Need help about this Essay

Sex Roles, Vol. 42, Nos. 3/4, 2000

No Sissy Boys Here: A Content Analysis of the Representation of Masculinity in Elementary School Reading Textbooks

Lorraine Evans University of Georgia

Kimberly Davies1 Augusta State University

In this study, we investigate the portrayal of gender characteristics in elemen- tary school reading textbooks. Over the past 25 years, most of the research on textbooks has focused on female roles and characteristics. In this research, we focus particularly on how males are portrayed. Using an evaluative instru- ment based on the Bem Sex Role Inventory, we examine first, third, and fifth grade literature textbooks. In particular, we analyze traits pertaining to masculine and feminine stereotypes. Our results show that despite publisher’s guidelines and Title IX, males are still primarily portrayed in a stereotypical light. Males are overwhelming shown to be aggressive, argumentative, and competitive.

In the schoolroom more than any other place, does the difference of sex, if there is any, need to be forgotten.

Susan B. Anthony, 1856

School is formally charged by law as a legitimate agent of socialization to educate our children (Richardson, 1977). Title IX of the Education Amendments Act of 1972 prohibits sex discrimination against students or employees in any federally funded program (Sandler, 1977). With its pas- sage, sexism became more than a topic of philosophical discussion; it became a legal issue in the schools of the United States. Title IX applied to virtually all aspects of student life: sports, testing, rules, regulations, and textbooks. 1To whom correspondence should be addressed at Department of Sociology, Augusta State University, 2500 Walton Way, Augusta, Georgia 30904-2200 (e-mail: [email protected]).

255

0360-0025/00/0200-0255$18.00/0  2000 Plenum Publishing Corporation

256 Evans and Davies

Publishing houses have recognized the importance of ensuring equal representation of all people in textbooks, and most of them have printed guidebooks emphasizing the effects that textbooks could have on children. In 1975, Macmillan, a major textbook publisher, stated that ‘‘children are not simply being taught mathematics and reading; they are also learning, sometimes subliminally, how society regards certain groups of people’’ (quoted in Britton & Lumpkin, 1977). Even earlier, in 1973, Ginn and Company went further, noting that, ‘‘Educational materials teach far more than information and a way of learning. In subtle, often unconscious ways, the tone and development of the content and the illustrations foster in a learner positive or negative attitudes about self, race, religion, regions, sex, ethnic and social class groups, occupations, life expectations, and life chances’’ (quoted in Britton & Lumpkin, 1977).

In addition to these publishers, scholars have also noted the importance of textbooks in teaching children gender behaviors. In a 1984 meta-analysis, Schau and Scott found that instructional materials affect students when it comes to sex equity. They identified and summarized over 40 studies which examined the effects on gender associations, sex-role attitudes, material preference, and comprehension among students. Their results included the effects of gender-biased language—the all-inclusive, generic, ‘‘man,’’ for example, leads to male gender associations overall, whereas unspecified language, e.g., ‘‘people,’’ reduces, although it does not eliminate, the ten- dency to think of men only. The studies reviewed by Schau and Scott demonstrated that among children, sex-equitable materials were associated with more flexible sex-role attitudes, whereas sexist material contributed to more sex-typed attitudes. Schau and Scott (1984) argue that for fulfill- ment of individual potential, a flexible gender-role attitude is necessary, and thus equitable portrayals of male and female characters with both masculine and feminine traits are needed in children’s textbooks.

PREVIOUS STUDIES

Clearly, school is a social experience in which social values and attitudes are transmitted, and textbooks are agents of this transmission. Previous research on textbooks has been based on the premise that books are a powerful tool in shaping children and their views of society during their formative years. It is therefore important that the content of our children’s textbooks be studied to reveal what messages are being conveyed through the authority of textbooks in the education of our youngest citizens.

While textbook publishers and researchers have noted the importance of displaying positive characteristics for both males and females in

Representation of Masculinity in Textbooks 257

textbooks, examinations of textbooks before and since the passage of Title IX have found little equality in the portrayal of males and females; the ‘‘hidden curriculum’’ continues. Scholars who examined textbooks published prior to 1980 found an underrepresentation of women and girls as main characters, extreme stereotyping of female characters when shown, very few depictions of women in occupational roles, and negative displays of feminine characteristics (Britton & Lumpkin, 1977; Oliver, 1974; Schnell & Sweeney, 1975; Weitzman & Rizo, 1974). In most studies published prior to 1980, male characterizations were not viewed as an area worthy of comment beyond noting that depictions of men and boys overshadowed that of women and girls. An exception is the study by Frasher and Walker (1972), who examined roles displayed by both males and females in early reading textbooks from four basal reading series and found males in only a narrow range of roles. According to Frasher and Walker, boys in these textbooks were shown as noisy and conveying a higher level of aggression through competition and assertiveness than girls in the textbooks. Few adult men were in fathering or caregiving positions, and ‘‘never did a boy of any age play with a doll or house, sew or pick flowers’’ (Frasher & Walker, 1972, p. 744). Frasher and Walker, however, did not compare the different series to determine if one publisher displayed men and boys in a more narrow range than another.

