Gender Bias in the Iranian High School EFL Textbooks

Gender bias is unfortunately still present in many societies especially the developing countries. Such prejudice is in most cases in favor of males and against females. While females nowadays comprise a great majority of the work force all around the world, they are still being looked upon as the weaker sex associated with stereotypical roles as stay-at-home mothers whose main task is to breed children, wash the dishes, etc. Such prejudice if present in a society gets manifested in its EFL/ ESL materials. The main purpose of the present study is to examine the extent and types of gender bias in two of the Iranian mostly used EFL textbooks at the high school level. Sexism will be investigated in five categories of visibility, firstness, generic masculine constructions, sex-linked occupations and activities. The findings are expected to raise awareness regarding sexist practices in EFL material developers, teachers and learners.


Introduction
The present study has been a by-product of another research currently being carried out by the same researchers, namely Differential Item Functioning (DIF) Analysis of the Iranian Entrance University Examinations. When important decisions are made based on test scores, it is critical to avoid bias, which may unfairly influence examinees' scores. Bias in a test is the presence of some characteristic of an item that results in differential performance for individuals of the same ability but from different ethnic, age, sex, academic, cultural or religious groups. Bias is often attributed to construct-irrelevant dimensions that differentially affect the test scores of different groups of examinees. Sadly so far, the analyses of the results in the study just referred to have revealed considerable DIF across different examinee groups.
As far as gender bias is concerned, which is the main focus of the present paper at the same time, the Iranian entrance university examinations is highly and unfairly biased against women and the main reason for that is undoubtedly the existence of gender-biased high school EFL materials. So the present paper is intended to call the attention of material developers and decision makers in Iran to take action against gender bias in high school EFL materials. Since users of any test, including entrance university examinations, must be confident that scores of all student subgroups can be interpreted in a similar way. The main concern here is perhaps the substantial difference in the mean test scores of different subgroups. Users of the instrument might worry whether such differences are caused by the differences in the skills measured or are an artifact of the instrumentation procedure. In other words, the question here is whether the test measures for various subgroups what it claims to measure! Studies similar in tone to the present one are not new in our country and already date back to almost two decades ago. However they might not have been forceful enough to be noticed and attended to by the authorities. The researchers here would like to ask their colleagues to take part in a serious campaign against gender bias in the Iranian high school EFL materials and replicate similar studies in different voices hoping the echo would get strong enough to be noticed by the decision makers. language often suggests an inherent male dominance and superiority in many fields of life. Sexist language is language that expresses bias in favor of one sex and thus treats the other sex in a discriminatory manner. The word sexism was originally coined to refer to ideas and practices that downgrade women relative to men (Cameron 1985). Nowadays, though, it is used to refer to practices that unfairly treat either sex.
One of the flourishing areas of sociolinguistics in the last decades has been that of language and gender motivated by feminist movements arguing over how language is used as an instrument to discriminate and dominate females (Key 1975;Lackoff 1975). Drawing from linguistic researches on gender representation in the English Language, a large number of studies have been directed towards the investigation of gender bias in language teaching materials, especially in EFL textbooks.
In the EFL context, within the classroom environment it is important that teachers pay attention to gendered language since language is not value-free, lifeless, or free of political bias (Beebe 1996in Cook, 2005. The attitudes repeatedly expressed in books and other media might gradually distort learners' perceptions regarding stereotypes and myth about either sex. Students learning a foreign language might internalize sexist male and female representations reflected in their textbooks. It is thus possible that exposure to gender-biased texts and materials may influence females' behavior in such a way as to restrict their social, behavioral, and linguistic roles. The present paper studies gender bias in two of the widely used Iranian EFL textbooks at the high school level compiled by the Textbook Curriculum Development and Planning Department of the Ministry of Education of the Islamic Republic of Iran. The premise of this investigation is that understanding sexism at the school level is an important, necessary, and often neglected aspect of curricular reform efforts and that such awareness would render more gender unbiased and valid tests. The focus of this study will be the investigation of gender bias in five categories of visual and textual presentation of males and females (visibility/ omission), firstness, masculine generic constructions, occupations, and activities performed by either sex.

