Boys Will be Boys, and Girls Will be. . .Hard to Find:
Gender Representation in Third Grade Basal Readers

Susan D. Witt, Ph.D.
School of Home Economics & Family Ecology
The University of Akron|
Email: susan8@uakron.edu

forthcoming in Education Research in 1997


Abstract
Gender bias in children's reading materials is an issue that has been examined periodically during the past twenty years. While there have been improvements in the way that males and females are portrayed in children's reading materials, in many instances biases are still prevalent. A recent study which analyzed sixteen basal readers from six publishers found that male characters in the books outnumbered female characters, biographies of males outnumbered biographies of females, and illustrations of males outnumbered illustrations of females. The only area in which females were represented more often than males is in authorship of the stories which are contained in the readers. It was found that female authors are more likely to fairly represent both male and female characters and to write characters who possess a balance of masculine and feminine traits, but that male authors write almost exclusively about males and masculine experiences and rarely portray males as androgynous.


Background Information
During the past twenty years there have been several studies which have examined the way males and females are portrayed in children's literature and children's basal readers (Purcell & Stewart, 1990; Grauerholz & Pescosolido, 1989; St. Peter, 1979; Weitzman, 1977; Women on Words and Images, 1972). Most of these studies found that male characters were written about much more frequently than female characters. Often, children's literature and basal readers contain words, pictures, and descriptions that indicate that it is more desirable to be male than it is to be female (Davis, 1984; Jacklin & Mischel, 1973). At a time when men and women are viewed as equals by much of the population, bias is still frequently found in the readers and literature that children are exposed to. For example, it has been found that in American children's books: - there are 2.3 males in the title for every female.

Because role models help a child develop, it is important that they not be too rigid or narrowly defined. And because obsolete, narrow role models make it difficult for children to grow into happy, productive adults, it is important that children be exposed to role models which are based on social reality (Purcell & Stewart, 1990). When a child spends nearly 20 years reading books that encourage males and masculine behaviors and ignore or discourage females and feminine behaviors, this is likely to play a major role in the development of gender attitudes (Rudman, 1984).

Children's literature and basal readers reach children at an early and impressionable age. Female characters in children's reading materials are sometimes so passive and have so little to do that they are simply colorless (Sadker & Sadker, 1994; Rudman, 1984). This lack of definition, as well as the stereotyping, of women, serves to reinforce females as less important and lets students come to understand that girls don't have as much fun as boys, girls can't have as many adventures as boys, and that girls need to rely on outside forces to rescue them from their problems while boys solve problems through their own cleverness and creativity (Jacklin & Mischel, 1973).

It has been suggested that textbooks account for anywhere from 75% to 90% of school instruction (Woodward & Elliott, 1992; Meece, 1987; Olson, 1980). It is evident that the books children use in school play an important role in their learning. Many of these books are filled with racial, ethnic, social class, and gender stereotypes, and influence children in negative ways. For example, an examination of 113 recently published books for children found that dependency themes which emphasize helpless behavior for females continue to be commonly used (White, 1986). As children read these books and assimilate the information about males and females contained therein, that information is added to what the child already knows about gender roles. When stereotyped or biased information is reinforced by parents, peers, and the media, it more easily becomes part of a child's belief system.

Looking beyond the text of children's readers, research has shown that photographs and illustrations can serve an instructional function in those readers. Illustrations may serve as examples or extend the content of the text (Duchastel & Waller, 1979). In many cases, however, illustrations may be inappropriate or have no relevance to the material being presented (Woodward, 1987). A study conducted by Purcell & Stewart, which was a replication of an earlier study of basal readers conducted by Women on Words & Images in 1972, found a more equitable portrayal of males and females, with numbers of male and female characters being almost equal (Purcell & Stewart, 1990). Even this study, however, found that males were still shown exhibiting mostly traditional masculine traits and getting to have the adventures, while female characters were frequently relegated to the sidelines much of the time. Another finding of this study was that 70% of the folktales found in basal readers were male focused and only 30% were female focused.

Because folktales are often chosen to be included in basal readers, this contributes to the overrepresentation of males and underrepresentation of females in the readers children are exposed to in school. In an effort to determine how males and females are portrayed in current basal readers, a content analysis was done using readers from six of the most popular publishers of children's readers. Because school is one of the primary socialization agents for children (Santrock, 1994), because basal readers are one of the teaching tools used in school, and because gender fairness should be one goal of an enlightened society, it is valid to know what message children are receiving from the readers they are exposed to during their school years. Teachers who wish to be gender fair would be wise to seek out reading materials which exhibit equitable treatment of both male and female characters. Unfortunately, those materials may be difficult to find.


