Learning and Instruction 19 (2009) 287e298 www.elsevier.com/locate/learninstruc
Separating academic and social experience as potential factors in epistemological development Michael Weinstock*, Hila Zviling-Beiser Department of Education, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel Received 6 March 2007; revised 15 February 2008; accepted 15 May 2008
Abstract Education has been regarded as the major factor in epistemological development. However, academic and other aspects of educational experience are difficult to disentangle. In the present study, 86 Israeli students in the same selective department differed by whether they had done mandatory army service before the university or would do it after. This difference allowed a comparison between epistemological levels of students with different social experiences but with similar academic experiences. Those with the socially diverse army experience were less likely to have absolutist thinking about everyday knowledge contexts. Those with more years of education had more sophisticated academiccontext epistemological beliefs in their discipline only. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Epistemological development; Epistemological beliefs; Social environment; Higher education
1. Introduction Perry’s (1970) original longitudinal work on epistemological development found shifts in undergraduates’ understandings about the nature of knowledge during their 4 years in college. In brief, students entered college holding understandings of knowledge in which a single claim had to be right and all competing claims had to be wrong. That is, true knowledge was understood to be certain and objective and, thus, experts could discover the truth. During the course of their undergraduate experience, the students developed relativist understandings of knowledge in which knowledge was first seen as radically subjective with no way to adjudicate between claims and no special status given to experts. This radical relativism began to give way among some of the students to an understanding that decisions could be made about the value of knowledge claims while also maintaining that knowledge was necessarily uncertain. Perry (1970) described nine discrete levels in this movement from belief that knowledge is objective to belief that knowledge is radically subjective to belief that knowledge has objective and subjective aspects. This movement has been commonly characterized as consisting of three overarching levels known as ‘‘absolutist,’’ ‘‘multiplist,’’ and ‘‘evaluativist’’ (Hofer & Pintrich, 1997; Kuhn & Weinstock, 2002). Analyzing extensive, open-ended interviews of the students repeated during their college careers, Perry suspected that the rich social and academic environment of college fostered these shifts in epistemological understanding. For the most part, students came from fairly homogenous communities and had traditional schooling. Socially, they had limited exposure to different worldviews and tended to believe in certain truth in religion, politics, values, and schooled knowledge. The lines of authoritydparents, clergy, community leaders, teachers, and principalsdwere well drawn. In contrast, the college experience, both in the dormitories and classrooms, filled with competing perspectives forced a confrontation with relativism. The experience of
* Corresponding author. Tel.: þ972 8 646 1872. E-mail address:
[email protected] (M. Weinstock). 0959-4752/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.learninstruc.2008.05.004
288
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
meeting people from different backgrounds with different values and beliefs and hearing professors disagree with one another and discuss best theoretical explanations to date shook the certainty of truth. Thus, Perry (1970) speculated, students had to develop new understandings of the sources of knowledge and learn to make commitments in the face of potentially overwhelming relativism. Most research on epistemological development in the decades following Perry has taken place within educational contexts. Many of these studies have found that educational level, and not age, seems to underlie epistemological development. Longitudinal studies limited to college (King, Kitchener, Davison, Parker, & Wood, 1983; Perry, 1970) or high school (Schommer, Calvert, Gariglietti, & Bajaj, 1997) have found a relationship between epistemological level and educational level. These findings, as well as cross-sectional studies of students within institutions, (for college, see Ryan, 1984; and for school-age, see Chandler, Boyes, & Ball, 1990; Leadbeater & Kuhn, 1989) give evidence that education influences epistemological development beyond what might be attributed to an effect of selection. The focus on academic contexts and educational level as a factor in development has extended one aspect of Perry’s pioneering work. In fact, the study of epistemological beliefs appears as a subfield of educational psychology even though epistemological understandings have been found to be related with the everyday reasoning of both students and non-students (Kardash & Scholes, 1996; Mason & Boscolo, 2004; Mason & Scirica, 2006; Stanovich, 1999; Weinstock & Cronin, 2003; Weinstock, Neuman, & Glassner, 2006). The pursuit of research on the academic aspect of epistemological beliefs has not followed up on Perry’s (1970) conjectures about the source of epistemological development to determine more specifically what occurs within the educational context. One possible factor in epistemological development found in the academic context is academic discipline (see Muis, Bendixen, & Haerle, 2006). There are several reasons to believe that disciplinary experience should be a factor in epistemological development: studies have found that people hold domain-specific epistemological beliefs (Hofer, 2000; Kuhn, Cheney, & Weinstock, 2000; Paulsen & Wells, 1998; Tabak & Weinstock, 2005), students’ epistemological beliefs differ by academic major (King, Wood, & Mines, 1990; Lonka & Lindblom-Yla¨nne, 1996), the nature of disciplines and the educational activities and emphases within the disciplines differ (Biglan, 1973; Donald, 1995; Donnelly, 1999; Kaartinen-Koutaniemi & Lindblom-Yla¨nne, 2008), and disciplines use discipline-specific methods of building knowledge (Lehman, Lempert, & Nisbett, 1988; Wineburg, 1991). However, there are few cross-sectional or longitudinal studies that test whether immersion in a major field of study does have a particular influence on domain-specific, or domain-general, epistemological development. In one cross-sectional study, Lonka and Lindblom-Yla¨nne (1996) found that advanced psychology and medical students were more relativistic than novices in the same fields, suggesting development with the number of years of schooling. Moreover, the psychology students were more relativist than the medical students, suggesting different courses of development depending on major. However, the psychology novices were also more relativistic than the medical novices. Perhaps, then, the differences appear not because of disciplinary experience, but because students tend to choose a major in which the nature of the discipline and the way it is taught suits their prior epistemological beliefs (Trautwein & Lu¨dtke, 2007a). What is lacking in the research on domain-specific epistemological beliefs and the influence of discipline on epistemology is a fusion that investigates epistemological beliefs about different disciplines over the course of several years of experience in a single discipline. Perry’s (1970) distinction between the social and academic experiences in the school context, and the possible role of social experience in particular, in epistemological development has received even less attention than the influence of disciplinary experience. Moreover, epistemological researchers have not looked to out-of-school contexts before or during university studies as a possible source of epistemological development. An intense, first immersion in a diverse social environment might prove to be more important in fostering epistemological development than the academic diversity characteristic of the American liberal arts college experience. A program peculiar to Israel provides a unique opportunity to tease academic experience from social experience. The majority of Israelis are required to perform army service following high school. However, some very capable students are eligible for a program in which they study at the university before performing the bulk of their service. These students are known as atuda’im (‘‘student soldiers’’). Thus, in the same department in a university, there are identifiable groups of students at the same educational level that differ by out-of-school experience. In addition, there are students of the similar age that differ by educational level. Whereas, groups of American undergraduates of different age levels at the same educational level, or of the same age but at different educational levels, could be assumed to differ on many selection characteristics, including academic ability, the discrepancies in age and educational level between the Israeli regular students and atuda’im can be attributed to classifiable characteristics of each groupdthat is, the regular students have all served in the army and the atuda’im have not. Whereas selection of the atuda’im is made according to academic ability in specified disciplines, it is possible to find a group that is comparable in academic ability and interest. Students at Israeli universities are accepted by departments that maintain their own entrance requirements. In the most competitive departments the regular students have high school grade point averages and entrance tests scores comparable to the atuda’im. For the sake of research, the atuda’im can be compared with regular students with the assumption that the cohorts differ by pre-university experience but not cognitive ability. The structure of university study in Israel, although similar to a number of systems in Europe, also provides a good context for investigating the influence of disciplinary studies on epistemological development. Unlike the typical American undergraduate experience studied by Perry (1970) and many other epistemological researchers who have looked at domain (see Muis et al., 2006), the
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
289
Israeli university is not a liberal arts institution. Whereas perhaps this is not optimal for promoting broad and varied perspectives on the nature of knowledge, it is an advantage for research purposes; it is reasonable to claim that one studying a major is truly immersed for several years in the thinking of that discipline with little exposure to thinking in other disciplines. 1.1. Conscription to the Israeli armed forces The army, as a hierarchical, regimented organization that restricts independence and autonomy, would not seem a likely place to look for the development of autonomous thinking. However, in Israel, where several years of military service is a fact of life for most youth of both sexes, discharged soldiers have said that the military experience promoted ‘‘a broader perspective on life and the world, a reevaluation of values and beliefs, and a more serious outlook on life’’ (Mayseless & Scharf, 2003, p. 7). This description echoes characteristics of the epistemological development of college students as described by Perry (1970) and others concerned with higher education and adult development (Baxter Magolda, 1992; Belenky, Clinchy, Goldberger, & Tarule, 1986). Notably, for many Israelis the army is their first immersion in a highly diverse social environment, just as college was the first such immersion for many of the Americans who have been the participants in research on epistemology. As Israeli military service is nearly universal for the Israeli youth, the army encompasses a broad cross-section of a multicultural society and is a defining experience shared by much of Israeli society, including university students. The uniqueness of the Israeli military experience, as one that is broadly shared and encompasses a broad cross-section of a multicultural society, affords the look at factors in epistemological development that we are exploring in this study. To attempt to compare students, who have done military service or been immersed in other socially diverse contexts, with students in other countries, who have not done military service, would be fraught with self-selection and systematic group difference problems. Such problems are minimized in this research because military service in Israel is not a choice, is universal and not selective, and includes students. Moreover, although there is no singular army experience, the range of activities of this age cohort in Israel is more classifiable than that of such cohorts elsewhere. Therefore, although the Israeli context is unique, it actually allows for greater generalizability of identifiable, common, key aspects of the army experience that might contribute to epistemological development such as the social diversity and exposure to multiple perspectives on truth and values in a social, non-academic environment. As for the universality and diversity of the Israeli Army, service is mandatory for all Jews, men and women, and all Druze and Circassian men.1 In addition, Muslim Arabs, Christian Arabs, Bedouins, and others may volunteer, so the army includes a broad representation of the population. Although service is not completely universal for Israeli youth, it is not a selective system (in which only a percentage of those in eligible areas are actually drafted) as it is in many of the decreasing number of systems of military conscription left in the world (Central Intelligence Agency, 2007; Haltiner, 1998; Jehn & Selden, 2002). Unlike other countries with systems of conscription, there is no waiver for students (Central Intelligence Agency, 2007). Thus, those who are drafted come from across social class and from a large percentage of the wide variety of subgroups that make up Israeli society. A major part of the army experience includes exposure to people from different cultural and religious backgrounds, both within the army and those involved in the ongoing regional conflicts. According to a recent report of the Israeli Defense Forces (Lior, 2007), 75% of those eligible for the draft are inducted into the army, a figure much higher than in all but a few countries in Europe in 2000 (Jehn & Selden, 2002) where, since then, a number of countries have ended the draft altogether (Central Intelligence Agency, 2007). Of the 25% of those of draft age in Israel not inducted, 11% were excused from service for religious reasons, 4% because of a criminal record, 4% because they were abroad, and the remainder for medical or mental health reasons. A negligible percentage of those of draft age refuse induction. Religious Jewish women may choose an alternative national service rather than military service, but 57% of women drafted are inducted into the military. Israelis are also in the military for a notably longer time than conscripted soldiers in other countries and more than every country in Europe. Men serve 3 years, and women serve for 2 years. The policy in Israel of conscripting women is unlike the policy of virtually any other country, including European countries (Central Intelligence Agency, 2007; Haltiner, 1998). Given that it is primarily a conscript military, there is a high cycling of inductees that requires soldiers to shift roles and locations during their service (Dar & Kimhi, 2001). Another aspect of the socially diverse experience is the chance for soldiers to encounter people while taking on different roles, learning and applying different skills, and taking on different roles of responsibility (Dar & Kimhi, 2001; Lieblich, 1989; Mayseless & Scharf, 2003). It is common that soldiers will be basically the same ages as or slightly younger than their commanders and unit leaders, who most often are also conscripts. In research with college graduates, Baxter Magolda (1998) also points to having responsibility and working with colleagues, superiors, and subordinates in the work environment as missing factors the higher education context that foster attainment of the higher reaches of epistemological development. 1 There are about 120,000 Druze in Israel (Israel Central Bureau of Statistics, 2007). The Druze are an offshoot of Islam, although generally not considered Moslems (Dana, 2003), found primarily in Lebanon, Syria, Israel, and Jordan. The Israeli Druze do not identify strongly as Arabs. They consider themselves to be loyal Israeli citizens (Amara & Schnell, 2004).There are about 3000 Circassians in Israel. Circassians (who refer to themselves as the Adyghe) are non-Arab Moslems. Their ancestors were exiled to the Ottoman Empire from the Caucasus region of Russia in the 19th century. A small group arrived in what is now northern Israel in the 1870s (Bram, 1999). Both the Druze and the Circassians have served in the Israeli military since 1948 and, at the request of their community leaders, have been subject to conscription since, respectively, 1957 (Dana, 2003) and 1958.
