Human Rights Education in Asian Schools Volume VII

Stereotypes in Textbooks and Teaching Materials in Hong Kong: A Literature Review*

Four local studies have investigated stereotyping in Hong Kong textbooks. Ho et al. (1997) examined the images of parents in Chinese-language textbooks at the primary level; Yau and Luk (1988) studied gender role stereotyping in Chinese history textbooks and social studies textbooks at the junior secondary level. Au (1992) also explored gender roles by examining textbooks used in three subjects (putonghua [Chinese language], social studies, and hygiene) at the primary level. Ma (1991) focused on studying the image and role of fathers in Chinese-language and social studies textbooks at the kindergarten and primary levels.

   The above studies all analyzed content. Texts (only words) and pictures are coded.
   These studies had a number of common findings.

   Physical context
   Mothers appeared more often in a household setting than fathers (Ho et al. 1997; Au 1992; Ma 1991). Outdoors, mothers are more likely to be found in markets or supermarkets (Ho et al. 1997; Yau and Luk 1988; Ma 1991).

   Housework
   Mothers are primarily responsible for housework. Ho et al. (1997) found that fathers are usually absent while mothers do housework. While fathers do housework, however, their wives are usually present and doing the same tasks. This result is supported by Au (1992) and Ma (1991). Mothers and fathers are also assigned different kinds of housework: fathers are usually found dusting and cleaning windows, while cooking, doing laundry, and shopping for food are left to the mothers (Ho et al. 1997; Ma 1991).

   Childcare
   Mothers mainly take care of children, particularly their daily lives and health (mothers accompanying children to school or to see a doctor, for instance) (Ho et al. 1997; Yau and Luk 1988; Ma 1991). Fathers' primary responsibilities are to take children to the library and department store (Ho et al. 1997) and to teach them (Ho et al. 1997; Ma 1991). Parents bring their children to the park or to visit friends and relatives (Ho et al. 1997).

   Division of labor
   Previous studies show that textbooks depict more working males than working females. Yau and Luk (1988) found that among 100 characters 73 were working males and 27 working females. Ma (1991) obtained a similar result. Males are portrayed as capable of working in a wide variety of sectors (Au 1992) and females mostly in education and low-paid jobs (Au 1992; Ma 1991). Males also occupy higher positions than females (Au 1992). Fathers are usually illustrated carrying briefcases and rarely in aprons, while mothers are never found with briefcases (Ma 1991).

   Activities
   Males are usually portrayed as more active and dominant than women. For example, Ho et al. (1997) found that fathers take photos of the whole family when visiting the park. Au (1992) found that fathers are shown as the primary source of income. Males are also accorded leadership, courage, curiosity for knowledge, determination, willingness to improve, motivation, rationality, and strength. Females are described as typically timid, sentimental, ignorant, emotional, and concerned with their appearance (Au 1992; Ma 1991).
   More famous men than women are shown in social studies textbooks.

   Visibility
   In general males are more visible than females. In Chinese history textbooks, males occurred 2,125 times, and females, 99. In social studies textbooks, the ratio of males to females is 61:31 (Yau and Luk 1988). Females are mostly found in household settings (Ho et al. 1997; Ma 1991).

   Use of generic nouns and pronouns
   Yau and Luk (1988) found that "he" is used to refer to the whole human race, and men's figures portray humankind.
   Previous studies show how the genders are usually portrayed as performing fixed roles, particularly occupational and family roles. No published studies, however, deal with stereotyping among other social groups (the disabled, older people, for instance). Previous studies have explored only a limited range of school subjects. For example, no science or English-language textbooks have been studied. Mathematics and English are important compulsory school subjects: the average Hong Kong student studies 16 years of English and 14 years of mathematics before entering university, so English and mathematics textbooks should be examined to present a more comprehensive study.


Overseas Studies on Stereotyping in Textbooks

Overseas studies are those not using data from Hong Kong. We identified one each from the People's Republic of China (PRC) (Nan 1992), Taiwan (Kong 1989), Singapore (Gupta and Lee 1990), and Israel (Gisnet 1988). The rest are Middle Eastern, European, and North American studies. This geographical and regional bias is noted in Sunderland's recent (2000b) special feature article on issues of language and gender in second- and foreignlanguage education (Sunderland 2000).
   Studies on gender stereotyping greatly outnumber those that deal with stereotyping of ethnicity and older people. No studies address stereotyping of disabilities or single-parent families and children. Only three studies deal with ethnic stereotyping (Baker 1996; Clarkson 1993; Stutzman 1993) and one study with stereotyping of older people (Stolley and Hill 1996).


Studies on Gender Stereotyping

Studies on gender stereotyping in textbooks abound, many carried out in the 1970s and 1980s. As our review indicates, and as Sunderland (2000) confirms, by far the most common methodology is that of content analysis. Much fewer studies used narrative analysis, and even fewer used linguistic analysis. No studies combine all three methodologies. We review studies using content and narrative analysis.
   Sunderland (1994b:55) identifies the most common six dimensions covered in the analysis of stereotyping in textbooks:

• Invisibility. Females are fewer than males.
• Occupational stereotyping in type and range of jobs. Women are not only fewer than men and have more menial jobs, but are also in roles that offer them a worse deal in the job market.
• Relationship stereotyping. Women are seen more often in relation to men than men are to women, usually in a relationship of flaunted heterosexuality or a perpetually happy nuclear family, and associated strongly with the domestic sphere.
• Personal characteristic stereotyping.
• Disempowering discourse roles for female characters. Women and girls speak less than men and boys, initiate less in mixed-sex dialogues, and exemplify different and less assertive language functions.
• "Degradation." Sexism is blatant, often to point of misogyny.

   We follow the dimensions suggested by Sunderland (1994b)^[1] and discuss how previous studies contribute to each dimension. Relationship stereotyping and disempowering discourse role will not be included here as they are more relevant to linguistic analysis.

   Visibility
   The first dimension we explore is gender visibility. Berelson (1952) suggests that significant absence can be a form of passive stereotyping. Previous studies have been interested in gender visibility in texts and visuals. A number of units have been proposed for counting characters' visibility:

• male and female appearance in texts (Gisnet 1988; Hartman and Judd 1978; Nan 1992; O'Donnell 1973; Porreca 1984);
• males and females as main characters in texts (Kortenhaus and Demarest 1993; O'Donnell 1973; Turner-Bowker 1996; U'Ren 1971);
• male and female appearance in story titles (Hellinger 1980; Kortenhaus and Demarest 1993; Turner-Bowker 1996); and
• male and female appearance in illustrations (Bazler and Simons 1991; Gisnet 1988; Kortenhaus and Demarest 1993; Porreca 1984; Stern 1976; Turner-Bowker 1996; U'Ren 1971).

