• No results found

Gender Bias in Science Secondary Education: Role Model Presentation in Japanese Science Textbooks

N/A
N/A
Protected

Academic year: 2023

Share "Gender Bias in Science Secondary Education: Role Model Presentation in Japanese Science Textbooks"

Copied!
36
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Gender Bias in Science Secondary Education:

Role Model Presentation in Japanese Science Textbooks

Tomoyuki Morimitsu (8795268)

Master thesis under supervision of

Dr. Ralph Meulenbroeks and Liesbeth de Bakker MSc

Utrecht University, Graduate School of Teaching

November 2022

(2)

Abstract

Although several measures have been taken to close the gender gaps in Japan, females are still underrepresented in the different fields of STEM. One possible reason is that female students do not have enough opportunities to see female role models.

This research studied gender bias in role model presentation in the textbooks for junior high school subject “science,” comprising four scientific disciplines: physics, chemistry, biology, and earth science. Science textbooks currently used in Japanese junior high schools are published by five private publishers. This study analyzed the number of male and female individuals in these textbooks statistically while focusing on the difference in the individuals’ potential to be role models.

The analysis revealed current Japanese junior high school science textbooks are all gender biased in terms of role model presentation, especially in terms of the occurrence of nonfictional science experts who are currently active or were active in the past. Additionally, the extent of bias varies between the five publishers and between the four scientific disciplines, but without any specific patterns or tendencies.

These results indicate that the current gender bias in the textbooks can be reduced by assigning more nonfictional female science experts who are currently active.

(3)

Table of Contents

Abstract...1

Introduction...3

Theoretical Background...6

Japanese science secondary education...6

Gender gap...7

Role models...9

Research question...11

Methods...11

Research sample: Science textbooks...11

Textbook analysis tool...12

Statistical analysis procedure...17

Results...18

Individuals and Gender...18

Existence...19

Relevance...23

Time...25

Discussion...26

Conclusion...26

Implications...28

Limitation...30

Future work...31

References...31

(4)

Introduction

The Global Gender Gap report 2021 ranked Japan 120th of 156 participating countries in gender equity, which is the lowest of all developed countries (World Economic Forum, 2021). The Government of Japan has taken measures to lessen the gap such as establishing a law for gender equity (Basic Act for Gender Equal Society, 1999), but evidently, gender equity is still far away in Japanese society. Looking into the academic research environment in Japan, there are also large gender gaps. The percentage of female academic researchers is only 16.9 % in 2020 (Statistics Bureau of Japan, 2020), while other OECD countries reach around 20-50% (OECD, n.d.). According to a survey among academic researchers, the gap is largest in the fields of science, technology, engineering and mathematics (STEM) in Japan: in universities and other research institutes, the proportions of female scientists in the field of science (including mathematics and technology in the classification of the survey) and engineering are merely 15.1% and 11.9%, respectively (Statistics Bureau of Japan, 2020).

Although there is no evidence of males being more capable at STEM subjects than females (O’Dea et al., 2018), various factors contribute to more female students leaving the STEM “stream” than their male counterparts (Clark Blickenstaff, 2005). The decrease of female students in the STEM education at various stages from primary to higher education, called the "leaky pipeline" (Figure 1), occurs in many countries (Dubois-Shaik & Fusulier, 2015). In the Japanese educational system, the first “leaking point” can be identified at the stage of high schools (upper secondary education, age 15-18). In many high schools, students select either humanities or science courses and choose one or two of the four scientific school subjects (physics, chemistry, biology and earth science). At this point, fewer female students select science courses than male students. In high schools offering both the science courses and the humanity courses, 41.5% of male students and 22.9% of female students select science courses (National Institute for Educational Policy Research [NIER], 2013). The

(5)

situation is especially pronounced for physics: a look-back survey for university students revealed that only 57.4% of female university students who were in the high school science courses took physics, compared to 86.9% for male students (Benesse Educational Research and Development Institute, 2005).

Figure 1

An Example of the Leaky Pipeline in Physics and Engineering in Ontario, Canada (Queen’s University, 2022)

Diminished popularity of science for female students in high school, especially one seen in physics, matters because that causes a decrease in the number of female students studying advanced science in university, which results in the low proportion of female scientists in STEM fields described above. Moreover, the lack of female scientists in STEM fields indicates that pursuing STEM careers is hard for even talented female students. The Japanese society has thus lost many opportunities for scientific innovation because of the

(6)

STEM careers and realize their full potential is necessary for further development of Japan, which lessens the gender gap in STEM.

Therefore, an approach to increase the proportion of female scientists in STEM fields is an attempt to reduce the number of female students that decide to drop the science subjects in high school. To do that, it is necessary to reach out to as many females students as possible and motivate them to choose science in junior high school (lower secondary education, age 12-15).

International studies indicate that both micro-level and macro-level cultural factors of STEM generate the gender gap in STEM (Cheryan et al., 2017). One of the main factors is the importance of exposure to role models that show students STEM careers are also for females (Science Council of Japan, 2020). This study therefore focuses on such role models.

Studies indicate that female science teachers are possible role models who can encourage female students to select STEM courses (Libertas Consulting, 2018). The proportion of female science teachers is only 33 % in junior high school in Japan (Ministry of Education, Culture, Sports, Science and Technology [MEXT], 2021). However, another possible source of role models at school can be found in the science textbooks that are used (MEXT, 2021).

Science textbooks can be used effectively for offering female students proper role models, considering all students are exposed to textbooks at school. However, role model presentation in Japanese science textbooks has not been investigated so far.