During the 1980s and 1990s, published research on sexism in elemen- tary school textbooks resumed focusing solely on the portrayal of girls. Scholars approached the issue of sexism in textbooks by asking questions such as the following about the books they examined: (1) What percentage of the main characters are female (Hitchcock & Tompkins, 1987; Rupley, Garcia, & Longnion, 1981)? (2) What are the occupations of female main characters (Hitchcock & Tompkins, 1987)? and (3) To what extent are characters gender stereotyped (Scott, 1981)?

While the research during the 1980s and 1990s continued to focus more on the portrayal of girls and women, the research that compares characterizations of females to the characterizations of males does give us some indication of how boys and men are portrayed. For instance, Scott (1981) examined the characteristics of male and female main characters in two 1978 basal series readers and found that males were more likely to be shown in a stereotypical light than were females. Scott, who did not compare the role behavior of characters across the two series, found, overall, that males were displayed with stereotypical masculine characteristics such as aggressiveness, competency, and independence, whereas girls were depicted as passive, nurturing, and dependent. In fact, most characters portrayed counter to their gender stereotypes were female (70%). According to Scott,

258 Evans and Davies

the portrayal of stereotypical males continues because sexism, for boys, has not yet been properly addressed, therefore roles depicting stereotypical masculinity are the norm for male characters.

Witt (1996) found similar results in her examination of 1993 and 1995 third grade textbooks from six publishers. While Witt’s examination is limited to third grade textbooks, she does include percentages of masculine and feminine traits by publisher. She found that while female characters possess a balance of masculine and feminine traits in third grade readers for five of the six publishers, male characters are depicted primarily in traditional roles with masculine characteristics by all six publishers. Witt concludes that any progress made in minimizing sexism in third grade basal readers relates only to the female depictions; males are still shown in a stereotypical light.

Purcell and Stewart’s (1990) examination of 62 readers from four basal series in use in three Texas cities in 1989 appears to cover more than one grade. However, they do not directly address the issue of the gender por- trayal of boys nor do they compare the publishers which they examined. Instead, Purcell and Stewart present the results of a content analysis of active and passive themes by gender in the readers they examined. While they note that ‘‘boys are still portrayed many times as being forced to deny their feelings to show their manhood’’ (Purcell & Stewart, 1990, p. 184), they do not attempt to interpret their results. However, their content analysis indicated that boys depicted in the textbooks they examined were involved in more imaginative, explorative, and adventurous play than were the girls.

Despite the fact that Vaughn-Roberson, Tompkins, Hitchcock, and Oldham (1989) do not compare their results across publishers, their exami- nation of first, second, fifth, and sixth grade basal reader textbook series for six publishers in 1985 or 1986 is the most comprehensive analysis of male characters in elementary school readers to date. Using Bem’s sex- role inventory characteristics, they found that male characters exhibited mostly masculine characteristics and were infrequently displayed with femi- nine traits. The most common feminine characteristic found in the boys, according to Vaughn-Roberson et al., was ‘‘sensitive to needs of others.’’ They conclude that while boys and girls are beginning to see some crossover in the display of gender characteristics in textbooks, ‘‘this suspected trend must be continually monitored to determine its strength’’ (Vaughn-Rober- son et al, 1989, p. 68).

In this study, we extend the work of Vaughn-Roberson et al. by examin- ing the display of gender among males in reading texts by two publishers in textbooks published and read by children during the 1990s. Research suggests that from a young age, boys adhere more closely to gender-appro- priate behaviors than girls (Henshaw, 1992). Thus, it is particularly impor-

Representation of Masculinity in Textbooks 259

tant to examine how textbooks display men and boys. Furthermore, while Vaughn-Roberson et al. (1989) include six publishers in their study, like most other scholars (e.g., Frasher and Walker, 1972; Purcell and Stewart, 1990; Scott, 1981), Vaughn-Roberson et al. do not make explicit compari- sons between the publishers to determine if one portrays males significantly different than the others. In this research, we set out to answer three questions. First, are males in textbooks portrayed in a manner in which they cross traditional boundaries of masculinity? In other words, are males depicted with both stereotypically masculine and feminine traits? Or are the males in elementary school reading textbooks more likely to be depicted with masculine characteristics only, as Witt found in her examination of 1993 and 1995 third grade readers? Second, do two of the leading publishers differ in how they portray males in their textbooks? Could it be that one publisher portrays males in a more stereotypical light than another? Third, does the portrayal of masculine and feminine traits in males vary by grade? Does grade level matter when examining portrayals of masculinity in ele- mentary school textbooks?

