Sage Journals: Discover world-class research

Abstract

The reform and opening-up of China have greatly improved the scale and quality of doctoral education for women. However, female doctors still face the “leaky pipeline” and the “unbreakable glass ceiling” in their development of academic careers. In this study, gender differences are investigated in doctoral graduates’ career choices, the level of educational institutions they attend, and their scientific research productivity after joining the institution. We analyzed the administrative data and scientific research publication information from ten years of doctoral graduates at a top research university in China. Results suggest that compared to their male counterparts, female doctors are more likely to pursue an academic career upon graduation, but they are also more likely to be employed in lower-level institutions as well as to publish Chinese scientific studies with lower influence and poorer quality. Moreover, gender differences in academic disciplines are heterogeneous. While academic career development for doctors in natural sciences is not gender-biased, female doctors in social sciences face the most significant challenges, and these results persist even after controlling for their scientific publications during graduate school. In other words, gender differences in academic career development are likely to result from gender symbols rather than differences in academic ability.

Keywords

doctoral graduates academic career development gender differences gender symbols and stereotypes research productivity

Introduction

Since the reform and opening-up of China, doctoral education has developed rapidly and played a crucial role in the country’s development of nature science and technology. Doctoral education for female students has historically progressed during this period (Yan & Shang, 2021). By 2021, the proportion of female doctorates nationwide reached 41.87%, an increase of 6.4% from 2012 (MOE, nd). Doctoral education plays a crucial role in China’s development of science and technology, and female doctors are steadily becoming an emerging force that enhances the country’s strategic development (Ren et al., 2013). In 2021, the Ministry of Science and Technology, the All-China Women’s Federation, as well as other 11 departments launched “A Series of Measures to Support Female Researchers in Playing a Greater Role in Science-technology Innovation”, indicating that female sci-tech (i.e., science-technology) talents are a vital part of China’s sci-tech innovation team as well as play an integral role in China’s sci-tech innovation initiatives. Of China’s scientific and technological endeavors, half is the work of female scientists (Ministry of Science and Technology of the People’s Republic of China, 2021).

While the number of female doctors has grown rapidly in recent years, they still face many difficulties and challenges in their career development. On the one hand, female doctoral students have to bear multiple pressures from society, family, and academic advisors in their graduate training (Centra & Gaubatz, 2000). Compared to their male counterparts, the deferral rate for female doctoral students is higher, but the amount of scientific research produced is lower when they are in graduate school (Li & Chen, 2021). This disadvantage will be extended to female doctors’ quality of employment, the amount of scientific research produced, as well as the promotion of professional titles. On the other hand, female doctors face an “unbreakable glass ceiling” in their development of academic careers. Although female doctors are more likely than male doctors to pursue academic careers in higher education institutions (Wang & Hao, 2006; Zhao & Shen, 2010), they tend to be employed in lower levels of higher education institutions (Jin et al., 2018). Upon joining the institutions, female doctors face the “Productivity Puzzle” as their scientific research productivity tends to be lower than that of male doctors. Female doctors often find it difficult to reach the highest positions of professional elites along with the improvement of their academic status, a phenomenon known as the “Leakey Pipeline” (Lin & Tong, 2012). In May 2023, the Chinese Academy of Sciences published statistics showing that females make up only 6% of the 824 academicians. A possible explanation for career development difficulties among female doctors lies in Sonnert and Holton’s “Deficit Model.” Typically, females are socialized to be gentle and dependent, leading to a deficiency in their ability development. Even if there is no difference in ability between females and males, gender as a symbol will cause females to be treated differently in career development. Although China has been committed to gender equality for many years and has repaired the formal structural barriers to gender discrimination, many hidden informal structural barriers still hinder female doctors' academic career development.

The purpose of this study is to identify whether the academic career development of female scholars differs significantly from that of their male counterparts, as well as the reasons behind the differences. This study empirically examines the differences and similarities in the academic career development of doctoral graduates across genders based on data from ten years of doctoral graduates at a top research university. Possible gender differences are explained by ability factors from an internal attribution perspective, and gender symbol signals from an external attribution perspective.

Literature review

Life development is irreversible. Gender differences in academic careers begin with differences in early life. After subtle “shaping,” labeling of stereotypes, and intensified discrimination, the gap between females’ academic career development and males’ is getting wider and wider in accumulating advantages.

Innate “gap” — Starting with Freud’s biological determinism

There is no doubt that there are innate physiological differences between genders. Freud’s (1987) biological determinism believed that men would be superior to women due to their physiological structure, but this statement was soon widely questioned by other scholars. While no consistent conclusion has been reached regarding whether males have gained career development advantages due to physiological advantages, the existence of gender differences in personality characteristics has been generally acknowledged by academia. It is harder for females to be in an advantageous position at work due to their dependence, avoidance of conflict, avoidance of taking credit, and non-complaining behaviors, as pointed out by Peterson (2018). In addition, females do not possess the same level of physical strength as men, which will also affect their ability to advance in their careers (Zhao & Bao, 2020). In terms of professional field preferences, males tend to choose interdisciplinary and more applied majors (Leahey, 2007), while females prefer to study family-related and health-related fields (Childress & Nault, 2019; Key & Sumner, 2019; Wullum Nielsen & Börjeson, 2019). Although these gender differences in personality and characteristics are innate and cannot be chosen, they will affect females’ academic career development.

