«A CROSS-CULTURAL STUDY OF THE RELATIONSHIPS BETWEEN EPISTEMOLOGICAL BELIEFS AND MORAL JUDGMENT AS A PSYCHOLOGICAL FOUNDATION FOR MORAL EDUCATION by ...»
be highly correlated (Hanks, 1985). Verbal ability has also been shown to correlate with DIT scores. A study of 360 Israeli students found moral reasoning scores correlated moderately with total aptitude scores; the strongest correlation found was with the verbal ability factor (Zeidner & Nevo, 1987).
In a study of 154 undergraduate students from a large midwestern university, Bendixen et al. (1998) performed a four-level hierarchical multiple regression in which gender was entered first, followed by age and education, then logical reasoning skill, and finally epistemological beliefs. Syllogisms were used to provide a measure of logical reasoning skill. Results of the regression analysis reported that logical reasoning skill was significant, accounting for 4% of additional sample variation in P scores.
There is general agreement that when the impact of higher education is assessed with critical thinking or moral development, intellectual ability must be considered as a potential confounding effect (Rest, 1979, 1994; Kitchener et al., 1984; Pascarella & Terenzini, 1991). Critical thinking, a subset of general intelligence, is strongly linked to moral development, and research utilizing the DIT to measure moral reasoning levels typically finds critical thinking and moral reasoning skills to be generally related (Fasco, 1994). A correlation as high as.41 between moral reasoning and critical thinking (as measured by the Cornell Critical Thinking Test) has been reported (Rest, 1990).
However, critical thinking has been found to be necessary but not sufficient for moral reasoning (Stewart & Pascual, 1992) and high cognitive ability also appears to be necessary but not sufficient for high moral judgment (Narvaez, 1993). This implies that moral development, while influenced by general intelligence, is a unique developmental
to moral reasoning skill increases. However, for college students, this has not been demonstrated with much success.
GPA as Related to the Development of Moral Judgment In DIT studies, there are several direct relationships with academic achievement and moral development such as evidence indicating that grade point average (GPA) is highly associated with moral judgment (Johnson, Insley, Motwani, & Zbib, 1993;
Pascarella & Terenzini, 1991). Johnson et al. investigated the relationship between students’ facility with business writing and moral judgment. Using a sample of 72 juniors and seniors, they found that GPA was the best predictor of DIT scores, accounting for 70% of the variance. Similarly, they also found a significant relationship between students’ grades on a series of writing assignments that were scored for writing mechanics, completeness, tone and design. Whether earning high scores on these aspects of good writing constitutes moral behavior is arguable.
Gender as Related to the Development of Moral Judgment Gilligan’s (1982) critique of Kohlberg's theory of moral reasoning and her assertion that two modes of moral reasoning (justice and care) exist have been the subject of debate within the field of psychology for more than 15 years. So far there is no evidence that there are two tracks of development, one for women and one for men.
Those sex differences that do exist appear to be differences in mode or style rather than structure (Colby & Kohlberg, 1987). Furthermore, there is abundant evidence that girls’ and women’s responses to Kohlberg’s hypothetical dilemmas are readily scorable by the
there are no sex differences in stages (Gibbs, Arnold, & Burkhardt, 1984; Walker, 1984).
More currently, Jaffe and Hyde (2000) conducted a meta-analysis to review quantitatively the work on gender differences in moral orientation. The meta-analysis revealed small differences in the care orientation favoring females (d = -.28) and small differences in the justice orientation favoring males (d =.19). Together, the moderator variables accounted for 16% of the variance in the effect sizes for care reasoning and 17% of the variance in the effect sizes for justice reasoning. These findings do not offer strong support for the claim that the care orientation is used predominantly by women and that the justice orientation is used predominantly by men.