METHODS

In order to investigate the portrayal of masculinity and femininity among male characters in textbooks, we conducted a content analysis of two major reading textbook basal series. Basal reading series are designed for developmental reading instruction, with a supplementary reading series for children with reading difficulties. We examined the 1997 Macmillan McGraw Hill basal series, Spotlight on Literacy, and the 1997 Silver Burdett Ginn series, the Literature Works. We selected these two reader series for two reasons. First, Macmillan McGraw Hill and Silver Burdett Ginn are two of the main publishers of basal readers in the United States (Schmitt & Hopkins, 1990) and second, they are the two series used in our city and we were particularly curious about what the children in our community were reading.

To examine a range of textbooks, we examined the first, third, and fifth grade books. We selected first grade books since they initialize the reading sequence, and third and fifth grade books because national stan- dardized testing takes place during these years. There were a total of six books in the Macmillan McGraw Hill series. There were 3 first-grade books, 2 third-grade books, and 1 fifth-grade book. In the Silver Burdett Ginn series, there were 4 first-grade books, 2 third-grade books, and 1 fifth-grade book, making a total of 7 in this set. Thus, we analyzed 13 books.

For the purpose of this study we analyzed only the written fiction

260 Evans and Davies

stories to determine how masculinity and femininity are introduced to young readers. While an examination of the pictures would be interesting, we believed that reliability would be stronger if we focused on the written word only. Fiction stories which we examined included fantasy, excerpts from novels, science fiction, folk tales, and fabricated real-life situations. We limited the analysis to fiction, as it is fabricated and thus may be more easily molded to reflect idealized versions of male and female. Furthermore, in many previous studies, scholars noted that they analyzed ‘‘stories’’ (Frasher & Walker, 1972; Purcell & Stewart, 1990; Scott, 1981; Vaughn- Roberson et al, 1989; Witt, 1996). Further, Purcell and Stewart (1990) categorized the stories into types (child, animal, adult, or folktale) and Scott (1981) divided hers into four groups by sex of main characters (male, female, male and female, and neither) and Frasher and Walker (1972) limited their stories analysis to those in which humans were the main characters. However, no definition of ‘‘story’’ is provided in any of these studies. The clarification of story type in this study should facilitate future analyses of basal readers. Altogether, there were 97 fiction stories in the 13 books.

We formulated an instrument to tabulate the personality traits of the main characters in the written stories in order to record how gender is depicted in these textbooks. We used Richardson’s (1983) guidelines and the Bem Sex Role Inventory (Bem, 1981) to develop our instrument. Rich- ardson, a sociologist, notes that much can be learned by studying personality traits of characters in children’s stories. While it may seem that males and females are equally represented, an analysis of character traits which reveals that the males all have stereotyped ‘‘masculine’’ personalities would indi- cate that the book has not eliminated sexism (Richardson, 1983, p. 76). Our final evaluation instrument included a total of 16 traits, 8 masculine and 8 feminine. The masculine traits included aggressive, competitive, argu- mentative, decisive, assertive, risk-taker, self-reliant, and adventurous. The feminine included nurturing, affectionate, tender, understanding, passive, impetuous, emotionally expressive, and panicky. We developed operational definitions for each of these terms by reading dictionary definitions and discussing what they meant broadly in our society. Table I includes a brief description for each of these masculine and feminine traits.

To be included in the tabulation of gendered personality characteristics, the written story had to include a reference to sex: ‘‘he’’ or ‘‘she.’’ Animal characters were included only if sex-typed. As a result, 82 of the 97 stories were analyzed for their portrayal of gender characteristics.

The main characters, those who were the focus of the story, served as the units of analysis in this examination of personality traits displayed in children’s textbooks. A character was considered a main character if he or

Representation of Masculinity in Textbooks 261

Table I. Definitions of Personality Traits as Used in Basal Reading Group Series

Personality Traits Definition

Masculine traits Adventurous Actively exploring the environment, be it real or imaginary Aggressive Actions and motives with intent to hurt or frighten; imparts

hostile feelings Argumentative Belligerent; verbally disagreeable with another Assertive Taking charge of a situation, making plans and issuing instruc-

tions Competitive Challenging to win over another physically or intellectually Decisive Quick to consider options/situation and make up mind Risk-taker Willing to take a chance on personal safety or reputaion to

achieve a goal Self-reliant Can accomplish tasks or handle situations alone with confi-

dence

Feminine traits Affectionate Openly expressing warm feelings; hugging, touching, holding Emotionally Allowing feelings to show, including temper tantrums, crying,

expressive or laughing Impetuous Quick to act without thinking of the consequences; impulsive Nurturing Actively caring and aiding another’s development, be it physi-

cally or emotionally Panicky Reacting to situation with hysteria; crying, shouting, running Passive Following another’s lead and not being active in a situation Tender Handling someone with gentle sensitivity and consideration Understanding Being able to see and comprehend a situation from another

person’s perspective; showing empathy

she was the protagonist. Main characters generally appeared throughout the story. Since these stories were written for young children, most of the stories had only one or two main characters.