Subtle “shaping” — Secret and real gender role socialization

The “gap” between female and male scholars will also continue to widen in the process of gender role socialization. Socialization theory believes that social status acquisition will be affected by social and cultural factors. Society, family, school, peers, and other groups pass on the gender concepts and norms that have been formed by society to individuals through formal education, family education, daily life, and work, enabling individuals to fulfill the gender roles required by society (Crespi, 2003; Gleitman et al., 2000). The gender role identity and expectations formed by an individual during the socialization process will have an impact on the individual’s career choices and expectations (Marini & Brinton, 1984; Zhu, 2012). The impact path of gender role socialization can be summarized as follows. On the one hand, it will affect individuals’ social role-playing expectations and further affect their concepts of work and making money (Betz & O’Connell, 1989; Granrose, 2005; Reskin & Padavic, 1994). On the other hand, it will drive men and women to seek employment based on their respective role positions and career preferences. Under the traditional social gender concept that “women work at home and men work outside”, male doctors prefer jobs with higher wages, while female doctors prefer stable jobs with low pressure, and enable them to spend time with their families (Lin & Tong, 2012). Gender role socialization explains the different preferences of males and females when choosing career types and reflects the problem of gender inequality. A family should be a shared property that requires both husband and wife to contribute time and energy; however, employers often place the responsibility of taking care of the family on females, even using it as an advantage for males. It is also an important reason why it is difficult for female doctors to obtain high-quality academic career development.

Unthinkingly “labels” — Emergence of gender stereotypes and symbolization

Both the above pieces of literature mainly explain gender differences from the perspective of females. Regarding how society spreads gender differences from the inside to the outside, and through what mechanism females are subjected to “absurd opinions” that they should not bear, the interaction between gender stereotypes and symbols in the sociology field provides an explanation.

Stereotypes refer to people’s innate stereotyped and fixed perceptions of different categories of people. This perception usually labels different categories of people in an extremely simple way and does not change the overall view of this group based on the specific person. Gender stereotypes refer to people’s generalizations and summaries about both genders' behavior and personality characteristics (Ma, 2000). Its existence has caused women in academic careers to bear labels that do not belong to them, such as “having to spend more time taking care of the family” and “worse scientific research ability” (Franco et al., 2021). It should be noted that compared with the actual differences in gender physiological characteristics, gender stereotypes are not supported by empirical data and may impose disadvantageous impressions that are not exclusively female on the entire female group.

The stereotypes currently imposed on females by society also play a role in gender symbolization. Females are generally symbolized as dependent, docile, and weak. Sociologist George Herbert Mead proposed the symbolic interaction theory, which believes that people understand others and themselves, form social relationships, as well as establish social structures through symbolic communication. In the widespread use of gender symbols and social interactions, social relations and social structures are affected; at the same time, both genders may further internalize the connotations behind the symbols during the symbolic interaction process, affecting the construction of females’ self-identities.

Intensified “discrimination” — Becker’s gender discrimination theory

Assuming that the explanations of “labels”, “gender symbols” and “gender stereotypes” are only at the “hidden level”, then the gender discrimination that arises from them is undoubtedly more irritating and harmful. Gary Becker’s discrimination theory believes that discriminatory behavior is caused by the unfair treatment of the weak by the strong and the negative stereotypes held by some groups against other groups. As a result of these negative impressions and unfair treatment, discriminatory behavior is formed as an inter-group control mechanism. In the event that discriminatory behavior occurs, the strong will gain certain benefits from the unfair treatment of the weak, thus encouraging the dominant group to perpetuate this discrimination and strengthening negative stereotypes of those who are disadvantaged. As a result, it achieves a cycle of self-satisfaction and dominance. In the academic labor market, males are generally considered to occupy the core position of resources in the academic community (Yuan, 2017). Men have more scientific research publications and better promotion opportunities than women (Schulze, 2015; Stack, 2004), while women are disadvantaged in terms of access to elite academic fields (Jin et al., 2018), wages (Lin & Tong, 2012), research publications (Sax et al., 2002; Shi & Chen, 2011), and promotions (Schulze, 2015). Based on the analysis of doctoral student survey data, researchers have found that 60% of females believed that they had been subjected to gender discrimination during the employment process, while only 1% of men believed that gender discrimination had taken place (Zhao et al., 2023).

Irreversible “snowball” — Analysis based on advantage accumulation and life cycle theory

The irreversibility of their emergence and development deserves more attention than the differences themselves. Merton’s theory of accumulated advantage believes that individual self-selection and social institutional selection affect people’s possibility of approaching opportunities in a particular field (Mendelsohn & Merton, 1989). When an individual’s role performance exceeds the standards required by the system, the process of accumulating advantages begins. In this process, the individual is provided with continuously expanding opportunities to further expand his or her access to the rewards associated with the job. As the direct bearers of population reproduction and participants in social production, females’ life cycles are closely linked to or even overlap with career development, and their reproductive and parenting functions have an important impact on career development (Tong & Pu, 2001). Academic committees pay special attention to females’ role in a family when recruiting young faculty members and are often unwilling to bear the additional cost of raising children that comes with hiring a female doctor. As a result, under the same conditions or when men are slightly worse off, they are more willing to hire male doctors (Wu & Jiang, 2015). Since elite institutions have relatively large resources, male doctors who are more likely to enter these institutions have a greater chance of gaining significant cumulative advantages. Some studies have found that this difference will affect their chances of entering elite institutions and various stages of future development, further widening the career development gap between female doctors and male doctors (Ying, 2009).