This study therefore aims to gain quantitative insight into the gender bias in role model presentation in junior high school science textbooks. Considering that gender gaps also exist in the subject selection of scientific disciplines in high school, particular focus is placed on the difference in role model presentation between the four scientific disciplines: physics, chemistry, biology and earth science.

(7)

Theoretical Background

In this study, role models provided by Japanese science textbooks for junior high schools are analyzed to reveal gender bias in role model presentation. As the key concepts of this study, this section describes existing gender gaps in the Japanese STEM education and the causes for gender gaps in STEM fields particularly focusing on role models. At the end of this section, the research question is provided based on the theoretical background discussed.

Japanese science secondary education

A general education track in Japan consists of primary education (age 6-12), lower secondary education (junior high school, age 12-15), upper secondary education (high school, age 15-18), and higher education (Figure 2). Based on the current education guidelines, in junior high school, all students learn physics, chemistry, biology and earth science in one school subject of “science”. Additionally, high schools offer either (1) three of the four basic subjects (basic physics, basic chemistry, basic biology, and basic earth science) or (2)

"Science and our daily life" and one of the four basic subjects. These are usually taken at the beginning of the first year of high school (MEXT, 2018). In many high schools, after finishing these basic subjects, students select either humanities or science course, and select one or two of the four advanced subjects (physics, chemistry, biology, and earth science).

Although the guidelines do not determine how many advanced subjects have to be selected, usually students in humanities courses select one and science courses select two out of the four scientific subjects. The choice for science subjects mainly has an impact on the preparation for university entrance exams: many humanities-related and science-related courses in universities respectively require candidates to take entrance exams for one (humanities) and two (science) of the four science subjects.

Figure 2

(8)

Japanese Educational System (Japan National Tourism Organization, 2022)

“Science” textbooks for junior high schools currently used in Japan are published by five different private publishers, which will be shown in the methods section. They are well- known textbook publishers in Japan that publish textbooks for different school subjects for primary to high school. Reflecting the structure of science school subjects described above, those science textbooks have textbook sections for the four scientific disciplines: physics, chemistry, biology and earth science as well as general parts that describe, for example, how to write observation notes and how to handle laboratory equipment. The design of the textbooks varies between the publishers, but all textbooks feature many individuals represented by photographs, illustrations, or in the text.

Gender gap

Gender gap in Japanese science secondary education

In Japan, the gender gap in general educational attainment has decreased: as of 2020, male and female high school enrollment rates are 95.3% and 95.7 %, and those of universities are 57.7 % and 50.9%, respectively (Gender Equality Bureau Cabinet Office of Japan, 2021).

Turning to science learning, the proportions of male and female students who select science courses are respectively 41.5% and 22.9% in high schools which offer both science courses

(9)

and humanity courses (NIER, 2013). These facts show the existence of a “leaking point” in high school where more female students prefer to not choose science as compared to male students. This in part causes the gender gap in STEM professional fields. In order to reduce the leakage from this leaking point in high school, motivating female students to study science is required in junior high school.

Factors that generate the gender gap in STEM fields

In developed countries, a range of cultural factors from micro-level to macro-level influences female students' attitudes towards STEM, which plays an important role in selecting STEM-related careers. Cheryan et al. (2017) proposed an influential model (Figure 3) that identifies several determining factors in explaining this situation. In this model, the three overarching factors are the masculine culture of STEM, insufficient early experience with STEM, and females tending to feel inferior to males in STEM. Since these factors interrelate, measures should be taken with respect to every factor to reduce gender bias in STEM professional fields. Additionally, the masculine culture of STEM causes stereotypes of people in the STEM fields, negative stereotypes of female abilities, and lack of female role models, all of which keep females away from STEM.

Figure 3

Factors That Generate Females' Lower Representation in Computer Science, Engineering, and Physics in the U.S. (Cheryan et al., 2017)

(10)

Role models

Of the factors described in Figure 3, this study focuses on the lack of proper role models for female students due to the masculine culture of STEM. A role model is "a person who someone admires and whose behaviour they try to copy" (Cambridge University Press, n.d.), from which students can take cues to imagine and create new "possible selves" (Conner

& Danielson, 2016).

There are many different forms of role model, for example, someone in one's surrounding environment such as a parent (Wiese & Freund, 2011) and a teacher (Master et al., 2014), someone famous, and someone who appears in newspapers, magazines, and textbooks (Good et al., 2010; Harrison, 2001; Hutchinson & Torres, 1994). All are possible role models for students in secondary education. Not every individual becomes a role model:

different factors such as relevance to the subject and specific context may influence whether students regard a person as a role model or not (Bakken, 2005).

Not having a proper role model in a (study) subject decreases female students' motivation to study the subject, which results in a low chance to have a successful career in the field (Opare, 2012). A U.S. study showed that female teachers are more helpful to reduce female students' concern about (the) negative stereotype(s) of STEM than male teachers (Master et al., 2014). Similarly in Japan, a survey for the second year of junior high school students says that female science teachers can be role models for female students. According to the survey, 33.8% of female students who are taught science or mathematics by female teachers think of themselves as a "science type," while only 22.5 % of female students who are taught both of the subjects by male teachers think of themselves in that way (Libertas Consulting, 2018). The necessity to give female students opportunities to be exposed to female role models has been recognized in Japan: the government offers an online portal site

(11)

for female students where they can be informed about female scientists and their research environments, and read their messages (Gender Equality Bureau Cabinet Office of Japan, 2022).

At school, female science teachers are a direct and powerful role models for female students, but they are underrepresented (MEXT, 2021). However, school textbooks are used by all students and can also be important sources of role models. A Dutch research example by Bax (2021) revealed a gender gap in role model presentation in Dutch physics and computer science textbooks for high school. Bax (2021) analyzed the textbooks using a textbook analysis tool for classifying individuals with different features. This method of classifying different individuals has been used in the case of psychology textbooks (Hogben

& Waterman, 1997) and enables individuals with various features to be analyzed from the perspective of potential to be role models.