The first author read each story and recorded the traits of these main characters. The traits were recorded based on the holistic portrayal of the main character or characters throughout the story, not just individual incidents in the story (Frasher & Walker 1972; Purcell & Stewart, 1990). Thus, if a character exhibited any of the relevant traits, the trait was checked once for the character, regardless of the number of times he or she expressed the character trait. The trait was coded only once partly because of the simplicity of many of the stories and also for accuracy of character portrayal. Stipulating each and every incident of one characteristic for a particular character would be redundant since we were examining the overall charac- ter portrayal within the written text. Thus, in our coding scheme, a character who was described as crying in two sentences was not considered twice as emotionally expressive as a character who was described as crying in only one sentence; each character would simply be regarded as emotionally expressive. Additionally, conducting our analysis in this fashion allows us to

262 Evans and Davies

compare our results with the results of other scholars in this field (Frasher & Walker 1972; Purcell & Stewart, 1990). To ensure reliability of coding, 16 Macmillan McGraw Hill series stories (nearly 20%) of the 82 stories were also coded by a second reader. The second reader agreed with the first reader just over 90% of the time, indicating a reliable coding system.

RESULTS

One hundred and thirty-two characters in 82 stories were analyzed for their portrayal of personality traits. Seventy-one (54%) of the characters were male and 61 (46%) were female, a relatively even split of the sexes compared to previous study results (Britton & Lumpkin, 1977; Oliver, 1974; Rupley et al., 1981; Schnell & Sweeney, 1975; Weitzman & Rizo, 1974). However, when the numbers were broken down into personality traits, a pattern emerged which reinforced the stereotyping of males.

Table II presents the percentage of main characters exhibiting mascu- line and feminine characteristics by sex. As can be seen in this table, chi-square analyses indicated that males were portrayed with traditionally

Table II. Percentage of Main Characters Exhibiting Mascu- line and Feminine Characteristics by Sex

Male Female Characters Characters

Personality Traits (n � 71) (n � 61)

Masculine traits Adventurous 26.7%(19) 19.6%(12) Aggressive** 23.9%(17) 4.9%(3) Argumentative* 21.1%(15) 6.5%(4) Assertive 28.1%(20) 21.3%(13) Competitive** 33.8%(24) 11.4%(7) Decisive 19.7%(14) 13.1%(8) Risk-taker 25.3%(18) 19.6%(12) Self-reliant 40.8%(29) 36.0%(22)

Feminine traits Affectionate* 18.3%(13) 33.3%(21) Emotionally 14.0%(10) 33.3%(21)

expressive** Impetuous 16.9%(12) 8.1%(5) Nurturing 15.4%(11) 16.3%(10) Panicky 15.4%(11) 21.3%(13) Passive** 8.4%(6) 29.5%(18) Tender* 7.0%(5) 19.6%(12) Understanding 23.9%(17) 29.5%(18)

*p � .05 (two-tailed test), **p � .01 (two-tailed test).

Representation of Masculinity in Textbooks 263

masculine characteristics more often than females. Males overall were shown as significantly more aggressive, argumentative, and competitive than females. Almost 24% of the males were aggressive, just over 21% were argumentative, and nearly 36% were competitive, as compared with 4.9%, 6.5%, and 11.4% for females, respectively. Surprisingly, the number of males displaying the stereotypical feminine traits of impetuous, nurturing, panicky, and understanding was not significantly different than the number of females displaying these same traits. Just 1% fewer males than females were described as nurturing and nearly 24% of the males were understand- ing. However, males were significantly less likely than females to be de- scribed as affectionate, emotionally expressive, passive, or tender. Only 8.4% of the males were passive, 7% were tender, 18.3% were affectionate and 14% were emotionally expressive. However, 29.5% of the females were described as passive, 19.6% were tender and one third (33.3%) were described as affectionate and emotionally expressive.

We also performed chi-square analyses to determine if the number of main characters exhibiting masculine and feminine personality traits dif- fered by sex within each textbook series (Table III). In the Silver Burdett Ginn (SBG) series, males were significantly more likely than females to be described as aggressive (�2 � 6.66, 1 df, p � .01) competitive (�2 � 7.07, 1 df, p � .01), and argumentative (�2 � 4.68, 1 df, p � .05) and less likely than females to be described as passive (�2 � 5.81, 1 df, p � .05). More than 30% of the males were portrayed as aggressive in the SBG series compared to only 6.4% of the females. Over 40% of the males and nearly 13% of the females were displayed as competitive, whereas 26.3% of the males and 6.4% of the females were described as argumentative. In contrast, only 5.2% of the males compared with 25.8% of the females were portrayed as passive in the SBG series.