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
290
All Israeli teenagers are given a pre-induction evaluation (Israel Defense Forces, 2007) that includes a personal interview to evaluate potential and an aptitude test. The scores from these evaluations determine eligibility for different units and functions in the army, including whether one is qualified for the atuda’im. Those eligible for the areas of study open to atuda’im in that year, must apply to and be accepted by the department as would any other applicant. There are no separate admissions and acceptance tracks for atuda’im. After one is accepted into a university department, then one can apply to do army service in the atuda’im, final acceptance to which includes an evaluation of psychometric scores and results of a personality interview. 1.2. Research questionsdhypotheses The above unique state of affairs allows for research to address the following questions: (1) Will students at higher educational levels be at higher epistemological levels regardless of pre-university experience? That is, will both the last-year atuda’im and the last-year regular students be at higher epistemological levels than the first-year atuda’im and the first-year regular students? (2) Will students of similar age and cognitive ability, but at different educational levels be at different epistemological levels? That is, will the last-year atuda’im be at a higher epistemological level than the first-year regular students, because they have more years of education? Or might these first-year regular students be at a higher epistemological level due to their army experience despite the fact they have less academic experience? (3) Will students who are at the same educational level but have different out-of-school experiences be at different epistemological levels? That is, will the first-year regular students be at the same epistemological level as the first-year atuda’im, and will the last-year regular students be at the same epistemological level as the last-year atuda’im, when they have the same educational experience but differ only by army experience? (4) Will disciplinary experience be reflected in changes in students’ epistemological beliefs in their own discipline, or in a dissimilar discipline, with more years of study? The aim of this study was to begin teasing apart variables that cannot be controlled for experimentally. While the study is basically exploratory, we assumed that a diverse social experience does matter in epistemological development, and, thus, we hypothesized that the regular-student group will be at a higher epistemological level than the atuda’im (Hypothesis 1). We also hypothesized that those with more years of higher education will be at a higher epistemological level regardless of student group (Hypothesis 2). In addition, we hypothesized that students with more years of higher education would show changes in their epistemological beliefs in their own academic discipline but not in a dissimilar discipline (Hypothesis 3). As out-of-school experience, such as the army, that is not related to an academic discipline should have either equal or no effect on disciplinary epistemological development, it was hypothesized that no significant differences in disciplinary beliefs would be found between regular students and atuda’im (Hypothesis 4). The conceptions of knowledge derived from two diverse contextsdthe everyday social context and the formal academic contextdthat might foster epistemological development provided the opportunity to compare the results of two instruments, which have been typically used to assess beliefs in relation to those contexts. Although the two instruments have been assumed to test aspects of the same general phenomenon (Duell & Schommer-Aikins, 2001), responses to items in epistemological assessments appear to be sensitive to specific contexts and domains (Kuhn et al., 2000; Trautwein & Lu¨dtke, 2007b; Weinstock, 2007). Therefore, we decided to use two instruments with the expectation that each might be sensitive, and thus help distinguish between, the academic contexts and social context that were being investigated as potential factors in epistemological development. One of the instruments (see Kuhn et al., 2000) comes from a family of instruments that focuses on epistemological understandings concerning discrepant knowledge claims in everyday contexts. Although the questions might concern questions of history and science that might be found in schools, the focus is on making sense of competing claims without any overt reference to the school context. Thus, this instrument was chosen as appropriate to test epistemological understandings that might have arisen from the types of competing perspectives that appear in the everyday social interactions with diverse people about any number of topics. The second instrument (see Hofer, 2000) was chosen because it was designed to address epistemological beliefs in a specifically academic context and across academic disciplines. Given that the instruments evoke epistemological thinking in dissimilar contexts, we assumed that no significant relationships between their scores would be found (Hypothesis 5). 2. Method 2.1. Participants The participants consisted of 86 students of two different statuses, each having two different educational levels. There were 44 atuda’im (21 first-year and 23 last-year) and 42 regular students (21 first-year and 21 last-year) from one of the leading engineering departments in Israel. The department is in a technological institute which does not have humanities or social science
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
291
majors. All of the participants were male, as are most of the regular students and almost all atuda’im in engineering departments and at the technological institute. Although studies have found different patterns of growth and types of expression in dimensions of epistemology by gender, assignments to general epistemological level do not tend to differ (Baxter Magolda, 1992; Belenky et al., 1986; Hofer, 2000; King & Kitchener, 2002; Kuhn et al., 2000; Weinstock & Cronin, 2003). Indications of the basic comparability of the atuda’im and regular students in this study, therefore, are that each first met the very high minimum requirements for admission to the particular department in the technology institute, and each were qualified to enter the army. To be admitted to the department, they all had to have scored highly at the most difficult level of the physics and mathematics bagrut (matriculation) exams in high school, and had notably high scores on the psychometric, postsecondary entrance exam. 2.2. Materials The participants completed two short, paper-and-pencil epistemological assessments. One (Kuhn et al., 2000), coming from research on the development of everyday, informal reasoning, invokes considerations of knowledge in informal contexts without making any specific reference to educational contexts. The other (Hofer, 2000), drawn from the field of educational psychology, invokes beliefs about knowledge and knowing in the academic context. As the research questions focus on the relative influence of these contexts in epistemological understanding, both assessments were used in order to capture possible different sensitivities to the contexts depending on the amount of experience in diverse informal and academic contexts that define the groups. 2.2.1. Discrepant Claims Epistemological Assessment task Kuhn et al. (2000) developed a paper-and-pencil task to allow assignment to absolutist, multiplist, or evaluativist epistemological level. The assessment consists of 15 items, each containing 2 discrepant knowledge statements (e.g., ‘‘Danny thinks that one book’s explanation for the outbreak of World War I is correct’’; ‘‘Jacob thinks a second book’s explanation for the outbreak of World War I is correct’’). The participant is then asked to indicate (a) whether only one of the statements may be right, or whether both could be right to some degree. If indicating that both could be right, the participant proceeds to indicate whether (b) one statement could not be more right than the other, or (c) one of them could be more right than the other. An indication that one must be right and the other wrong was designated as an absolutist response. An indication that both are equally right was designated as a multiplist response. An indication that although both could be right to a degree but one could be determined to be more right was designated as evaluativist. The 15 items include 3 items each in 5 domains: taste, aesthetics, values, physical truth, and social truth. The example given above is from the domain of social truth. An example of a taste item concerned whether or not a soup was spicy. The aesthetics domain concerned issues such as which song, book, or painting was better. The values domain concerned issues such as whether it is all right to lie. An example of a physical truth item was concerned with discrepant explanations of how the brain works. The modal response to the three items in each domain was used to assign the epistemological level for that domain. For all of the participants across all of the domains, epistemological level could not be assigned in a small percentage (5%) of cases because either there was no modal level or there was missing information. The assessment was designed as a distillation of other more extensive, interview-based epistemological assessments that focus on discrepant accounts of knowledge as a means of evoking epistemological understandings (King & Kitchener, 1994; Leadbeater & Kuhn, 1989). These in turn were meant to distill the experience reported by the participants in Perry’s (1970) open-ended interviews. As explained by Perry, life in 4 years of college was characterized by confrontations with discrepant knowledge and values claims. The shorter assessment used in this study was earlier found to be effective in assigning individuals to epistemological level. When assessed with both this instrument and an extended epistemological interview, 73% of a sample of 33 college students was assigned to the same epistemological level, and 94% was assigned to either the same or an adjacent level (Kuhn et al., 2000). The present use of the instrument had low, but acceptable reliability in terms of internal consistency (Cronbach’s a ¼ 0.60). It should be noted that differences by years of schooling have been found among adolescents with the use of this assessment (Kuhn et al., 2000; Weinstock et al., 2006) and this attests to its discriminant validity. Although the assessment is designed to distinguish between knowledge domains and assignment to these three levels, because of the small number of participants of the present study there was an insufficient distribution across the levels in three of the domains to allow for statistical tests by domain. Consistent with other research using the instrument (Kuhn et al., 2000; Weinstock et al., 2006), the vast majority were multiplist in the domains of taste (91% atuda’im and 85% regular students with just one absolutist across groups) and aesthetics (84% atuda’im and 90% regular students with just one absolutist across groups). As there was so little variability in these domains, they would be uninformative regarding possible group differences, and, thus, they were not included in the statistical analyses. The other three domains were used to create cross-domain scores. The scores consisted of the ratio of domains in which one gave predominantly absolutist, multiplist, and evaluativist responses. These scores thus represented a relative tendency to think as an absolutist, multiplist, or evaluativist when taking the domains of values, physical truth, and social truth into consideration.