Role model presentation in science textbooks for Japanese junior high school, which may influence female students' participation in science at high school, has not been studied.

Research question

As described above, female students tending to avoid science courses in high school contribute to the low representation of female in STEM fields. Therefore, motivating female students to study science in junior high school and increasing the number of female students who choose science courses in high school are necessary. Furthermore, junior high school science textbooks have the potential to be sources of female role models because all students study all of the four scientific disciplines using these science textbooks. With this in mind, for exploring science textbooks currently used in junior high in Japan, this study has the following research question:

What is the extent of gender bias in role model presentation in physics, chemistry, biology and earth science textbook sections of Japanese junior high school science textbooks?

(12)

Additionally, for answering this research question, the following sub-questions are set:

(1) What is the extent of differences in role model presentation in Japanese junior high school science textbooks between physics, chemistry, biology and earth science textbook sections?

(2) What is the extent of differences in role model presentation in Japanese junior high school science textbooks from different publishers?

Methods

In order to answer the aforementioned research question, all science textbooks for junior high school in Japan were explored. Using a textbook analysis tool designed for this analysis, individuals in the textbooks were classified and analyzed quantitatively.

Research sample: Science textbooks

The analysis was conducted on all science textbooks currently used in Japanese junior high school published by five different private publishers (Table 1). There are 4622 pages in total in 15 textbooks, three from each of the five publishers (one for each grade). Hereafter, the names of the five publishers are referred to as publisher A, B, C, D, and E.

Table 1

List of Science Textbooks Analyzed in This Study

Publisher Textbook name Edition

Dainihon Tosho (publisher A)

Rika no Sekai 1 2021

Rika no Sekai 2 2021

Rika no Sekai 3 2021

Gakkou Tosho (publisher B)

Kagaku 1 2021

Kagaku 2 2021

Kagaku 3 2021

Keirinkan (publisher C)

Science 1 2021

Science 2 2021

Science 3 2021

Kyoiku Shuppan (publisher D) Chugaku Rika 1 2021

Chugaku Rika 2 2021

(13)

Chugaku Rika 3 2021 Tokyo Shoseki (publisher E)

Atarashii Kagaku 1 2021 Atarashii Kagaku 2 2021 Atarashii Kagaku 3 2021

Textbook analysis tool

The analysis tool for this study (Table 2) was adapted from a tool used in a similar gender bias in role model presentation textbook analysis (Bax, 2021). Of the categories used in (Bax, 2021), “Firstness” and “Education level” are not used because they do not fit this study. Besides, this study focuses on differences between publishers and between the four scientific disciplines in the school subject "science," so individuals were also categorized according to in which part of the textbooks they appear. Each category of the tool is now described in detail.

Table 2

Analysis Tool for Textbook Exploration (based on Bax, 2021)

Category Option

Individual 1, 2, 3… (Unique for each individual)

Gender Male/ Female/ Unknown

Textbook section General/ Physics/ Chemistry/ Biology/ Earth science Occupation Any mentioned or presumed occupation/ Unknown

Existence Fictional/ Nonfictional

Relevance Relevant/ Irrelevant

Time Historical/ Contemporary

Publisher Dainihon Tosho (A)/ Gakkou Tosho (B)/ Keirinkan (C)/

Kyoiku Shuppan (D)/ Tokyo Shoseki (E) Additional remark If any.

Individual

Each individual in the textbooks appearing in photographs, illustrations, or the text was given a unique code. Both text and images in textbooks influence readers (Peeters et al., 2010), but the influences of a person in a photo without further explanation and a person

(14)

described only in text are difficult to compare, so no classification was made based on whether individuals are in a photo, illustrations or the text. Additionally, in cases when individuals could not be classified (e.g., in the case when they are in a group photo or only a hand is shown in a picture), these cases were excluded from the study.

Gender

Each individual was categorized with either male, female or unknown. The gender

"other" is not covered in this study. With the gender of each individual, the numbers of gendered individuals were compared. This method was derived from an American History textbook analysis (Chick, 2006). Those whose gender could not be determined from the description, such as "Student A" appearing only in text, were categorized as “Unknown” and excluded from the analysis.

Textbook section

The options here are general, physics, chemistry, biology, and earth science. These categories indicate in which textbook section an individual appears. If an individual appears on a page that is not categorized into any of the four scientific disciplines, the individual was categorized "General." These categories were used to make a comparison in the way gendered individuals appeared in the textbook sections. Individuals appearing on non- content-based textbook pages (e.g., writers and designers of the textbooks in the publication data pages) were excluded from the analysis.

Occupation

In this category, the occupation of an individual is noted if it is explicitly described or can be derived from the content in which the individual is presented. Figure 4 is an example of the latter case: a photograph of a female in the physics textbook section of the second grade publisher B textbook. The photo carries no further text explanation about her.

However, it can be presumed that she is working on physics-related experiments from the

(15)

photograph, so she was categorized as “working within the STEM (physics) field.”

Individuals whose occupations were not identified were categorized "Unknown." The categorized occupations were used for determining the aspect of "Relevance," another classification category of the tool.