For the Macmillan McGraw Hill (MMH) series, chi-square analyses revealed that male characters were significantly more likely than females to be portrayed as risk-taking (�2 � 6.02, 1 df, p � .05) and assertive (�2 � 4.95, 1 df, p � .05) and significantly less likely than females to be described as affectionate (�2 � 4.92, 1 df, p � .05), emotionally expressive (�2 � 4.45, 1 df, p � .05), and passive (�2 � 4.09, 1 df, p � .05). Looking at the stereotypically masculine traits, we see that 33.3% of the males were described as assertive and 36.3% were portrayed as taking risks, whereas 10% of the females were displayed with each of these two characteristics in the MMH series. As for stereotypically feminine traits, three traits were found to be statistically significantly different between male and female characters. Just over 15% (5 of 33) of the males were displayed as affection- ate, 9% were shown to be emotionally expressive, and 12.1% were displayed as passive. In comparison, 40% of the female characters in the MMH

264 Evans and Davies

Table III. Percentage of Main Characters Exhibiting Masculine and Feminine Characteristics by Sex and Textbook Series

Male Characters Female Characters

Personality SBGinn MMHill SBG MMHill Trait (n � 38) (n � 33) (n � 31) (n � 30)

Masculine traits Adventurous 23.6%(9) 30.3%(10) 25.8%(8) 13.3%(4) Aggressivea 31.5%(12) 15.1%(5) 6.4%(2) 3.3%(1) Argumentativeb 26.3%(10) 15.1%(5) 6.4%(2) 6.6%(2) Assertivec,d 23.6%(9) 33.3%(11) 32.2%(10) 10.0%(3) Competitivea 42.1%(16) 24.2%(8) 12.9%(4) 10.0%(3) Decisive 26.3%(10) 12.1%(4) 18.1%(6) 6.6%(2) Risk-takerd 15.7%(6) 36.3%(12) 29.0%(9) 10.0%(3) Self-reliant 36.8%(14) 45.4%(15) 35.4%(11) 36.6% (11)

Feminine traits Affectionated 21.0%(8) 15.1%(5) 29.0%(9) 40.0%(12) Emotionally 18.4%(7) 9.0%(3) 38.7%(12) 30.0%(9)

expressived

Impetuous 13.1%(5) 21.2%(7) 9.6%(3) 6.6%(2) Nurturing 18.4%(7) 12.1%(4) 16.1%(5) 16.6%(5) Panicky 13.1%(5) 18.1%(6) 22.5%(7) 20.0%(6) Passiveb,d 5.2%(2) 12.1%(4) 25.8%(8) 33.3%(10) Tender 10.5%(4) 0.3%(1) 22.5%(7) 16.6%(5) Understanding 26.3%(10) 21.2%(7) 38.7%(12) 20.0%(6)

Note: SBG, Silver Burdett Ginn; MMH, Macmillan McGraw Hill. There are no significant difference in portrayal of traits between the two series for male characters. aThere is a significant difference between male and female characters at the .01 level between series for female characters for the Silver Burdett Ginn series for this trait.

bThere is a significant difference between male and female characters at the .05 level between series for female characters for the Silver Burdett Ginn series for this trait.

cThere is a significant difference at the .05 level for the trait assertiveness between series for female characters for this trait.

dThere is a significant difference between male and female characters at the .05 level between series for female characters for the Macmillan McGraw Hill series for this trait.

series were affectionate, 30% were emotionally expressive, and 33.3% were described as passive.

Interestingly, while we found differences between the two publishers in their portrayals of males and females, a chi-square analysis between the two series revealed no significant differences in any of the characteristics which the males displayed. Each of the traits which we examined were just as likely or unlikely to be portrayed by males in the Silver Burdett Ginn series as in the Macmillan McGraw Hill series.

When we compared the portrayal of masculine and feminine character- istics between the male and female characters by grade, we found no difference between sexes in the first grade readers (Table IV). However, when we compared the characters for the fifth and third grade readers, chi-

Representation of Masculinity in Textbooks 265

Table IV. Percentage of Main Characters Exhibiting Masculine and Feminine Characteristics by Sex and Grade

Male Characters Female characters

1st Gradea 3rd Grade 5th Grade 1st gradea 3rd grade 5th grade Trait (n � 12) (n � 33) (n � 26) (n � 11) (n � 27) (n � 23)

Masculine traits Adventurousb 50.0%(6) 9.0%(3) 38.4%(10) 18.0%(2) 11.1%(3) 30.4%(7) Aggressivec 16.6%(2) 27.0%(9) 23.0%(6) 9.0%(1) 0.0%(0) 8.6%(2) Argumentatived 16.6%(2) 15.1%(5) 30.7%(8) 0.0%(0) 0.0%(0) 17.3%(4) Assertive 25.0%(3) 33.0%(10) 26.9%(7) 18.0%(2) 18.5%(5) 26.0%(6) Competitivea 16.6%(2) 33.0%(10) 46.1%(12) 0.0%(0) 3.7%(1) 26.0%(6) Decisive 16.6%(2) 18.2%(6) 23.0%(6) 9.0%(1) 7.4%(2) 21.7%(5) Risk-taker 25.0%(3) 24.0%(8) 26.9%(7) 0.0%(0) 18.5%(5) 30.4%(7) Self-reliant 33.3%(4) 42.4%(14) 42.3%(11) 18.0%(2) 37.0%(10) 43.4%(10)