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
292
2.2.2. Discipline-Focused Epistemological Beliefs questionnaire This questionnaire comprises 27 items (Hofer, 2000). Participants are instructed to think about an academic discipline in answering the questions. Example items are: ‘‘If scholars try hard enough, they can find the answers to almost anything’’, ‘‘In this subject, most work has only one correct answer’’, and ‘‘If you read something in a textbook for this subject, you can be sure it’s true’’. Responses were given on a five-point Likert-type scale indicating the level of agreement with such statements. Responses ranged from 1 (representing a strong absolutist belief) to 5 (representing a strong relativist belief). Whereas one pole of the scale is absolutism, the other pole represents relativist or subjectivist responses that characterize aspects of multiplism and evaluativism without distinguishing between the two. In the present study, each participant filled out the questionnaire twice, once when instructed to answer according to the discipline of physics and the other time according to the discipline of history. Hofer (2000) followed this treatment, with different disciplines, in order to assess domain-specific beliefs. Although the intent was to perform a factor analysis to confirm theoretical epistemological dimensions (Hofer, 2000), there were too many items for the number of participants in each group to allow for such an analysis, so the average of all of the items in each use of the questionnaire was used as an indicator of epistemological orientation in each of the disciplines. The questionnaire was translated into Hebrew and back-translated into English. Both versions of the questionnaire had acceptable reliability in terms of internal reliability (for physics, Cronbach’s a ¼ 0.72; for history, Cronbach’s a ¼ 0.74). 2.2.3. Personal information questionnaire In addition to the epistemological assessment, each participant completed an information sheet that asked for personal information regarding religious observance, country of birth, parents’ education levels, and age. 2.3. Procedure The students were presented with the study and asked to participate at the end of class time. Those who chose to participate remained in the class and took 15e20 min to complete the assessments. Those who completed the assessments received a small reward, a candy bar, for their participation. Almost all of the students approached participated. 3. Results Hypotheses 1 and 2, namely that the regular students would tend to be less absolutist than the atuda’im, and that the students with more years of study would tend to be less absolutist than those with less years of study regardless of status of student, were tested with both instruments. Two-way ANOVAs were used to test the hypothesis that the epistemological assessment scores would vary according to student status and year of study. Correlations were used in order to test for relationships between the responses on the two instruments. The significance tests used an alpha level of 0.05. 3.1. Comparability between groups For the tests of the hypotheses to be meaningful, the assumed basic comparability of the atuda’im and regular students had to be established. For reasons of privacy, the department was able to provide the scores of the whole population of its students (N ¼ 288) from which the sample was drawn without identifying the students who participated in the study. Table 1 shows the psychometric and bagrut scores as well as the grade point average by group. The t-tests showed no significant differences between atuda’im and regular students in the department in their psychometric scores, t(284) ¼ 1.21, p ¼ 0.23. In the latest report on the distribution (National Institute for Testing and Evaluation, n.d.), both means would fall in the top 5% of all, and three-quarters of the students in each group (atuda’im with 681 and above, and regular students with 682 and above) would be in the top 10% of scores. Table 1 Mean psychometric and bagrut scores and grades by status of student. Scores
Student status Atuda’im (n ¼ 67)
Psychometric Bagrut Grades
Regular (n ¼ 221)
M
SD
Range
M
SD
Range
706.57 93.28 85.50
42.18 3.06 6.67
580e780 83.40e97.80 71.30e97.90
702.33 89.55 84.37
38.50 5.18 5.22
518e780 69.00e98.50 73.90e97.00
The maximum possible psychometric score is 800. The maximum possible bagrut or grade average is 100.