Figure 4

An Example of an Individual in a Photo in Publisher B Textbook: A Female Handling Laboratory Equipment

Existence

One factor that can influence whether a person becomes a role model for readers is whether or not the person is fictional. That is, whether the person physically exists or has existed in the readers’ world or is imaginary (Drury et al., 2011). For instance, Dr. Shinya Yamanaka, a well-known and still active Japanese scientist, who was awarded the Nobel Prize in 2012 for his discovery of iPS cells, is a better role model than a fictional scientist. He provides a concrete image of what a real scientist is like. Considering this, the individuals in the textbooks were categorized as being either fictional or nonfictional characters. These

(16)

categories were used to analyze the influence of the low proportion of females in STEM professional fields on the gender gap in role model presentation in the textbooks.

Relevance

Not all of the individuals in the textbooks can be considered role models who encourage students to choose science paths. Fictional individuals in the textbooks do not have any detailed personalities. Most of them do not even have a name. Therefore, fictional individuals in the textbooks are considered to have a lower potential to be role models.

However, nonfictional characters do not automatically classify as role models, either. In the physics textbook sections of publisher E textbook for the first grade, for example, there is a photograph of Naomi Osaka, a well-known female professional tennis player. This photograph shows her hitting a ball and provides an example of an object in motion under force. So in terms of physics content her presence is relevant. However, she is not relevant as a role model for those who want to pursue a physics-related career path. Hence, the existence of Naomi Osaka in a textbook does not contribute to increasing female role model presentation regarding physics-related careers. As a comparison, Marie Curie is much more likely to be a role model in physics-related careers because she is a well-known female scientist with prominent accomplishments in physics and chemistry. Individuals in textbooks must at least be active or have been active in that discipline in order to be possible role models in a particular scientific discipline.

Considering these differences in the potential to become role models for a specific scientific discipline, nonfictional individuals in the textbooks were classified into “relevant”

and “irrelevant” based on their occupation: an individual was regarded as “relevant” if the occupation of the individual is related to expertise in a scientific discipline where the individual appears, and “irrelevant” if not. Individuals without any explanation about their expertise appearing in a in general part in the textbooks were excluded from this analysis

(17)

because this study aims to analyze the difference in role model presentation between the four scientific disciplines.

However, since the determination of relevance may have some ambiguities, this part of the coding was subjected to second coding to check the reliability: Cohen’s Kappa is 0.88 for the coding of 328 nonfictional individuals in the textbooks of publisher A, indicating near-perfect agreement.

Time

In addition to whether or not individuals in textbooks are nonfictional and whether or not they can be tied to expertise in a scientific discipline, it is also important that individuals are in the readers’ existing world in order to be regarded as role models (González-Pérez et al., 2020; Lockwood, 2006). For example, Dr. Shinya Yamanaka, a Japanese Nobel Prize laureate still active today, is more likely to be a role model than Galileo Galilei who lived in the 17th century.

From this perspective, nonfictional relevant individuals were classified into

"historical" and "contemporary." Those who were still active when the readers of the textbooks were born are considered "contemporary." With these categories, the influence of historical underrepresentation of females in STEM fields is taken into account.

Publisher

Individuals in the textbooks were grouped by publisher using this category in order to compare the difference in role model presentation between the textbooks from the five different publishers.

Statistical analysis procedure

In order to analyze how the individuals appear in the textbooks statistically, the analysis program R studio (R version 4.1.2) was used. To answer the research question, the differences in the number of males and females per different category from the analysis tool

(18)

were analyzed by textbook sections with the following null hypothesis: the frequencies of appearance of males and females are the same (50%). The observed differences in the number of males and females were determined significant when the p value < 0.05 in the binomial test.

Considering that (1) the focus is given to the differences between the four scientific disciplines and between the five publishers, and that (2) the categories of “existence” and

“relevance” influence one’s potential to be a role model, the following empirical hypotheses were tested.

 The numbers of male and female individuals are not the same in each publisher’s textbooks.

 The numbers of fictional male and female individuals are not the same in each textbook section.

 The numbers of nonfictional male and female individuals are not the same in each textbook section.

 The numbers of nonfictional relevant male and female individuals are not the same in each textbook section.

Results

In the following parts, the results of the analysis of individuals in Japanese junior high school science textbooks are presented. An overview of the results is firstly given in the section of “Individuals and Gender,” followed by the results per the categories of

“Existence,” “Relevance,” and “Time.”

Individuals and Gender

The total number of individuals who appeared in the textbooks is 8403, of which 4508 (53.6 %) are males, 3385 (40.3 %) are females, and 510 (6.1 %) are unknown. Table 3 shows

(19)

the number of males and females in the textbooks by publishers. Each of the five publishers has significantly more males than females (p<0.05 in the binomial test).

Most of the individuals in the textbooks exist in illustrations or photographs to help readers understand learning contents by, for example, visualizing scientific phenomena (Figure 5). The names and occupations of the individuals are often not mentioned, except for those who are explicitly introduced as scientists. This is probably because the information about who the people in photographs and illustrations exactly are, is usually not necessary to explain learning contents.

Table 3

The Total Number of Male and Female Individuals in the Textbooks

Publisher Male Female Binomial test

A 786 510 p<0.05

B 1212 1108 p<0.05

C 717 565 p<0.05

D 685 467 p<0.05

E 1108 735 p<0.05

Note. The number of individuals in the second and third columns is a cumulative total of the three volumes for the first to the third grades. The right-most column shows the results of the binomial test for the differences between the number of males and females by publishers.

Figure 5

(20)

An Example of an Individual in Publisher C Textbook: A Photo for Explaining a Moving Object on Ice With Low Friction

Existence

Fictional individuals

Figure 6 provides the number of fictional male and female individuals in the five publishers’ textbooks, which are shown by textbook sections: general information, physics, chemistry, biology, and earth science. In any textbook, the majority of the fictional individuals are students and teachers who appear throughout the textbooks to explain learning contents.