It can be seen from this that the academic career development of doctors of different genders may be affected by both individual abilities and gender discrimination in the workplace. Many studies have discussed the phenomenon of gender differences in career choice and development. There are, however, few empirical studies focused on the academic career field, and among those studies, most focus on the differences among genders in scientific research productivity, title promotions, and salaries. There is little in-depth analysis of the possible reasons for the gender difference, and it is difficult to answer questions such as: Whether gender differences in academic career development are due to differences in ability or gender symbols? Is there a distinction between academic disciplines in terms of gender differences?

Data and method

Data and variables

Doctoral graduates are the most likely to develop into leading scientific research talents in the future and are the best candidates for academic careers. Thus, in this study, ten years of doctors who graduated from a top university in China from 2008 to 2017 are studied, and two types of data are used to answer the research questions. The first type of data is administrative data, which does not only include a series of background characteristics of individual doctoral students, such as gender, ethnicity, admission method, admission department, graduation year, and graduation job destination, but also includes a series of characteristic variables of his mentor, such as position, and academic experience. The second type of data is the Chinese and English scientific research productivity data of doctoral students collected from the China National Knowledge Infrastructure and Web of Science websites, respectively. Compared with previous studies that mostly used questionnaire surveys to understand scientific research productivity, the data in this article have the advantage of being more objective and accurate. A total of 7,415 samples were analyzed, of which 3,651 chose academic careers upon graduation.

This study focuses on three types of dependent variables: (1) whether doctoral students choose an academic career when they graduate (binary variable, 1 if they choose an academic career, zero if they do not); (2) the level of the institution where the academic profession is located (divided into four categories: non-double world-class universities, double world-class universities while excluding Tsinghua University and Peking University, overseas universities, as well as Tsinghua University and Peking University); (3) scientific research productivity after entering the academic career (including the quantity, quality and influence of Chinese and English publications)¹. Gender is the primary explanatory variable, with females representing 36.6% of the total sample. Other control variables include a series of variables representing the characteristics of individual doctoral graduates and their academic advisors. Considering that different majors and graduation years may also have an impact on doctors’ academic career choice and development, this study also adds the fixed effect of major and year of graduation. The description and distribution of relevant variables are presented in Table 1.

Table 1.

Description of Variable Used.

Variables	Description	Mean	SD	Min	Max
Gender	female = 1; male = 0	0.366	0.482	0	1
Program duration: 3-year	Program duration	0.051	0.219	0	1
Program duration: 4-year	Program duration	0.392	0.488	0	1
Program duration: 5-year	Program duration	0.505	0.500	0	1
Program duration: 6-year	Program duration	0.052	0.222	0	1
Party membership	Yes = 1; No = 0	0.545	0.498	0	1
Ethnicity: Han	Yes = 1; No = 0	0.946	0.226	0	1
Academic advisor in talent program	Yes = 1; No = 0	0.149	0.357	0	1
Academic advisor is overseas returnee	Yes = 1; No = 0	0.274	0.446	0	1
Academic advisor with administrative positions	Yes = 1; No = 0	0.117	0.322	0	1
Nature sciences	Discipline classification	0.441	0.497	0	1
Engineering	Discipline classification	0.172	0.378	0	1
Liberal arts	Discipline classification	0.149	0.356	0	1
Social sciences	Discipline classification	0.237	0.426	0	1
Graduation year:2008	Graduation year	0.052	0.222	0	1
Graduation year:2009	Graduation year	0.084	0.277	0	1
Graduation year:2010	Graduation year	0.081	0.273	0	1
Graduation year:2011	Graduation year	0.100	0.300	0	1
Graduation year:2012	Graduation year	0.113	0.316	0	1
Graduation year:2013	Graduation year	0.120	0.325	0	1
Graduation year:2014	Graduation year	0.111	0.314	0	1
Graduation year:2015	Graduation year	0.113	0.317	0	1
Graduation year:2016	Graduation year	0.121	0.326	0	1
Graduation year:2017	Graduation year	0.106	0.308	0	1
Number of English publications in graduate school	Scientific research productivity	0.448	1.593	0	34
Quality of English publications in graduate school	Scientific research productivity	1.932	10.89	0	283.3
Influence of English publications in graduate school	Scientific research productivity	15.37	96.00	0	2561
Number of Chinese publications in graduate school	Scientific research productivity	3.797	8.916	0	219
Quality of Chinese publications in graduate school	Scientific research productivity	3.347	8.665	0	212.5
Influence of Chinese publications in graduate school	Scientific research productivity	45.90	143.9	0	4226
Number of English publications after graduation	Scientific research productivity	0.749	3.309	0	136
Quality of English publications after graduation	Scientific research productivity	4.087	22.98	0	1008
Influence of English publications after graduation	Scientific research productivity	18.40	160.1	0	11,000
Number of Chinese publications after graduation	Scientific research productivity	4.174	15.04	0	598
Quality of Chinese publications after graduation	Scientific research productivity	3.478	11.02	0	283.4
Influence of Chinese publications after graduation	Scientific research productivity	26.07	128.0	0	6886
Academic career	Yes = 1; No = 0	0.548	0.498	0	1
Level of institutions employed	Level of institutions	1.589	1.054	1	4