   Regardless of the origins of the textbooks, males are more visible in texts and visuals than females. For example, Gisnet (1988) found that in Israeli English textbooks, 89% out of all children in pictures are boys. Overall, women constitute less than 33% of all the characters. Hellinger (1980) found that females are excluded from the titles of stories, and females appear in less than 10% of the titles of stories in English-language textbooks used in German schools. The male-to-female ratio in US Teaching English as a Second or Other Language (TESOL) materials was 63:37 (Hartman and Judd 1978). Nan (1992) found that 82 out of 93 characters in mainland Chinese-language textbooks were male.
   Fortunately, a number of studies have found that female visibility has improved over recent decades (see Sunderland et al. [2001]). For example, in their study of gender fairness in US high school chemistry textbooks, Bazler and Simons (1991) found that textbooks used in the 1990s included more females, unnamed characters, and unnamed youth characters in illustrations than those published in the 1970s. Still, the genders remain imbalanced.
   Kortenhaus and Demarest (1993:225) also found a similar result in children's picture books published in the US after the 1970s. Before then, texts tended to "contain four time as many boys as girls in titles, more than twice as many... in central roles, almost twice as many pictures." Textbook writers are correcting gender bias, although slowly and not across all school subjects.
   However, merely counting how many times the genders appear provides little insight into how they are portrayed in texts. Characters do not always have identical attributes. We should also consider how the genders are depicted in other dimensions such as occupational roles, family roles, and personalities.

   Occupational and family roles
   The second dimension commonly found in previous studies is that of occupational and family roles assigned to the genders.
   Researchers refer to horizontal and vertical dimensions of occupational roles. The horizontal dimension refers to the range of occupations that both genders occupy. The vertical dimension refers to the rank of the occupation, for example, bank managers belong to a higher rank than bank tellers.
   Previous studies show females in fewer occupations than males. Gisnet (1988:65) examined an Israeli English textbook, Here We Are, and found "the roles fulfilled by women include a mother, grandmother, kiosk worker, singer, cook, secretary, fortune teller, laundry clerk, reporter, and a traffic policewoman who is frightened by the traffic." Males, however, can be a "doctor, owner of a large department store, film producer, salesman, baker, truck driver, fireman, detective, acrobat, clown, soldier" (ibid.).
   Hartman and Judd (1978) found that women can only be students, bank employees, nurses, stewardesses (sic), salesgirls, and housewives. If women ever appear as congresswomen or lawyers, in the view of the authors, these are token gestures.
   Deliyanni-Kouimtzi (1992) found that after principles of gender equality were applied in education, Greek primary-school books now contain more working women. However, only 13.6% of women are portrayed as paid workers as opposed to 53.2% of men.
   O'Donnell (1973) also found that US primary-level social studies textbooks depict more working men than women. Of all male characters 83% were employed while only 17% of all female characters were paid workers. Porreca (1984) found that there were nearly six working males to one working female in US textbooks for English as a second language.
   Women in PRC Chinese-language textbooks (Nan 1992) also play less important work roles. In the textbooks that Nan examined, all scientists, artists, experts, and scholars are male. Porreca (1984) also found that the most frequently mentioned occupations for male characters were president (111 times), writer (59), teacher (43), and policeman and explorer (41 each). The most frequently mentioned occupations for female characters were teacher (28), actress (22), doctor (16), and secretary (13). In a related textbook study, O'Donnell (1973) found 72 males in highly paid and prestigious occupations, with only 1 female in such work.
   Not only do fewer females work than do males, but females also work in fewer occupations. The genders work in stereotypical occupations: we have yet to see a female president, truck driver, or fireman, or to see a male secretary or, surprisingly, captain or flight attendant.
   Family roles have also been frequently studied. U'ren (1971:221) found that primary-level textbooks portrayed mothers' lives as humdrum and uninteresting:

   Primary texts present the mother figure as a pleasant, hardworking, but basically uninteresting, individual: her life offers little excitement; she has no effect upon the world beyond her family, and even within the family her contribution is limited to housekeeping and cooking. Often she is merely a propman [sic] for the story; she enters a scene only to place a cake on the table and then disappears, or she plays foil to her husband by setting him up for his line. It is mother who asks what can be done and invites a speech from father.

   First, females are more likely to be restricted to the kitchen than males (Byrne 1978; Gaff 1978) and more likely to be found at home (Deliyanni-Kouimtzi 1992). Second, females are usually assigned fixed and settled domestic tasks such as cooking, baking, cleaning, polishing, mending, sewing, and washing, while males do the painting, gardening, repairing, and taking out the garbage (Hartman and Judd 1978; Stern 1976). O'Donnell (1973) found that males in primary-level social studies textbooks never washed dishes or cooked. Mothers also took care of children while fathers took the children for outdoor activities (O'Donnell 1973).
   After the gender equality program was implemented in Greek education, male characters were found to do more house tasks than they used to (Deliyanni-Kouimtzi 1992). However, women were still not portrayed as working at home nor as paid workers outside the home. Consequently, the roles of women became blurred.
   To conclude, we may cite Stern (1976:296): "Mothers do go to work throughout the world, but the textbook writers do not accept this reality."

   Attribution
   Previous studies have examined attributions assigned to textbook characters. U'ren (1971:223) found that textbooks give boys physical strength and mental qualities.

   The emphasis on masculine strength extends beyond physical qualities. Males of all ages are pictured as having greater mental perseverance and moral strength than females. Not only are females more often described as lazy and incapable of independent thinking or direct action but they are also shown as giving up more easily. They collapse into tears, they betray secrets; they are more likely to act upon petty or selfish motives. This last somewhat contradicts girl's typical representation as helpmates both to adults and to males to their own age.

   Females are portrayed in Isralei English textbooks as "unintelligent, careless, incompetent and gossipy" (Gisnet 1988:63), or as "incompetent, confused, inefficient, and not to be taken seriously" (Gisnet 1988:65). In British children's readers, girls are seen as weak, submissive, passive, and home based (Byrne 1978). In a more recent study, Turner-Bowker (1996) found that females in children's literature were still generally accorded weakness, passivity, negative evaluation, and femininity, while the male attributes included strength, activity, positive evaluation, and masculinity. Females were most frequently described as beautiful, frightened, worthy, sweet, weak, and scared, while men were big, horrible, fierce, great, terrible, furious, brave, and proud (Turner-Bowker 1996).
   Females showed typical "female" behaviors in German English-language textbooks (Hellinger 1980): emotionality, passivity, and limited intellectual ability. In French-language textbooks, the female characters are "positively unsympathetic" (Gaff 1978:74) while the male characters are much more agreeable. In US TESOL materials, females are overemotional, easily frightened, easily angered, spend too much time talking on the phone and getting dressed, and easily convinced that their goal is to find a wealthy husband (Hartman and Judd 1978). Mothers in US children's picture books are ineffectual and overwhelmed by problems, while their husbands are capable and have a take-charge manner (Kortenhaus and Demarest 1993).
   Last, Vaughn-Roberson et al. (1989) found that male characters in basal readers displayed traditional masculine traits: self-reliance, independence, and willingness to take risks.