Feminine traits Affectionateb,d 50.0%(6) 6.0%(2) 19.2%(5) 36.0%(4) 29.6%(8) 39.1%(9) Emotionally 25.0%(3) 9.0%(3) 15.3%(4) 27.0%(3) 25.9%(7) 47.8%(11)

expressivee

Impetuousd 16.6%(2) 21.2%(7) 11.5%(3) 9.0%(1) 0.0%(0) 17.3%(4) Nurturing 16.6%(2) 15.1%(5) 15.3%(4) 18.0%(2) 22.2%(6) 8.6%(2) Panicky 0.0%(0) 24.2%(8) 11.5%(3) 9.0%(1) 22.2%(6) 26.0%(6) Passivec,f 8.3%(1) 0.0%(0) 19.2%(5) 27.0%(3) 37.0%(10) 21.7%(5) Tender 16.6%(2) 6.0%(2) 3.8%(1) 27.0%(3) 14.8%(4) 21.7%(5) Understanding 16.6%(2) 21.2%(7) 30.7%(8) 9.0%(1) 33.3%(9) 34.7%(8)

aThere are no significant differences between 1st grade male and female characters. bThere is a significant difference at the .01 level between grades for male characters for this trait. cThere is a significant difference between 3rd grade male and female characters at the .01 level for this trait.

dThere is a significant difference between 3rd grade male and female characters at the .05 level for this trait.

eThere is a significant difference between 5th grade male and female characters at the .05 level for this trait.

f There is a significant difference at the .05 level between grades for male characters for this trait.

square analyses revealed some statistically significant differences. In fifth grade texts, males were significantly less likely (15.3%, �2 � 6.04, 1 df, p � .05) to be displayed as emotionally expressive than were females (47.8%). In the third grade readers, chi-square analyses revealed several statistically significant differences between the portrayal of male and female characters. Males in the third grade readers were significantly more often portrayed as aggressive (�2 � 8.66, 1 df, p � .01), argumentative (�2 � 4.46, 1 df, p � .05), competitive (�2 � 7.01, 1 df, p � .01), and impetuous (�2 � 6.48, 1 df, p � .05) and less often displayed as affectionate (�2 � 5.93, 1 df, p � .05) and passive (�2 � 14.66, 1 df, p � .01) than their female counterparts. Twenty-seven percent of the males were displayed as aggressive, 15.1% were argumentative, 33% were competitive, and 21.2% were impetuous, whereas none of the female characters were shown as aggressive, argumen- tative or impetuous and only one (3.7%) female character in the third grade readers was displayed as competitive. In contrast, 29.6% of the females in

266 Evans and Davies

the third grade readers were shown being affectionate, whereas only 6% of the males in these readers were shown as affectionate. Finally, none of the males in the third grade readers were displayed as passive as compared with 37% of the female characters.

DISCUSSION

In our content analysis of two major reading textbook basal series, we found that males were depicted with both stereotypically masculine and feminine traits. However, we did not find a great crossing of traditional boundaries of masculinity. Masculine traits still outnumbered feminine traits among male characters. In comparing the two series, we found that neither publisher portrayed the males in a more stereotypical light than the other. A comparison of the readers by grade levels, however, led to the discovery that gender was more dichotomized in the third grade readers than in the first or fifth grade readers.

Thorne (1993) reported, from her research about children’s behavior in schools, that boys and girls recognize acceptable behavior for their gender at early ages. She noted that this is often reinforced, perhaps unwittingly, by teachers and school staff. Girls are more group-oriented, whereas boys maintain their argumentative and aggressive behaviors. Boys’ aggressive behaviors are thought to be natural or unchangeable and are often excused with the rationale that ‘‘boys will be boys.’’ As noted by Thorne, however, these behaviors are, in fact, socially reinforced by such rationales, and as this reading textbook analysis reiterates, socially sanctioned by the schools themselves.

In one story from a fifth grade book, for example, the display of male aggressiveness is noteworthy. In this story, a boy wants to be in charge of the fair project; he is the biggest and looks at his raised fist while glancing at the other children to signify no one was to argue. No one did. In other stories, the adult males are shaking their fists at other males, shouting and often chasing them. Unfortunately, it is often little boys causing the trouble, so there is the double impact of out-of-control youths and angry men.

Males acting or behaving in socially acceptable ways rarely display feminine traits. As Witt (1996) notes, society pays lip service to wanting males to engage in a variety of behaviors, but in actuality only masculine traits are acceptable. The infrequency of feminine traits displayed in the textbooks studied reiterates to boys that these types of characteristics are not pertinent to male behavior. The basic definition of masculinity is having qualities appropriate to, or usually associated with, man—that which is not female.

Representation of Masculinity in Textbooks 267

While aggressiveness may be viewed as negative at times, it reaffirms the ‘‘normal’’ male stereotype and is often excused and accepted as ‘‘natu- ral’’ masculine behavior, whereas the open display of traits from the oppo- site end of the gender dichotomy raises questions concerning male sexuality. There is a collapse of sexuality onto gender behaviors, especially for boys, manifesting into the overall avoidance of the incorporation or obvious portrayal of feminine traits among males. Our results reflect this uneasiness. Looking at Table II, one can see that fewer than one fifth of the male characters display seven of eight feminine traits we examined.