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
293
A significant Levene’s test for equality of variances indicated that the variances in the two groups’ bagrut scores were not equal, F ¼ 9.51, p ¼ 0.002. The test for the equality of the means of the groups’ bagrut scores reflects the modification of the t-test that takes into account the violation of the assumption of equal variances. A significant difference was found between the groups in their bagrut scores, t(195.67) ¼ 7.08, p ¼ 0.0001, Cohen’s d ¼ 0.88,2 with the atuda’im having a higher mean score. However, the means for both the atuda’im and regular groups are much higher than the average (Lavy, 2004) and enabled them to be admitted into a very selective program, in which, according to their university grades, both groups were performing equally well in their university studies. As shown in Table 1, there was no significant difference in their grade point averages, t(91.79) ¼ 1.27, p ¼ 0.21. Again, the t-test reflects a significant Levene’s test indicating that equal variances could not be assumed, F ¼ 8.66, p ¼ 0.004. Neither student status nor year of study was found to be associated with the personal variables of religious observance, country of birth, and parents’ education levels. These tests of the academic and social characteristics of the participants indicate that the student status groups were essentially comparable across a range of variables. 3.2. Responses to the epistemological assessments 3.2.1. Discrepant Claims Epistemological Assessment task Table 2 shows the mean ratios of domains in which students were at each epistemological level according to their status, and Table 3 shows the mean ratio of domains in which students were at each epistemological level according to year in studies. We used 2(student status) 2(year in studies) ANOVAs to test for main and interaction effects that might explain the variance in tendency toward absolutism, multiplism, or evaluativism. There was a main effect of student status, F(1, 82) ¼ 6.20, p ¼ 0.15, partial h2 ¼ 0.07. There was neither any main effect of year in studies, F(1, 82) ¼ 0.13, ns, nor interaction of student status with year in studies, F(1, 82) ¼ 0.12, ns. However, it should be noted that Levene’s test of homogeneity for the ANOVA was significant, F(3, 82) ¼ 5.97, p ¼ 0.001, indicating unequal variances. In particular with a small sample, this violation of the assumptions of the ANOVA might make the test of the model inaccurate (Ananda & Weerahandi, 1997; Bathke, 2004). As Levene’s tests with each student status and year of studies showed the unequal variances only with student status, F ¼ 15.61, p ¼ 0.0001, a t-test was also used to test for differences in the tendency toward absolutism according to student status as regular students or atuda’im. The appropriate modified t-test was used and the difference by student status was confirmed, t(72.44) ¼ 2.54, p ¼ 0.013, Cohen’s d ¼ 0.55.3 The atuda’im tended to have higher percentages of domains in which they were absolutist (M ¼ 0.17, SD ¼ 0.24) than the regular students (M ¼ 0.06, SD ¼ 0.15). As indicated by the ratios shown in Table 2, the tendency of the atuda’im toward absolutism was relatively higher to the regular students, but neither group of students was particularly absolutist. The ANOVAs testing differences in the percentages of domains in which the participants were either multiplist or evaluativist were nonsignificant and there were no main effects or interactions between the variables of the study. 3.2.2. Discipline-Focused Epistemological Beliefs questionnaire Hypothesis 3 was that the students would differ by year in studies in their epistemological beliefs in their own field, as measured by the physics-focused questionnaire, but not in a dissimilar field, as measured by the history-focused questionnaire. Hypothesis 4 was that no differences in disciplinary epistemological beliefs would be found according to student status. Table 4 shows the means of the responses by student status and year in studies. The 2(student status) 2(year in studies) ANOVA showed a main effect of the year in studies on the physics-focused epistemological beliefs, F(1, 82) ¼ 4.99, p ¼ 0.028, partial h2 ¼ 0.06, although, as indicated by the effect size, this did not account for a lot of the variance. Nonetheless, the last-year students appeared to be more relativistic than the first-year students. However, there was no main effect of student status, F(1, 82) ¼ 0.02, ns, nor an interaction between student status and year in studies, F(1, 82) ¼ 0.93, ns. As regards the history-focused epistemological beliefs, the 2(student status) 2(year in studies) ANOVA showed no main effects of either student status, F(1, 82) ¼ 1.12, ns, or of year in studies, F(1, 82) ¼ 1.84, ns. Also, there was no interaction between them, F(1, 82) ¼ 0.10, ns. No significant correlation was found between physics-focused and history-focused epistemological beliefs (Pearson’s r ¼ 0.10). Nevertheless, a paired-sample t-test showed that there were no significant differences in the mean scores of the physics-focused questionnaire (M ¼ 3.36, SD ¼ 0.31) and the history-focused questionnaire (M ¼ 3.44, SD ¼ 0.36), t(83) ¼ 1.54, ns. 3.3. Relations between epistemological assessment scores It was predicted (Hypothesis 5) that no relation would be found between the Discrepant Claims Epistemological Assessment task and the Discipline-Focused Epistemological Beliefs questionnaires. Of course, what was actually tested in the statistical analysis is that there was a relationship between the two. The predicted lack of significant relation does not confirm that there 2 3
0.80 is considered a large effect size (Cohen, 1988). 0.50 is considered a medium effect size (Cohen, 1988).
294
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
Table 2 Mean ratio of domains in which participants were at each epistemological level by student status. Epistemological level
Student status Atuda’im
Absolutist Multiplist Evaluativist
Regular
M
SD
M
SD
0.17 0.31 0.51
0.24 0.32 0.34
0.06 0.36 0.57
0.15 0.35 0.35
Table 3 Mean ratio of domains in which participants were at each epistemological level by year in studies. Epistemological level
Year in studies First
Absolutist Multiplist Evaluativist
Last
M
SD
M
SD
0.13 0.36 0.52
0.23 0.35 0.37
0.11 0.32 0.57
0.19 0.33 0.33
Table 4 Means of responses to the Discipline-Focused Epistemological Beliefs questionnaire by student status and year in studies. Group
Discipline-focused epistemological beliefs Physics
History
M
SD
M
SD
Student status Atuda’im Regular
3.38 3.37
0.30 0.36
3.40 3.48
0.37 0.36
Year in studies First Last
3.29 3.45
0.31 0.33
3.38 3.49
0.40 0.32
would be no relation, but would demonstrate lack of support for the presumption in the literature that the questionnaires are indeed assessing the same construct. The following presents several comparisons between the two measures. Hypotheses 2 and 3 both suggested that higher epistemological levels would appear with more years of study. This did not turn out to be the case when assessed with the Discrepant Claims Epistemological Assessment task. This empirical finding, as well as the rationale stated for Hypothesis 5, namely that no relation would be found between the measures, raises the question of whether there appears to be any relations between epistemological understandings about knowledge claims outside of a clearly defined academic context and the measures of epistemological beliefs specifically concerning academic disciplines. Pearson’s correlations were used to examine if the scores of either the physics-focused or the History-Focused Epistemological Beliefs questionnaires were correlated with the ratios of domain-specific responses at each epistemological level derived from the Discrepant Claims Epistemological Assessment task. No correlations were found. As the Discrepant Claims Epistemological Assessment task was designed to distinguish between domains, ideally it could produce domain-specific data parallel to that of the Discipline-Focused Epistemological Beliefs questionnaire. However, because there were too few absolutists in the domain of social truth (including history) as measured in the Discrepant Claims Epistemological Assessment task, only the domain of physical truth assessed by this task could be used to explore the relationship between the responses to the two epistemological instruments. The two instruments are not exactly parallel in that the Discipline-Focused Epistemological Beliefs questionnaire does not make a distinction between multiplism and evaluativism. The contrast with absolutism in the poles of the items is relativist beliefs that focus on the subjectivist understandings common to multiplism and evaluativism. In the five-point scale, it is not assumed that one pole is absolutist, a middle position is multiplist, and the second pole is evaluativist.4 Thus, in order to make an interpretable statistical comparison between the two
4 In fact, it has been argued (Muis et al., 2006; Weinstock, 2007) that with some of the items evaluativism would represent a less extreme response than multiplism.