Model

In order to study whether there are gender differences in the academic career choices of doctoral graduates, this study mainly focuses on two factors. First, both students’ personal characteristics, as well as academic advisors’ characteristics may affect their employment choices. Further, many studies have shown a positive relationship between students' scientific research productivity in graduate school and whether they pursue academic careers. Therefore, this study uses a stepwise regression method to gradually control the personal characteristics of students, academic advisors’ characteristics, and scientific research productivity in graduate school. The model of multinomial logistic regression used in this study is as follows

{A ca}_{i} = β_{0} + β_{1} {gender}_{i} + β_{2} {ind}_{i} + β_{3} {fac}_{i} + β_{4} {pub}_{i} + γ + η + ε_{i}

{gender}_{i}

is the gender of student_i;

{ind}_{i}

is student_i’s personal characteristics, including the type of doctoral degree, ethnicity, and party membership;

{fac}_{i}

is the characteristics of the student_i’s academic advisor, including whether the academic advisor belongs to the talent program, whether the advisor is an overseas returnee, and whether the advisor has an administrative position;

{pub}_{i}

is the scientific research productivity of student_i in his or her graduate training, including the quantity, quality (impact factor) and influence (number of citations) published in both Chinese and English; γ and η refer to the fixed effect of major year of graduation, respectively.

An individual’s academic path is in part determined by the level of their educational institution once they decide to enter academics. To analyze whether there are gender differences in the levels of higher education institutions where doctoral graduates work this study takes doctoral graduates who choose academic occupations as the analysis objects. The model of ordered logistic regression used in this study is as follows

{A ca_C}_{i} = β_{0} + β_{1} {gender}_{i} + β_{2} {ind}_{i} + β_{3} {fac}_{i} + β_{4} {pub}_{i} + γ + η + ε_{i}

A young faculty member’s scientific research productivity will be an important factor in selecting and promoting them after choosing an academic career and enrolling in a higher education institution. To examine whether there are gender differences in scientific research productivity after engaging in academic careers, this study further constructs the following multiple regression model

{Pub_lx}_{i} = β_{0} + β_{1} {g ener}_{i} + β_{2} {ind}_{i} + β_{3} {fac}_{i} + β_{4} {pub}_{i} + β_{5} {A ca_C}_{i} + γ + η + ε_{i}

It should be noted that the scientific research productivity of doctor graduates after they engage in academic careers is influenced by the level of institutions they enroll in and the academic resources they acquire. We thus control for the influence of the quality of employment institutions on one’s scientific research productivity by adding the variable indicating one’s employment institution level ${A ca_C}_{i}$ in the above regression model.

To test whether there is academic discipline heterogeneity in gender differences, this study also conducts analyses on doctoral students in science, engineering, liberal arts, and social sciences based on the above model. The models are not described in detail here.

Results

Looming Shadow of gender role socialization: The academic career preferences of female doctors

It can be seen from the regression results in Table 2 that the probability of female doctoral students choosing an academic career after graduation is 12.1% higher than that of males (column 1). After controlling for personal characteristics, the probability of women choosing an academic career is still 8.6% higher than that of men. Further, after controlling for the characteristics of academic advisors and the fixed effects of graduation grade and major, the difference is 8.5% (column 3). Finally, after further controlling for scientific research productivity while studying, results suggest that the probability of women choosing an academic career is still 8.6% higher than that of men (column 4).

Table 2.

Gender Differences in Doctoral Students’ Academic Career Choices.

Variable	(1)	(2)	(3)	(4)
Female	0.121**	0.086*	0.085*	0.086*
Female	(0.049)	(0.050)	(0.050)	(0.050)
Party membership		−0.029	−0.024	−0.034
Party membership		(0.049)	(0.049)	(0.049)
Ethnicity: Han		0.012	0.004	−0.004
Ethnicity: Han		(0.105)	(0.105)	(0.105)
Academic advisor in talent program			0.163**	0.182**
Academic advisor in talent program			(0.073)	·(0.073)
Academic advisor is overseas returnee			0.162***	0.159***
Academic advisor is overseas returnee			(0.056)	(0.056)
Academic advisor with administrative positions			−0.313***	−0.328***
Academic advisor with administrative positions			(0.081)	(0.081)
Program duration		Yes	Yes	Yes
Fixed effects of graduation year		Yes	Yes	Yes
Fixed effects of major		Yes	Yes	Yes
Scientific research productivity in graduate school				Yes
Constant	0.149***	0.245	0.243	0.109
Constant	(0.029)	(0.175)	(0.177)	(0.180)
N	7,415	7,415	7,415	7,415

Two main factors influence an individual’s personal career choice: one is an individual’s career choice preference; and the other is whether individual abilities match professional requirements. The above regression results show that at the same level of scientific research ability, females are still more likely to enter academic careers. This not only shows that females’ abilities can fully meet the requirements of academic careers but also illustrates females’ preference for academic careers. One possible reason is that academic careers are more flexible, controllable, and stable than other careers, and can better meet females’ family needs. From another perspective, the preference of female doctors for academic careers reflects the socialization needs of females’ roles, which can prove that gender role socialization has a subtle impact on the career choices of males and females. This division of gender roles will make females more inclined to choose careers that are more compatible with family responsibilities when faced with career choices. However, this does not mean that females’ career choices are only subject to social factors. Personal abilities and personal career choice preferences are also critical factors.