In the textbooks of the four publishers other than publisher E, the difference in the number of fictional males and females in each section is not statistically significant in the binomial test (p>0.05). This means that the four publishers designed their textbooks so that about the same number of fictional males and females appear in each textbook section.

Regarding publisher E textbooks, there are more fictional females than males in physics and earth science textbook sections, and more fictional males than females in chemistry and biology textbook sections (p<0.05 in the binomial test). However, these gender gaps in

(21)

numbers are due to fictional female teachers explaining the learning contents of physics and earth science sections, and fictional male teachers explaining it in the chemistry and biology sections. Considering the difference in the number of fictional males and females in publisher E textbooks as a whole is not statistically significant (p>0.05 in the binomial test), the gaps seen in publisher E textbooks are considered to be just a choice of where to put males and females while keeping an overall gender balance in fictional individuals.

In terms of comparison between publishers and between textbook sections, the way male and female fictional individuals appear varies but there seem to be no distinctive tendencies.

Figure 6

Fictional Male and Female Individuals in the Five Different Textbooks

(22)

Note. These five graphs describe the number of fictional males (blue) and females (orange) in the five publishers’ textbooks, which are shown by general information, physics, chemistry, biology, and earth science textbook sections. The number of individuals in the graphs is a cumulative total of the three volumes (first to third grade) for each publisher. Blue dotted circles indicate textbook sections where the differences between the number of fictional males and females are statistically significant (p<0.05) in the binomial test.

Nonfictional Individuals

Figure 7 describes the number of nonfictional males and females in the five publishers’ textbooks, which are shown by textbook sections: general information, physics, chemistry, biology, and earth science. The occupations of the individuals in this category are, for instance, scientists, sports people, and teachers.

It is evident that in many textbook sections there are more nonfictional males than females (p<0.05 in the binomial test), which contrasts with the fictional category where males and females are well balanced in number. In this nonfictional group, the use of chronological tables which feature important historic individuals contributes significantly to increasing the gender gap in the textbooks of publisher A, C, D and E: in these chronological tables, important scientific discoveries and their discoverers, almost all of whom are male scientists, are introduced. In the case of publisher C and D, individuals in the chronological tables were assigned to the "General" section because the scientific disciplines where the individuals worked are not explicitly mentioned. In contrast, the chronological tables of publisher A and E textbooks introduce historical scientists together with their scientific disciplines, so the individuals were counted in one of the four scientific disciplines.

However, there are no significant differences (p>0.05 in the binomial test) in the number of nonfictional males and females in physics and biology textbook sections of

(23)

publisher B; physics and geology textbook sections of publisher C; and physics and earth science textbook sections of publisher D. This seems to be a coincidence because it is unlikely that these three publishers have a specific reason to keep the gender balance in the number of individuals only in some parts of the textbooks.

Figure 7

Nonfictional Male and Female Individuals in the Five Different Textbooks

Note. These five graphs describe the number of nonfictional males (blue) and females (orange) in the five publishers’ textbooks, which are shown by general information, physics, chemistry, biology, and earth science textbook sections. The number of individuals in the graphs is a cumulative total of the three volumes (first to third grade) for each publisher. Blue dotted circles indicate textbook sections where the differences between the number of nonfictional males and females are statistically significant (p<0.05) in the binomial test.

(24)

Relevance

Nonfictional individuals in the textbooks categorized as “relevant” are shown in Figure 8. They are counted by textbook sections: general information, physics, chemistry, biology, and earth science. In all four scientific disciplines of the five publishers’ textbooks, males are highly predominant (p<0.05 in the binomial test). Furthermore, the figure also shows that many of the nonfictional females in the textbooks shown in Figure 7 are not categorized as “relevant.” Additionally, as with the fictional (Figure 6) and nonfictional (Figure 7) categories, there are differences between publishers and between textbook sections in the number of males and females, but differences vary and do not follow a specific pattern or lean towards a certain tendency.

Figure 8

Nonfictional Relevant Male and Female Individuals in the Five Different Textbooks

(25)

Note. These five graphs describe the number of nonfictional relevant males (blue) and females (orange) in the five publishers’ textbooks, which are shown by physics, chemistry, biology, and earth science textbook sections. The number of individuals in the graphs is a cumulative total of the three volumes (first to third grade) for each publisher. Blue dotted circles indicate textbook sections where the differences between the number of nonfictional males and females are statistically significant (p<0.05) in the binomial test.

What is noteworthy in publisher A, C, and E textbooks is that they introduce female experts in detail. For instance, the third grade publisher A textbook has an interview with a female researcher in the chemistry textbook section. Although the female researcher is neither a discoverer of the learning contents nor necessary for understanding the contents, she explains her research theme and motivation, and gives readers advice on how to become a researcher in the interview. Similarly, the other two publishers’ textbooks also have interviews with female scientific experts. Their fields of expertise differ, but all interviewees described their research themes as fascinating, which indicates they are there to serve as role models.

Time

Focusing on historical relevant individuals, it can be seen that large gender gaps in each section of the textbooks exist (Figure 9). This is due to the fact that almost all of the historical figures are male scientists who made scientific discoveries or inventions relevant to the learning contents. In the "Electricity" unit in the physics section, for example, all of the five publishers introduce a male physicist called André-Marie Ampère, explaining the unit of electric current as it is named after him. Other than that, publisher A and E textbooks have a chronological table where scientists are mentioned with their field of expertise. Most of the scientists in the tables are male, which increases the gender gap in the number of historical

(26)

relevant individuals present in the textbooks. Additionally, the textbooks of publisher A, C, D and E have a list of Japanese Nobel Prize laureates. All are male and many of them are historic figures, which also widens the gender gaps.

