Book · January 994 citations 110 reads 2,264 authors
particular, and will attempt to show that empirical evidence in support of
Download 5.72 Mb. Pdf ko'rish
|
1994 Book DidacticsOfMathematicsAsAScien
particular, and will attempt to show that empirical evidence in support of these hypotheses is totally lacking or indirect and unconvincing. I will ar- gue, furthermore, that such claims may work against the increased participa- tion of women in the mathematical sciences. In a subsequent section, I will then discuss the empirical evidence for the existence of gender-related differences in mathematics achievement, with the overall conclusion that at the primary and secondary levels these differ- ences are small, where they exist at all, and are getting smaller. My thesis, then, in light of the evidence I examine, is that proposals for differential didactics by gender represent an unwarranted and potentially harmful initiative for the solution of a problem that in any case is disappear- ing at the school level. 2. HYPOTHESES ON DIFFERENTIAL LEARNING I first discuss two hypotheses arising directly from research in mathematics education, and then examine some hypotheses about gender differences in 304 knowledge acquisition in general that stem from the work of Gilligan (1982) and that of Belenky, Clinchy, Golderberg, and Tarule (1986). 2.1 Rote Versus Autonomous Learning Kimball (1989) examined some 150 studies on gender-related differences in mathematics achievement and noted a marked contrast between differences in classroom grades and differences in the results of standardized tests. Girls were consistently reported to perform better on classroom tests than on standardized tests. Furthermore, girls, on average, were reported to outper- form boys on classroom tests, but to underperform them on standardized tests. It is very problematic to compare the size of such differences, because standardized achievement tests and classroom tests differ in their psychome- tric properties, but these results have nevertheless given rise to hypotheses relating either to learning styles or to the possible inherent bias against fe- males in some standardized tests. To explain why gender differences in performance on standardized tests are more pronounced than those on classroom tests, perhaps the most signif- icant learning-style explanation is that of "rote versus autonomous learning" put forward by Fennema and Peterson (1985). Their hypothesis is that girls have an advantage in classroom examinations, because they tend to take a rule-following and rote-learning approach, whereas boys get higher grades on external standardized tests (and eventually outstrip girls in mathematical understanding), because they have a more autonomous approach to learning. Because the concepts of "autonomous" and "rote" learning have not been operationalized, it is not possible to observe such behaviors directly. Kimball (1989) evaluated this hypothesis, however, by examining over 30 published research studies that had investigated variables that could be con- sidered to be related to an autonomous or a rote style of learning. If girls en- gage less often than boys in extracurricular activities related to mathematics, for example, they might tend to rely on rote learning and memorization. Similarly, if boys display more rebellious attitudes, they might well be more autonomous learners. Kimball also examined evidence on the existence of links between these two presumed learning styles and performance on both classroom and standardized tests. In assessing this hypothesis, Kimball concluded that there would be a need for more evidence "before we can evaluate its potential to explain sex- related differences in classroom and standardized achievement measures" (p. 206). To date, there is still no convincing evidence that girls adhere to a style of learning that can be branded rote as opposed to autonomous, nor any evidence that either of these presumed learning styles might be directly linked to achievement in mathematics. GILA HANNA 305 2.2 Novelty Versus Familiarity Another hypothesis offered to explain gender differences in mathematics achievement is that of "novelty versus familiarity," which suggests that boys are confident and motivated to do well when confronted with new and challenging tasks such as those met in mathematical studies, whereas girls are less confident and often feel lost in such situations. This hypothesis, too, is prompted by the observation that girls tend to do well on classroom tests that cover material explicitly taught by the teacher and thus familiar to the students, but not as well on standardized tests, which may present unfamil- iar content. The extensive study mentioned earlier led Kimball (1989) to the addi- tional conclusions that (a) girls are not disadvantaged when the tests reflect classroom content, (b) girls do display avoidance and lack of confidence in the presence of novel tasks, and (c) girls have been observed to be less ac- tive in classrooms that emphasize competition (which is likely to require an- swers to novel problems). She