The label of ‘‘sissy’’ is given to boys who avoid tough aggressive play, including sports, and those who exhibit any ‘‘weakness’’ through affection or tenderness. This is an obvious manifestation of the devaluation of feminine traits. Studies done in the 1970s by Green were targeted toward these ‘‘sissy’’ effeminate boys, as he believed they were at risk of becoming transsexuals as adults (Burke, 1996). ‘‘At-risk’’ behavior for boys included having too much interest in domestic chores, wanting to interact with girls in quiet play rather than with boys in active, aggressive play, and expressing a distaste for overtly masculine pastimes (like playing war or sports). De- spite the focus on sex equity in educational material in the last two decades, none of these ‘‘at-risk’’ behaviors were evident in this textbook analysis. Gender behavior for boys is still relatively static. Our results coincide with Witt’s conclusion that ‘‘male characters are being portrayed in the same way they were 20 years ago’’ (Witt, 1996, p. 316).

Furthermore, chi-square analyses for the different grade levels indicated that the most stereotypical portrayals of male and female characters in the written texts occurs in the third grade readers (the year Witt examines in her 1996 article). Male characters in the third grade readers were significantly more likely to be aggressive, competitive, and argumentative, three stereo- typically masculine characteristics which may be viewed in a more negative light than the other stereotypical masculine traits we were examining such as being decisive and adventurous. Similarly, female characters were more likely than male characters to be described as affectionate and passive in the third grade texts. Why are the third grade textbooks the most polarized? Why are we more likely to see these relatively more negative stereotypical masculine traits among males in the third grade texts? What could this mean to the socialization of young boys and girls at this time in their development? Most third graders are 8 or 9 years old. Could it be that we see these strong messages of masculine-typed and feminine-typed roles for boys and girls, respectively, right before boys and girls are often encouraged to begin notic- ing one another as potential heterosexual partners?

Also interesting (and unexpected) when making comparisons between the male and female characters in the third grade texts is the fact that males

268 Evans and Davies

in the third grade texts were more likely than their female counterparts to be impetuous—a characteristic which we had defined as feminine. What are we to make of this? In thinking about why we saw more males than females being portrayed as impulsive in the third grade texts, we wondered whether it could be the result of no males being depicted as passive, or perhaps there was something about the ages of the main characters in these stories? It remains a curiosity. If nothing else, the differences in textbooks by grade levels demonstrate the need for examinations of textbooks at different levels.

Our examinations of two different textbook publishers did not reveal that males were portrayed differently in the two series. However, as we noted above, our comparison of male and female characters in the 1997 Silver Burdett Ginn (SBG) and Macmillan McGraw Hill (MMH) textbooks reveals that within the SBG series, males were more aggressive, competitive, and argumentative than the females and within the MGH series, males were more risk-taking and assertive than the females. Thus, it seems that the SBG stories are more likely to feature males (as compared with females) in a less positive stereotypical masculine light than do the MMH stories. Because we know that children are influenced by what they read, differences between the textbook publishers could prove to be important for school boards, parents, and society as we select our textbooks.

Gender behavior as displayed by girls and women is relatively fluid. It seems that girls who possess or exhibit male traits are not ostracized by society but complimented, and many women reflect positively on the label of tomboy (Thorne, 1993). The move between traditional masculine and feminine traits is more acceptable for girls, but they usually do not incorpo- rate the aggressiveness displayed in males. This is also reflected in our findings. Females were more often portrayed similarly to males when we look at the masculine characteristics of assertiveness and self-reliance, rather than the more negative or extreme masculine attributes of aggressive- ness and competitiveness.

CONCLUSION

Even though our results indicate a greater numerical equality of males and females displayed in textbooks as compared with previous studies (Britton & Lumpkin, 1977), the manner in which males and females are depicted through personality traits is still sexist. Males are overwhelmingly more often portrayed as aggressive, argumentative, and competitive, whereas females are more likely to be characterized as affectionate, emo- tionally expressive, and passive. These findings contrast with the expecta-

Representation of Masculinity in Textbooks 269

tions that publishing house guidelines established in their efforts to create nonsexist literature in textbooks.

Textbook publishers and their watchdogs seem to be trying to erase sexism through expanding female characters with the inclusion of masculine traits, but feminine traits in males need to be valued and recognized, too. True equality in which all people can reach their individual potential will never be achieved if only one side of the continuum is affected by change. Both boys and girls need role models to help them develop all aspects of their personality. Our society needs citizens who are flexible, who will choose activities based on individual ability rather than societal expectations (Katz, 1996). Just as girls should be able to have a career and play profes- sional sports, boys should be able to be nurturing and raise children. Though recent events such as the shootings at Columbine High School in Colorado have not been fully examined, there are some who suggest that these crimes and others occur because males know few outlets for their emotions. Is it a total coincidence that we see males in textbooks being aggressive, competitive, and argumentative and boys in our schools as aggressive?

When discussing gender characteristics in textbooks, Schau and Scott (1984) define sexist and sex-equitable literature. Sexist literature includes stereotypical portrayals of males and females, whereas sex-equitable may be measured on a continuum which ranges from sex-fair, characters portraying both traditional and nontraditional traits, to sex-affirmative, which has the emphasis on characters incorporating nontraditional traits. It seems that textbooks have yet to become sex-equitable, much less sex-affirmative.

Research, past and present, has yet to demonstrate that reading text- book publishers are adequately addressing the needs for nonsexist material, even as established in their own guidelines. Schau and Scott’s (1984) text analysis revealed that while there were some sex-affirmative materials avail- able directed toward females, males were still portrayed in a more sexist light. Our research reiterates the fact that males in textbooks are more bound by traditional standards than are females.

Despite the focus on gender equity in textbooks over the past 25 years, the research demonstrates that more attention needs to be directed toward the portrayal of boys in textbooks. If we want girls and boys to develop to their full potential, we need to teach all children, not just girls, that it is acceptable to display a wide variety of traits, those traditionally described as masculine as well as those which are labeled feminine.

ACKNOWLEDGMENTS

This article stems from an Honor’s thesis completed in April 1998 at Augusta State University. A version of this article was presented as part

270 Evans and Davies

of the MOST student roundtables at the August 1998 American Sociology Association Meeting in San Francisco. The authors thank Beth Fanning, Sian Mile, Allen Scarboro, Helga Shugart, and two anonymous reviewers for their comments on this and/or previous versions of this article and Sabina Widner for her psychological and methodological insights.

REFERENCES

Bem, S. (1981). Bem Sex Role Inventory professional manual. Palo Alto, CA: Consulting Psychologists Press.

Britton, G. E., & Lumpkin, M. C. (1977). For sale: Subliminal bias in textbooks. Reading Teacher, 31, 40–45.

Burke, P. (1996). Gendershock. New York: Anchor Books. Frasher, R. (1982). A feminist look at literature for children: Ten years later. In M. Sheridan

(Ed.), Sex stereotypes and reading Washington, DC: International Reading Association. Frasher, R., & Walker, A. (1972). Sex roles in early reading textbooks. Reading Teacher,

25, 741–749. Henshaw, A. (1992). Skippings for girls: Children’s perceptions of gender roles and gender

preferences. Educational Research, 34, 229–35. Hitchcock, M., & Tompkins G. E. (1987). Basal readers: Are they still sexist? Reading Teacher,

41, 288–292. Katz, P. A. (1996). Raising feminists. Psychology of Women Quarterly, 20, 323–340. Oliver, L. (1974). Women in aprons: The female stereotype in children’s readers. Elementary

School Journal, 74, 253–259. Purcell, P., & Stewart, L. (1990). Dick and Jane in 1989. Sex Roles, 22, 177–185. Richardson, L. (1981). The dynamics of sex and gender. Boston: Houghton Mifflin. Richardson, L. (1983). Guidelines for evaluation of elementary textbooks. In L. Richardson &

V. Taylor (Eds.), Feminist frontiers: Rethinking sex, gender, and society. Reading, MA: Addison-Wesley.

Rupley, W., Garcia, J., & Longnion, B. (1981). Sex role portrayal in reading materials: Implications for the 1980’s. Reading Teacher, 35, 786–790.

Sandler, B. (1977). Title IX: Antisexism’s big legal stick. In Taking the sexism out of education. Washington, DC: U.S. Government Printing Office.

Schau, C., & Scott, K. (1984). Impact of gender characteristics of instructional materials: An integration of the research literature. Journal of Educational Psychology, 26, 183–193.

Schmitt, M. C., & Hopkins, C. J. (1990). Metacognitive theory applied: Strategic reading instruction in the current generation of basal readers. Paper presented at the Annual Meeting of the National Reading Conference, Miami, November 1990.

Schnell, T. R., & Sweeney, J. (1975). Sex role bias in basal readers. Elementary English, 52, 737–742.

Scott, K. (1981). Whatever happened to Jane and Dick?: Sexism in texts reexamined. Peabody Journal of Education, 58, 135–140.

Thorne, B. (1993). Genderplay. New Brunswick, NJ: Rutgers University Press. Vaughn-Roberson, C., Tompkins, G., Hitchcock, M., & Oldham, M. (1989). Sexism in basal

readers: An analysis of male main characters. Journal of Research in Childhood Education, 4, 62–68.

Weitzman, L., & Rizo, D. (1974). Biased textbooks: Images of males and females in elementary school textbooks in five subject areas. Washington, DC: National Foundation for the Improvement of Education.

Witt, S. D. (1996). Traditional or androgynous: An analysis to determine gender role orienta- tion of basal readers. Child Study Journal, 26, 303–318.