1994) also found that students in highly integrated service-learning courses made statistically significant increases in prosocial reasoning (reasoning reflecting empathy, concern for societal justice, importance of contractual obligations) during a semester. However, their analysis did not provide for a control group; so it is unclear how much of the growth is attributable to service-learning.