Previous Studies
Since 1970s, researchers worldwide have shown interest in the study of gender bias in ESL/ EFL materials. Graham (1975) was among the first to investigate sexism in a school children's dictionary. She found that the nouns used to describe male activities were seven times as many as the nouns used for female activities. In addition, the visibility of males to females was four to one.
In  analyses of a standard dictionary, it was revealed that the visibility of men to women was three to one and that there were five times as many words for things named after males than after females. And the words referring to females were mainly passive compared against the active words used to refer to males. Coles (1977) studied five series of educational materials in which as the results firmly indicated males outnumbered females. Furthermore, females were mainly depicted as stay-at-home housewives and mothers whereas males appeared in a variety of activities and jobs ranging from drivers to doctors.
Hoomes (1978) studied twenty eight high school anthologies. The results indicated that the visibility of females to males was one to three and that men outnumbered women by far in activities and occupations. was concerned.
Johansson and Bachelder (2005) studied four EFL series in terms of four typologies including dialogue initiation, turn taking, number of characters, and number of words used. In the first series Time , men were overrepresented in three out of the four typologies studied. In the second series Wings (2001-03), men overrepresented women in two out of the four typologies. In the next two series Happy (2004-06) and What 's up (2004-07), women overrepresented men in three out of four and all typologies respectively. Ansary and Babaii (2005) performed two types of analyses to examine the manifestation(s) of sexist attitudes and values in two textbooks (Right Path to English I & II). In their quantitative analysis, they investigated sex visibility in both texts and illustrations and female/male topic presentation in dialogs and reading passages. And in their qualitative analysis, they studied sex-linked job possibilities, sex-based activity types, stereotyped sex, firstness and masculine generic conception. Results revealed that the books studied were biased in terms of all the categories investigated.
Lee and Collins (2006) studied gender representation of English textbooks in Hong Kong. Of the twenty recent and earlier books studied, only 6 were reported in their study since their investigations were still on the way. The examined sexism in seven categories including omission (visibility), roles, masculine generic constructions, titles, order of appearance, pictorial representation of both genders and their roles and activities. The comparison of sexism in recent and earlier textbooks indicated that textbook writers had become more conscious regarding sexist practices in some categories. Their study however differed from other studies since they included the investigation of semantic as well as domestic and social roles in their study. Based on the Systemic-Functional concept of transitivity and its primary semantic categories of participant and process (Lock 1996;Halliday 2004), they investigated gender bias across five semantic processes (material, verbal, relational, mental and existential) and five categories of participants. Close examination of various semantic roles revealed subtle gender stereotyping and semantic role shifts in recent versus earlier textbooks (Lee and Collins 2006).

Mineshima (2008) studied gender representations in an EFL textbook Birdland Oral Communication I to investigate
how it portrays the two genders. She examined sexism in three categories including number of females versus males, number of their utterances and firstness. In all categories studied, females were under-represented and overloaded with traditional stereotypical roles whereas males appeared only as cooperative. Mukundan and Nimehchisalem (2008) studied the representation of gender in some English language textbooks in Malaysia. The results indicated significant gender bias against both sexes with males outnumbering females in texts and illustrations and males mainly associated with negative traits. However, discourse wise, males dominated females. They spoke more, talked first, and were more visible. Vogli (2009) investigated gender representation in New Success at First Certificate in six categories of omission, occupation, F/M stereotypes, masculine generic constructions, firstness and adjectives associated to each sex. Despite the efforts the writers had made to portray women and men equally, the findings indicated that men were upfront and women followers. Sometimes they (females) were protagonists in a 'man's world' [but that looked like sugar coat] (Vogli 2009).

Instruments
The instruments used in this study consisted of the 2010-11 Iranian English textbooks for the second and third grades of high school compiled by the Textbook Curriculum Development and Planning Department of the Ministry of Education of the Islamic Republic of Iran.
Book II consisted of seven lessons and Book III of six lessons, each beginning with a new words section, accompanied by a reading passage. After the reading text, comprehension questions followed in forms of True/False, Complete the sentences, etc. Then, there came the speaking and writing sections in different exercise patterns. Other sections which could be noticed were language functions, pronunciation practice, vocabulary review or drill and a vocabulary list.
Both at the beginning and the end of each book, there were some review exercises which were excluded from the gender bias analyses in the present study because they were intended to check the mastery of the students over the EFL materials covered so far since the beginning of the Iranian EFL instruction (junior high school).

Design
The present study is mainly a descriptive research including quantitative and qualitative investigations of the visibility (frequency and nature) of the two genders across five categories including omission, firstness, masculine generic constructions, occupations and activities in two of the Iranian EFL textbooks currently practiced at the high school level.