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
295
instruments, those initially assigned to the multiplist and evaluativists levels with the Discrepant Claims Epistemological Assessment task were combined into a single relativist level. An independent samples t-test, in which the mean scores on the Physics-Focused Epistemological Beliefs questionnaire of those assigned as absolutists on the Discrepant Claims Epistemological Assessment task (M ¼ 3.26, SD ¼ 0.31) were compared with the mean scores of the relativists (M ¼ 3.41, SD ¼ 0.33). No significant difference was found, t(79) ¼ 1.59, ns. As differences in the Physics-Focused Epistemological Beliefs questionnaire were found between students in different year in studies (as reported in Section 3.2.2), associations between the levels in the domain of physical truth as determined by the Discrepant Claims Epistemological Assessment task were also tested. Maintaining the comparison between the absolutist and a relativist level, a Fisher’s Exact test did not find a significant association between year in studies and epistemological level. That is, neither the first- nor last-year students were more likely to be either absolutist or relativist. However, a Fisher’s Exact test ( p ¼ 0.035) indicated an association between student status and epistemological level, that was not found using the PhysicsFocused Epistemological Beliefs questionnaire, such that atuda’im were more likely to be absolutist in the domain of physical truth than the regular students. Twenty-nine percent of the atuda’im was absolutist compared with 10% of the regular students. 4. Discussion The results with each instrument confirmed parts of the hypotheses. The analysis on the scores of the Discrepant Claims Epistemological Assessment task, referring not specifically to academic knowledge contexts, showed that group status, but not the year in studies, was associated with epistemological level, thus supporting Hypothesis 1 but not Hypothesis 2. In the DisciplineFocused Epistemological Beliefs questionnaires that assessed beliefs in a specifically academic context, the year in studies, but not group status, was related to epistemological level, but only in the case of physics, a discipline included in their studies. This supported Hypothesis 3. That the instruments produced partly different results was somewhat unexpected in that there was a difference by year in studies only with the physics-focused epistemological beliefs but not, as assumed in Hypothesis 2, with the assessment of epistemological levels using the Discrepant Claims Epistemological Assessment task. Much prior epistemological research has shown general epistemological development with years of study (Baxter Magolda, 1992; King & Kitchener, 1994; Kuhn & Weinstock, 2002; Perry, 1970). However, the divergent, but complementary findings actually serve to underscore the main assumption behind the study: experience in social and academic contexts may make distinct contributions to epistemological development. Social and academic experience might involve different levels of engagement with particular dimensions of epistemological understanding. These results suggest that a key difference between the groups, namely army experience, may play a role in epistemological development, at least concerning knowledge in everyday, not explicitly academic contexts. This suggests that diverse social experience, a greater part of the background of regular students than of the atuda’im, would be a factor to investigate as a possible source of epistemological differences. That year in studies was found to be related to epistemological beliefs in physics but not history (Hypothesis 4) suggests that academic engagement in a discipline is related to epistemological beliefs about that discipline and possibly not to beliefs about other disciplines not studied. This finding is consistent with a number of studies investigating the domain-specificity of epistemological beliefs (see Muis et al., 2006). However, the finding also suggests that epistemological thinking in an academic area may not carry over when considering knowledge claims outside the academic context. The year in studies was related only with epistemological beliefs within the students’ domain of study, and was not related to their responses to the Discrepant Claims Epistemological Assessment task in non-academic contexts. Conversely, those with the greater experience with social diversity (namely the regular students) tended to be less absolutist in their responses to discrepant knowledge claims concerning the physical sciences when not presented in an academic context, whereas no differences were found between the regular students and the atuda’im when their epistemological beliefs were assessed with the overtly academically oriented Physics-Focused Epistemological Beliefs questionnaire. An implication of these findings is that academic experience alone may not produce personal epistemological understandings that concern everyday problems. This has implications for how different types of educational contexts with greater or lesser inclusion of social experience might impact on everyday reasoning. The basis for the hypotheses was the assumption that a factor underlying the findings of previous research showing the epistemological development of college students was that the college experience provides the type of diverse environment, socially as well as academically, that fosters epistemological development. In Israel, the army, rather than college, serves as the first immersion for many high school graduates in an environment with a wide variety of people from different backgrounds. Moreover, the diverse academic experience that characterizes the liberal arts studies of many college students was greatly restricted among the sample of Israeli students. Like all other Israeli universities, most of these students’ courses were in just one or two disciplines. The experience of the sample contrasts sharply with direct and indirect exposure to a range of knowledge domains typical of students in the liberal arts curricula of American colleges. This might explain why there were differences by year in studies only with the Discipline-Focused Epistemological Beliefs questionnaire focusing on physics and not history. Furthermore, as the students were in an applied, scientific field, they may not have focused on problems in knowing (Donald, 1995; Donnelly, 1999; Paulsen & Wells, 1998), such as uncertain, discrepant claims of ill-structured problems that are more salient in history and more the focus of the Discrepant Claims Epistemological Assessment task.
296
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
Taking both types of measures together, this study suggests that particular aspects of experience in an academic environment, such as the type of discipline and cross-discipline interactions, rather than schooling in general might account for the earlier found relationships between educational and epistemological level. But even more, it suggests that experience in a socially diverse environment may be a significant contributor to epistemological development. The paucity of academic diversity and social diversity in the academic context in this study highlights the possible role of diverse social experience in explaining the group differences found in this study. Future research could test for interactions between diverse social and academic experience by using samples of regular students and atuda’im who study at a general university rather than a technological institute, thus increasing the possibility of social interactions between sexes and people studying a wider variety of the academic disciplines. The results also point to an issue of assessment of epistemological thinking or beliefs. The two instruments used in the present study were chosen as state-of-the-art instruments that come from two different approaches to personal epistemology. This treatment would have been consistent with the literature (see Duell & Schommer-Aikins, 2001) if, by using each one, the results had supported each other. Although the explanation of the divergent results with each instrument (Hypothesis 5), namely that the instruments measure epistemological thinking in different contexts, is consistent with the assumptions of the study and reasons for having chosen to use measures from different approaches, it is also possible that the responses diverged because the instruments measure different epistemological dimensions, or different epistemological constructs altogether. In addition, the instruments differ in how they operationalize the constructs that they do measure. The Discrepant Claims Epistemological Assessment task engages participants in problems in order to evoke underlying epistemological understanding with responses distributed into three categories. The Discipline-Focused Epistemological Beliefs questionnaire, on the other hand, asks for a self-report of beliefs. The responses are distributed on a spectrum of na€ıve (absolutist) to sophisticated (relativist) beliefs. The results from the instruments in the current study do not cancel each other out and each provides support for a different hypothesis. But as these two types of epistemological instruments have not been compared previously, that there is a lack of consistency between them underscores the need for epistemology researchers to carefully examine the assumptions of epistemological assessments and constructs. 