Sample
The sample for this study consisted of sixteen third grade level basal readers from six publishers. Third grade basal readers were selected because there are more and varied stories within books at this level than there are in first or second grade readers. Also, children in the third grade are more cognitively developed than children in the younger grades. By the third grade, students are firmly entrenched in the school system and have internalized the stated rules and regulations of school life and have also been exposed to the unstated norms of school society. At this particular age and developmental level, children are able to take pleasure from reading and can internalize the messages of the books that they read (Piaget & Inhelder, 1969; Piaget, 1954).

One of the tasks faced by children at this developmental level is to absorb the norms of school and the larger culture (Erikson, 1964). At about the third grade level, children are exhibiting a global judgment of self worth; one influence on a child's sense of self occurs through the books they are exposed to in school (Bee, 1995). The publishers chosen for this study were Harcourt, Brace, Jovanovich, Inc., D. C. Heath and Company, Macmillan/McGraw-Hill Publishing, Inc., Open Court Publishing Company, Scott Foresman and Company, and Silver, Burdett, Ginn. Names and publishing dates of the individual readers are listed in Figure 1. Books were obtained from the Curriculum Center of the Summit County Board of Education in Cuyahoga Falls, Ohio and the Reading Center at the University of Akron in Akron, Ohio. In the case of all publishers except D. C. Heath, a complete series of readers was obtained. Only one of the two books in the D. C. Heath series was available for analysis.

Method
Over the course of several weeks, counts were made of all characters written about in the readers. Individual counts were made of male and female children, male and female adults, illustrations of males and females, biographies of males and females, and male and female authors represented in the readers. Also, a master list of behaviors exhibited by all characters in the readers was compiled. These behaviors were then rated by three individuals to determine whether they could be classified as masculine or feminine. This behavior list made it possible to determine how many masculine traits and how many feminine traits both the male characters and the female characters exhibited.

These 567 characters exhibited a total of 5,147 behaviors which were rated by three individuals, with agreement among the raters as to masculinity or femininity on all but seven behaviors. The instrument used to determine masculine and feminine traits of characters was adapted from the Bem Sex Role Inventory (BSRI), which was developed as a means of determining levels of masculinity, femininity, or androgyny (Bem, 1974). A listing of the traits from the BSRI is found in Figure 2.

Characters were determined to be masculine, feminine, or androgynous based on the following calculations.
1. All masculine traits for male characters in a book were counted.
2. All feminine traits for male characters in a book were counted.
3. All masculine traits for female characters in a book were counted.
4. All feminine traits for female characters in a book were counted.
5. Masculine traits for male characters were divided by the number of male characters to get a male masculine score (MM).
6. Feminine traits for male characters were divided by the number of male characters to get a male feminine score (MF).
7. Masculine traits for female characters were divided by the number of female characters to get a female masculine score (FM).
8. Feminine traits for female characters were divided by the number of female characters to get a female feminine score (FF).

The results of these calculations were then analyzed in the following manner:
1. If the MM score was greater than the MF score and the FM score was greater than or equal to the FF score, males were considered to have a traditional masculine orientation and females were considered to have an androgynous orientation.
2. If the MM score was less than or equal to the MF score and the FM score was less than the FF score, males were considered to have an androgynous orientation and females were considered to have a traditional feminine orientation.

These scores were then used to determine if there was a statistically significant relationship between the number of male and female authors in the readers and the exhibition of masculine and feminine traits by the characters.

Results
Results of this study showed that male characters outnumbered female characters, illustrations of males outnumbered those of females, and biographies of males outnumbered biographies of females for all publishers in the sample. In all the books in the sample, male characters were portrayed as predominately masculine and exhibited few feminine traits. No male characters were classified as androgynous. Female characters, however, were portrayed in an androgynous manner in the books of five of the publishers, and as traditionally feminine for one publisher, Open Court. Findings of the study are detailed below.

Male and Female Characters
Table 1 shows the ratio of male to female adult characters and male to female child characters for the six publishers. As indicated by the figures in the table, all publishers were found to have higher ratios of male characters than female characters. Publishers have indicated that they are making efforts to publish basal readers which are more gender fair, and to some extent this is true. The findings of this study, for example, showed that female characters are more likely to be classified as androgynous (possessing a balance of masculine and feminine traits) than as traditionally feminine.

However, the ratios in the table indicate that even though publishers may be making efforts to publish basal readers which are gender fair, when looking at the counts of male and female characters, there are still many more male characters within the readers than there are female characters.