Stereotypes under gender symbols: Gender inequality in the academic labor market

The regression results in Table 3 suggest that among the group of doctoral students who choose academic careers, the level of the institution where female doctors enter is significantly 0.341-level lower than that of male doctors. Even after controlling for personal characteristics and academic advisor’s characteristics, as well as graduation grades and major fixed effects, the level of the institution where female doctors enter is still significantly lower than that of men by 0.277-level. It means that even when men and women are similar in various other characteristics, women are still at a disadvantage in the competition in the academic labor market. We further control the scientific research productivity during their doctoral training to examine whether this difference is due to women’s weaker scientific research ability. Still, results suggest that the level of institutions where women enter is significantly lower than that of men by 0.272-level. From one aspect, it suggests that gender discrimination may exist in the academic labor market. One possible explanation for this discrimination is gender stereotypes. Gender stereotypes emphasize the stereotyped and simplified impressions formed by people based on fixed cognitive models of gender roles. This stereotype does not equate to the fact that this gender difference exists between individuals of different genders. The underestimation of women’s scientific research capabilities and the over-focus on women’s family responsibilities are common gender stereotypes in the current academic labor market. Another possible explanation is that men have an advantageous position in the job market. Compared with academic careers, there are many other job options with high pay and high status. Therefore, when men only get teaching positions in lower-level schools, they often There are other non-academic careers available. In contrast, women have fewer choices. Therefore, women who obtain teaching positions in lower-level schools are more likely to accept this teaching position because they prefer academic careers.

Table 3.

Gender Differences in Doctoral Students’ Education Institutions Employed.

Variable	(1)	(2)	(3)	(4)
Female	−0.341***	−0.273***	−0.277***	−0.272***
Female	(0.075)	(0.078)	(0.078)	(0.078)
Party membership		−0.135*	−0.116	−0.133*
Party membership		(0.075)	(0.076)	(0.076)
Ethnicity: Han		−0.340**	−0.341**	−0.337**
Ethnicity: Han		(0.154)	(0.154)	(0.156)
Academic advisor in talent program			0.155	0.187*
Academic advisor in talent program			(0.107)	(0.108)
Academic advisor is overseas returnee			0.196**	0.181**
Academic advisor is overseas returnee			(0.084)	(0.085)
Academic advisor with administrative positions			0.098	0.037
Academic advisor with administrative positions			(0.130)	(0.132)
Program duration		Yes	Yes	Yes
Fixed effects of graduation year		Yes	Yes	Yes
Fixed effects of major		Yes	Yes	Yes
Scientific research productivity in graduate school				Yes
N	3,651	3,651	3,651	3,651

Gender symbols in scientific research publications: Gender segregation in Chinese publications

Gender symbols in scientific research publications refer to the existence of gender stereotypes in the process of scientific research publication, which leads to inequality between men and women in publishing papers and obtaining academic status under the same conditions. The regression results in Tables 4 and 5 show no significant difference between women’s English scientific research productivity and men’s. However, after controlling for relevant characteristic variables and scientific research productivity, the quality of women’s Chinese scientific research productivity is significantly lower than that of men. This shows that although women are not inferior to men in the number of Chinese publications, they are not as good as men in the quality of their publications. There are several possibilities for the existence of this gender segregation. First, reviewers and editors are more stringent about women publishing papers. Second, gender stereotypes and prejudices against women cause them to face greater challenges in scientific research publications. Third, it is easier for men to establish informal relationships. Men have more opportunities to travel to other places and participate in social activities such as academic conferences; therefore, it is easier for men to establish informal relationships with journal reviewers and editors compared to women. The impact of such gender symbols on scientific research publications not only affects women’s career prospects and academic reputation but also affects the fairness and development of the entire scientific research field.

Table 4.

Gender Differences in Doctors’ English Publications.

	After graduation: Number				After graduation: Quality				After graduation: Influence
Variable	Control gender	Control characteristics	Control output	Control employment	Control gender	Control characteristics	Control output	Control employment	Control gender	Control characteristics	Control output	Control employment
Female	−0.176	−0.042	0.007	0.024	−0.760	−0.216	−0.098	0.027	−0.312	2.587	3.007	3.301
	(0.132)	(0.132)	(0.112)	(0.112)	(0.916)	(0.918)	(0.791)	(0.791)	(7.077)	(7.158)	(6.797)	(6.807)
Personal and advisor’s characteristic		Yes	Yes			Yes	Yes			Yes	Yes
Fixed effect of graduation year		Yes	Yes			Yes	Yes			Yes	Yes
Fixed effect of major
research Productivity in doctoral training		Yes	Yes			Yes	Yes			Yes	Yes
Level of institutions			Yes				Yes				Yes
N	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651

Table 5.

Gender Differences in Doctors’ Chinese Publications.