4.1. Limitations The present study had a small sample, thus limiting the power of the results. The numbers of available atuda’im are limited, but in future research more might be recruited and more than one university might be tapped. This would have the added advantage of enabling the comparison of atuda’im in a technological institute where there are only pure and applied science majors, with atuda’im in a more diverse educational context where they might have more contact with a more ethnically diverse and genderbalanced student body studying a wider range of knowledge disciplines. Another problem with the small sample is that it was impossible to test for non-academic experiences other than the army that might be related with epistemological development. For instance, some students may have worked or traveled in between high school and army service or between army service and the start of their university studies. Although we expect that the army service would be the most significant non-academic factor, a large sample would allow better control of other possibilities. The small sample also meant that scoring of the measures had to be modified, thus losing information. In particular, the discipline-focused epistemological beliefs variables may have adequately represented a general epistemological orientation, but the theory behind the questionnaire is that there are distinct dimensions that could be measured (Hofer, 2000). It is possible that the differences found by year in studies on the physics-focused questionnaire did not include all of the four proposed dimensions of the certainty, simplicity, source, and justification of knowledge. It is also possible that there were differences in one or more of these dimensions on the history-focused questionnaire, but that differences were diluted in the global score that was used. Aside from the methodological limitation, it would be of great theoretical interest to see to what degree the development of particular epistemological dimensions are influenced by disciplinary experience, and whether the same or different dimensions develop depending on the domain. Another possible limitation is group differences. Although a best effort was made to ensure that the primary difference between the groups would be army service by matching the groups by academic ability, institutional context, and discipline, and by testing and ruling out religious observance, country of origin, and parents’ educational level, there might be personality differences between those who would enter the army before or after their studies. Perry (1970) found that authoritarian personality had no bearing on the epistemological development of college students. Nevertheless, as the participants in this study were chosen according to group membership rather than by random selection, it would be prudent to control for various personality types in future research. It should be emphasized that several possible group differences were controlled because both groups were selected by their university department because of comparably high promise and interest in the same field. Another difference could be age. But age is indistinguishable from the groups of interest. The regular students are older than the atuda’im, and the last-year students within each student group are older than the first-year students. We assume that age represents a type of experience accumulated over the years, and thus emphasize this aspect of the differences between the groups. Studies involving people of various educational and occupational backgrounds of all ages (Kuhn et al., 2000; Leadbeater & Kuhn, 1989; Weinstock & Cronin, 2003) have found little relationship between age and epistemological development. The studies that do show
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
297
increases with age have samples with very common experience (e.g., King & Kitchener, 1994; Perry, 1970) suggesting that it is the type of specific experience, such as education, and not just the increase of age that is the relevant factor. However, to rule out simple chronological age effects in relation to the issues posited in the present study, it would be worthwhile to compare matched age groups with different experiences, such as first- and last-year atuda’im and same age soldiers. If the groups were at the same epistemological development between the two ages, then age, rather than specific experience would be implicated. A final limitation on the research is that, although it suggests a possible causal relationship, it cannot claim one. A longitudinal study that would track these two groups, as well as a group of academically comparable soldiers before they entered the university as regular students would help make bolster such claims. However, because random assignment to groups would be impossible, even results from a longitudinal study could only be suggestive. It should be noted that a comparison between the atuda’im and same-aged soldiers would also help control for age. 4.2. Conclusions In the present study, the samples in this research and their academic contexts were closely matcheddprobably more closely matched than college students with different majors and than college students with same-aged adults not in college. Nevertheless, a non-academic life experience (army service) common to one group but not the other, was related with significant differences in how students in the same university department understood the nature of knowledge and knowing. Within an academic environment, the quality of social and academic diversity might vary within a college depending on major, living arrangements, and involvement in activities having implications for the epistemological development of the students. The study is meant primarily to address theoretical issues. Thus, we have highlighted its implications for considering the sources of epistemological development in both social and academic experience, and how these types of experience may have different relationships with different aspects of epistemological understanding. Because of the divergent findings of the two instruments, we hope that it also prompts reevaluation about the meaning of the instruments and the constructs they are purported to measure. However, the study also may have practical implications, which we make cautiously because of the size of the sample. We do not suggest that countries adopt mandatory 2- to 3-year military service, nor do we hope that they ever would have the need to do so. However, the social experience of students either before or during their undergraduate years might be enhanced to allow for exposure to social diversity and the broadening of perspectives outside of the academic realm. The present research was designed to be the first step in a research project that more carefully controls for possible explanations of the differences found between the groups. The constructs of academic and socially diverse experience are broadly defined. Nevertheless, this is an advance on much epistemological research that has suggested that years of schooling or postsecondary education are important factors in epistemological development without further specifying what occurs in schooling that might be relevant. The present study makes a first, small step in teasing apart the assumed social and academic contribution to epistemological development. More than providing definitive conclusions, the findings suggest the value of continuing this effort. References Amara, M., & Schnell, I. (2004). Identity repertoires among Arabs in Israel. Journal of Ethnic and Migration Studies, 30, 175e193. Ananda, M. M. A., & Weerahandi, S. (1997). Two-way ANOVA with unequal cell frequencies and unequal variances. Statistica Sinica, 7, 631e646. Bathke, A. (2004). The ANOVA F test can still be used in some balanced designs with unequal variances and nonnormal data. Journal of Statistical Planning and Inference, 126, 413e422. Baxter Magolda, M. B. (1992). Knowing and reasoning in college: Gender-related patterns in students’ intellectual development. San Francisco: JosseyeBass. Baxter Magolda, M. B. (1998). Developing self-authorship in young adult life. Journal of College Student Development, 39, 143e156. Belenky, M. F., Clinchy, B. M., Goldberger, N. R., & Tarule, J. M. (1986). Women’s ways of knowing: The development of self, voice and mind. New York: Basic Books. Biglan, A. (1973). Relationships between subject-matter characteristics and the structure and output of university departments. Journal of Applied Psychology, 57, 204e213. Bram, C. (1999). Circassian re-immigration to the Caucasus. In S. Weil (Ed.), Routes and roots: Emigration in a global perspective (pp. 205e222). Jerusalem: Magnes. Central Intelligence Agency. (2007). The world factbook. Retrieved February 5, 2008 from. https://www.cia.gov/library/publications/the-world-factbook/index.html. Chandler, M., Boyes, M., & Ball, L. (1990). Relativism and stations of epistemic doubt. Journal of Experimental Child Psychology, 50, 370e395. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum. Dana, N. (2003). The Druze in the Middle East: Their faith, leadership, identity and status. Brighton, UK: Sussex Academic Press. Dar, Y., & Kimhi, S. (2001). Military service and self-perceived maturation among Israeli youth. Journal of Youth and Adolescence, 30, 427e448. Donald, J. G. (1995). Disciplinary differences in knowledge validation. New Directions for Teaching and Learning, 64, 7e17. Donnelly, J. (1999). Interpreting differences: the educational aims of teachers of science and history, and their implications. Journal of Curriculum Studes, 31, 17e 41. Duell, O. K., & Schommer-Aikins, M. (2001). Measures of people’s beliefs about knowledge and learning. Educational Psychology Review, 13, 419e449. Haltiner, K. W. (1998). The definite end of the mass army in Western Europe? Armed Forces and Society, 25, 7e36. Hofer, B. K. (2000). Dimensionality and disciplinary differences in personal epistemology. Contemporary Educational Psychology, 25, 378e405. Hofer, B. K., & Pintrich, P. R. (1997). The development of epistemological theories: beliefs about knowledge and knowing and their relation to learning. Review of Educational Research, 67, 88e140.