In the case of D. C. Heath, male child characters outnumbered female child characters by a more than two to one ratio. Open Court and Silver, Burdett, Ginn had a better than two to one ratio of male adult to female adult characters. Harcourt, Brace, Jovanovich, Macmillan/McGraw-Hill, and Scott Foresman were more equitable in their numbers of male and female characters, but still had more male characters than female. None of the publishers had a greater number of female characters than male characters.

Illustrations and Biographies
Table 2 offers ratios of illustrations featuring male characters to illustrations featuring female characters. As indicated by the figures in the table, for all the readers in the sample, illustrations featuring males outnumbered illustrations featuring females, in the case of D. C. Heath by a more than two to one margin.

The publishers that showed fairly close ratios of male to female illustrations were Harcourt, Brace, Jovanovich and Macmillan/McGraw-Hill, with 1.3 to 1 and 1.1 to 1 ratios, respectively. In addition to D. C. Heath showing more than a two to one ratio of male to female illustrations, two other publishers, Open Court and Silver, Burdett, Ginn, showed nearly a two to one ratio of male to female illustrations. Overall, all the publishers in the sample were consistent in having greater numbers of males than females, both as characters in the readers and as portrayed in illustrations.

Although biographies were not featured prominently in any of the reading series, when biographies were used within the readers, there were more than twice as many biographies of males as there were of females. In this area it would appear that there was more equity five years ago than there is now. Findings of an earlier study showed that 46% of the biographies found in children's basal readers were about the lives of males and 54% were about the lives of females (Purcell & Stewart,1990). The finding of this current study seems to indicate a shift toward favoring the male experience as far as publishers' choice of biographies is concerned, with 69% of the biographies focusing on males and 31% focusing on females.

Male and Female Authors
In all sixteen of the basal readers in the sample, female authors outnumbered male authors, in most cases by about a two to one margin. Table 3 shows the ratios of female to male authors. One of the more interesting findings of this study concerned differences in the treatment of male and female

characters in the readers by male and female authors. To determine if there was a statistically significant relationship between the number of male and female authors and the portrayal of males and females as predominately masculine, predominately feminine, or androgynous, correlation coefficients were computed. Table IV shows the results of these computations.

The function of the correlation coefficient is to indicate the strength of a relationship so that predictions can be made (Walsh, 1990). If it is agreed that having basal readers which are gender fair and which show a balance of traits for both males and females is beneficial for students, then the information in Table 4 could be used to justify the inclusion of even more female authors in the readers that are used in schools.

The correlations of .78 and .62 indicate that there is a significant relationship between the number of male authors and male trait scores. However, the correlations of .28 and .40 are weak and not significant, indicating that there is no relationship between the number of male authors and the female trait scores. This indicates that the male authors in these readers are more concerned with the portrayal of male characters as masculine than their portrayal of female characters at all.

As shown by the regression line in Figure 3 for male masculine and male feminine traits by male authors, it is apparent that as the number of male authors increases, there is a corresponding increase in the number of male masculine traits.

Although the number of male feminine traits also increases, the disparity between male masculine and male feminine traits is predicted to grow wider, indicating even more disparity between male masculine and male feminine traits. This wide disparity between male masculine and male feminine traits illustrates the prediction that as more male authors would be added to the readers, the more frequently male characters would continue to be portrayed as traditionally masculine rather than having a balance of traits (androgynous). The insignificant correlations between number of male authors and female masculine and female feminine traits show that, in this sample, females are not seriously considered by male authors at all.

Female authors, on the other hand, emerge as consistently and fairly considering all traits for both male and female characters. These similar correlations across the board indicate that female authors are more likely to write stories populated with characters who have a balance of traits, whether the characters are male or female. Thus, the more female authors included in the readers, the more likely the books would have an androgynous gender role orientation for male and female characters. Figure 4 shows the regression lines for the correlation coefficients found in Table 3 for female authors and all traits.

As shown in Figure 4, the prediction can be made that as more female authors are included in the readers, the more likely it is that characters would continue to be portrayed in a gender fair manner. As the number of female authors represented in the basal readers increases, the number of male and female characters exhibiting a balance of masculine and feminine traits would also increase.

When comparing Figure 3 with Figure 4, it is easy to see that a fairer portrayal of gender roles is found with female authors for both male and female characters. Rather than showing an ever-widening disparity between masculine and feminine traits, as in Figure 3, Figure 4 paints a picture of traits which are in close proximity to each other, thus indicating a balanced portrayal of male and female characters and masculine and feminine traits. In short, what this finding shows is that female authors in this sample make efforts to treat both male and female characters in a gender fair way, leading to a more androgynous orientation for basal readers. However, when even a few male authors are represented in the readers, the scales are tipped in favor of a traditional masculine view, with much greater emphasis placed on male characters and the male experience, to the detriment of female characters.