Academic Major as Related to the Development of Moral Judgment The DIT has been used to measure differences in the moral reasoning of college students across academic disciplines (Cummings, Days, & Maddux, 2001; Icerman, Karcher, & Kennelley, 1991; King & Mayhew, 2002; Paradice & Dejoie, 1991; Jeffrey, 1993; Ponemon & Gabhart, 1994; Snodgrass & Behling, 1996; St. Pierre, Nelson, & Gabbin, 1990; Zeidler & Schafer, 1984). Variability of moral reasoning scores within certain disciplines has also been observed (Icerman et al., 1991; Jeffrey, 1993; Paradice & Dejoie, 1991).
Several studies have attempted to measure differences in moral reasoning between academic disciplines (St Pierre et al., 1990; Snodgrass & Behling, 1996), yielding inconclusive results. For example, St Pierre et al. (1990) found that accounting majors and students majoring in other business disciplines (i.e., finance, information systems, hotel/restaurant management, management, marketing and international business)
psychology, math and social work. Snodgrass and Behling (1996), by contrast, found no significant differences in the moral reasoning levels between business and non-business majors (i.e., arts and humanities, social sciences, natural sciences and undeclared).
Summary There is considerable evidence that moral reasoning is significantly related with cognitive abilities such as logical reasoning (Bendixen et al., 1998), moral comprehension (Rest et al., 1999), IQ (Hanks, 1985), and verbal ability (Zeidner & Nevo, 1987). Previous research also suggests that moral reasoning level is affected by certain background variables such as age (Kurtines, 1982; Rest, 1986; Rykiel, 1995), gender (Bendixen et al., 1998: Jaffe & Hyde, 2000), education (Davison, 1979; Kitchener et al., 1984; Rest, 1979; Rest & Thoma, 1985; Rest et al., 1999), GPA (Johnson et al., 1993;
Pascarella & Terenzini, 1991), and academic major (St Pierre et al., 1990).
This study investigated whether epistemological beliefs are related to moral reasoning over and above the effects of other critical variables (i.e., age, education, gender, syllogistic reasoning skill, GPA, and academic major) in Korean and U.S. college students and whether differences between the two groups exist. Therefore this study extends the work of Rest and associates (1986, 1999) by studying the effects of epistemological beliefs, age, education, gender, basic reasoning skill, GPA, and academic major on the acquisition of moral reasoning ability.
This correlational study is cross-sectional and non-experimental in nature utilizing subjects where random sampling is not possible. The study utilizes a measure of student moral reasoning ability as the criterion variable. All-possible regressions using two criteria (R2 and the Mallows Cp) examine the predictive probability of: (a) students’ epistemological beliefs, and (b) other critical variables such as age, education, gender, syllogistic reasoning skill, grade point average (GPA), and academic major.
This dissertation explores whether cultural patterns exist in the relationship among epistemological beliefs and moral reasoning. Three research questions guide my work: (a) What are similarities and differences in the relationships among five epistemological beliefs and moral reasoning between Korean and U.S. college students?
(b) Will epistemological beliefs be related to moral reasoning over and above the effects of other critical variables (i.e., age, education, gender, syllogistic reasoning skill, GPA, and academic major) in each respective group? and (3) Which of five epistemological beliefs explains the greatest amount of sample variation in the performance on the Defining Issues Test in each respective group?
A total of 481 undergraduate college students from Korea and the United States were involved in this study. All the participants were undergraduate students taking education courses at their respective institution. The Korean sample consisted of 267
(n = 49), Chuncheon National University of Education (n = 118), and Incheon National University of Education (n = 100). Responses by 24 subjects were eliminated; twenty two for failure to pass the internal check for consistency of responses on the Defining Issues Test, and two for returning the questionnaires substantially incomplete. Of the remaining 243 subjects, subjects’ ages ranged from 18-38 with the average age being 22.1 years.
Regarding subjects’ gender, 151 (62.1%) were female and 92 (37.9%) were male. The average GPA was 3.33. The sample included 44 freshmen (18.1%), 26 sophomores (10.7%), 125 juniors (51.4%), and 48 seniors (19.8%). In the Korean sample, all the subjects came from ethics education (33.7%), mathematics education (26.7%), social science education (23%), education (6.2%), computer education (5.8%), and Korean language education (4.5%).