298
M. Weinstock, H. Zviling-Beiser / Learning and Instruction 19 (2009) 287e298
Israel Central Bureau of Statistics. (2007, April). Ha’achlusiah ha’Druzit b’Yisrael [The Druze Population of Israel]. Jerusalem: Author. Israel Defense Forces. (2007). Ha’atuda ha’academit: Maydon lashanat ha’limudim tashsach. [The student soldier: Information booklet for academic year 2007e 2008]. Tel-Aviv: Author. Jehn, C., & Selden, Z. (2002). The end of conscription in Europe? Contemporary Economic Policy, 20, 93e100. Kaartinen-Koutaniemi, M., & Lindblom-Yla¨nne, S. (2008). Personal epistemology of psychology, theology and pharmacy students: a comparative study. Studies in Higher Education, 33, 179e191. Kardash, C. M., & Scholes, R. J. (1996). Effects of preexisting beliefs, epistemological beliefs, and need for cognition on interpretation of controversial issues. Journal of Educational Psychology, 88, 260e271. King, P. M., & Kitchener, K. S. (1994). Developing reflective judgment: Understanding and promoting intellectual growth and critical thinking in adolescents and adults. San Francisco: Jossey-Bass. King, P. M., & Kitchener, K. S. (2002). The Reflective Judgment Model: twenty years of research on epistemic cognition. In B. K. Hofer, & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 37e61). Mahwah, NJ: Erlbaum. King, P. M., Kitchener, K. S., Davison, M. L., Parker, C., & Wood, P. K. (1983). The justification of beliefs in young adults: a longitudinal study. Human Development, 26, 106e116. King, P. M., Wood, P. K., & Mines, R. A. (1990). Critical thinking among college and graduate students. The Review of Higher Education, 13, 167e186. Kuhn, D., Cheney, R., & Weinstock, M. (2000). The development of epistemological understanding. Cognitive Development, 15, 309e328. Kuhn, D., & Weinstock, M. (2002). What is epistemological thinking and why does it matter? In B. K. Hofer, & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 121e144) Mahwah, NJ: Erlbaum. Lavy, V. (2004). Do gender stereotypes reduce girls’ human capital outcomes? Evidence from a natural experiment. (Working Paper No. 10678). Cambridge, MA: National Bureau of Economic Research. Leadbeater, B., & Kuhn, D. (1989). Interpreting discrepant narratives: hermeneutics and adult cognition. In J. Sinnott (Ed.), Everyday problem solving (pp. 175e 190). New York: Praeger. Lehman, D. R., Lempert, R. O., & Nisbett, R. E. (1988). The effects of graduate training on reasoning: formal discipline and thinking about everyday life events. American Psychologist, 43, 431e443. Lieblich, A. (1989). Transition to adulthood during military service. Albany, NY: State University of New York Press. Lior, R. (2007, July 17). Reva m’bnai noar ainam mitgayasim. [A quarter of the youth are not drafted]. Channel 10 News. Retrieved February 5, 2008 from. http://news. nana10.co.il/Article/?ArticleID¼501271&;TypeID¼1&sid¼126. Lonka, K., & Lindblom-Yla¨nne, S. (1996). Epistemologies, conceptions of learning, and study practices in medicine and psychology. Higher Education, 31, 5e24. Mason, L., & Boscolo, P. (2004). Role of epistemological understanding and interest in interpreting a controversy and in topic-specific belief change. Contemporary Educational Psychology, 29, 103e128. Mason, L., & Scirica, F. (2006). Prediction of students’ argumentation skills about controversial topics by epistemological understanding. Learning and Instruction, 16, 492e509. Mayseless, O., & Scharf, M. (2003). What does it mean to be an adult? The Israeli experience. New Directions for Child and Adolescent Development, 100, 5e20. Muis, K. R., Bendixen, L. D., & Haerle, F. C. (2006). Domain-generality and domain-specificity in personal epistemology research: philosophical and empirical questions in the development of a theoretical model. Educational Psychology Review, 18, 3e54. National Institute for Testing and Evaluation (n.d.). Interpreting scores on the psychometric. Test. Retrieved February 5, 2008 from https://admin.nite.org.il/ gradeFiles/Guides/English_guide.pdf Paulsen, M. B., & Wells, C. T. (1998). Domain differences in the epistemological beliefs of college students. Research in Higher Education, 39, 365e384. Perry, W. (1970). Forms of intellectual and ethical development in the college years. New York: Holt. Ryan, M. P. (1984). Monitoring text comprehension: individual differences in epistemological standards. Journal of Educational Psychology, 76, 248e258. Schommer, M., Calvert, C., Gariglietti, G., & Bajaj, A. (1997). The development of epistemological beliefs among secondary students: a longitudinal study. Journal of Educational Psychology, 89, 37e40. Stanovich, K. E. (1999). Who is rational? Studies of individual differences in reasoning. Mahwah, NJ: Erlbaum. Tabak, I., & Weinstock, M. P. (2005). Knowledge is knowledge is knowledge? The relationship between personal and scientific epistemologies. Canadian Journal of Science, Mathematics and Technology Education, 5, 305e328. Trautwein, U., & Lu¨dtke, O. (2007). Epistemological beliefs, school achievement, and college major: a large-scale longitudinal study on the impact of certainty beliefs. Contemporary Educational Psychology, 32, 348e366. Trautwein, U., & Lu¨dtke, O. (2007). Predicting global and topic-specific certainty beliefs: domain-specificity and the role of the academic environment. British Journal of Educational Psychology, 77, 907e934. Weinstock, M. (2007, August). Content-rich and content-poor measures of epistemological understanding. In J. Elen (Chair). Assessing epistemological beliefs. Symposium conducted at the 12th Biennial Conference for Research on Learning and Instruction, Budapest, Hungary. Weinstock, M., & Cronin, M. A. (2003). The everyday production of knowledge: individual differences in epistemological understanding and juror reasoning skill. Applied Cognitive Psychology, 17, 161e181. Weinstock, M. P., Neuman, Y., & Glassner, A. (2006). Identification of informal reasoning fallacies as a function of epistemological level, grade level, and cognitive ability. Journal of Educational Psychology, 98, 327e341. Wineburg, S. S. (1991). Historical problem solving: a study of the cognitive processes used in the evaluation of documentary and pictorial evidence. Journal of Educational Psychology, 83, 73e87.