	After graduation: Number				After graduation: Quality				After graduation: Influence
Variable	Control gender	Control characteristics	Control output	Control employment	Control gender	Control characteristics	Control output	Control employment	Control gender	Control characteristics	Control output	Control employment
Female	−0.571	−0.658	−0.528	−0.622	−0.719*	−0.801**	−0.583*	−0.653*	−7.524	−8.112	−7.434	−7.972
	(0.473)	(0.472)	(0.446)	(0.446)	(0.400)	(0.395)	(0.340)	(0.339)	(5.588)	(5.577)	(5.013)	(5.017)
Personal and advisor’s characteristic			Yes				Yes				Yes
Fixed effects of graduation year		Yes	Yes			Yes	Yes			Yes	Yes
Fixed effects of major
Scientific research productivity in graduate school			Yes				Yes				Yes
Level of institutions
N	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651	3,651

Disciplinary heterogeneity of gender symbols—different responses to gender symbols in different disciplines

The impact of gender symbols may show different manifestations due to the heterogeneity of discipline characteristics. According to Table 6, female doctoral graduates from different disciplines have heterogeneous preferences for academic careers, the level of teaching institutions, and scientific research productivity. In specific, there is no significant difference between male and female doctors in science majors in terms of academic career choice, institution level, and scientific research productivity after entering the academic career. Female doctoral graduates in engineering majors are significantly lower than males at the level of institutions they enter. Female doctors majoring in liberal arts are significantly less likely to pursue an academic career after graduation and have significantly lower scientific output after pursuing an academic career compared to their male counterparts. After entering academic careers, female doctoral graduates in social sciences are more likely to enter academic careers than their male peers, while they are significantly lower than men in terms of institution level and scientific research productivity.

Table 6.

Doctors’ Scientific Research Productivity From Different Disciplines.

Variable	Academic career	Level of institution	Number of English publications	Quality of English publication	Influence of English publication	Number of Chinese publications	Quality of Chinese publication	Influence of Chinese publication
Nature sciences: Female	0.037	−0.167	0.108	0.914	13.49	0.237	0.208	5.094
	(0.077)	(0.118)	(0.253)	(1.800)	(16.170)	(0.350)	(0.406)	(5.504)
N	3,269	1,536	1,536	1,536	1,536	1,536	1,536	1,536
Engineer: Female	0.113	−0.665***	−0.176	−2.099	−9.400	0.929	1.511	12.01
	(0.133)	(0.203)	(0.220)	(1.693)	(6.923)	(2.456)	(0.998)	(8.686)
N	1,278	578	578	578	578	578	578	578
Liberal arts: Female	−0.332**	−0.062	−0.0317**	−0.125**	−0.230	−1.886**	−0.706**	−5.892**
	(0.132)	(0.198)	(0.015)	(0.056)	(0.156)	(0.881)	(0.348)	(2.860)
N	1,107	650	650	650	650	650	650	650
Social science: Female	0.450***	−0.451**	0.020	0.003	−0.792	−1.721**	−2.830***	−28.97*
	(0.102)	(0.177)	(0.013)	(0.052)	(0.900)	(0.693)	(0.989)	(17.090)
N	1,761	887	887	887	887	887	887	887

Note. All models controlled personal and advisor characteristics, scientific research productivity in graduate school, and graduation year fixed effects. The regression of various indicators published in Chinese and English also controlled scientific research productivity in graduate school.

In short, the results suggest that the impact of gender symbols differs in different academic disciplines. Female doctors majoring in science are less affected by the impact of gender symbols on their academic careers, and they have no gender disadvantages before and after entering academic careers. Women majoring in engineering and humanities are more or less impacted by gender symbols before and after entering academic careers, causing them to be disadvantaged in certain aspects of academic career development. Those most affected by gender symbols are female doctors in social sciences. Although they are more likely to enter academic careers, they face disadvantages in the hierarchy of teaching institutions and scientific research publications after entering academic careers.

Conclusion and discussion

In recent years, the number of females obtaining higher education opportunities, getting doctorates, and entering academic research has increased rapidly in China. In 2021, the proportion of female doctors nationwide has reached 41.87%, an increase of 6.4% from 2012 (MOE, n.d.)— an obvious result of China’s educational reform and social progress. Despite the increased participation of females in academic fields, however, they still face many difficulties and challenges in their career development. Existing research has attributed this to internal and external factors, of which internal factors mainly involve gender differences in abilities and personality traits. This view is related to biological determinism as well as gender role socialization theory. However, even after controlling internal factors such as scientific research ability, our results suggest that gender differences persist in academic career choice and development, illustrating the limitations of internal attribution.

External factors mainly come from a society’s gender stereotypes and gender discrimination. The existence of gender symbols and stereotypes makes females in academia often bear labels such as “need to spend more time taking care of the family” and “worse scientific research ability.” These labels undoubtedly pose a major obstacle to females’ academic career development, leaving them at a competitive disadvantage even if their abilities are the same as their male counterparts. Additionally, Becker’s gender discrimination theory further reveals the social mechanism behind this phenomenon. In the current academic labor market, the male preference for “similar abilities, giving priority to men” prevails. This preference not only stems from social structure and culture but also reflects the recognition of gender symbols and stereotypes in the academic labor market to a certain extent. This phenomenon of male preference has undoubtedly exacerbated the dilemma of females in academic career development. This study confirms that females are currently in a disadvantaged position in the academic labor market and academic career development. Meanwhile, this study reveals that this disadvantaged position is not mainly due to the difference in scientific research capabilities between genders; it is the gender symbols and stereotypes conveyed by the word “female” that are the decisive factors why females face many difficulties in academia.