   Degradation and disempowerment
   What Sunderland (1994b) refers to as degradation and disempowerment are also addressed in previous studies. Degradation essentially is making the characters look worthless, such as making them the butt of jokes. Disempowerment is withdrawing power and status from characters by failing to acknowledge their worth or contribution.
   Hartman and Judd (1978) found that girls are frequently the butt of offensive jokes such as: "Like a woman, it (the weather) is very changeable" and "Never underestimate a woman--if you're not talking about her weight or her age."
   U'ren (1971:222) found that US primarylevel textbooks rarely recognize girls' achievements: "Few girls receive any community publicity for contributions they have made, though they may be praised by members of their immediate family for work done in the house or at school. In thirty textbooks only two girls receive public acclaim."
   Hellinger (1980:273) found that in German English-language teaching textbooks, the "presence of intellectual or other achievements of women are ignored, downgraded or described as exceptional."
   Several studies focus on history textbooks (see Cairns and Inglis [1989]; Commeyras and Alvermann [1996]; Hartman and Judd [1978]). Females rarely appear in history textbooks and, if they do, have no power or authority: "The history of women only received specific attention in three textbooks and only marked emphasis in one. An immediate response to this result is to say that it simply reflects our limited knowledge of women in the past" (Cairns and Inglis 1989:225).
   Trecker (1971) makes the point:

Texts omit many women of importance, while simultaneously minimizing the legal, social and cultural disabilities which they faced. The authors tend to depict women in a passive role and to stress that their lives are determined by economic and political trends (cited in Commeyras and Alvermann 1996:34).

   Reese (1994) found that a number of key topics in women's history were omitted from the examples given in social science textbooks: for example, the traditional roles of Indian women, and women's lives and activities during the Early Republic period.

   Other areas
   Apart from visibility, occupational and family roles, attributions, and degradation and disempowerment, previous studies have also investigated the following:

• Participation of women in events (Hellinger 1980). Males participated more frequently in events than females. Hellinger (1980) found that females participated in only 30% of all stories in German Englishlanguage textbooks.
• Instrumental and independent, passive and dependent. Kortenhaus and Demarest (1993) found that boys were more instrumental and independent while girls were more passive and dependent.
• "Firstness." Porreca (1984) found that the female-to-male "firstness" ratio was approximately 1:3, that is, males are three times more likely to appear first in textbooks than females.
• Dressing. Knopp (1980) examined how females were dressed in textbooks used in West Germany and the former East Germany, and found that females in East German textbooks were more likely to wear trousers than those in West German textbooks. Still, only 28% of females in East German textbooks wore pants, and only 7% of females in West German textbooks.

   We agree with Sunderland (2000) that gender stereotyping is pervasive, especially in language-teaching materials. Our literature review corroborates her analysis and extends it to school subjects other than language education.


Studies on the Stereotyping of Ethnicity

Baker (1996:36) examined geography textbooks written by Jesse Olney and used in 1828 for "common schools" in the US, and found that ethnic groups were generalized: "National characters were considered to be innate and immutable and almost on a par with racial characteristics --the traits of any national group were considered to operate within its racial traits."
   Stutzman (1993) also found that Indians were overly generalized in US textbooks as simple and primitive.
   Clarkson (1993) found that the ethnic groups included in Australian mathematics textbooks were not as diverse as the Australian population, with only 8% of all characters being other than Anglo-Australian.
   African-American women, who are doubly disadvantaged, were barely included in US history textbooks.


Studies on the Stereotyping of Older People

Only one study examined the presentation of older people in textbooks. Stolley and Hill (1996) studied US undergraduate sociology textbooks and found that older people occupy only 3.6% of all the sampled pages. Older people are associated only with changing relationships, retirement, widowhood, and changing demographics of US society. Older people belonging to the doubly disadvantaged groups are underrepresented: 88% of all represented older people are white, "have few financial worries, rarely work, and, if coupled, are never gay" (Stolley and Hill 1996:41).


Inadequacies of Previous Studies

Stereotyping in elementary-, secondary-, and university-level textbooks has attracted attention from scholars for more than two decades. The studies we reviewed are valuable as they inform what we can examine and explore in educational materials and how. The studies, however, have a number of inadequacies.
   First, the number of studies on stereotyping among the disabled, older people, ethnic groups, and single-parent families is remarkably low. We have not been able to locate any studies that investigate stereotyping of the disabled and single-parent families. A possible explanation is the extremely low visibility of these two groups in textbooks, making the issue one of significant omission rather than commission.
   Representations of the disabled and older people should be examined in detail. We should look at not only visibility and manner of representation, but also family and occupational roles, and personality traits assigned. In short, equal attention should be given to a range of characteristics associated with people represented regardless of their gender, age, and ethnicity.
   The second reservation is the lack of accounting for generic or non-gendered characters in previous studies, which in no way denies the significance of tabulating the frequency of occurrence of male and female characters. Equally important, however, is the frequency of occurrence of generic characters and gendered characters. Generic characters are manifested through the use of explicitly coined nongendered terms such as "they," "we," "you," "people," "doctors," and so on. If generic characters outnumber gendered characters, leaving it to teachers and students of textbooks to interpret for themselves the gender of the characters, content analytical studies should be linked to studies that explicitly seek reader or user response to such sample materials, for example, via the use of focus group interviews or structured questionnaires, as we did in this study in Hong Kong.
   Third, as Sunderland (2000) and Sunderland et al. (2001) convincingly argue, content analyses of texts and focus group and questionnaire analyses of teacher responses and beliefs should be correlated with descriptive studies of actual teacher and student discourses and activities in classrooms. Teachers will not necessarily interpret stereotyped or non-stereotyped texts in either a supportive or critical and contradictory manner. How teachers and students interpret and deal with, and respond to, examples of stereotyping thus becomes a key datum. This is the case with overtly stereotyped texts, those where stereotyping is covert and hidden, and even where textbooks are progressive and counter-stereotypical. Such classroom observation remains a highly desirable follow-up research theme in Hong Kong.
   Fourth, a robust and comprehensive framework should be designed for content analysis. Previous studies have concentrated on gender stereotyping and neglected the influences of variables such as subject area. A number of previous studies have also failed to give an account of the coding criteria and associated working definitions of some of the terms used, making it difficult to understand how the investigators reached their results. For example, Nan (1992) suggests that male characters have attributed to them a larger number of "fine" qualities than female characters, but does not define "fine." It is also not clear whether the personality traits attributed to the characters are as interpreted by the researchers or conceived by the textbook authors.
   Finally, a number of complementary methodologies should be interconnected in a single study: content analysis; focus group and questionnaire-mediated qualitative and quantitative response data; linguistic analysis (analysis of lexico-grammatical features, metaphor, narrative, and discourse and interaction); and interview data from organizations and groups. We have adopted this multiple methodological approach in this study.