thus concluded that there is some empirical evidence to support the hypothesis of novelty versus familiarity. Kimball's conclusion is guarded, however: "Although there is some theoretical and empirical support for this hypothesis, more empirical evidence is necessary to evaluate it" (p. 208). This hypothesis is challenged at a very basic level, however, by the results of a more recent study. According to Felson and Trudeau (1991), the hypothesis of novelty versus familiarity is not a useful one, because gender differences appear to depend upon the mathematical topic covered by the tests. Girls have been shown to outperform boys on certain standardized mathematics tests, indicating that the novelty of the task in itself is not a barrier to their achievement. 2.3 Separate Versus Connected Knowing The hypothesis of "separate versus connected knowing" takes its inspiration and its wording from the work of Carol Gilligan (1982) on moral choices. In her book, In a Different Voice, she presents a model of moral development in which women are contextually focused, relationship-oriented, responsi- ble, and caregiving, whereas men are abstract, individualistic, and dominat- ing. Gilligan suggests that women have a special edge on moral issues, in that they are prepared to make moral judgments on the basis of a concern for relatedness among people rather than on the basis of abstract principles of justice. From her analysis she also concludes that women have a "con- nected" way of grasping situations, whereas that of men is "separate." Though it is not immediately clear what implications this distinction might have for knowledge acquisition, it has been embraced, seemingly rather uncritically, by several researchers and mathematics educators. Brown (1984), for example, found Gilligan's analysis relevant to mathemat- ics education and turned "towards a consideration, in a rather global way, of GENDER AND INSTRUCTION 306 GILA HANNA how it is that a Gilliganish perspective on morality might impinge on the study of mathematics" (p. 12). From the idea that women, with their "different voice," learn best when connections are made explicit, others have gone on to suggest a pedagogy for mathematics education specific to women. In their view, the successful teaching of females would necessitate, among other steps, injecting the mathematics curriculum with a high degree of context, emphasizing con- nectedness among concepts and situations, and de-emphasizing the teaching of general principles (Buerk, 1985; Lee, 1989). Instructional strategies along these lines may well turn out to be effective for both men and women. But there would seem to be little basis for the thesis of "connected knowing" advanced in their support. Those educators who so quickly attempted to translate this thesis into a rationale for differen- tial didactics appear to have ignored the controversy engendered by Gilligan's analysis as a whole. What has come to be called the "Gilligan de- bate" consists of over 100 articles published during the period 1982 to 1992 in books and learned journals. As reported by Davis (1992), this debate among psychologists, sociologists, and human rights theorists, most of whom are self-declared feminists, has subjected Gilligan's model of moral development to extensive and often devastating criticism. Some of Gilligan's critics have pointed out serious flaws in her methodol- ogy, questioning, for example, the validity of findings based only upon a sample composed primarily of Harvard students. Others have questioned the very possibility of a specifically female morality based upon care. Yet other critics have found the interpretation of her results to be wanting. It has been pointed out, for example, that the "ethic of care," which Gilligan observed in her female subjects, and which she interpreted as a sign of moral maturity and a source of strength specific to women, could quite validly be inter- preted as a normal reaction by any oppressed group, male or female, to their oppression. Some critics have also contended that the indecisiveness in moral issues that Gilligan observed in her female subjects is not necessarily an indication of superior moral standards. It could well be seen as an sign of subordina- tion, and one not necessarily associated with gender, but perhaps with social position. As Davis (1992) reports, many critics have made the further point that sociological perspectives are more useful than psychological ones as explanations for perceived gender differences in moral development. Gilligan's work may be important as a critique of sexist bias in the litera- ture on human development. Because her study was of very modest propor- tions and her data open to alternative interpretations, however, her female- male dichotomy remains far from convincing. In any case, it would seem reckless to make uncritical application to the teaching of mathematics, as some appear to have done, of her contention that women grasp situations in distinctive ways. 307 2.4 Women's Ways of Knowing This hypothesis for gender differences in cognition is