For higher education institutions such as universities and scientific research institutions, more attention should be paid to the examination of applicants’ scientific research capabilities and comprehensive qualities, rather than evaluating applicants through symbols and stereotypes such as gender. The result is not only a reflection of fair treatment of individuals, but also a necessary path to promote gender equality and the healthy development of the academic field. Our research is consistent with many gender equality theories, such as liberal gender equality theory and social constructionist gender equality theory.

Instead of using gender as the main criterion for evaluation, gender discrimination should be eliminated, gender equality should be achieved, as well as individual abilities and contributions should be emphasized. To achieve this goal, we need to challenge and change existing gender stereotypes. This not only requires reforms at the social and policy levels but also provides more support and guarantees to female scholars, such as establishing fairer recruitment procedures and providing more flexible working time arrangements. Meanwhile, changes need to be made at the cultural and conceptual levels to challenge and eliminate stereotypes and gender symbols about females. Finally, a society’s awareness and understanding of gender issues also need to be deepened. This includes deepening the public’s understanding of gender discrimination, increasing the academic community’s attention to gender equality, and increasing employers’ enforcement of gender equality. Only in this way can we truly promote the progress of gender equality and achieve fairness and justice in academia.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article:

This work was supported by the Ministry of Education Liberal Arts and Social Sciences Youth Fund Project “Academic Career Choice and Changing Trends of Doctoral Students in World-Class Universities: The Perspective of Academic Labor Market Changes” 21YJC880055); National Natural Science Foundation of China General Program “Research on the Structure and Career Development of Teachers in Higher Education Institutions in China” (71974007); Peking University Lanyuan Academy Academic Fund Project “Cultivation of Female Scholars and Academic Career Development”.

ORCID iD

Shuyi Zhao

Note

References

Betz

O'Connell

(1989). Work orientations of males and females: Exploring the gender socialization approach. Sociological Inquiry, 59(3), 318–330. https://doi.org/10.1111/j.1475-682x.1989.tb00109.x

Centra

J. A.

Gaubatz

N. B.

(2000). Is there gender bias in student evaluations of teaching? The journal of higher education, 71(1), 17–33. https://doi.org/10.1080/00221546.2000.11780814

Childress

Nault

J. F.

(2019). Encultured biases: The role of products in pathways to inequality. American Sociological Review, 84(1), 115–141. https://doi.org/10.1177/0003122418816109

Crespi

(2003). Gender socialization within the family: A study on adolescents and their parents in Great Britain. Department of Sociology Catholic University of Milan.

Franco

M. C.

Helal

Cenci

M. S.

Moher

(2021). The impact of gender on researchers’ assessment: A randomized controlled trial. Journal of Clinical Epidemiology, 138(1), 95–101. https://doi.org/10.1016/j.jclinepi.2021.05.026

Freud

(1987). New introduction to Psychoanalysis [M]. Translated by Gao Juefu (pp. 157–160). Beijing Press.

Gleitman

Fridlund

A. J.

Reisberg

(2000). Basic Psychology[M]. W. W. Norton & Company, Inc.

Granrose

C. S.

(2005). Employment of Women in Chinese Cultures. Edward Elgar Publishing. https://doi.org/10.4337/9781845428068

Harris

G. T.

(1990). Research output in Australian university economics departments: an update for 1984-88. Australian Economic Papers, 29(55), 249–259. https://doi.org/10.1111/j.1467-8454.1990.tb00520.x

10.

Jin

Wang

Lin

(2018). The academic career destination and level of female engineering doctorates—an analysis based on Tsinghua University’s doctoral graduates from 2005 to 2014 (03) (pp. 1–7). Graduate Education Research.

11.

Key

E. M.

Sumner

J. L.

(2019). You research like a girl: Gendered research agendas and their implications. PS: Political Science & Politics, 52(4), 663–668. https://doi.org/10.1017/s1049096519000945

12.

Leahey

(2007). Not by productivity alone: How visibility and specialization contribute to academic earnings. American Sociological Review, 72(4), 533–561. https://doi.org/10.1177/000312240707200403

13.

Chen

(2021). Is the academic performance of female doctoral students better or worse than that of male doctoral students? analysis based on national survey data of doctoral graduates [J]. Graduate Education Research, 7(01), 63–71. https://doi.org/10.19834/j.cnki.yjsjy2011.2021.01.10

14.

Lin

Tong

(2012). The career of female professionals from a gender comparative perspective. Proceedings of the forum “research on the development Laws and Policies of female talents in China” at the 2012 China sociology Annual conference, July 2012, 2012:4–29.

15.

Liang

Zhou

(2016). Social capital, collaboration and the “research productivity puzzle”—An analysis based on the experience of teachers at Chinese research universities. Peking University Education Review, (02), 133–156. https://doi.org/10.19355/j.cnki.1671-9468.2016.02.009

16.

(2000). Gender stereotypes and gender education. Education review, 1(06), 35–37.

17.

Marini

M. M.

Brinton

M. C.

(1984). 11 Sex typing in occupational socialization. SPONS AGENCY 199.

18.

Mendelsohn

Merton

R. K.

(1989). Robert K. Merton: The celebration and defense of science. Science in Context, 3(1), 269–289. https://doi.org/10.1017/s0269889700000806

19.