Stereotyping in a multiple research perspective

   Overview
   A comprehensive account of stereotyping in educational materials requires the integration of a range of different analytical methods, each contributing its perspective to the overall study. We employ five methodologies, including a range of qualitative and quantitative methods:

• content analysis of key variables in sampled texts;
• linguistic analysis (including lexico-grammatical, metaphor, narrative, and interaction analysis of sampled texts);
• ethnographic analysis (focus group discussions and studies of participant response to stimuli);
• survey analysis (questionnaire-mediated studies of participant beliefs and responses to stimuli); and
• structured interviews (responses to proposals from key stakeholders).

   Content analysis of selected texts
   Content analysis provides an objective, quantified description of how frequently selected social groups and attributes associated with them occur in the sampled educational materials. Occurrence is determined through the quantification of images (pictures, illustrations, photographs) and uses of language (words and phrases, characters). We coded 69,957 distinct references to mostly human characters from the entire corpus of 289 textbooks and examination papers. The analytical framework allows identification of four independent variables (subject, level, publisher, and mode); five major dependent variables (gender, old age, disability, ethnicity, and single parenthood and childhood); and a range of other dependent variables (family role, occupation, economic status, interest, attire, social status, achievement, failure, and public participation). The framework further allowed identification of stereotyping in one or more of the above areas, actively (by overt reference) and passively (by covert or no references). The framework is exploratory in not specifying a priori patterns of stereotyping but enabling them to be revealed through analysis of co-variation. Inspection of the analytical framework and its application on the data allows us to claim that this study is more extensive than previous studies on stereotyping.

   Linguistic analysis of selected texts
   Linguistic analysis studies how language use may reflect stereotyping by examining words, phrases, clauses, dialogues, and narratives. As language can be used metaphorically, its figurative uses should be studied to see how they reflect stereotypical attitudes and beliefs. Linguistic analysis complements content analysis and is crucial to research of stereotyping by providing quantitative information on how characters are referred to and allowing us to speculate on the relationships between characters (who does what to whom) and between characters and material objects (who stands in which relationship to what). Linguistic analysis also allows us to focus on the contextual meanings of individual clauses, that is, the environment where particular grammatical stretches of language occur. In short, linguistic analysis allows us to explore stereotyping in linguistic choice or discourse structure. We explore six areas of linguistic analysis:

• frequency of occurrence as manifested through language forms and collocations,
• generic nouns and generic pronouns,
• metaphors,
• transitivity,
• turn-taking and interactions, and
• narratives.

   The database to analyze frequency of occurrence, collocation, generic noun and pronouns, and metaphors was constructed from a sample of English-language texts, which could be processed using available corpus analytical tools.^[2] Selected Chinese- and English-language texts were also analyzed. For transitivity we concentrated on the grammar section of selected English-language teaching textbooks. For turntaking and interactions we drew on a sample of English- and Chinese-language textbooks, focusing on the analysis of dialogues. Finally, for narrative analysis we focused on primary-level Chinese-language textbooks and junior secondary-level English-language textbooks, as they typically include stories.

   Ethnographic analysis of teachers' and students' perceptions of stereotypicality
   Focus group interviews complement questionnaires. Specifically, the opinions of various participants can be recorded, extracted, highlighted, and interpreted through interviews. The rationale is to obtain as diverse a range of opinions as possible rather than to gauge the popularity of a narrow range of opinions. Although opinions obtained from focus groups do not lend themselves to generalization in the manner of quantitative accounts, they do provide in greater detail important variety to understandings of stereotyping. Six student focus groups from primary and secondary levels and from both genders, and 10 teacher focus groups from different disciplines and from primary and secondary levels, participated. Students drew tasks, opened discussions and story-telling, and performed response tasks. Teachers were involved in picture and textbook sample response tasks, and guided discussion tasks.

   Survey analysis of teachers' perceptions
   The questionnaire generates quantitative or numeric data while the focus group yields rich qualitative data in the form of the interviewees' voices and drawings. Questionnaire data lend themselves to the task of generalization while the interviewees' voiced perceptions are best suited to more detailed description and explanation. Both perspectives are necessary to build a complete picture of stereotyping in Hong Kong educational materials. Taken together, both modes of analysis augment the content analysis of textbooks and corresponding linguistic analysis. The questionnaire was designed to gauge respondents' awareness of the extent of stereotyping in the main subject being taught and with reference to the main textbooks being used, all of which are also investigated as part of the content analysis. Items in the questionnaire were derived partly with reference to the outcomes of the content and linguistic analyses. Other items focused on respondents' judgements of the perceived stereotypicality of selected textbook stimuli in relation to the main variables. Finally, the remaining items concerned demographic information about the respondents, especially their teaching experience.

   Structured interview analysis
   An important part of the project brief was to consult with a range of publishers and educational organizations involved in the development and approval of educational materials--textbooks, examination materials, and audiovisual educational programs. The purpose was to assess the awareness of organizations and institutions of the general issues raised in the project, and to gauge the extent to which staff were already modifying stereotyping in the educational materials that they were either responsible for producing or were involved in evaluating. At the same time, the project team wished to test reactions to a draft set of guidelines to avoid stereotyping in educational materials, and to modify them in the light of stakeholder discussions. Structured interviews were carried out with a selection of English- and Chinese-language publishers who produced textbooks for the primary and secondary levels in Hong Kong, and across a range of subject disciplines. Interviews were also conducted with representatives of the Hong Kong Examinations Authority (HKEA), Hong Kong Special Administrative Region (HKSAR) Curriculum Development Institute, and Educational Television (ETV). Such interviews were augmented by inspection of relevant documents on promoting non-stereotypicality in educational materials.

   Interconnecting different research perspectives
   Here we provide some snapshot examples of the ways that the different perspectives complement each other in respect of three of the major variables involved in this research study: gender, age and disability.