Procedure
In the present textbook gender bias research, sexism has been studied across five categories including visibility, firstness, masculine generic constructions, occupations and activities associated with either sex in the following order.
First, the visibility of females and males in both texts and illustrations was investigated. Female/male firstness --the number of times when males or females were presented first in texts -was also recorded. To witness whether masculine generic constructions were truly generic (including both sexes) or merely male referenced, the association between all masculine generic nouns and pronouns and their referents was determined as well.
In addition, occupations in which females and males were depicted were identified and compared against. Finally, the type of illustrated and textual activities in which females and males participated was investigated.

Visibility
One of the main manifestations of sexism in textbooks has always been female/ male visibility also called omission. When females do not appear as often as males, the implicit message is that women are not as important as men, or that their accomplishments are not as worthwhile to mention as men's or that they themselves as human beings are not important enough to be included (Porreca 1984:706).
In Iran, according to census statistics, the number of females in recent years has consistently surpassed that of males. And for the textbooks here investigated to be authentic and sex fair, they should reflect such demographic pattern. To see whether this is reflected in the Iranian EFL textbooks or not, the number of textual and visual mentions of male and female characters was counted both in the second and third grade EFL Iranian textbooks at the high school level. Tables 1 and 2 below illustrate the findings.
As indicated in Table 1, it was found that there were 52 male (80%) and 13 female (20%) characters in Book II illustrations, and 76 female (18.8) and 329 male (81.2) characters in Book II texts. Things were not any better in Book III.
In Book III illustrations, out of 59 depictions, 37 (62.7) belonged to males and only 22 (37.3) to females. And in texts, the condition even got worse. Out of 229 mentions, only 38 (16.6) concerned females.
Both tables as presented here reinforce the assumption that women are underrepresented in terms of visibility in number in Iranian high school EFL textbooks in a way that does by no way reflect the reality.

Firstness
Firstness refers to the mention of men first in texts before women as if 'the masculine gender is more worthy than the feminine' (Poole 1946:21;cited in Eckert and McConnell-Ginet 2003:34).
In Iran, men in most cases come ahead of women in most social practices. When walking together for instance, women usually follow their men. In parties and public ceremonies, men are served first. In Iran, it is almost 'Gentlemen first, ladies next!' To investigate gender firstness in Iranian high school textbooks, the present study examined all the instances in the two textbooks in which the two genders were mentioned together and checked which appeared first. Here are the findings.
In Book II, for 12 (85.71%) first mentions of males, there were only 2 (14.28) first mentions of females.
In Book III, there was one out of one instance of male firstness. As indicated in Tables 3 and 4, men in majority of the cases are mentioned first. Such subordination of females as mentioned earlier is deeply rooted in the Iranian culture.

Masculine Generic Constructions
One other manifestation of sexism is the use of the masculine as generic rooted in the grammar of the English language itself. In most cases where words like man, person, etc. and pronouns such as everybody, somebody, etc. are used in EFL textbooks, people hardly ever conceptualize females when reading such constructions (Porreca 1984). Quite recently, however, masculine generics are being avoided in EFL/ ESL textbooks. One strategy is to use paired pronoun expressions such as he/she, him/her, his/her, and s/he.
To investigate this issue, all occurrences of generic constructions were identified and analyzed to see whether masculine generic constructions were truly generics (including both females and males) or exclusively male referenced.
In Book II (page 87), there was only one case of a masculine generic construction that was definitely male referenced as presented below:

Occupation
The depiction of males and females in traditionally stereotypical biased occupations can be another reflection of sexism in EFL textbooks. Table 5 below lists various occupations female and male characters were engaged in as far as texts and pictures were concerned in book II.
Of the 19 (not considering the number of times an occupation was repeated) occupations either mentioned in texts or illustrated in pictures in Book II, 17 different jobs such as engineers, bus drivers, police officers, doctors and bosses belonged to men and in only 2 women were stereotypically depicted (either as teachers or dress makers).
And in Book III, the ratio of appearance of females to males in both texts and illustrations (ignoring repetitions of the same occupation) was 3 to 13. Here again, men were depicted in a variety of manly jobs including wrestlers, pilots, hunters, etc. And women played their ever traditional roles as nurses, teachers, and only once as doctors.
For the word, inventor, there was only one male mention displaying men as ever achievers.