In contrast to the historical group, the contemporary group includes both those with historic scientific achievements (e.g., Dr. Shinya Yamanaka, a Japanese Nobel Prize laureate still active today) and those without such achievements (e.g., an unnamed male scientist in a photo without any further explanation). Both are dominated by males, which contribute to large gender gaps seen in contemporary relevant individuals.

Figure 9

Nonfictional Relevant Historical and Contemporary Individuals in the Five Different Textbooks

(27)

Note. In these graphs, nonfictional relevant individuals in Figure 8 are separately described according to whether they are historical or contemporary. Light blue, light orange, dark blue, and dark orange respectively represent historical males, historical females, contemporary males and contemporary females. Again here the differences between the number of nonfictional relevant (historical and contemporary) males and females are statistically significant.

Discussion Conclusion

We now revisit our research questions. The main question is: “What is the extent of gender bias in role model presentation in physics, chemistry, biology and earth science textbook sections of Japanese junior high school science textbooks?” Additionally, the two sub questions are: “What is the extent of differences in role model presentation in Japanese junior high school science textbooks between physics, chemistry, biology and earth science textbook sections?” and “What is the extent of differences in role model presentation in Japanese junior high school science textbooks from different publishers?”

The results described above firstly show that science textbooks analyzed here are designed to be well gender balanced in general in the number of fictional individuals, whereas large gender gaps exist in the number of nonfictional individuals. The results show the existence of large gender gaps in the number of nonfictional relevant individuals, who have a high potential to be role models. The gaps are seen in both historical and

contemporary categories, but the features of the members in these two categories differ from one another: the historical group consists of discoverers and inventors well-known for their scientific achievements, whereas the contemporary group includes both those with and without such achievements. Regarding these gender gaps, the answer to the first sub question is, the differences in the number nonfictional relevant males and females between the four

(28)

textbook sections vary but do not follow a specific pattern or lean towards a certain tendency.

Furthermore, the answer to the second sub question is, the differences in the number of nonfictional relevant males and females between the five publishers also vary but do not follow a specific pattern or lean towards a certain tendency.

Based on these results for the sub questions, the following answer can be formulated to the main question: science textbooks currently used in Japanese junior high schools, which were published by five private publishers, are all gender biased in terms of role model presentation by relevant (historic and contemporary) individuals. The extent of the bias differs between the publishers and between textbook sections, but there are no specific patterns or tendencies in how the bias appears.

Implications

Publishers can freely decide what kind of fictional people to use in their textbooks without being influenced by the gender ratio of actual scientists or other factors. Hence, it can be stated that the observed balance in the number of fictional males and females in the textbooks indicates that all of the five publishers recognize the necessity of reducing gender gaps in science textbooks, at least to some extent.

However, there are many more nonfictional relevant males than nonfictional relevant females in the four scientific textbook sections of the five publishers’ textbooks. One factor that causes this gender gap is that most of the (mainly historical but also some contemporary) scientists, whose scientific achievements are related to the learning contents of “science”, are male. The gender balance of people who are chosen because of their achievements is difficult to control, and gender gaps due to the existence of such male scientists may be inevitable.

Even if there is a need to lessen the existing gender gap in the number of “relevant” figures in a textbook, for example, refraining from mentioning Isaac Newton while discussing classical mechanics may not be appropriate. Since such male scientists are dominant in the historical

(29)

group, there is little room to decrease the gender gap in the number of relevant individuals there.

On the other hand, in the contemporary group, all textbooks have relevant individuals whose achievements are not emphasized, such as a person just handling experimental equipment in a photograph. Such individuals are interchangeable, and they do not have to be males. It is here that significant room for improvement is available. Considering that the historical relevant individuals are biased toward males, bringing more contemporary relevant females to the fore in the textbooks is a possible countermeasure to lessen gender gaps in the number of nonfictional relevant individuals. Additionally, contemporary relevant individuals have the highest potential to be role models, so this approach may contribute significantly to decreasing gender bias in role model presentation in science textbooks.

However, the number of individuals is not the only factor to be considered when providing role models. The textbooks of publisher A, C, and E are devised for providing role models differently from simply increasing the number of female relevant individuals: these publishers introduced detailed interviews with female science experts. It is plausible that individuals who are introduced through an interview can attract more readers’ attention than those who are just briefly mentioned or are shown in a picture. Therefore, introducing interviews with female experts can be considered another effective countermeasure to reduce gender bias in role model presentation other than just reducing the gap in actual numbers of appearance of male and female figures.

In terms of comparison between the four scientific disciplines, the way male and female individuals distribute varies, whereas no specific tendencies can be determined. This is probably because the four disciplines are taught in one single school subject “science” and are supported by one single textbook which presented four different textbook sections, one per discipline. This result shows that none of the four disciplines of the textbooks are

(30)

particularly inferior to the others in terms of providing female role models, which indicates that the way the textbooks provide role models most likely will not be an important factor when female students select which of the four scientific disciplines to take in high school. It should be noted that since the textbooks analyzed here have introduced as recent as 2021, most of the users of the textbooks have not yet officially selected which of the four scientific disciplines to learn at high school. This means that these textbooks cannot be a cause of the observed difference in female students’ scientific subject preferences in high school.

In Japan, textbooks are allowed to be used in schools only after passing the textbook authorization process conducted by the Textbook Authorization Research Council, an advisory body of the Minister of Education, Culture, Sports, Science and Technology (MEXT, n.d.). The science textbooks analyzed in this study, which have been used since 2021, were authorized in 2019. In the authorization process, the council reviews and judges if the textbooks applied by private publishers satisfy the Textbook Authorization Standard.