   Gender
   The content analysis of gender indicates that genderization (male versus female, and genderized versus non-genderized appearances and reference) in terms of frequency of occurrence varies across subjects, educational levels, publishers, and mode. Genderization, like gender, has repeatedly been shown to be involved in the above relationships as well as in those involving an additional attribute such as family role, occupation, economic status, interest, attire, social status, achievement, failure, and public participation. The dominance of nongendered over gendered characters in the database of textbook content reflects the usefulness the category "non-genderized" under the variable gender.
   The linguistic analysis confirms this variation and indicates that in terms of the variable frequency of occurrence, male references predominate; in respect of collocation females appear more dependent on males than vice versa; in terms of the variable genderization, Chinese-language materials differ from English-language materials in being more gender-specifically referenced, although this may in part be due to the internal grammatical structure of Chinese. In terms of metaphorical usage both English- and Chineselanguage materials associate maleness with what may be loosely characterized as "strong and active" qualities or persons, and femaleness with "weak and responsive" counterparts. In terms of transitivity there is greater balance between the genders, although there is some stereotypicality in terms of genderized behavioral processes. In variable turn-taking and interaction, both genders are equally visible but differentiated by topic, and in terms of where they work and who initiates talk (male) and responds (female). Finally, the narrative analysis shows that Chinese-language materials reinforce traditional role distributions in families, while English-language materials take a more balanced view. Chineselanguage materials are more likely than Englishlanguage materials to genderize the distribution of moral precepts: what knowledge is provided to children by which gender in the family and how family problems are solved and by whom.
   The ethnographic analysis of teachers' and students' perceptions further corroborated genderization, although awareness by primaryand secondary-school students of women in a nontraditional and non-stereotypical distribution of role and occupation was marked and related in part to their own family experiences or public exposure in the media. Among primary- and secondary-school teachers stereotypical genderization was generally recognized but conditioned by the need to contextualize such genderization in relation to subject matter, to the actual gendered distribution of roles and occupations in Hong Kong, and to the level of awareness of their students. Avoiding genderization in classroom practice was desirable, although again conditioned by the degree of students' awareness. Teacher education was insufficiently directed to countering stereotypical genderization.
   The survey analysis of teacher perceptions indicated that teachers did not always perceive the extent of stereotypicality and counterstereotypicality of textbooks as evidenced by the content and linguistic analyses. Such features were over- and underestimated. Among the main variables, gender was generally assumed to be presented in a more balanced way, while primary-school teachers seemed more sensitive to stereotypical genderization and counterstereotypicality than did secondary-school teachers in the sample.
   The structured interview analysis with major stakeholders indicated that among English-language publishers genderization was a key issue and materials were explicitly devised to avoid stereotypicality and to promote counterstereotypicality. In-house guidelines reinforced this policy. Chinese-language publishers were less aware--none except one had an in-house policy--and more reliant on awaiting clear guidelines from the Education Department. The publishers also markedly wished to condition the absolute avoidance of genderized stereotypicality by considering historically determined social roles. The HKEA and ETV saw avoidance of genderization as a priority, and the Curriculum Institute has a clearly defined policy of avoiding genderization and promoting counter-stereotypicality of gender.

   Age
   The study of age and older people in the content analysis indicates that older people appear less often than younger people in textbooks in every subject save economics and integrated sciences (where they appear equally) at every level of schooling; in all materials except for those published by Longman and Ling Kee (where they appear equally) and Pilot (where only older people are featured); and in the written mode (although appearing equally in the visual mode). Apart from economic status and occupation, age is rarely represented along with other attributes such as family role, interest, attire, social status, achievement, failure, and public participation. The low frequencies of occurrence translate into difficulties in making statistical generalizations and, therefore, highlight the need for other, more qualitative research methodological perspectives.
   The linguistic analysis provides marked evidence of the imbalance in representation between older and younger people in that where older people appeared (very rarely) they were stereotyped as less physically capable than younger people, less active, more likely to need help and assistance; gender and age combine in stereotypes of distribution of family roles, with grandparents providing moral precepts and knowledge, although gendered in relation to theme and focus.
   The ethnographic analysis of teachers' and students' perceptions indicated that students saw older people were typically as needing assistance and, in some cases, as needing to respond favorably to such assistance. Their attire was stereotypically depicted and gendered. Among teachers, role, task, and occupational distribution in relation to age could not be determined in absolute terms but relative to such persons' age and strength, and also in terms of the current distribution of such roles, tasks, and occupations in Hong Kong.
   The survey analysis of teachers' perceptions indicated that respondents in general were not aware of age stereotypicality or, rarely, counterstereotypicality in the materials.
   The structured interview analysis with major stakeholders indicated that English-language publishers, HKEA, and ETV need to be more aware of age stereotypicality in actual presence and in occupation and role, and to counter this in their educational materials. HKEA has explicitly recognized this need. Older people are both rare and likely to be stereotypically portrayed or, in the case of older single people, absent. Chinese-language publishers felt that while the increased and differentiated representation of older people was important, to an extent their representation in educational materials was constrained by the life experiences of students and the difficulty of determining what constituted an "older person." As with gender, the Curriculum Institute wishes to increase the non-stereotypical and counter-stereotypical representation of older people in educational materials and to encourage project work that will engage students in exploring the active extension of older people in their social roles and occupations.

   Disability
   The study of disability in the content analysis of educational materials reveals that disabled people are barely represented. Only 73 (or 0.1% of the total number of characters in the entire corpus) indicate disability. Only physically disabled characters are featured in visual materials, while written materials also refer to sensory impairment and developmental disability. As with the variable age, the low frequencies of occurrence translate into difficulties in making statistical generalizations and, therefore, highlight the need for other, more qualitative research methodological perspectives.
   The linguistic analysis of the corpus data indicates that the word "disabled" only occurs five times in the whole corpus and the word "disability" only twice. In relation to the variable of transitivity, disabled people only occur twice in the sample corpus, where they are seen as passive, needing support and help. They do not appear as fully functioning social actors. In terms of turn-taking and interaction, disabled people do not feature explicitly, and are almost invisible, as they are in the selected extracts for narrative analysis.
   The ethnographic analysis of teachers' and students' perceptions provides further evidence of this relative invisibility and marginalization of the disabled. Lack of contact with disabled people affected primary-level students' willingness to identify with, or to know much about, disabled characters in stories, although secondary-level students' drawings showed friendliness and helpfulness toward the blind. The stories and responses from secondary-level students evaluated disabled people as needing assistance but also as lacking enthusiasm and confidence. The students recognized the difficulties faced by the disabled, especially women, who are to be admired, but the students had little personal experience with them. Teachers are aware that disabled people in educational materials are never seen as actively contributing to society but as needing assistance. Teachers felt that this absence may be due to an unwillingness by the disabled to permit representation in materials and were unsure of how to deal with class discussions on disability.
   The survey analysis of teachers' perceptions indicated an awareness of the almost complete absence of disabled people from textbooks.
   The structured interview analysis indicated that, as with the issues surrounding the representation of older people, educational materials had not represented disability appropriately. Among English-language publishers, HKEA and ETV were aware of the need to increase the representation of a range of disabilities in educational materials and to show disabled people as actively contributing to family and social life. Chinese-language publishers reported difficulty in obtaining images of disabled people acceptable to the social service organizations concerned with the disabled, but believed that educational materials should heighten the positive presence of the disabled to motivate students to be supportive. As with the other variables discussed here, the Curriculum Institute believed that educational materials should greatly increase their positive representation of the disabled and involve students in project work highlighting the active contributions of the disabled to society.
   The above examples show how adopting an integrated, multiple research perspective not only provides a richer and more grounded account of the variables involved and the issues to which they relate, but also that particular research methodologies offer mutually corroborative and distinctive contributions to that overall account.