Activities
Both men and women should be shown cleaning, cooking, making household repairs, doing laundry, washing the car, and taking care of the children … Males as well as females can be fearful, weak, mechanically inept, and illogical … Males can be polite, cooperative, inactive, or neat. Because such characteristics are shared by males and females in reality, textbooks that classify them as 'masculine' or 'feminine' are misrepresenting reality (Scott 1981:139 in Mineshima 2008.
Contrary to the above generous quote, in Iran, men work out and women usually stay in, taking hold of the house choirs; taking care of the children, cooking meals, cleaning the house, washing the dishes, etc. The Iranian woman whether working out or staying at home is solely in charge of the household and his man if kind enough may cooperate sometimes.
To examine the presentation of females and males in Books II and III texts and illustrations, the activities in which females and males were depicted were located and studied. Tables 7 and 8 below display the findings.
As can be seen in Table 7, texts and pictures in Book II did reinforce traditionally gendered roles: there were only females engaged in various kinds of household chores. More importantly, of the 40 (not considering the number of repetitions) different activities illustrated in Book II, women were represented in only 13 including cooking the dinner, making the tea, etc.
In Book III as presented below in Table 8, the picture was not much different.

Discussion and Conclusions
The present study was an attempt to re-examine the second and third Iranian EFL textbooks at the high school level to investigate how they portray the two genders both in texts and illustrations after almost two decades of attention to the avoidance of gender biased materials. In every category there was evidence that sexism is still vividly practiced in Iranian EFL materials.
Although females comprise over half the population of the Islamic Republic of Iran and are recently quite active in a variety of social roles, they suffered from low visibility both in texts and illustrations. It seems that the Islamic Iranian culture tends to leave no room for female visibility. A woman would better be invisible from the public eye as much as possible. Within the family, in many cases she is not welcome in men gatherings. Out, she still gets called using her eldest son's name (by her husband) rather than her own to avoid visibility. Upon her death, on her funeral notice her photo usually gets replaced with a flower.
Since textbooks reflect the social, cultural and religious ideologies and perceptions of their writers, it is no doubt there is not much room for women visibility in the Iranian EFL textbooks at the high school level.
Regarding firstness, again in majority of cases men were mentioned before women as is the case in the present Iranian culture especially in traditional families. In the two books studied here the masculine generic was in addition solely male referenced. Considering occupation, gender inequality was perceived even more. In the Islamic Republic of Iran today, women comprise a high percentage of the work force, and a great number of women are involved in highly skilled, professional jobs. However, in two Iranian high school textbooks investigated, this was not realistically depicted. Women were mainly illustrated in traditional sex-linked occupations such as teaching and doing household chores. This is far from the reality and equity. Currently women are proving more capable of the traditionally marked occupations, they are even seen as bus drivers in bigger cities where globalization is on the way! Regarding the depiction of males/ females in activities, women were overall stereotyped as stay-at-home mothers and mainly engaged in household choirs including room cleaning, making the tea, and baking the cake and only sometimes had the chance of studying or watching TV. They were usually depicted as over-emotional and careless creatures that cried after TV films (Book II, p. 44) and more often than not lost things and only then busied themselves finding them (Book II,pp. 8 & 34). Whereas Men were mostly busy playing football or Ping-Pong, reading newspapers, fixing the car, swimming, finding a new job, buying different things, etc. Males as breadwinners were always involved in buying things. The verb buy thus always collocated with a male agent.
Overall, the two textbooks investigated overrepresented male characters both linguistically and visually in frequency and order of occurrence, occupation, stereotypical activities, and the linguistic manifestation of masculine generic referents.
Sexism, it seems, is so deeply ingrained in our culture, our language, and our subconscious that it is difficult for us to avoid it in the production of language teaching materials (Florent et. al. 1989: 183 in Vogli 2009). And the question is how come gender bias reviews and critiques of the Iranian EFL materials throughout all these years have been left unnoticed and for almost 20 years or so the very same unmodified biased high school materials have been published and republished over and over again without the slightest change.

Implications
One implication of this study might be to raise awareness and consciousness in EFL material developers regarding gender prejudiced materials so as to help them initiate modifications regarding such inequities after almost 20 years of publishing the same gender-bias loaded materials with no change. Material developers and curriculum designers should pay attention to and consider the guidelines of gender-fair material development. There are several working guidelines in literature including On Balance (Florent et. al., 1994), Teachers, in addition, if made aware can deal with gender-biased materials logically and present them in an unbiased way (Sunderland et. al., 2001). They can select their own teaching materials with more care if allowed and if needed 'where the input of the course book reflects sexist practices or attitudes, … comment on them. Critical reading and listening may thus become regular part of the suggested pedagogy' (Sunderland 1992:87). They should also beware not to attribute any specific roles, domestic or societal to either gender so as not to inculcate any preconceptions in the learners (Mineshima 2008