Regarding the requirement for gender equality, the standard only requires that the purpose of education shall be consistent with the goals of education stipulated in Article 2 of the Basic Act on Education: fostering the values of equality between males and females (Basic Act on Education, 2006). Additionally, there are no instructions for revision requiring gender equality in the comments sent by the Council to the five publishers during the authorization process in 2019 (MEXT, 2020). It can be assumed that the gender balance in the textbooks revealed in this study has been implemented by the publishers voluntarily.

All the five publishers seem to consider gender balance in their textbooks to some extent without being forced to do so by the Council. However, at the same time, the absence of more detailed requirements on gender equality in the textbook authorization standard means that the government does not expect more from science textbooks as sources of female role models than what they are doing already. Junior high school is a stage of compulsory

(31)

education, so science textbooks used in that stage reach all students there. Therefore, science textbooks have a big potential to be used effectively and actively to fill the gender gap in the STEM career paths. To do so, it would at least be necessary to amend the standard and describe further specific requirements regarding gender balance in role model presentation in textbooks.

Limitation

This quantitative study analyzed the differences in the number of male and female individuals in science textbooks focusing on role model presentation and revealed the

existence of gender bias there. However, whether or not each “relevant” individual identified here really works as a role model was not analyzed in detail. In addition to the features of individuals characterized by the categories listed in Table 2, the way they are described may also influence their potential to become role models. Furthermore, some may not regard an individual in a textbook as a role model even if the individual is a good role model for others.

Moreover, the influence on students’ decision from role models given by textbooks and those from other factors cannot be evaluated separately. For these reasons, the influence of role models in school textbooks on students’ course choice is hard to measure.

Future work

Considering the aforementioned limitation of this study, getting further meaningful results about role models by analyzing these science textbooks in more detail seems to be difficult. Instead, in order to reduce existing gender bias in STEM, it would be good to consider an improved design of science textbooks which would be more suitable for

providing female role models using the results of this study. In doing so, it would be effective to interview junior high school teachers and students to learn how they use science textbooks.

At the same time, the Textbook Authorization Research Council should require that the publishers consider providing more female role models in the next edition of science

(32)

textbooks in order to realize improved science textbooks. Science textbooks should become natural sources of proper role models.

References

Bakken, L. L. (2005). Who Are Physician–Scientists’ Role Models? Gender Makes a Difference.

Academic Medicine, 80(5).

https://journals.lww.com/academicmedicine/Fulltext/2005/05000/Who_Are_Physician_Scien tists__Role_Models__Gender.20.aspx

Basic Act for Gender Equal Society, Act No.78 of 1999. (1999).

http://www.japaneselawtranslation.go.jp/law/detail/?

ft=1&re=01&dn=1&x=0&y=0&co=01&ia=03&ja=04&ky=

%E7%94%B7%E5%A5%B3%E5%85%B1%E5%90%8C&page=5 Basic Act on Education, Act No.120 of 2006. (2006).

https://www.japaneselawtranslation.go.jp/en/laws/view/2442

Bax, J. (2021). Gender bias in secondary education - An explorative study on the availability of role models for physics and computer science students in Dutch high schools [Utrecht University]. https://dspace.library.uu.nl/handle/1874/403321

Benesse Educational Research and Development Institute. (2005). A look-back survey on course choice for university students. https://berd.benesse.jp/koutou/research/detail1.php?id=3170 Cambridge University Press. (n.d.). Role model. Cambridge Dictionary. Retrieved January 15,

2022, from https://dictionary.cambridge.org/dictionary/english/role-model

Cheryan, S., Ziegler, S., Montoya, A., & Jiang, L. (2017). Why are some STEM fields more gender balanced than others? Psychological Bulletin, 143(1), 1–35.

https://doi.org/10.1037/bul0000052

Chick, K. (2006). Gender Balance in K-12 American History Textbooks. Social Studies Research and Practice, 1.

(33)

Clark Blickenstaff, J. (2005). Women and science careers: leaky pipeline or gender filter? Gender and Education, 17(4), 369–386. https://doi.org/10.1080/09540250500145072

Conner, L. D. C., & Danielson, J. (2016). Scientist role models in the classroom: how important is gender matching? International Journal of Science Education, 38(15), 2414–2430.

https://doi.org/10.1080/09500693.2016.1246780

Drury, B., Siy, J., & Cheryan, S. (2011). When Do Female Role Models Benefit Women? The Importance of Differentiating Recruitment From Retention in STEM. Psychological Inquiry, 22, 265–269. https://doi.org/10.1080/1047840X.2011.620935

Dubois-Shaik, F., & Fusulier, B. (2015). Academic Careers and Gender Inequality: Leaky Pipeline and Interrelated Phenomena in Seven European Countries.

http://garciaproject.eu/wp-content/uploads/2015/12/GARCIA_working_paper_5.pdf Gender Equality Bureau Cabinet Office of Japan. (2021). The White Paper on Gender Equality

2021. https://www.gender.go.jp/about_danjo/whitepaper/r03/zentai/pdf/r03_print.pdf Gender Equality Bureau Cabinet Office of Japan. (2022, January). Messages from senpai.

https://www.gender.go.jp/c-challenge/senpai/index.html

González-Pérez, S., Cabo, R., & Ibáñez, Dr. M. (2020). Girls in STEM: Is It a Female Role-Model Thing? Frontiers in Psychology, 11, 2204. https://doi.org/10.3389/fpsyg.2020.02204

Good, J. J., Woodzicka, J. A., & Wingfield, L. C. (2010). The Effects of Gender Stereotypic and Counter-Stereotypic Textbook Images on Science Performance. The Journal of Social Psychology, 150(2), 132–147. https://doi.org/10.1080/00224540903366552

Harrison, A. G. (2001). How do Teachers and Textbook Writers Model Scientific Ideas for Students? Research in Science Education, 31(3), 401–435.

https://doi.org/10.1023/A:1013120312331

Hogben, M., & Waterman, C. K. (1997). Are All of Your Students Represented in Their

Textbooks? A Content Analysis of Coverage of Diversity Issues in Introductory Psychology

(34)

Textbooks. Teaching of Psychology, 24(2), 95–100.

https://doi.org/10.1207/s15328023top2402_3

Hutchinson, T., & Torres, E. (1994). The textbook as agent of change. ELT Journal, 48(4), 315–

328.