Discussion
Publishers have made some strides in their portrayal of female characters in basal readers. Rather than showing girls and women with traditionally feminine traits, most of the publishers in this sample seem to have realized that females as well as males can be assertive, decisive, and exhibit leadership qualities. This is a positive step for young students in that they can come to see the female characters in their readers in more varied, active roles rather than simply as passive beings waiting to be rescued by the more powerful, adventurous male characters.

Because these cognitively-developing students are assimilating environmental information around them all the time, they will be able to read about girls and women who possess a balance of traits and engage in a variety of behaviors and make this information a part of themselves. This is a positive thing. These strides, however, must be tempered with the knowledge that while females are being treated in a balanced way within the readers, male characters are being portrayed in the same way they were 20 years ago, with little in the way of any variety in behaviors.

While there is much that is positive about many of the traditional masculine behaviors and traits, students are being done a disservice when, in their reading, they rarely come across a male character who exhibits traits such as sensitivity or empathy. Also, the fact that male characters outnumbered female characters, illustrations of males outnumbered those of females, and biographies of males outnumbered biographies of females - sometimes by a large margin - shows a disposition on the part of these publishers toward placing more emphasis on males than on females. This emphasis could be taken to mean that there is more value seen in being male than female.

Whether this is done volitionally or not is not clear, but it does seem that the old idea of "boys only want to read about boys, but girls will read anything" is alive and well. In the interest of gender fairness, it is concluded that female authors are more likely to offer a better mix of male and female characters, with a wider variety of traits and behaviors, than male authors. Female authors were found to write stories that included both males and females and to give boys as well as girls opportunities to exhibit a balance of behaviors.

Based on the data gathered in this study, if publishers are interested in producing readers which are fair to both genders, they would do well to almost exclusively use female authors. While this may sound like discrimination against male authors, it could be said that male authors discriminate, albeit unknowingly, against female characters. The findings of this study confirmed that males are given much more attention and appear more prominently than females in basal readers. However, because readers from only six publishers were analyzed, it would be beneficial for a larger study to be conducted which would include representation from additional publishers. Also, this study concentrated on third grade basal readers, and future research might broaden the sample to include readers from several grades.


References

Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162.

Davis, A. (1984). Sex differentiated behaviors in nonsexist picture books. Sex Roles, 11, 1-16.

Duchastel, P. & Waller, R. (1979). Illustrating instructional texts. Educational Technology, 19, 20-25.

Erikson, E. (1964). Childhood and society, 2nd ed. New York: Norton.

Grauerholz, E. & Pescosolido, B. (1989). Gender representation in children's literature: 1900 to 1984. Gender & Society, 3, 113-125.

Jacklin, C. & Mischel, H. (1973). As the twig is bent - sex role stereotyping in early readers. School Psychology Digest, 2, 30-37.

Meece, J. L. (1987). The influence of school experience on the development of gender schemata. New Directions for Child Development, 38, 57-73.

Olson, D. (1980). On the language and authority of textbooks. Journal of Communication, 30, 186-196.

Piaget, J. (1954). The construction of reality in the child. New York: Basic Books.

Piaget, J. & Inhelder, B (1969). The psychology of the child. New York: Basic Books.

Purcell, P. & Stewart, L. (1990). Dick and Jane in 1989. Sex Roles, 22, 177-185.

Rudman, M. (1984). Children's literature: An issues approach. New York: Longman Publishing.

Sadker, M. & Sadker, D. (1994). Failing at fairness: How America's schools cheat girls. New York: Charles Scribner's Sons.

St. Peter, S. (1979). Jack went up the hill. . .but where was Jill? Psychology of Women Quarterly, 4, 256-260. Santrock, J. (1994). Child development, 6th ed. Madison: Brown and Benchmark.

Walsh, A. (1990). Statistics for the social sciences. New York: Harper & Row, Publishers.

Weiss, D. (1991). The great divide. New York: Poseidon Press.

Weitzman, L. (1977). Sex role socialization in picture books for preschool children. American Journal of Sociology, 77, 1125-1150.

White, H. (1986). Damsels in distress: Dependency themes in fiction for children and adolescents. Adolescence, 21, 251-256.

Women on Words and Images (1972). Dick and Jane as victims. Princeton: Women on Words and Images.

Woodward, A. (1987). Textbooks: Less than meets the eye. Journal of Curriculum Studies, 19, 511-526.

Woodward, A. & Elliott, D. (1992). Teacher professionalism, school reform, and textbooks. Educational Horizons, Summer, 176-180.


Back to Susan's Home Page