References

   Abraham, J. 1989. Teacher ideology and sex roles in curriculum texts. British Journal of Sociology of Education, 10(1):33-51.
   Alcon, E. 1994. The role of participation and gender in non-native speakers' classroom interaction. Working Papers on Language, Gender and Sexism, 4(1):51-68.
   Bacon, S. 1992. The relationship between gender, comprehension, processing strategies, and cognitive and affective response in second language learning. Modern Language Journal, 76:160-178.
   Bacon, S. and M. Finneman. 1992. Sex differences in self-reported beliefs about foreign language learning and authentic oral and written input. Language Learning, 42(4):471-495.
   Baker, S. 1996. Jess Olney's innovative geography text of 1828 for common schools. Journal of Geography, 95(1):32-38.
   Bazler, J.A. and D.A. Simonis. 1991. Are high school chemistry textbooks gender fair? Journal of Research in Science Teaching, 28(4):353-362.
   Beebe, J. 1996. Sexist language and English as a foreign language: A problem of knowledge and choice. The Language Teacher, 22(5):7-10, 19.
   Bell, R. and J. Hay. 1987. Differences and biases in English language examination formats. British Journal of Educational Psychology, 57:212-220.
   Beswick, C. 1976. Mixed or single sex for French? Audio-Visual Language Journal, pp. 34-38.
   Biber, D., S. Conrad, and R. Reppen. 1998. Corpus linguistics: Investigating language structure and use. Cambridge: Cambridge University Press.
   Boyle, J.P. 1987. Sex differences in listening vocabulary. Language Learning, 37(2):273-284.
   Bruce, K. 1986. Sex stereotyping in English language coursebooks--is it still an issue? TESOL France News, 6(1):46-48.
   Bugel, K. and B. Buunk. 1996. Sex differences in foreign language text comprehension: the role of interests and prior knowledge. Modern Language Journal, 80(1):15-31.
   Butt, D., R. Fahey, S. Feez, S. Spinks, and C. Yallop. 2000. Using functional grammar: An explorer's guide (2nd ed.). Sydney, Australia: Macquarie University, National Centre for English Language Teaching and Research.
   Byrne, E.M. 1978. Women and Education. London: Tavistock.
   Byrnes, F. 1994. Including gender in EFL/ESL teacher education programmes. In: J. Sunderland (ed.) (1994a).
   Cairns, J. and B. Inglis. 1989. A content analysis of ten popular history textbooks for primary schools with particular emphasis on the role of women. Educational Review, 41(3):221-226.
   Cameron, D. 1994. Problems of sexist and non-sexist language. In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education. New York: Prentice Hall, pp. 26-33.
   Cameron, D. 1998. The Feminist Critique of Language (2nd ed.). London: Routledge.
   Cameron, L. and G. Low (eds.). 1999a. Researching and applying metaphor. Cambridge: Cambridge University Press.
   Cameron, L. and G. Low. 1999b. Survey article: Metaphor. Language Teaching, 32:77-96.
   Carroll, D. and J. Kowitz. 1994. Using concordancing techniques to study gender stereotyping in ELT textbooks. In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education New York: Prentice Hall, pp. 73-82.
   Cerezal, F. 1994. Lenguaje y discriminación generica en libros de texto de inglés. Miscelánea, 10.
   Chavez, M. 2000. Teacher and student gender and peer group gender composition in German foreign language classroom discourse: an exploratory study. Journal of Pragmatics, 32:1019-1058.
   Cheshire, J. and N. Jenkins. 1991. Gender issues in the GCSE oral English examination: Part 2. Language and Education, 5(1):19-40.
   Cincotta, M.S. 1978. Textbooks and their influence on sex-role stereotype formation. BABEL: Journal of the Australian Federation of MLTS Associations, 14(3):24- 29.
   Clarkson, P. 1993. Gender, ethnicity and textbooks. The Australian Mathematics Teacher, 49(2):14-16.
   Clausen, J. 1982. Textbooks and (in)-equality: a survey of literary readers for elementary and intermediate German. Die Unterrichtspraxis, 15:244-253.
   Cochran, E. 1992. Towards degendered English in the ESL classroom. Working Papers on Language, Gender and Sexism, 2(2):27-35.
   Cochran, E. 1996. Gender and the ESL classroom. TESOL Quarterly, 3(1):159-162.
   Cohen, J. 1968. Weighted kappa: Nominal scale agreement with provision for scaled disagreement or partial credit. Psychological Bulletin, 70(4):213-220.
   Commeyras, M. and D.E. Alvermann. 1996. Reading about women in world history textbooks from one feminist perspective. Gender and Education, 8(1):31-48.
   Woman's and Youth Division, Commonwealth Secretariat. 1995. Gender Bias in School Text Books. London.
   Crabb, P.B. and D. Bielawski. 1994. The social representation of material culture and gender in children's books. Sex Roles, 30(1/2):69-79.
   Curriculum Development Council. 2000. Learning to Learn (Key Learning Area: English Language Education). Hong Kong: Curriculum Development Council.
   Curriculum Development Council. 2000. Learning to Learn (The Way Forward in Curriculum Development). Hong Kong: Curriculum Development Council.
   Dallas, D. and L. Pelham. 2001. New Welcome to English (2nd ed.). Hong Kong: Longman Hong Kong Education.
   Davies, B. 1992. Education for sexism: a theoretical analysis of the sex/gender bias in education. Educational Philosophy and Theory, 21(1):1-19.
   Davies, B. 1992. A feminist post-structuralist analysis of discursive practices in the classroom and playground. Discourse, 1(1):49-66.
   Deliyanni-Kouimtzi, K. 1992. "Father is out shopping because Mother is at work..." Greek primary school reading texts as an example of educational policy for gender equality. Gender and Education, 4(1/2):67-79.
   Dendrinos, B. 1992. The EFL Textbook and Ideology. Athens: N.C. Grivas Publications.
   Dupont, B. 1981. Unequal Education: A Study of Sex Differences in Secondary-school Curricula. Paris: UNESCO.
   Ehrlich, S. 1997. Gender as social practice: Implications for second language acquisition. Studies in Second Language Acquisition, 19(4):421-426.
   Equal Opportunities Commission. 2001. Disability Discrimination Ordinance Code of Practice on Education for Public Consultation. Hong Kong: Equal Opportunities Commission.
   Everitt, B.S. 1992. The Analysis of Contingency Tables (2nd ed.). London: Chapman & Hall.
   Florent, J. and C. Walter. 1989. A better role for women in TEFL. ELT Journal, 43(3):180-184.
   Florent, J., K. Fuller, J. Pugsley, C. Walter, and A. Young. 1994. On balance: Guidelines for the representation of women and men in English Language Teaching Materials. In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education. Hemel Hempstead: Prentice Hall.
   Freudenstein, R. (ed.). 1978. The role of Women in Foreign Language Textbooks: A Collection of Essays. Collections d'Etudes Linguistiques, No. 24. Gand. Federation Interbnationale des Professeurs de Langues Vivantes.