Ministry of Science and Technology of the People’s Republic of China . (2021, July 21). “Several measures to support female scientific and technological talents to play a greater role in scientific and technological innovation” Policy Interpretation. https://www.most.gov.cn/xxgk/xinxifenlei/fdzdgknr/fgzc/zcjd/202107/t20210720_175987.html

20.

Peterson

J. B.

(2018). 12 rules for life: An antidote to chaos. Random House Canada.

21.

Ren

(2013). Research on factors influencing employment of doctoral students based on gender differences. Journal of Ocean University of China, 2013(06), 45–50.

22.

Reskin

B. F.

Padavic

(1994). Women and men at work (pp. 3–97). Pine Forge Press.

23.

Sax

L. J.

Hagedorn

L. S.

Arredondo

Dicrisi III

F. A.

(2002). Faculty research productivity: Exploring the role of gender and family-related factors. Research in Higher Education, 43(4), 423–446. https://doi.org/10.1023/a:1015575616285

24.

Schulze

(2015). The gender wage gap among PhDs in the UK. Cambridge Journal of Economics, 39(2), 599–629. https://doi.org/10.1093/cje/bev001

25.

Shi

Chen

(2011). An exploration into the mystery of scientific research productivity of female scientific and technological personnel from the perspective of gender differences—an empirical investigation based on Hubei Province. Journal of China Women's University, 23(02), 60–65.

26.

Stack

(2004). Gender, children and research productivity. Research in Higher Education, 45(8), 891–920. https://doi.org/10.1007/s11162-004-5953-z

27.

The Ministry of Education of the People’s Republic of China (MOE) . (nd). Number of Female Students of Schools by Type and Level. https://www.moe.gov.cn/jyb_sjzl/moe_560/2020/quanguo/202108/t20210831_556360.html

28.

Tong

(2001). Theoretical framework for studying the career development of urban women. Women's Studies Series, 4(03), 4–10.

29.

Wang

Hao

(2006). Beyond the Ivory Tower: A review of the latest trends in Australian PhD employment expectations. Liaoning Education Research, 4(07), 86–89. https://doi.org/10.16697/j.cnki.xdjygl.2006.07.030

30.

Jiang

(2015). Gender differences in the employment process and career selection outcomes of doctoral students in Beijing universities. China Science and Technology Forum, 1(02):125–129. https://doi.org/10.13580/j.cnki.fstc.2015.02.023

31.

Wullum Nielsen

Börjeson

(2019). Gender diversity in the management field: Does it matter for research outcomes? Research Policy, 48(7), 1617–1632. https://doi.org/10.1016/j.respol.2019.03.006

32.

Yan

Shang

(2021). The development of female doctoral education since the reform and opening up: Achievements, motivations and prospects.01 (pp. 84–93). Degrees and Graduate Education. https://doi.org/10.16750/j.adge.2021.01.013

33.

Ying

(2009). Gender differences and career achievements of doctoral graduates. Women's Studies, 1(06), 38–42.

34.

Yuan

(2017). An analysis of gender differences in scientific research productivity of university teachers and its causes - based on the experience analysis of education teachers at a research-oriented university. Higher Education Exploration, 6(03), 5–12.

35.

Liang

(2019). Are returnee teachers really better than local teachers?—Evidence from the education discipline of research universities. Education and Economics, (01), 75–86.

36.

Zhao

Bao

(2020). Reconstruction of the research paradigm on women’s academic career development—research on female college teachers from a multi-dimensional perspective. Educational Academic Monthly, 05(02), 67–76. https://doi.org/10.16477/j.cnki.issn1674-2311.2020.05.009

37.

Zhao

Shen

(2010). An empirical analysis of doctoral students’ employment preferences. Degrees and Graduate Education, 05(04), 53–56. https://doi.org/10.16750/j.adge.2010.04.015

38.

Zhao

Shen

Zhu

(2023). Women should not give way to men? ——a study on gender differences in elite academic positions obtained by engineering PhDs. Journal of East China Normal University, 1(05), 84–97. https://doi.org/10.16382/j.cnki.1000-5560.2023.05.007

39.

Zhu

Yang

Jing

Zhu

Zhang

(2012). Gender differences in salary expectations among college students: A perspective based on gender role socialization. Cardiovascular Diabetology, 11(03): 90–97. https://doi.org/10.1186/1475-2840-11-90

Ability difference or gender symbolism? An empirical research on gender differences in academic career development of doctoral graduates in China

Abstract

Keywords

Introduction

Literature review

Innate “gap” — Starting with Freud’s biological determinism

Subtle “shaping” — Secret and real gender role socialization

Unthinkingly “labels” — Emergence of gender stereotypes and symbolization

Intensified “discrimination” — Becker’s gender discrimination theory

Irreversible “snowball” — Analysis based on advantage accumulation and life cycle theory

Data and method

Data and variables

Model

Results

Looming Shadow of gender role socialization: The academic career preferences of female doctors

Stereotypes under gender symbols: Gender inequality in the academic labor market

Gender symbols in scientific research publications: Gender segregation in Chinese publications

Disciplinary heterogeneity of gender symbols—different responses to gender symbols in different disciplines

Conclusion and discussion

Footnotes

Declaration of conflicting interests

Funding

ORCID iD

Note

References