The U.S. sample consisted of 214 undergraduate students who were studying at the University of Georgia. Twenty three sets of responses were eliminated; seventeen for failure to pass the internal check for consistency of responses on the Defining Issues Test, and six for returning the questionnaires substantially incomplete. Of the remaining 191 subjects, subjects’ ages ranged from 17-49 with the average age being 21.2 years.
Regarding subjects’ gender, 142 (74.3%) were female and 49 (25.7%) were male. The average GPA was 3.37. The sample included 20 freshmen (10.5%), 48 sophomores (25.1%), 82 juniors (42.9%), and 41 seniors (21.5%). In the U.S. sample, early childhood education, social science education, science education, and mathematics education were chosen by approximately 71.2% of respondents, while the remaining 28.8% chose a
Because I had to manage various subjects from different places at the same time, I needed individuals to help me collect data in Korea. I contacted professors and friends in
to administer the instruments for me. As a result, three moral education professors and four graduate students were involved in this study. In terms of my explanations of this study, I only provided instructions for administering the instruments. I restricted other information as much as possible to reduce the likelihood of any experimental bias. At the University of Georgia I administered all the instruments.
After securing permission from appropriate faculty members, the researcher (or research assistants) addressed the classes and asked for students to volunteer for the study following procedures approved by the University of Georgia Institutional Review Board.
The consent document (see Appendix E) was reviewed and students were assured of their anonymity and the confidentiality of their responses. They were also informed that if they chose not to participate they would not incur negative consequences. On a voluntary basis three Korean students and 86 students at UGA chose not to participate. For those students who chose to participate, they could either complete the questionnaires in a specially provided room or complete the instrument at home and return it to the researcher (or research assistants).
To be consistent with Bendixen et al. (1998), procedures were the same as the Bendixen et al. study. Participants were asked to indicate the degree to which they agreed with each statement on the Epistemic Beliefs Inventory (EBI), using the 5-point scale.
Mean completion time was approximately five minutes. Participants next completed the 12-item syllogism test by circling the most plausible option from the four possible responses. Mean completion time was about 10 minutes. After completing the brief
The Defining Issues Test (DIT) Rest (1979) formulated the Defining Issues Test (see Appendix A) as an instrument to assess a subject’s level of moral reasoning (Rest, 1979, 1986; Rest et al., 1999). The DIT presents each subject with a set of stories containing a moral dilemma and a list of statements reflecting possible considerations for deciding how to solve the dilemma. The subjects are asked to rate the statements in importance and then rank the four statements they see as the most important considerations.
Principled moral reasoning is moral reasoning at Stages 5 and 6. Rest (1979) devised a P score (sum of weighted ranks given to Stage 5 and Stage 6 items) as a measure of principled moral reasoning. This score is interpreted as “the relative importance a subject gives to principled moral considerations in making a decision about moral problems” and is “the most utilized and sensitive index from this instrument” (Rest, 1990, p. 4.2). The P score gives a percentage type score varying between.00 and
1.00. Thus, a score of.65 would indicate that the individual utilized principled moral reasoning 65% of the time, while a score of.20 would indicate that the individual utilized Level 5 or 6 moral reasoning 20% of the time.
A summary of seven studies concluded that the test-retest reliabilities for the P are in the high.70s or.80s (Rest et al., 1974). Internal consistency measurements are also satisfactory. A 1974 study of 160 subjects yielded an alpha of.77 for the P index (Rest et al., 1974). A more recent sample of 1,080 subjects also yielded an alpha of.77 for the P
computed. The entire instrument was utilized for the present research because, “although the DIT can be divided into two sets of three stories each, the two sets should not be considered alternative forms of the DIT” (Rest, 1990, p. 5.3).