Japan National Tourism Organization. (2022, January). JAPANESE EDUCATIONAL SYSTEM.

https://education.jnto.go.jp/en/school-in-japan/japanese-education-system/

Libertas Consulting. (2018). Research on students’ attitudes toward supporting female students’

course choice in STEM. https://www.gender.go.jp/research/kenkyu/pdf/girls-course_h29.pdf Lockwood, P. (2006). “Someone Like Me can be Successful”: Do College Students Need Same-

Gender Role Models? Psychology of Women Quarterly, 30, 36–46.

https://doi.org/10.1111/j.1471-6402.2006.00260.x

Master, A., Cheryan, S., & Meltzoff, A. (2014). Reducing adolescent girls’ concerns about STEM stereotypes: When do female teachers matter? Revue Internationale de Psychologie Sociale, 27, 79–102.

Ministry of Education, Culture, Sports, Science, & Technology. (n.d.). Japan’s School Textbook.

Retrieved October 27, 2022, from

https://www.mext.go.jp/en/policy/education/elsec/title02/detail02/sdetail02/1383719.html Ministry of Education, Culture, Sports, Science, & Technology. (2018). Education guidelines.

https://www.mext.go.jp/a_menu/shotou/new-cs/1407074.htm

Ministry of Education, Culture, Sports, Science, & Technology. (2020). Comments for Authorization.

https://www.mext.go.jp/a_menu/shotou/kyoukasho/kentei/1416452_00001.htm

Ministry of Education, Culture, Sports, Science, & Technology. (2021). School teacher statistics survey 2019.

https://www.mext.go.jp/b_menu/toukei/chousa01/kyouin/kekka/k_detail/1395309_00001.htm

(35)

National Institute for Educational Policy Research. (2013). A Study of Science Course Choice in Junior High Schools and High Schools: Final Report.

O’Dea, R. E., Lagisz, M., Jennions, M. D., & Nakagawa, S. (2018). Gender differences in individual variation in academic grades fail to fit expected patterns for STEM. Nature Communications, 9(1), 3777. https://doi.org/10.1038/s41467-018-06292-0

OECD. (n.d.). Main Science and Technology Indicators. Retrieved January 15, 2022, from https://stats.oecd.org/Index.aspx?DataSetCode=MSTI_PUB#

Opare, P. B. (2012). That Female Higher Education Faculty in Select Science, Technology, Engineering, and Mathematics (STEM) Fields Perceive as Being Influential to Their Success and Persistence in Their Chosen Professions. STEMPS Theses & Dissertations.

https://doi.org/10.25777/m44g-4q33

Peeters, M., Churchwell, M., Mauro, L., Cappelletty, D., & Stone, G. (2010). A student-inclusive pharmacotherapeutics textbook selection process. Currents in Pharmacy Teaching and Learning, 2. https://doi.org/10.1016/j.cptl.2009.12.004

Queen’s University. (2022, January). Faculty of Engineering and Applied Science.

https://engineering.queensu.ca/women-in-engineering/research

Science Council of Japan. (2020). The current status and issues of gender balance in the STEM fields. https://www.scj.go.jp/ja/info/kohyo/pdf/kohyo-24-h200605.pdf

Statistics Bureau of Japan. (2020). Report on the Survey of Research and Development.

https://www.stat.go.jp/data/kagaku/kekka/kekkagai/pdf/2020ke_gai.pdf

Wiese, B. S., & Freund, A. M. (2011). Parents as role models: Parental behavior affects

adolescents’ plans for work involvement. International Journal of Behavioral Development, 35(3), 218–224. https://doi.org/10.1177/0165025411398182

World Economic Forum. (2021). The Global Gender Gap report 2021.

https://www.weforum.org/reports/global-gender-gap-report-2021

(36)

Referenties

GERELATEERDE DOCUMENTEN

While I will use the case study method to understand how cognitive values can be applied in theory appraisal and the epistemic benefits that non-cognitive values can provide

The prospective Indonesian foster parents could perform adoption if they meet all requirements as stipulated in the prevailing rules and regulations. The parents must be in

Nadat we hier even naar gekeken hadden, deelde ik het opdrachtenblad uit waarbij leerlingen weer hun observaties moesten noteren en moesten beredeneren wat de link tussen het

In this thesis I discuss the works of three writers commonly associated with the genre of feminist science fiction: Ursula Le Guin (The Left Hand of Darkness), Joanna Russ (The

In hoeverre de kleinere afzetmarkten en de GMO-vrije teelt mogelijkheden bieden voor droog te oogsten sojabonen is nu niet

Het stratigrafisch onderzoek toonde aan dat de middeleeuwse pakketten een groot aantal kuilen afdekken. Voor een gedetailleerd overzicht van het vondstenmateriaal, kan

(It is important to note that constrained K-means on itself does not learn a metric, ie, the side- information is not used for learning which directions in the data space are

There was a fair correlation of the Flanker number of completed trials on all conditions (Table 2) and accuracy on the congruent and incongruent condition (Table 2), with MMSE