   Gaff, R. 1978. Sex stereotyping in modern language teaching --An aspect of the hidden curriculum. British Journal of Language Teaching, 20:71-78.
   Galanes, G. J., K. Adams, and J.K. Brilhart. 2000. Communicating in groups: Applications and skills (4th ed.). Boston: McGraw Hill.
   Gass, S. and E. Varonis. 1986. Sex differences in nonnative speaker-nonnative speaker interactions. In: R. Day (ed.) Talking to Learn: Conversation in Second Language Acquisition. New York: Newbury House.
   Gisnet, J. 1988. Sexism in English by Television textbooks. English Teachers' Journal (Israel), 36:60-66.
   Goffman, E. 1981. Forms of talk. Philadelphia: University of Pennsylvania Press.
   Goh, C. and K.P. Foong. 1997. Chinese ESL students' learning strategies: A look at frequency, proficiency and gender. Hong Kong Journal of Applied Linguistics, 2(1):39-53.
   Graci, J. 1989. Are foreign language textbooks sexist? An exploration of modes of evaluation. Foreign Language Annals, 22(5):477-486.
   Graham, A. 1975. The making of a non-sexist dictionary. In: B. Thorne and N. Henley (eds.) Language and Sex: Difference and Dominance. Rowley, Massachusetts: Newbury House.
   Gupta, A.F. and A.S.Y. Lee. 1990. Gender representation in English language textbooks used in the Singapore primary schools. Gender and Education, 4(1):29-49.
   Haccou, P. and E. Meelis. 1992. Statistical analysis of behavioural data. Oxford: Oxford University Press.
   Hall, K. and M. Bucholtz (eds.). 1995. Gender Articulated: Language and the Socially Constructed Self. New York: Routledge.
   Halliday, M.A.K. 1971. Linguistic function and style: An inquiry into the language of William Golding's The Inheritors. In: S. Chatman (ed.) Literacy style: A symposium. London: Routledge, pp. 330-368.
   Halliday, M.A.K. 1994. An Introduction to Functional Grammar (2nd ed.). London: Arnold.
   Hartman, P. and E. Judd. 1978. Sexism and TESOL materials. TESOL Quarterly, 12(4):383-392.
   Hays, W.L. 1994. Statistics (5th ed.). Orlando, Florida: Harcourt Brace & Company. Hellekant, J. 1994. Are multiple choice tests unfair to girls? System, 22(3):349-352.
   Hellinger, M. 1980. For men must work, and women must weep: Sexism in English language textbooks used in German schools. Women's Studies International Quarterly, 3:267-275.
   Henessey, M. 1984. Gender and pedagogical dictionaries. In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education. Hemel Hempstead: Prentice Hall.
   Hill, P. 1980. Women in the world of ELT textbooks, 1979-1980. EFL Gazette ,18/19: June/July.
   Hoey, M. 1996. A clause-relational analysis of selected dictionary entries: contrast and compatability in the definitions of 'man' and 'woman'. In: C. Caldas-Coulthard and M. Coulthard (eds.) Test and Practices: Readings in Critical Discourse Analysis. London: Routledge.
   Holmes, J. 1987. Sex differences and language use in the ESL classroom. In: B.K. Das (ed.) Communication and Learning in the Classroom Community. Singapore: SEAMEO Regional English Language Centre.
   Holmes, J. 1989a. Is sex relevant in the ESL classroom? Language Issues, 3(1):14-18.
   Holmes, J. 1991. Language and gender. Language Teaching, 24(4):207-220.
   Hong Kong Examinations Authority. 1999. Hong Kong Certificate of Education Examination Annual Report 1999. Hong Kong: Hong Kong Examination Authority.
   Jacobs, G., Q. Zhuo, P. Jocson, C. Ong, and M. Austria. 1996. Asian second language educationists' views on gender-inclusive English. Working Papers on Language, Gender and Sexism, 6(1):75-94.
   Jaworski, A. 1983. Sexism in textbooks. British Journal of Language and Teaching, 21(2):109-113.
   Jenkins, N. and J. Cheshire. 1990. Gender issues in the GCSE oral English examination: Part 1. Language & Education, 4(4):261-292.
   Jette, R.D. 1990. Participation: Activities, influence or both? A conceptual clarification and field test. Ann Arbor: University of Michigan Press.
   Jones, M.A., C. Kitetu, and J. Sunderland. 1997. Discourse roles, gender and language textbook dialogues: Who learns from John and Sally? Gender and Education, 9(4):469-490.
   Judd, E. 1983. The problem of applying sociolinguistic findings to TESOL: The case of male/female language. In: N. Wolfson and E. Judd (eds.) Sociolinguistics and Second Language Acquisition. Rowley, Massachusetts: Newbury House.
   Kanaris, A. 1999. Gendered journeys: Children's writing and the construction of gender. Language and Education, 13(4):254-268.
   Kaye, P. 1989. 'Women are alcoholics and drug addicts', says dictionary. ELT Journal, 43(3):192-195.
   Kendall, P.C. and C. Hammen. 1995. Abnormal psychology. Boston: Houghton Mifflin.
   Kennedy, G. 1998. An introduction to corpus linguistics. London: Longman.
   Kitetu, C. 1998. An examination of physics classroom discourse practices and the construction of gendered identities in a Kenyan secondary school. PhD thesis. UK: Lancaster University.
   Kortenhaus, C.M. and J. Demarest. 1993. Gender role stereotyping in children's literature: An update. Sex Roles, 28(3/4):219-232.
   Krippendorff, K. 1980. Content analysis: An introduction to its methodology. Beverly Hills: Sage.
   Leech, G. 1974. Semantics. Harmondsworth: Penguin.
   Lesikin, J. 1998. Determining social prominence: A methodology for uncovering gender bias in ESL textbooks. College ESL, 8(1):83-96.
   Lindsey, M.P. 1989. Dictionary of mental handicap. London: Routledge.
   Lo, E., S. Lit, and F. Cheung. 2001. Stereotypes in Junior Secondary English Textbooks in Hong Kong. TESLHK, 6:4-5.
   Lyons, J. 1977. Semantics (Vol. 2). Cambridge: Cambridge University Press.
   Mannheim, C. 1994. Learners' attitudes to EFL textbook sexism. In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education. Hemel Hempstead: Prentice Hall.
   Markham, P. 1998. Gender and the perceived expertness of the speaker as factors in ESL listening recall. TESOL Quarterly, 22(3):397-406.
   Myers, K. (ed.). 1992. Genderwatch! Cambridge: Cambridge University Press.
   Mills, S. (ed.). 1994. Gendering the Reader. Hemel Hempstead: Harvester Wheatsheaf.
   Mills, S. 1995. Feminist stylistics. London: Routledge.
   Nan, N. 1992. Sex discrimination in education. Chinese Education and Society, 25(1):44-47.
   Nelson, C. 1999. Sexual identities in ESL: queer theory and classroom enquiry. TESOL Quarterly, 33(3):371- 391.
   O'Donnell, R.W. 1973. Sex bias in primary social studies textbooks. Educational Leadership Research Supplement, November 1973.
   O'Neill, R. 1994. 'The case of Julia Frost' or 'The lure of the stereotype.' In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education. Hemel Hemsptead: Prentice Hall.
   Orne, M.T. 1963. On the social psychology of the psychological experiment: with particular reference to demand characteristics and their implications. American Psychologist, 17(10):776-783.