known by the title of the book Women's Ways of Knowing by Belenky et al. (1986), who contend that women have distinct ways of knowing that have arisen from their life experiences as women. The authors also criticize existing theories of cogni- tive development, mainly for their reliance on studies that observed male students only. In their own research, they do make use of the epistemologi- cal work of Perry (1970), however, claiming to have fleshed out those areas that they believe to be sketchy or missing in his model of intellectual devel- opment. The book reports on a thorough study of 135 women from various social groups. All the data, in the form of interview notes and responses to a series of open-ended questions, deal with the role of gender in the self-per- ception of these women, in their vision of their own future development, in their formal learning experiences, and in their ways of making moral judg- ments. A major conclusion of the authors is that the women they studied felt voiceless and alienated in the course of their education, an education that the authors believe to have been tailored to the needs of men, and that, by implication, this feeling is common to women in general. The authors further conclude that women have a distinctive way of ac- quiring and constructing knowledge, and claim to have identified five "epistemological categories" proper to female cognition. They state that women follow a sequence from dependence to autonomy through five spe- cific stages: silence knowing (the experience of being voiceless), received knowing (the capacity of receiving knowledge from external authorities), subjective knowing (the perception of knowledge as privately intuited), pro- cedural knowing (the ability to apply knowledge), and constructed knowing (the ability to create knowledge). Although this book is not about mathematics, it has had a great deal of influence in the field of gender and mathematics. Seeing in it convincing ev- idence for a conflict between the ways in which mathematics is presently taught and the ways in which they think women learn best, some researchers have concluded that mathematics education must take into account distinct masculine and feminine ways of learning. Becker (1991) and Damarin (1990), for example, have attempted to spell out in some detail the implications of this book for mathematics teaching (referring to their proposals as a "feminist perspective" on the teaching of mathematics). Damarin suggests that "women learn abstractions (such as mathematics principles) best if statements of rules are preceded by quiet ob- servation, by listening to others, and by personal experiences that women can relate to the abstractions" (p. 148): in other words, if teaching conforms to the five stages suggested by Belenky et al. Becker states that the ideas in this book have major "implications for how to encourage girls and women to pursue mathematics" (p. 1). Although she recognizes that the case for a specifically feminine cognitive strategy is not GENDER AND INSTRUCTION 308 GILA HANNA proven, conceding that "research can help provide evidence to support or re- fute the hypothesis of women's ways of knowing in mathematics" (p. 3), she nevertheless proceeds to suggest some implications of this hypothesis for the teaching of mathematics. She believes, for example, that the third stage of knowing postulated by Belenky et al., subjective knowing, "is a very powerful one for the knower and brings in women's intuitive way of know- ing" (p. 4). The "women's ways of knowing" hypothesis elaborated by Belenky et al. has taken its place among the many schemes of cognitive development that compete for attention. There does not appear to be any evidence in their study or elsewhere, however, that this scheme has an advantage in any gen- eral sense over others, nor even that it is particularly useful in understanding how women learn. Belenky et al. have certainly not proved that there are cognitive differences between male and female learners, as they claim. (It is not at all clear that one can even design a study to prove or disprove such a contention.) Those who argue for an intrinsically feminine way of understanding mathematics, most of them feminists and all of them well-intentioned, are actually doing a disservice to education and to other women. (In other con- texts, their views would quickly be labeled as "sexist.") In reinforcing the traditional view of women as caregivers who are better at personal relations than abstract ideas, they run the risk of portraying women as fundamentally unsuited for science. In suggesting that the traditional male-female di- chotomies (such as logic vs. intuition, aggression vs. submission, and rigor vs. creativity) are valid and ingrained, if not inherent, they run the risk of perpetuating existing stereotypes, legitimizing gender differences in mathe- matics achievement, and providing a rationale for the relatively low partici- pation of women in scientific pursuits in general. 3. GENDER AND ACHIEVEMENT I will now turn my attention to studies of gender differences in mathematics achievement, examining meta-analyses of research papers published in the last 20 years as well as some international achievement surveys and national studies. All indicate rather clearly that gender differences in mathematics achievement are rapidly disappearing. A meta-analysis is a synthesis of several studies with more or less similar designs, in which the results of the studies are analyzed to yield summary measures about the overall statistical significance and the effect size of a given outcome. Meta-analyses thus combine information across multiple empirical studies that measured the same outcome, and can provide clear and concise effect-size measures of that outcome. In these studies, the term "achievement" refers to the results of standard- ized tests or textbook tests. Such tests are designed to measure competence in a general sense, but, in point of fact, most of the test items measure only 309 the ability to recall facts and the ability to apply concepts to the solution of relatively short problems. It should be pointed out that both classroom tests and standardized tests have recently come under severe criticism. The view of learning as mastery of factual knowledge implicit in the design of such tests is today considered incomplete, and is inconsistent with the view cur- rently held by many cognitive and educational psychologists that learning is active and constructive. However, assessment instruments capable of mea- suring adequately what is now referred to as "authentic" mathematical com- petence have yet to be designed. In the meantime, we have to rely on studies that used testing instruments and psychometric methods pervasive through- out the 1970s and 1980s. Fortunately these studies do give us reliable in- formation on significant aspects of mathematical mastery. 3.1 Meta-Analyses Hyde, Fennema, and Lamon (1990) examined about 100 studies pub- lished in the years 1967 to 1987 that used standardized mathematics tests and reported on gender differences in achievement. Their meta-analysis in- dicated that, in elementary and middle school, there were no gender differ- ences, that small gender differences favoring males emerged in high school and in college, and that the magnitude of these gender differences had de- clined over a 20-year period. As a measure of the magnitude of gender differences in the general popu- lation, the authors derived an effect size (d metric), defined as the mean for males minus the mean for females, divided by the mean within-sexes stan- dard deviation. Effect sizes were calculated as a function of the cognitive level (e.g., computation, concepts, or problem solving), as a function of the mathematical content (arithmetic, algebra, geometry, calculus, or mixed), as a function of age, as a function of ethnicity, and, finally, as a function of the selectivity of the sample (population selected according to level of perfor- mance). When the data were analyzed by cognitive level, all effect sizes proved to be small; for computation and for concepts, the small effects were in the girls' favor; while for problem-solving and mixed levels, they were in the boys' favor. Looking at the data by mathematics content, effect sizes were again very small for all topics. When examined by age, the data showed a "slight female superiority in performance in the elementary and middle school years. A moderate male superiority emerged in the high school years . . . and continued in the college years . . . as well as in adulthood" (p. 149). The researchers concluded that their meta-analysis provided "little support for the global conclusions" of previous studies that boys outperform girls in mathematics achievement (p. 151). Another meta-analysis by Friedman (1989) investigated 98 studies done between 1974 and 1987, comprising journal articles, doctoral dissertations, and large nationwide assessments carried out in the United States. The au- GENDER AND INSTRUCTION 310 thor concluded that "the mean random effects model . . . is minute . . . in- dicating that we cannot say with 95% confidence that a sex difference exists in the general United States population of school-age youth" (p. 204). The analysis also showed that "the sex difference in favor of males is decreasing over short periods of time" (p. 205). Friedman comments that the finding that "the average sex difference is now very small . . . should have considerable practical import" (p. 206), presumably in the expectation that it would help to dispel the widespread perception that boys outperform girls in school mathematics. A third meta-analysis conducted by Feingold (1988) reviewed research done over the previous 27 years and concluded that the magnitude of gender differences in cognitive abilities had declined markedly over that period. Though the achievement gap at the upper levels of high-school mathematics had remained constant, gender differences in verbal reasoning, abstract rea- soning, spatial relations, numerical ability, and other areas of cognitive ability had declined precipitously. 