   Osugi, K., K. Sadakane, Y. Shimogouchi, and K. Takahashi. 1990. Sex bias in Japan-published monolingual English dictionaries. JALT Journal, 12(2) (November)
   Porreca, K.L. 1984. Sexism in current ESL textbooks. TESOL Quarterly, 18(4):705-724.
   Poulou, S. 1997. Sexism in the discourse roles of textbook dialogues. Language Learning Journal, 15:68-73.
   Prechter, S. 1999. Women's rights-children's games: Sexism in learners' dictionaries of English. Multilingua, 18(1):47-68.
   Psathas, G. 1995. Conversation analysis: The study of talkin-interaction (Qualitative Research Methods, Vol. 35). Thousand Oaks: SAGE.
   Pugsley, J. 1992. Sexist language and stereotyping in ELT materials. Working Papers on Language, Gender and Sexism, 2(2):5-13.
   Reber, A.S. 1985. The penguin dictionary of pscyhology. London: Viking.
   Rees-Parnell, H. 1976. Women in the world of Kernel Lessons Intermediate. ARELS Journal , 2(2).
   Reese, L. 1994. Gender equity and texts. Social Studies Review, 33(2):12-15.
   Schermerhorn, R.A. 1970. Comparative ethnic relations: A framework for theory and research. New York: Random House.
   Schmitz, B. 1975. Sexism in French language textbooks. In: R. Lafayette (ed.) The Cultural Revolution in Foreign Language Teaching. Skokjie, Illinois: National Textbook Company.
   Schmitz, B. 1984. Guidelines for reviewing foreign language textbooks for sex bias. Women's Studies Quarterly, XII(3):7-9.
   Stern, R.H. 1976. Review article: Sexism in foreign language textbooks. Foreign Language Annals, 9:294-299.
   Stolley, K.S. and A.E. Hill. 1996. Presentations of the elderly in textbooks on marriage and family. Teaching Sociology, 24:34-45.
   Stutzman, E. 1993. American Indian stereotypes: The truth behind the hype. An Indian education curriculum unit. Oregon: Coos County Indian Education Coordination Program.
   Sunderland, J. 1994a. Exploring Gender: Questions and Implications for English Language Education. Hemel Hempstead: Prentice Hall.
   Sunderland, J. 1994b. Introduction to Quadrant II. In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education. New York: Prentice Hall, pp. 55-66.
   Sunderland, J. 1994c. Pedagogical and other filters: The representation of non-sexist language change in British pedagogical grammars. In: J. Sunderland (ed.) Exploring Gender: Questions and Implications for English Language Education. New York: Prentice Hall, pp. 92-103.
   Sunderland, J. 2000a. New understandings of gender and language classroom research: Texts, teacher talk and student talk. Language Teaching Research, 4(2):149- 173.
   Sunderland, J. 2000b. Review article: Issues of language and gender in second and foreign language education. Language Teaching, 33:203-223.
   Sunderland, J., F. Abdul Rahim, M. Cowley, C. Leontzakou, and J. Shattuck. 2001. From bias 'in the text' to 'teacher talk around the text': An exploration of teacher discourse and foreign language textbook texts. Linguistics and Education, 11(3):251-286.
   Swann, J. and D. Graddol. 1998. Gender inequalities in classroom talk. English in Education, 22(1):48-65.
   Talansky, S. 1986. Sex role stereotyping in TEFL teaching materials. Perspectives, XI(3):32-41.
   Turner-Bowker, D.M. 1996. Gender stereotyped descriptors in children's picture books: Does "curious Jane" exist in the literature? Sex Roles, 35(7/8):461- 488.
   U'Ren, M.B. 1971. The image of woman in textbooks. In: V. Gornick and B.K. Moran (eds.) Woman in sexist society: Studies in power and powerlessness. New York: Basic Books, pp. 218-225.
   Van Leeuwen, T. 1996. The representation of social actors. In: C. Caldas-Coulthard and M. Coulthard (eds.) Texts and Practices: Readings in Critical Discourse Analysis. London: Routledge, pp. 32-70.
   Vaughn-Roberson, C., G.E. Tompkins, M.E. Hitchcock, and M. Oldham. 1989. Sexism in Basal readers: An analysis of male main characters. Journal of Research in Childhood Education, 4(1):62-68.
   Way, E.C. 1991. Knowledge Representation and Metaphor. Dordrecht: Kluwer Academic Press.
   Weber, M. 1968. Economy and society: An outline of interpretative sociology. Berkeley, CA: University of California Press.
   Weber, R.P. 1990. Basic content analysis (2nd ed.). Newbury Park, CA: Sage.
   Whylde, J. 1983. Sexism in the Secondary Curriculum. New York: Harper, Row.
   Wichmann, A., S. Fligelstone, T. EcEnery, and G. Knowles (eds.). 1997. Teaching and language corpora. London: Longman.
   Willeke, A.B. and R.H. Sanders. 1978. Walter ist intelligent und Brigitte ist blond: dealing with sex bias in language texts. Unterrichtspraxis, 11:60-65.
   Yepez, M. 1994b. An observation of gender-specific teacher behaviour in the ESL classroom. Sex Roles, 30(1/2):121-133.
   Jiang, Li-Ly. 1989. Discussion on the development of gender roles in children's books. Proceedings of the Conference on Gender Roles and Changing Society. Taipei: Taipei Population and Gender Studies Center Women Research Program, National Taiwan University and National Tsing Hua University Department of Humanities and Social Sciences (in Chinese language).
   He, W.-J., G.-L. Chang, M.-Y. Huang, and S.-J. Wang. 1997. Research on the image of parents in Chinese textbooks for the elementary schools in Hong Kong. Journal of Basic Education, 7(1) (in Chinese language).
   Cyu, Jie-Jhen. 1992. Gender roles research on the textbooks of elementary schools in Hong Kong. Hong Kong: Hong Kong Institute of Asia-Pacific Studies, The Chinese University of Hong Kong (in Chinese language).
   Youli, L.-L., H.-J. Lu. 1988. Gender roles research on the History and Society textbooks of junior high schools in Hong Kong. Hong Kong: The Centre for Hong Kong Studies of the Institute of Social Studies, The Chinese University of Hong Kong (in Chinese language).
   Ma, Cuei-Fen. 1991. The report on the father image in the textbooks of elementary schools and kindergartens in Hong Kong. Hong Kong: Hong Kong Christian Service (in Chinese language).


Notes

   1. Sunderland (2000b) provides a valuable, if necessarily brief, overview of gender bias in foreign- and second-language teaching textbooks in one section of her valuable overall review of language and gender in foreign- and second-language education for the abstracting journal, Language Teaching.
   2. Our corpus analysis was limited to English-language texts because the only corpus analytical tool available to us (Wordsmith) is restricted to English.

* This is chapter 2 of Final Report: Research on Content Analysis of Textbooks and Teaching Materials in Respect of Stereotypes (Hong Kong: Centre for English Language Education Communication, City University of Hong Kong, 2001). The research project was sponsored by the Equal Opportunity Commission of Hong Kong.

---

• « Insensitivity of Pakistani School Education to Religious Diversity of the Nation*
• Shin-Eiken and Peace Education in English-Language Classrooms »
• Human Rights Education in Asian Schools