3.2 International Surveys An analysis of data collected in 1981 to 1982 by the Second International Mathematics Study (SIMS), which compared mathematics achievement in 20 countries at age 13, has shown not only that gender differences vary widely from country to country but also that they are smaller than differ- ences among countries (Hanna, 1989). In some countries, girls outper- formed boys in one to three of the five subtests; while in others, it was boys who did better on some of the subtests. In 5 of the 20 countries studied, no gender-related differences were observed. The more recent International Assessment of Educational Progress (IAEP) studies carried out in 1988 and 1991 also concluded that there are no marked gender differences in mathematics achievement among 13-year-old students. The first IAEP study encompassed 12 student populations from nine countries. The findings were that "boys and girls were performing at about the same level in 10 of the 12 populations assessed. Only in Korea and Spain do boys at this age achieve significantly higher in mathematics than do girls." (Lapointe, Mead, & Phillips, 1989, p. 18). The second IAEP study surveyed the mathematical performance of 13- year-old students in 20 countries, as well as that of 9-year-old students in 14 countries. The results indicated that there were few statistically significant differences in performance between the genders. One of the findings was that "the patterns of performance for males and females at age 9 . . . are not the same as those seen at age 13." More precisely, where small gender dif- ferences did exist in favor of boys, they were found in some countries at age 9 and in other countries at age 13 (Lapointe, Mead, & Askew, 1992, p. 86). GILA HANNA 311 3.3 National Studies In their 4-year "Girls and Mathematics" study of English schools, Walden and Walkerdine (1982, 1985) and their research team not only administered achievement tests but also carried out extensive classroom observation and interviews with students and teachers. Though Walden and Walkerdine have serious reservations about achievement tests, they nevertheless drew the fol- lowing conclusions from their test data: (a) There are few statistically sig- nificant sex differences in the elementary and 1st-year secondary tests; (b) in the 4th-year secondary test, girls consistently outperform boys; (c) there is no support in the data for the commonly held view that girls perform bet- ter only on computation or low-level items; and (d) despite the high perfor- mance of girls,". . . it turned out subsequently to be considerably more dif- ficult for girls to be entered for O level mathematics than boys" (1985, p. 46). The results of a study conducted in France over the period 1985 to 1989 indicate that gender differences in mathematics achievement among 12- and 15-year-old students are nonexistent or very small. Girls outperformed boys on some tasks and did as well as boys on others, while on yet other tasks, boys did better than girls. Baudelot and Establet (1992) conclude that, on the basis of their results, it is difficult to uphold the prevalent belief that girls are inferior to boys in mathematics achievement. 4. CONCLUSION The recent studies discussed here show that girls are not underachievers in school mathematics. On average, they perform as well as boys on most of the mathematics tests; on some tests, they outperform boys, whereas, on others, boys have the edge. When one considers their level of achievement in light of the observations made in many studies that boys often get more attention and time from teachers, that girls tend to have less confidence in their ability to do mathematics, and that, when it comes to mathematics achievement, parents often have lower expectations of their daughters, one must conclude that girls have benefited from undifferentiated mathematics instruction at least as much as boys. Girls have made enormous strides in mathematics achievement at the secondary level and are pursuing mathematics at the postsecondary level in increasing numbers. This is no reason, of course, for researchers to ignore those gender differences that persist. But, in the past decade, we have seen far-reaching proposals for a differential didactics at the school level resting upon alleged differences in cognition between boys and girls. Are not both the validity and the relevance of this radical solution clearly undermined by the achievements of girls in mathematics over this very decade in the face of well-recognized obstacles? In any case, we have not seen good evidence for differences in cognition. What case has been made that women have "a dif- ferent voice" or a monopoly on "connected knowing?" And if differences in GENDER AND INSTRUCTION 312 GILA HANNA Baudelot, C., & Establet, R. (1992). Allez les filles! Paris: Seuil. Becker, J. R. (1991). Women's ways of knowing in mathematics. Paper presented at the in- vited symposium of the IOWME, Assisi, Italy. Belenky, M. F., Clinchy, B. M., Golderberg, N. R., & Tarule, J. M. (1986). Women's ways Download 5.72 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling