On the effect of gender on secondary school students’ causal attributions to choose or abandon physics & chemistry

Abstract

Secondary school students’ early choices related to staying in the science track define their future decisions to choose chemistry at college. This investigation aims at analyzing the role of gender in students’ causal attributions to choose or abandon chemistry when it first becomes optional in the Spanish educational system. Our analyses uncovered a relevant effect of gender in the students’ decision, boys being more likely to choose physics & chemistry when they face, for the first time, the possibility of continuing or opting out the subject. Besides, students’ causal attributions to the subject relationship with mathematics and to friends are affected by gender regardless of the students’ level of motivation. In turn, there is a gender effect in attributions to friends and media only in the case of highly-motivated students. A multinomial logistic regression model revealed that gender is a strong predictor of the students’ decision. The regression model also uncovered a significant interaction effect between gender and attributions to the subject relationship with mathematics, girls becoming less likely to choose physics & chemistry when the latter increase. Our results highlight the need of working on the students’ and families’ stereotypes and propose gender-balanced teaching models to close the gap between girls’ and boys' attitudes, motivation, and anxiety towards mathematics in the context of physics & chemistry teaching and learning.

---

Introduction

The society of the 21st century is facing scientific and technological challenges of unprecedented relevance. This fact not only reinforces the need of citizens with enough scientific knowledge and understanding to be competent in such model of society (Roth and Lee, 2004; Gil-Pérez and Vilches, 2005; Blanco-López et al., 2015; OECD, 2016), but also requires qualified professionals working in STEM fields (Science, Technology, Engineering and Mathematics). This way, science education plays a key role in children's and teenagers’ schooling (see, for example, European Commission, 2004; Rocard et al., 2007; Dillon, 2009). Nevertheless, STEM subjects are not always mandatory along the entire secondary education and the decrease in the number of students taking them when they become optional has become an alarming matter to work on (see, for example, Lyons, 2006a,b; Oon and Subramaniam, 2010; Ulriksen et al., 2010; Bennett et al., 2013).

Besides, a gender gap in STEM has been observed at all levels, from the very first moment when students have to make their decision on pursuing the STEM academic track or not, to the professional level, when they enter STEM-based job positions in both industry and academia (see, for example, Isaacs, 2001; Jacobs, 2005; Cheryan et al., 2017; Wang and Degol, 2017). However, this gap depends significantly on different aspects, as for example the culture or the specific discipline. Regarding the former, several recent reports claim that cultural factors are more important than individual ones (see, for example, Miner et al., 2018; Sachdev, 2018; Aldén and Neuman, 2022). Nevertheless, paradoxically the largest STEM gaps appear in countries with higher levels of gender equality, as found by Stoet and Geary (2018). In relation to the particular discipline, whereas fields like physics or engineering are mainly occupied by men, women are more numerous in health-related disciplines (Mau, 2003; Jacobs, 2005; Smyth and Hannan, 2006; Su and Rounds, 2015; Delaney and Devereux, 2019; Kang et al., 2019). Although chemistry is generally considered as a “neutral” field (see, for example, Smyth and Hannan, 2006), some recent studies suggest that women tend to choose chemistry more than men at high school and university levels (Avargil et al., 2020), or alternatively, that chemistry is still considered as a masculine subject (Cousins and Mills, 2015).

It has also to be acknowledged that this gender gap in STEM has significantly narrowed in the last three to four decades, which makes the underrepresentation of women in most STEM fields less important than in the end of the last century (Jacobs, 2005; Eccles, 2011; Wang and Degol, 2017). The number of girls entering STEM degrees at university is, of course, influenced by the subject choice at high school (Delaney and Devereux, 2019). Interestingly, already at high school, the gender gap seems to be less significant when the students are closer to finish secondary education (Sahin and Waxman, 2020). In fact, as pointed out by Isaacs (2001), the problem is the low rates of enrolment of girls in such studies, and not that they abandon the courses once they are in college.

These gender differences in science at all levels are still a matter of concern, making them a popular field of research. In particular, unravelling the underlying reasons for such situation is of special interest, as well as increasing the participation of female students in science (Wang and Degol, 2017). In a previous work, we studied the students’ causal attributions to choose or abandon chemistry when it first becomes optional in the Spanish educational system, i.e. at the age of 15–16, and we highlighted the effect of the students’ science motivation on these attributions (Ardura et al., 2021). Herein we will tackle the investigation of the effect of gender on these attributions. This way, we aim at getting a deeper knowledge into the gender differences existing in the subject physics & chemistry in the very first moment when chemistry becomes optional in Spain, as this is normally the triggering decision of the gender gap also existing in higher academic and/or professional levels. In the theoretical framework, we will first briefly summarize the key variables uncovered by previous research on chemistry choice and then we will focus on the mediating effect of gender on these variables.

Theoretical framework

Relevant variables in the choice of chemistry

The choice of science studies in general, and of chemistry in particular, is affected by a number of different factors, often even interrelated (Bennett et al., 2013; Sha et al., 2015). This makes the understanding of the decision rather complex and has actually led to extensive research in the last decades (see, for example, Cerinsek et al., 2013; Potvin and Hasni, 2014; Nugent et al., 2015; Palmer et al., 2017; Shirazi, 2017; Reinhold et al., 2018; Ardura et al., 2021; Shwartz et al., 2021). These underlying variables that explain the choice of chemistry can be mainly divided into intrinsic (e.g. motivation, interest, prior academic achievement or gender) and extrinsic (e.g. participation in out-of-school activities, support from family and friends or the influence of media) to the students (Palmer et al., 2017). The extrinsic factors are related to what is known as science capital (Archer et al., 2014), which in our case collates all kinds of economic, social and cultural capital or resources that might give value for students to support and enhance their participation in science.

One of the most important intrinsic factors is motivation towards science, as it has a strong effect on the persistence in science courses when they become optional, both at secondary and at tertiary education. Nevertheless, motivation comprises several traits and not all of them contribute equally neither to the overall motivation towards science, nor to the students’ future preferences for scientific studies (Glynn et al., 2011; Salta and Koulougliotis, 2015, 2020; Schumm and Bogner, 2016; Ardura and Pérez-Bitrián, 2018). Additionally, motivation plays a key role on students’ causal attributions to opt for or out chemistry (Ardura et al., 2021). Besides, interest in science (Bøe, 2012; Cerinsek et al., 2013; Sha et al., 2015) and attitudes towards science (Osborne et al., 2003) also influence students’ future choices.

Students’ achievement in prior courses is also an extremely important factor affecting students’ choice of staying in science (Palmer et al., 2017), yet previous grades in scientific subjects seem to be more significant than final grades (Anderhag et al., 2013). This prior achievement is somehow related to the students’ perceived difficulty of the subject, both being related and therefore having a key influence in their decision of pursuing scientific studies at all levels (House and Telese, 2017; England et al., 2019). These two factors are also influenced by the whole learning experience, which also by itself plays a pivotal role in the decision. In particular, teachers and their methodologies in the classroom are of key importance to promote students’ future persistence in science. The most effective pedagogical practices are those that put the focus on the student and promote their active learning and, consequently, they help retain students in the science track (Broman and Simon, 2015; McDonald, 2016; Shirazi, 2017). Nevertheless, school science is usually less authentic than real science (Osborne et al., 2003; Braund and Reiss, 2006) and, therefore, it must be complemented by out-of-school activities (Braund and Reiss, 2006). Of course, both kind of science-learning activities influence students’ future decisions (Sha et al., 2015), yet out-of-school learning opportunities are, in general, more positive for improving motivation, interest, and attitudes towards science (Potvin and Hasni, 2014). This eventually influences their future career choices (Dabney et al., 2012). Related to this, also the mathematical ability of students is a key factor for them to succeed in science (Wagner et al., 2002; Lewis and Lewis, 2007; Cooper and Pearson, 2012; Scott, 2012; Xu et al., 2013).

As extrinsic factors, the influence of different people and external stimuli play a decisive role although, for example, they do not seem to be as important as personal factors such as self-efficacy, according to Avargil et al. (2020). Parents have a fundamental influence in students’ science behaviors (Lyons, 2006b; Stake, 2006; Buday et al., 2012; Vedder-Weiss and Fortus, 2013; Ing, 2014; Mujtaba and Reiss, 2014; Nugent et al., 2015; Simpkins et al., 2015; Sha et al., 2016; Halim et al., 2018), as they can positively impact their achievement, interest, and involvement in science (Halim et al., 2018), which eventually boosts their motivation and therefore helps them opt for future science studies (Simpkins et al., 2015; Sha et al., 2016). Parents’ level of education (Stokking, 2000; Maltese and Tai, 2011; Harackiewicz et al., 2012; Anderhag et al., 2013) or even their income (Moakler Jr. and Kim, 2014; Avargil et al., 2020) have also been discussed to be related to the choice of science. The importance of the relationships among classmates and friends, so crucial during adolescence due to their key role in the development of the personality, also affects the reciprocal influences in the attitudes and motivation towards science of students (Osborne et al., 2003; Vedder-Weiss and Fortus, 2013). Nevertheless, the impact of peers on career aspirations is complex and not clearly understood yet (Wang and Staver, 2001). In any case, good teachers are more influential than parents and peers (Cerinsek et al., 2013).

Gender effects on the choice of chemistry

The lower number of women in STEM in comparison to men is a complex problem, in which many factors play a role (Blickenstaff, 2005; Brotman and Moore, 2008; Kelly, 2016; Lv et al., 2022). As this requires a holistic strategy to approach a solution, the effect of gender on students’ decision of choosing science in general, and chemistry in particular, has been thoroughly studied at different levels in the last few decades. As already stated in the introduction, secondary school boys seem to be more likely to choose mathematics or physics than girls. In turn, biology and health-related degrees are preferred by girls (Mau, 2003; Jacobs, 2005; Smyth and Hannan, 2006; Su and Rounds, 2015; Delaney and Devereux, 2019; Kang et al., 2019). Of course, high school is a decisive moment in the choice of science, as the subject selection and the environment at secondary education influence future enrolment at university (Legewie and DiPrete, 2014; Jacob et al., 2020). Anyway, gender disparities in college STEM can be found not only in participation, but also in achievement or affective measures (Eddy and Brownell, 2016), although the choice of a particular educational pathway does not seem to be different for males and females (Guo et al., 2015). In the following, the influence of gender in the most important factors involved in the choice of science studies will be summarized.

Motivation and attitudes. In several studies, no significant or little differences between genders regarding the choice of physics and chemistry were found (Stokking, 2000; Sheldrake et al., 2017; Salta and Koulougliotis, 2020). In contrast, distinct gender effects were found in motivational variables, depending on the students’ choice or withdrawal of physics and chemistry the first time they must make the decision. For example, motivation seems to be typically driven by intrinsic interest in the case of women (Kelly, 2016). Extrinsic motivation is significantly important and, among those who did not study physics and chemistry anymore, boys were more motivated towards a future career in science than girls, whereas such effect was not present in those who did study such subject (Ardura and Pérez-Bitrián, 2018). However, a higher career motivation in girls, as well as a higher intrinsic motivation, was found by Salta and Koulougliotis (2015). Interestingly, although Ardura and Pérez-Bitrián (2018), Ardura and Galán (2019) or Salta and Koulougliotis (2015) did not find any gender effect on self-efficacy, Cheryan et al. (2017) suggested that the gender difference in self-efficacy, which is lower for females than for males (Huang, 2013), was one of the reasons accounting for the gender inequality in university STEM fields. Similarly, Kelly (2016) or Fisher et al. (2020b) also highlighted how the underestimation of girls’ abilities led them to display lower confidence and self-efficacy than boys. Nevertheless, the interaction between gender and motivation does not seem to play a decisive role in explaining students’ future preferences for physics and chemistry at the secondary level.

Although there is no agreement in the literature, girls and boys also have different attitudes towards participation in science. This has led not only to the underrepresentation of women in science degrees, since the way science was traditionally presented generated more favorable attitudes in boys than in girls, but also, consequently, in science-related jobs (Brotman and Moore, 2008). In fact, already during the high school years, the percentage of females interested in a STEM career decreased, in contrast to that of males, which remained stable (Sadler et al., 2012). In a recent study (Moore and Burrus, 2019), it was found that the most predictive components for STEM majors and career choice, namely attitudes and interests, are more important in the case of females. Nevertheless, Akpınar et al. (2009) already found a better attitude towards science learning and significant higher interest in science in the case of females. Interestingly, they also found that one of the reasons behind the decline in students’ attitudes while grade levels increased was the belief that science is a difficult subject, which can definitely affect future learning experiences.

Difficulty and achievement. It is well-known from several recent studies, that girls outperform boys on science at different educational levels (Acar et al., 2015; Eddy and Brownell, 2016). Interestingly, according to the study performed by Delaney and Devereux (2019) in Ireland, gender gaps are smaller among high-achieving students. Different explanations have been given for these differences in achievement in science, including the importance of noncognitive factors (Turner and Lindsay, 2003), the utility value of science (Acar et al., 2015), the higher achievement motivation of girls (Fischer et al., 2013) or different socialization experiences (Adamuti-Trache and Sweet, 2013). It is also important to mention that low grades in scientific subjects definitely help students decide not to continue in science, this effect being more important for girls (Ost, 2010).

Stereotypes, science identity and belonging. One of the most important factors responsible for the gender gap in STEM is the gender-related stereotypes and from that, the different sense of belonging of both genders to science (Lane et al., 2012; Cheryan et al., 2017; Wang and Degol, 2017; Fisher et al., 2020b). Gendered perceptions of academic fields of course influence the choice of STEM studies in a number of ways, as some of them are considered to be more masculine or more feminine (Farenga and Joyce, 1999; Brotman and Moore, 2008; Simon et al., 2017; Beutel et al., 2019; Makarova et al., 2019). Whereas the former are associated to objectivity or rigor and include normally STEM fields, the latter seem to have more to do with arts, humanities, and health care, where subjectivity or occupational values are more important. In fact, this gender segregation by field of study seems to be present even after the socioeconomic modernization experienced in the last decades (Charles and Bradley, 2009).

Interestingly, Bond (2016) found that televised stereotypes are more powerful in activating the existing gender schema, contrary to counter-stereotypes. However, it is fair to say that, despite usually females and minority characters are still underrepresented in some STEM television programs addressed to children, there are a few emerging strengths including representation of characters as equal regardless of their gender, especially in young characters (Aladé et al., 2020).

These stereotypes significantly affect the sense of belonging and the science identity, which have been found to play a key role in undergraduate student persistence and in the gender gap. This situation is especially negative for females, for whom affective domains are of extreme importance (Hazari et al., 2010; Good et al., 2012; Lewis et al., 2016; Rainey et al., 2018; Makarova et al., 2019; Fisher et al., 2020a, 2022). In fact, developing a strong science identity seems to be especially decisive for girls during middle and high school (Vincent-Ruz and Schunn, 2018). In this regard, role models are influential, as it happens, for example, with the simplest case of female teachers in science courses positively influencing girls (Bottia et al., 2015).

Future employment and occupational values. Occupational and lifestyle values are also among the strongest predictors of gender differences in choosing a scientific degree (Eccles and Wang, 2016). A main factor to point out here is that in general, unlike boys, who study science mainly because they enjoy it or are good at it, girls seem to be more focused on getting a specific job, which might be in other different fields (Eccles, 2011; Ogunde et al., 2017). However, jobs in science or technology seem to attract more boys than girls, according to Sjøberg et al. (2010). In fact, it seems that completing a degree is a predictor of female employment in a scientific or technological profession, contrary to the case of males (Kimmel et al., 2012). Anyway, STEM education must be connected to future professional training and careers, also by pointing out the gender parity (Mann and DiPrete, 2013).

Mathematics. They key role of mathematics in influencing the gender gap existing in STEM choice cannot be overlooked. One of the reasons why girls are less likely to choose math or physics is the way other people provide information about these subjects, as it eases the development of lower self-perceptions in comparison to boys regarding their future role in these fields (Eccles, 2011). In fact, self-concept in mathematics seems to be higher in males than in females (Sax et al., 2015; Seo et al., 2019; Mejía-Rodríguez et al., 2021).

Interestingly, it would be the lack of mathematical confidence that seems to be responsible for the high dropout rate of women, rather than a lack of mathematical ability (Ellis et al., 2016), math ability self-concept being actually one of the strongest predictors of the gender differences in the likelihood of entering STEM careers (Eccles and Wang, 2016). In fact, these significant gender differences in math self-efficacy seem to appear in the late adolescence (Huang, 2013). Nevertheless, according to Bench et al. (2015), it is indeed not the underestimation of their abilities in women, but the greater men's overconfidence, which accounts for the greater intention of the latter to stay in math fields. It is also interesting to mention, that according to the study by Good et al. (2012), women are protected from stereotypes when they know that math ability can be acquired, which eventually leads to a higher sense of belonging and the intention to stay in math.

A lower math self-efficacy is usually connected to a higher math anxiety, and both can predict a low retention in STEM (Ashcraft, 2002). In this regard, women show much more math anxiety than men, and especially spatial anxiety (Sokolowski et al., 2019). Finally, the role of parents in math stereotype beliefs is also remarkable, as these are easily transferred to their female children through their intrusive support with math homework (Bhanot and Jovanovic, 2005). Additionally, Gunderson et al. (2012) pointed out that adults can communicate negative math attitudes to girls especially, not only through their stereotypes and expectancies, but also through their own math anxiety.

Influential people: friends, family and teachers. Again, the effect that different people can have on the choice of the STEM track differs for both genders. As described above, one of the problems is that students do not picture themselves in a future science career, a fact that can be enhanced by social support from teachers, family, and friends in both genders (Buday et al., 2012). Interestingly, similar encouragement from important others to both genders may explain why both have similar expectations for a science career (Stake, 2006). However, it is actually a different social support, which is argued by Kelly (2016), that might be decisive for women to continue in STEM. In fact, it is very important for women to have an integrated support network (Miller-Friedmann et al., 2018).

Parents’ influence in career aspirations and encouragement cannot be overseen and seems to be different for both genders (Taskinen et al., 2016). According to Adya and Kaiser (2005), parents, and particularly fathers, have a decisive influence on girls’ choice of technological careers. However, direct support by parents can also impact negatively girls’ self-concept, which is definitely linked to the future choice of STEM (Ertl et al., 2017).

Peers are also of extreme importance, not only from a supportive point of view, but also because students may adjust their choices to those of their friends. In fact, Raabe et al. (2019) could observe that the presence of girls in science classes influences other girls to persist. Similarly, Robnett and Leaper (2013) pointed out that belonging to a mixed-gender group of friends who value science make it easier for girls to see science as gender neutral and therefore stay.

Interestingly, according to the study by Griffith (2010) in the U.S., role models seem to influence the choice of STEM courses at college. This way, female students are more likely to choose STEM courses at institutions with higher percentage of female STEM graduate students. A similar observation was done by Bottia et al. (2015), who found that, contrary to the case of males, the proportion of female math and science teachers at school has a strong influence on female students’ probability of finishing a STEM degree, and especially for female students with the highest math skills.

Research questions

As stated above, women's underrepresentation in STEM has attracted the interest of practitioners and researchers over the last decade (Cheryan et al., 2017; Wang and Degol, 2017; Holman et al., 2018; Fisher et al., 2020b). Besides, some previous studies stressed the relevance of secondary-school students’ choices in their future intentions of continuing in the science pipeline at the university level and, consequently, in the workforce once they finish their academic trajectory (Jacob et al., 2020). In previous studies, we found an interesting effect of motivation on students’ choices (Ardura and Pérez-Bitrián, 2018) and the existence of three different motivational profiles towards physics & chemistry for secondary school students’: highly motivated, average motivated (at risk of abandonment) and low motivated students (Ardura and Pérez-Bitrián, 2019). Besides, in a related investigation, we undertook the study of the students’ causal attributions to choose or abandon physics & chemistry when it first becomes optional in the educational system as a function of students’ motivational levels (Ardura et al., 2021). Interestingly, in our logistic regression model, whereas attributions to the effect of the family and to the teacher and classroom methodology were found to be a common predictor of the decision for all students, other types of attributions were significantly dependent on the motivation level. In particular, for students with an average level of motivation, attributions to the subject's relationship with mathematics and to the effect of friends were crucial in their decision, whilst the effect of media was a significant predictor only in the case of highly-motivated students. With the objective of following up our previous research, this paper aims to study the effect of gender on our previous model on students’ causal attributions to choose or abandon physics & chemistry in the first instance the student can opt this subject out. To achieve this goal, three research questions guided our study:

RQ1. Are there any gender differences in the secondary school students’ causal attributions to choose or abandon physics & chemistry as a function of their motivation?

RQ2. Is gender a predictor variable of the students’ choice?

RQ3. Are there any interactions between students’ gender and their attributions to choose or abandon physics & chemistry when it turns into an optional subject in secondary schools?

Methods

Sample and context

The context of the data gathering was the subject physics & chemistry in the last year of the Spanish compulsory education, as this is the first time the students have the chance to choose physics & chemistry or not. At this point, both disciplines are taught together within the same subject. A total of 1060 students were included in the sample through a convenient sampling based on the accessibility of the students and the schools. The students belonged to 5 private and 10 public Spanish schools. It is important to bear in mind that secondary education is compulsory for all the students in Spain, so the private schools are also supported by founds from the Spanish government (charter schools). The schools were situated in medium-sized Spanish cities in areas with an overall middle-class socio-economic status. Besides, no relevant cultural diversity was present in the schools. Given the purposes of this investigation, the sampling included both students who had chosen (n = 695, 65.6%, PC students) and who had abandoned (n = 365, 34.4%, nonPC students) the subject. The average age of the students was 15.03 years. All the students classified themselves either as boys (n = 522, 49.2%) or girls (n = 538, 51.8%). Regarding gender differences in the choice of physics & chemistry, the percentage of boys who chose the subject (68.8%) was statistically higher (χ² = 4.69, p = 0.03) than the percentage of girls (62.5%).

Data collection procedure

After reaching an agreement to carry out the study with the schools, the aims of the study and the procedure for the data collection were explained to the teachers and the students. Both students and their families gave their informed consent to be involved in the data gathering. The students’ participation was volunteer and anonymity during this and subsequent stages of the investigation was ensured. The data collection took place at the schools and was supervised by the teachers of each classroom, who had been previously trained for this task. The schools chose between an online or a paper-based survey to collect the students’ answers depending on their resources. All the data was collected from students using a self-reported survey. Therefore, in our investigation participants self-reported their gender identity.

Instruments and variables

Motivation towards physics & chemistry. Students’ motivational traits were measured using the Spanish version of the Science Motivation Questionnaire II (SMQII) (Glynn et al., 2011). This version was adapted to measure motivation towards physics & chemistry in secondary-school students (Ardura and Pérez-Bitrián, 2018). Within the context of Bandura's Social Cognitive Theory, this instrument measures five motivational traits (intrinsic motivation, grade motivation, career motivation, self-efficacy, and self-determination) through a set of 25 items and a 5-point Likert-type scale (from 0 to 4; from never to always). Validity evidence for the same sample of participants was provided in a previous study (Ardura and Bitrián, 2018, 2019), using both exploratory and confirmatory factor analyses, which showed strong evidence of factorial validity. Given the aims of this and the previous investigations, invariance testing supported group comparisons. Besides, the five motivational subscales rendered good levels of reliability. A detailed description of the items of the instrument and their psychometric properties including evidence for validity, reliability, and invariance tests for the same sample of students are available in previous studies (Ardura and Pérez-Bitrián, 2018, 2019; Ardura et al., 2021).

Students’ causal attributions. Two instruments to gather information about the students’ causal attributions to choose or abandon physics & chemistry were designed ad hoc for both PC students (PC Causal Attributions Questionnaire, PC CAQ) and nonPC students (nonPC Causal Attributions Questionnaire, nonPC CAQ), respectively. These two instruments were created to measure the students’ levels of attributions to six different reasons to opt for or out physics & chemistry and were based on a thorough review of previous studies (see, for example Cerinsek et al., 2013; Ing, 2014; Mujtaba and Reiss, 2014; Stake, 2006; Vedder-Weiss and Fortus, 2013). To allow the comparison between the two groups of students, the items were written in parallel to collect the same information from each set of students. For instance, one of the items to measure the attribution to the subject's relationship with mathematics was ‘I chose physics & chemistry because I like mathematics’ in the questionnaire designed for students who had opted for the subject (PC students) whereas the item was presented as ‘I opted out physics & chemistry because I do not like mathematics’ in the case of nonPC students. Consequently, both instruments comprised the same number of items and, in light of the factor validation analyses, accounted for a six-attribution-correlated model: students’ perceived difficulty of the subject (4 items), teacher and classroom methodology (10 items), subject's relationship with mathematics (3 items), and the effects of friends (3 items), family (4 items) and media (3 items). In a previous paper, we presented the results of the validity and reliability analyses of both instruments for the same sample of students used in the present paper (Ardura et al., 2021). A five-point Likert-type scale from 0 (strongly disagree) to 4 (strongly agree) was used to gather students’ answers. Confirmatory factor analyses (CFA) corroborated the factorial structure for both genders in each sub-scale (see Table 5 in the Appendix).

Since the purpose of this paper is to investigate the effect of gender on students’ attributions, testing invariance of the models for gender was carried out to support this comparison in both instruments. To this aim, we defined four models (configural, metric, scalar, and conservative), as suggested previously enforcing the following subsequent constrictions for the two groups of students (Rocabado et al., 2020): factor loadings (metric invariance), loadings and item intercepts (scalar invariance), factor loadings, item intercepts, and error variances (conservative invariance).

The assessment of measurement invariance is also based on the comparison between these four models, the cut-off values for unequal sample sizes being: ΔCFI < 0.01; ΔSRMR < 0.03; ΔRMSEA < 0.015, for metric invariance, and ΔCFI < 0.01; ΔSRMR < 0.01; ΔRMSEA < 0.015 for scalar and conservative invariance (Chen, 2007; Rocabado et al., 2020). Given that the models held for the two groups of students in each instrument, configural invariance was met. Our computations also confirmed metric, scalar, and conservative invariance by gender to a reasonable extent according to the cut-off values for the complete instruments (see Table 6 in the Appendix).

The same analyses for the six attributions considered separately for both instruments confirmed, as well, a reasonable measurement invariance behavior (see Table 6 in the Appendix). Even though the cut-off values stated above were not strictly met, it must be kept in mind that previous simulation studies in measurement invariance have shown that it is difficult to state statistical standards and, consequently, sound judgment and expertise should accompany purely statistical decisions (Byrne, 2001; Chen, 2007).

Procedure and analyses. A previous study using the same sample of students revealed the presence of three different motivational profiles among the students (Ardura and Pérez-Bitrián, 2019). In the first place, students who abandoned the subject when it becomes optional in the Spanish educational system showed the lowest motivation in the five studied motivational traits. Besides, our computations uncovered the existence of two different students’ motivational profiles among those who chose the subject at this point, which were characterized as average and high motivated towards physics & chemistry. These three groups found in previous studies were used in the present investigation.

First, descriptive and inferential analyses for the gender-wise comparison of students’ attributional mean values were undertaken by means of a two-way Multivariate Analysis of the Variance (MANOVA), followed by Tuckey post hoc analyses using the Bonferroni correction. The size effect for the effect of gender in the pair–wise comparisons for each attribution was estimated using the Cohen's d statistic, with the reference values for this statistic being: <0.20 very small, 0.20–0.49 small, 0.50–0.79 moderate, >0.80 large (López-Martín and Ardura, 2023). For the MANOVA and the follow-up ANOVA, omega squared was used to compute the effect size to interpret the mean empirical differences found, with the cut-off values being: <0.01 very small, 0.01–0.05 small, 0.01–0.05 moderate, and >0.14 large (López-Martín and Ardura, 2023). Students’ causal attributions to choose or abandon physics & chemistry were taken as the dependent variables and gender and motivation acted as factors. This approach helped the study of the simple effects of gender within each motivational group and the interaction effect between gender and all the causal attributions. Second, to model attributions and their interactions with gender in the students’ choices, a multinomial regression approach was used. Therefore, the predictor variables were gender and the six students’ attributions, and the outcome variable was belonging to one of the three sets of students (high, average, or low motivation). Besides, the interaction between gender and the six attributions was also included in the model. To assess the contribution of each predictor and the interaction terms, the odds ratio was computed. This number can be interpreted as the change in the odds of pertaining to one of the three sets of students when there is a change of one unit in the predictor variables. To ease the interpretation of the effect of gender, the odds ratios were accompanied by the computations of the changes in the probabilities when students’ gender changes in the case of a modeled nonPC student. The calculation of these probabilities was made using the multinomial regression coefficients (Field, 2013). All computations were carried out using SPSS and AMOS (Arbuckle, 2010).

Results

First, we will report the descriptive and mean difference analyses for the simple effect of gender on students’ attributions. Given the aforementioned effect of motivation on students’ attributions (Ardura et al., 2021), the results were disaggregated using the three groups yielded by the cluster analysis (i.e., highly-motivated, average-motivated, and low-motivated students). The first two groups comprised students who opted for the subject and the third group students who abandoned it. Second, the results of the multinomial logistic regression model including both the attributions and the interactions between the attributions and gender will be presented.

Descriptive and mean difference analyses

Two-way MANOVA rendered significant main effects for motivation (F(12, 2098) = 30.34, p < 0.001, Wilks’ Λ = 0.73) and gender (F(6, 1049) = 9.43, p < 0.001, Wilks’ Λ = 0.95) on the students’ causal attributions. Besides, a significant interaction between motivation and gender was found (F(12, 2098) = 2.87, p = 0.001, Wilks’ Λ = 0.97). The main effect for motivation was reported in our previous study (Ardura et al., 2021). In this work we focus on the effect of gender and its interaction with motivation. Table 1 shows the mean values of the six causal attributions disaggregated by gender and the statistics for the ANOVA follow up comparisons within each attribution. Gender has a main effect on attributions to friends, media and family, these attributions being higher for boys than for girls. In the rest of the attributions, no significant gender-wise differences were found.

Table 1 Results from the analyses of the main effect of gender on the students’ causal attributions

	Boys	Girls	F	p	ω ^2
Teacher: teacher and classroom methodology; Difficulty: perceived difficulty of the subject; Math: subject's relationship with mathematics; Friends: effect of friends; Media: effect of media; Family: effect of the family.
Teacher	1.64	1.53	2.53	0.112	<0.01
Difficulty	2.11	2.04	0.93	0.336	<0.01
Math	2.22	2.13	1.17	0.279	<0.01
Friends	0.79	0.41	51.95	<0.001	0.05
Media	1.80	1.60	9.82	0.002	<0.01
Family	1.35	1.20	6.85	0.009	<0.01

---

The results for the mean values of the six attributions studied in this investigation disaggregated by motivation and gender are collected in Table 2. To assess the gender differences in each of the three motivational levels, simple effect tests were carried out using the Bonferroni correction (see Table 2). Regardless of the students’ motivational level, the largest effect sizes in the three motivational groups were found in attributions to friends, boys presenting higher levels in this attribution than girls. For instance, highly-motivated boys doubled the mean score of girls (see Table 2). Statistically significant mean differences were also found in the case of the attributions to the subject's relationship with mathematics. Interestingly, our analyses showed a different trend in this attribution for students who chose physics & chemistry and those who opted it out. Highly- and average-motivated boys presented higher levels in this attribution (2.77 and 1.90, respectively) than their gender counterparts (2.38 and 1.59, respectively). However, in the group of low-motivated students, girls (2.44) outscored boys (1.98) in this attribution. In turn, attributions to media and family rendered significant gender differences only in the instance of highly-motivated students. In both cases, boys scored higher than girls (see Table 2).

Table 2 Mean difference analysis by motivation and gender for the six causal attributions under study

	PC/highly motivated				PC/average motivated				nonPC/low motivated
	Boys (n = 128)	Girls (n = 148)	p	d	Boys (n = 231)	Girls (n = 188)	p	d	Boys (n = 163)	Girls (n = 202)	p	d
Teacher: teacher and classroom methodology; Difficulty: perceived difficulty of the subject; Math: subject's relationship with mathematics; Friends: effect of friends; Media: effect of media; Family: effect of the family.a p < 0.05.b p < 0.01.
Teacher	M = 1.91	M = 1.90	0.968	0.01	M = 1.76	M = 1.50	0.024^a	0.23	M = 1.24	M = 1.16	0.491	0.07
	SD = 1.09	SD = 1.30			SD = 1.11	SD = 1.16			SD = 1.05	SD = 1.11
Difficulty	M = 2.37	M = 2.21	0.236	0.16	M = 1.85	M = 1.68	0.139	0.17	M = 2.10	M = 2.22	0.306	−0.09
	SD = 1.01	SD = 1.00			SD = 1.04	SD = 0.99			SD = 1.29	SD = 1.33
Math	M = 2.77	M = 2.38	0.008^b	0.34	M = 1.90	M = 1.59	0.01^a	0.26	M = 1.98	M = 2.44	<0.001^b	−0.34
	SD = 1.04	SD = 1.26			SD = 1.20	SD = 1.15			SD = 1.30	SD = 1.36
Friends	M = 0.89	M = 0.42	<0.001^b	0.55	M = 0.93	M = 0.58	<0.001^b	0.38	M = 0.54	M = 0.24	<0.001^b	0.42
	SD = 0.98	SD = 0.72			SD = 0.94	SD = 0.88			SD = 0.79	SD = 0.63
Media	M = 2.41	M = 2.10	0.011^a	0.31	M = 1.47	M = 1.35	0.220	0.13	M = 1.50	M = 1.34	0.127	0.15
	SD = 1.01	SD = 1.01			SD = 0.94	SD = 0.95			SD = 1.05	SD = 1.04
Family	M = 1.72	M = 1.50	0.033^a	0.21	M = 1.37	M = 1.27	0.282	0.11	M = 0.94	M = 0.82	0.223	0.10
	SD = 1.03	SD = 1.01			SD = 0.92	SD = 0.94			SD = 0.91	SD = 0.72

---

Follow-up ANOVA were used to explore the possible interactions between students’ gender and the different students’ attributions to choose or abandon physics & chemistry. The analyses rendered only one significant interaction between gender and the attribution to the subject's relationship with mathematics (F = 12.769, p < 0.001, ω² = 0.04), with the rest of the interactions being non-significant (see Table 3).

Table 3 Results from the analyses of the interaction between the students’ causal attributions and gender

	F	p	ω ^2
Teacher: teacher and classroom methodology; Difficulty: perceived difficulty of the subject; Math: subject's relationship with mathematics; Friends: effect of friends; Media: effect of media; Family: effect of the family.
Teacher × gender	1.083	0.339	<0.01
Difficulty × gender	1.886	0.152	<0.01
Math × gender	12.769	<0.001	0.04
Friends × gender	0.751	0.472	<0.01
Media × gender	0.760	0.468	<0.01
Family × gender	0.517	0.597	<0.01

---

As shown in Fig. 1, this significant effect between the gender of the students and their motivational level indicates that boys’ and girls’ attributions were affected differently depending on their motivation for physics & chemistry. For those students who chose the subject when it became optional in the Spanish educational system, the level of attributions to the subject's relationship with mathematics decreased in boys from 2.77 (highly-motivated students) to 1.90 (average-motivated students). In the case of girls, a decrease was also found from 2.38 (highly-motivated students) to 1.59 (average-motivated students). However, while the level of this attribution remained fairly constant from average (1.90) to low-motivated boys (1.98), a significant increase was found for girls who abandoned the subject (from 1.59 to 2.44).

Fig. 1 Interaction graph of gender and motivation in attributions to the subject's relationship with mathematics. The red and blue lines correspond to boys and girls, respectively.

Multinomial logistic regression model

As mentioned above, in a previous study we reported the results of multinomial logistic regression analysis to study the role of students’ attributions on their decision about choosing or abandoning physics & chemistry (Ardura et al., 2021). We now focus on the effect of including the variable gender in the model, as a predictor of the students’ choice and the assessment of the interactions between this variable and the students’ attributions. To ease the comparison between the two models, we include the main features of both in Table 4. As Model I was presented and deeply described in our previous work (Ardura et al., 2021), we will focus on the description of how the inclusion of gender affected this model to render Model II.

Table 4 Multinominal logistic regression results^a

	Model I^b		Model II
	B(SE)	OR (95% CI)	B(SE)	OR (95% CI)
Teacher: teacher and classroom methodology; Difficulty: perceived difficulty of the subject; Math: subject's relationship with mathematics; Friends: effect of friends; Media: effect of media; Family: effect of the family.a Only the significant interactions between gender and the attributions are included in the table.b Model developed by Ardura et al. (2021).c The reference category was nonPC for motivation and female for gender.*p < 0.05, **p < 0.01, ***p < 0.001.
PC high motivated vs. nonPC^c
Intercept	−2.30 (0.26)***		−3.24 (0.38)
Teacher	0.45 (0.09)***	1.56 (1.30, 1.88)	0.44 (0.09)***	1.55 (1.29, 1.88)
Difficulty	−0.46 (0.10)***	0.63 (0.52, 0.77)	−0.46 (0.10)***	0.63 (0.51, 0.77)
Math	0.13 (0.08)	1.13 (0.97, 1.33)	0.48 (0.12)***	1.62 (1.27, 20.6)
Friends	−0.08 (0.12)	0.92 (0.72, 1.18)	−0.68 (0.13)	0.93 (0.73, 1.20)
Media	0.65 (0.09)***	1.92 (1.60, 2.31)	0.65 (0.09)***	1.93 (1.60, 2.33)
Family	0.73 (0.11)***	2.08 (1.68, 2.57)	0.73 (0.11)***	2.08 (1.68, 2.59)
Gender	—	—	1.60 (0.41)***	4.97 (2.25, 11.33)
Gender × Math	—	—	−0.60 (0.15)***	0.54 (0.41, 0.73)
PC average motivated vs. nonPC^c
Intercept	0.26 (0.18)		−0.29 (0.23)
Teacher	0.53 (0.09)***	1.69 (1.43, 2.01)	0.52 (0.09)***	1.70 (1.42, 2.00)
Difficulty	−0.58 (0.09)***	0.57 (0.47, 0.67)	−0.58 (0.09)***	0.55 (0.47, 0.66)
Math	−0.21 (0.07)**	0.81 (0.71, 0.93)	−0.01 (0.09)	1.00 (0.83, 1.21)
Friends	0.37 (0.11)***	1.44 (1.17, 1.78)	0.32 (0.11)**	1.40 (1.11, 1.71)
Media	−0.04 (0.09)	0.96 (0.81, 1.13)	−0.05 (0.09)	0.95 (0.80, 1.12)
Family	0.50 (0.99)***	1.64 (1.35, 1.95)	0.50 (0.10)***	1.65 (1.36, 2.01)
Gender	—	—	0.61 (0.29)**	1.84 (1.04, 3.26)
Gender × Math	—	—	−0.47 (0.12)***	0.66 (0.52, 0.85)

---

The logistic regression model was built using the group of students who abandoned the subject (nonPC) as the reference category. Thus, our results allow the comparison of the weights of the predictors included in the model for the groups of highly- and average-motivated students who chose the subject once it becomes optional in the Spanish educational system.

The inclusion of gender in Model I rendered a statistically significant (χ² = 368.03; p < 0.001) model. The predictors’ power was assessed using the values of R² of Nagelkerke and R² of Cox and Snell and the model reached values of 0.33 and 0.29, respectively. The model was able to predict 55.7% of cases in the sample, 42.5% being the proportional-by-chance that predicts an improvement of 25% (Petrucci, 2009). Therefore, the classification power of our model exceeds the percentage expected in a random assignation of the students to the three groups. The results of the logistic regression model for each predictor are shown in Table 4. Once gender was included in the model, it became the strongest predictor of the students’ choice both for highly-motivated (B = 1.60, p < 0.001; OR = 4.97) and average-motivated (B = 0.61, p < 0.01; OR = 1.84) students. Therefore, for highly-motivated students, the probability of choosing physics & chemistry for a modeled nonPC student with average values in the causal attributions would raise from 0.104 to 0.365 when gender changes from female to male. The same prediction for average-motivated students increases the probability to a lesser extent: from 0.493 to 0.642.

Besides, the inclusion of gender in the model altered Model I only in the predictability of attributions to the subject's relationship with mathematics. First, in the case of highly-motivated students this attribution became significant (B = 0.48, p < 0.001; OR = 1.62), although it was non-significant in Model I (B = 0.13, p = 0.26; OR = 1.13). Second, in the case of average-motivated students this attribution was a significant predictor in Model I (B = −0.21, p < 0.01; OR = 0.81), whereas it turned into a non-significant predictor once gender was included in the model (B = −0.01, p = 0.54; OR = 1.00). Finally, only the students’ attribution to the subject's relationship with mathematics interacted significantly with gender for both highly-motivated (B = −0.60, p < 0.001; OR = 0.54) and average-motivated students (B = −0.47, p < 0.001; OR = 0.66). Therefore, as attributions to the subject's relationship with mathematics increases, girls become less likely to choose physics & chemistry than boys, this effect being more pronounced in the case of highly-motivated students.

Discussion and conclusions

The present investigation aimed to study the role of gender on students’ causal attributions to choose or abandon physics & chemistry when it becomes optional in Spanish secondary schools as a function of the students’ motivational level. Our study comprised three research questions.

Our first research question was: are there any gender differences in the secondary school students’ causal attributions to choose or abandon physics & chemistry as a function of their motivation? Even though attributions to friends was the least preferred explanation of the students’ choice, it showed the largest effect size despite the students’ motivational level, boys presenting higher average scores than girls. This result is in line with a previous study that demonstrated that students can be influenced by their friends’ preferences, girls being affected to a lesser extent than boys (Raabe et al., 2019). Moreover, the same study found that boys were more likely to show preferences for a STEM subject than girls and that the major social fondness for STEM came from same-gender friends. Therefore, given our findings, this effect may be contributing to both the choice of the subject by highly- and average-motivated students and its abandonment by their low-motivated counterparts.

Considering our results, gender effects on attributions to media were motivational dependent. On the one hand, highly-motivated boys present higher attributions to media than girls. This trend may be related to the fact that females are underrepresented in STEM-focused TV shows compared to males (Aladé et al., 2020). Besides, stereotypes could be playing an important role in this effect since there is still a sizable dissociation between women and STEM in people's mindsets (Bond, 2016; Cheryan et al., 2017). On the other hand, although the same tendency was found in average- and low-motivated students, the mean differences were not significant, perhaps due to the low levels of this attribution in the case of these students.

Similar behavior was found in attributions to family, since the attributional levels are higher in the case of boys, being significantly different for highly-motivated students. It has been reported that supplying information about STEM to parents improves their positive perceptions about this field and affects their children's future decisions about taking STEM courses, triggering their communication about the topic (Harackiewicz et al., 2012). Moreover, a recent study uncovered that parents’ expectations and encouragement predict boys’ STEM self-efficacy, which, in turn, is strongly correlated to STEM career expectations (Lv et al., 2022).

Given our results, highly-motivated students’ families may be better informed about STEM options. However, given the gender differences found, stereotypes may be also affecting the image that students get from their families, as previous studies suggest that families may lower females’ STEM-specific self-concept (Ertl et al., 2017), which in turn, may negatively impact the girls’ uptake in the science track. Moreover, it is worth noting that specific counseling for girls may be counterproductive, as it can be interpreted as a reference to a supposed lack of ability to succeed in science (Bhanot and Jovanovic, 2005; Ertl et al., 2017).

Considering our results, the subject's relationship with mathematics was a more important attribution for boys as a reason to opt for the subject than for girls in the groups of highly- and average-motivated students. However, in the case of low-motivated students (i.e., those who abandoned physics & chemistry), girls presented a significantly higher level in this attribution as a possible reason to opt out the subject. Therefore, the subject's relationship with mathematics seems to be less important for girls who choose the subject but more important in the decision to leave it. This fact may be related to gender effects previously found in several important constructs in the domain of mathematics, as girls seem to display lower levels of math self-concept (Sax et al., 2015; Seo et al., 2019; Mejía-Rodríguez et al., 2021) and self-efficacy (Huang, 2013), but higher levels of math anxiety (Huang et al., 2019).

Our second research question was: Is gender a predictor variable of the students’ choice? Given the results of our logistic regression model, boys are more likely to choose the subject when it becomes optional, backing the previous findings on the underrepresentation of females in STEM both at secondary (Jacob et al., 2020) and tertiary levels (Holman et al., 2018; Wang and Degol, 2017), as this decision usually causes the permanent abandonment of the science pipeline. Remarkably, the effect of gender is almost three times larger in highly motivated students compared to their average-motivated counterparts. Many factors have been identified by previous studies to explain the effect of gender on students’ decision (Fisher et al., 2020b). Gender differences in science identity (Hudson and Matthews, 2012; Michell et al., 2017) and belonging (Fink et al., 2020) may be playing an important role, as stereotypes could be more established in the case of the group of highly-motivated students, who have shown a clearer motivation for a future career in the STEM field than those with average motivation. Besides, in the case of the latter, both the students’ attributions to opt for the subject and their motivational levels are lower (Ardura and Pérez-Bitrián, 2019) and, consequently, their science identity may be less delineated. In this vein, previous studies found that, concerning their choices in STEM, developing a science identity in middle school is particularly crucial for girls (Vincent-Ruz and Schunn, 2018). Some previous studies uncovered the relevance of STEM self-efficacy on students’ intentions to persist in the science track (Britner, 2008; Palmer et al., 2017; Shwartz et al., 2021).

Finally, our third research question was: are there any interactions between students’ gender and their attributions to choose or abandon physics & chemistry when it turns into an optional subject in secondary schools? Our regression model predicted only one significant interaction, which takes place between the students’ gender and attributions to the subject's relationship with mathematics. Regardless of the students’ motivation for the subject being high or average, as the level of this attribution increases it is less likely for girls to choose the subject. Despite students’ gender, the subject's relationship with mathematics displayed the highest level of all the attributions investigated. Therefore, the fact that physics & chemistry involves the use of mathematics seems to be a good reason for students to choose or leave the subject. However, the significant interaction found reveals that, for high- and average-motivated girls, the subject's relationship with mathematics is a less important reason to decide than for boys. On the other hand, for low-motivated girls, the presence of mathematical contents in the subject could trigger its abandonment. These findings may be related to the aforementioned individual math-related traits such as math self-efficacy or math anxiety, which can be impacting girls’ interest in scientific careers (Huang et al., 2019). In particular, secondary school girls have been found to be more anxious about mathematics (Devine et al., 2012) and have lower levels of math self-efficacy emerging in late adolescence (Huang, 2013). Stereotypes about mathematics can be again playing a relevant role in this case as, for instance, there is previous evidence that high-achieving students are more likely to picture a mathematician as a male (Sánchez-Aguilar et al., 2016).

Educational implications

Given the relevance of the secondary school students’ decisions in future choices related to continuing in the STEM track, our study focused on the effect of gender on the choice of physics & chemistry in Spain. In view of our findings, two main educational implications can be drawn. First, gender differences found in students’ attributions for their choice may be stereotype-driven, as the influence of friends, family, and media seem to be biased by students’ gender. Thus, we would recommend teachers and schools to work on STEM stereotypes not only with students but also with their families. Since at a school level it is not feasible to act over media stereotypes in this matter, using TV programs as a tool to enhance students’ critical thinking about this matter could be a way of working on this topic. As families are also one of the reasons the students use to explain their choice and previous studies found that stereotypes are also present in this context, schools should try to involve parents in the actions they put into practice to address this issue.

Second, the interaction between gender and students’ attributions to mathematics are pointing to the importance of girls’ perceptions about mathematics. In this way, following previous studies on the role of gender in math self-efficacy and math anxiety, it seems important for schools and teachers to design specific gender-balanced actions to even girls’ and boys’ perceptions about mathematics. To this aim, it is important to bear in mind that attitudes and interest are more predictive of the students’ intentions of persisting for girls than for boys (Moore and Burrus, 2019). Thus, one possible way of promoting mathematics among girls could be enhancing the students’ perceived utility value of this subject as a key tool to study physics & chemistry.

Limitations and prospect

The results presented in this work must be understood with several limitations in mind which, in turn, lead to some prospective research for future studies in the field. First, the instruments used for the data gathering were self-reports. Thus, the results may be biased by social desirability or inaccurate self-knowledge of the students. Given the size of the sample used, we expect these possible biasing effects to be mitigated overall. Further research using a mixed-methods approach could be appropriate to get deep into the reasons for the choice of the students. Second, an ex-post-facto approach taken in this study does not allow to study casual relationships among variables, as the analyses are based on correlations. For this reason, further experimental analyses should be undertaken. Third, considering the findings of our study, stereotypes seem to be playing a relevant role in students’ decisions. Therefore, future research should get further insights into their effect on students’ perceptions, attitudes, motivations, and decisions related to science. Given that stereotypes can occur in younger students than those who were the subject of this study (Buckley et al., 2021), it could be useful to undertake new research involving primary school students. Forth, our results point to the possible relevance of constructs like science identity and belonging. Consequently, further studies are needed to explore the role of these individual traits on students’ future choices as a function of students’ gender. Finally, since some previous studies uncovered cultural differences affecting the problem we tackled in this paper or variables related to it (Toma et al., 2019; Rüschenpöhler and Markic, 2020), further research should explore the effects of culture on the uptake of students in the chemistry track.

Conflicts of interest

There are no conflicts to declare.

Appendix

Table 5 Fit indexes for the confirmatory factor analyses (CFA) of the total and individual subscales of the PC CAQ and the nonPC CAQ by gender

	χ ^2/df		CFI		RMSEA		SRMR
	PC	nonPC	PC	nonPC	PC	nonPC	PC	nonPC
CAQ: total scale
Boys	2.54	3.02	0.91	0.89	0.07	0.07	0.04	0.05
Girls	1.98	2.68	0.93	0.90	0.06	0.06	0.05	0.06
Subscale: attributions to the teacher and classroom methodology
Boys	2.07	2.63	0.91	0.87	0.06	0.06	0.07	0.08
Girls	2.72	2.89	0.90	0.89	0.07	0.05	0.06	0.07
Subscale: attributions to the students’ perceived difficulty of the subject
Boys	3.06	3.01	0.88	0.89	0.04	0.05	0.05	0.06
Girls	2.14	2.67	0.94	0.91	0.05	0.06	0.05	0.07
Subscale: attributions to the subject's relationship with mathematics
Boys	1.73	2.48	0.92	0.89	0.05	0.06	0.07	0.08
Girls	2.01	2.67	0.90	0.91	0.06	0.05	0.06	0.07
Subscale: attributions to the effect of friends
Boys	3.18	3.04	0.89	0.88	0.06	0.06	0.08	0.08
Girls	2.25	2.83	0.88	0.88	0.05	0.06	0.07	0.08
Subscale: attributions to the effect of the family
Boys	1.03	2.58	0.91	0.89	0.06	0.05	0.08	0.07
Girls	2.47	2.78	0.92	0.90	0.05	0.04	0.06	0.08
Subscale: attributions to the effect of media
Boys	2.76	3.02	0.94	0.90	0.06	0.06	0.07	0.08
Girls	1.85	2.56	0.89	0.91	0.05	0.06	0.06	0.09

---

Table 6 Invariance tests of the total and individual subscales of the PC CAQ and the nonPC CAQ

Invariance	p		ΔCFI		ΔRMSEA		ΔSRMR
	PC	nonPC	PC	nonPC	PC	nonPC	PC	nonPC
CAQ: total scale
Metric	0.271	0.382	0.003	0.005	0.011	0.013	0.008	0.006
Scalar	0.121	0.238	0.012	0.009	0.012	0.015	0.010	0.012
Conservative	0.095	0.154	0.017	0.014	0.016	0.019	0.011	0.016
Subscale: attributions to the teacher and classroom methodology
Metric	0.431	0.341	0.013	0.010	0.007	0.005	0.007	0.005
Scalar	0.352	0.259	0.015	0.019	0.009	0.010	0.011	0.013
Conservative	0.238	0.198	0.017	0.023	0.012	0.014	0.015	0.021
Subscale: attributions to the students’ perceived difficulty of the subject
Metric	0.562	0.356	0.013	0.008	0.009	0.006	0.009	0.007
Scalar	0.412	0.219	0.016	0.010	0.010	0.012	0.010	0.009
Conservative	0.361	0.285	0.019	0.018	0.011	0.024	0.012	0.015
Subscale: attributions to the subject's relationship with mathematics
Metric	0.487	0.241	0.010	0.009	0.006	0.004	0.006	0.005
Scalar	0.405	0.201	0.011	0.012	0.009	0.006	0.008	0.006
Conservative	0.253	0.164	0.015	0.018	0.017	0.015	0.017	0.014
Subscale: attributions to the effect of friends
Metric	0.389	0.294	0.005	0.005	0.008	0.011	0.008	0.010
Scalar	0.256	0.207	0.007	0.013	0.005	0.014	0.011	0.013
Conservative	0.159	0.101	0.013	0.021	0.013	0.026	0.014	0.021
Subscale: attributions to the effect of the family
Metric	0.651	0.348	0.005	0.007	0.008	0.005	0.009	0.010
Scalar	0.307	0.204	0.012	0.009	0.011	0.012	0.006	0.014
Conservative	0.206	0.117	0.018	0.025	0.014	0.017	0.011	0.020
Subscale: attributions to the effect of media
Metric	0.459	0.345	0.014	0.016	0.009	0.004	0.007	0.005
Scalar	0.394	0.295	0.013	0.020	0.012	0.010	0.008	0.009
Conservative	0.185	0.152	0.017	0.026	0.021	0.013	0.013	0.017

---

Acknowledgements

The authors would like to acknowledge the students, teachers, and schools involved in the data gathering for their help. We are also thankful to the Colegio Oficial de Químicos de Asturias y León and the Asociación de Químicos del Principado de Asturias for their help in our first contact with the schools. A. P.-B. thanks the Spanish Ministerio de Educación, Cultura y Deporte for a grant (FPU15/03940).

Notes and references

1. Acar Ö., Türkmen L. and Bilgin A., (2015), Examination of Gender Differences on Cognitive and Motivational Factors that Influence 8th Graders’ Science Achievement in Turkey, Eurasia J. Math. Sci. Technol. Educ., 11(5), 1027–1040.
2. Adamuti-Trache M. and Sweet R., (2013), Academic Effort and Achievement in Science: Beyond a Gendered Relationship, Res. Sci. Educ., 43(6), 2367–2385.
3. Adya M. and Kaiser K. M., (2005), Early determinants of women in the IT workforce: a model of girls’ career choices, Inf. Technol. People, 18(3), 230–259.
4. Akpınar E., Yıldız E., Tatar N. and Ergin Ö., (2009), Students’ attitudes toward science and technology: an investigation of gender, grade level, and academic achievement, Proc. Soc. Behav. Sci., 1(1), 2804–2808.
5. Aladé F., Lauricella A., Kumar Y. and Wartella E., (2020), Who's modeling STEM for kids? A character analysis of children's STEM-focused television in the U.S, J. Child. Media, 15(3), 338–357.
6. Aldén L. and Neuman E., (2022), Culture and the gender gap in choice of major: An analysis using sibling comparisons, J. Econ. Behav. Organ., 201, 346–373.
7. Anderhag P., Emanuelsson P., Wickman P.-O. and Hamza K. M., (2013), Students’ Choice of Post-Compulsory Science: in search of schools that compensate for the socio-economic background of their students, Int. J. Sci. Educ., 35(18), 3141–3160.
8. Arbuckle J. L., (2010), SPSS (version 19.0) [Computer program], Chicago: SPSS.
9. Archer L., Dewitt J. and Willis B., (2014), Adolescent boys’ science aspirations: masculinity, capital, and power, J. Res. Sci. Teach.,51(1), 1–30.
10. Ardura D. and Galán A., (2019), The interplay of learning approaches and self-efficacy in secondary school students’ academic achievement in science, Int. J. Sci. Educ., 41(13), 1723–1743.
11. Ardura D. and Pérez-Bitrián A., (2018), The effect of motivation on the choice of chemistry in secondary schools: adaptation and validation of the Science Motivation Questionnaire II to Spanish students, Chem. Educ. Res. Pract., 19(3), 905–918.
12. Ardura D. and Pérez-Bitrián A., (2019), Motivational pathways towards academic achievement in physics & chemistry: a comparison between students who opt out and those who persist, Chem. Educ. Res. Pract., 20(3), 618–632.
13. Ardura D., Zamora Á. and Pérez-Bitrián A., (2021), The role of motivation on secondary school students’ causal attributions to choose or abandon chemistry, Chem. Educ. Res. Pract., 22(1), 43–61.
14. Ashcraft M. H., (2002), Math Anxiety: Personal, Educational, and Cognitive Consequences, Curr. Dir. Psychol. Sci., 11(5), 181–185.
15. Avargil S., Kohen Z. and Dori Y. J., (2020), Trends and perceptions of choosing chemistry as a major and a career, Chem. Educ. Res. Pract., 21(2), 668–684.
16. Bench S. W., Lench H. C., Liew J., Miner K. and Flores S. A., (2015), Gender Gaps in Overestimation of Math Performance, Sex Roles, 72(11–12), 536–546.
17. Bennett J., Lubben F. and Hampden-Thompson G., (2013), Schools That Make a Difference to Post-Compulsory Uptake of Physical Science Subjects: Some comparative case studies in England, Int. J. Sci. Educ., 35(4), 663–689.
18. Beutel A. M., Burge S. W. and Borden B. A., (2019), Masculinity and Men's Choice of College Major, Gend. Issues, 36(4), 374–391.
19. Bhanot R. and Jovanovic J., (2005), Do Parents’ Academic Gender Stereotypes Influence Whether They Intrude on their Children's Homework? Sex Roles, 52(9), 597–607.
20. Blanco-López Á., España-Ramos E., González-García F. J. and Franco-Mariscal A. J., (2015), Key aspects of scientific competence for citizenship: A Delphi study of the expert community in Spain, J. Res. Sci. Teach., 52(2), 164–198.
21. Blickenstaff J. C., (2005), Women and science careers: leaky pipeline or gender filter? Gend. Educ., 17(4), 369–386.
22. Bøe M. V., (2012), Science Choices in Norwegian Upper Secondary School: What Matters? Sci. Educ., 96(1), 1–20.
23. Bond B. J., (2016), Fairy Godmothers > Robots: The Influence of Televised Gender Stereotypes and Counter-Stereotypes on Girls’ Perceptions of STEM, Bull. Sci. Technol. Soc., 36(2), 91–97.
24. Bottia M. C., Stearns E., Mickelson R. A., Moller S. and Valentino L., (2015), Growing the roots of STEM majors: female math and science high school faculty and the participation of students in STEM, Econ. Educ. Rev., 45, 14–27.
25. Braund M. and Reiss M., (2006), Towards a More Authentic Science Curriculum: The contribution of out-of-school learning, Int. J. Sci. Educ., 28(12), 1373–1388.
26. Britner S. L., (2008), Motivation in High School Science Students: A Comparison of Gender Differences in Life, Physical, and Earth Science Classes, J. Res. Sci. Teach., 45(8), 955–970.
27. Broman K. and Simon S., (2015), Upper Secondary School Students’ Choice and Their Ideas on How to Improve Chemistry Education, Int. J. Sci. Math. Educ., 13(6), 1255–1278.
28. Brotman J. S. and Moore F. M., (2008), Girls and science: A review of four themes in the science education literature, J. Res. Sci. Teach., 45(9), 971–1002.
29. Buckley C., Farrell L. and Tyndall I., (2021), Brief Stories of Successful Female Role Models in Science Help Counter Gender Stereotypes Regarding Intellectual Ability among Young Girls: A Pilot Study, Early Educ. Dev., 33(4), 555–566.
30. Buday S. K., Stake J. E. and Peterson Z. D., (2012), Gender and the Choice of a Science Career: The Impact of Social Support and Possible Selves, Sex Roles, 66(3), 197–209.
31. Byrne, B. M., (2001), Structural equation modeling with AMOS: Basic concepts, applications, and programming, Mahwah, NJ: Lawrence Erlbaum Associates.
32. Cerinsek G., Hribar T., Glodez N. and Dolinsek S., (2013), Which are my Future Career Priorities and What Influenced my Choice of Studying Science, Technology, Engineering or Mathematics? Some Insights on Educational Choice—Case of Slovenia, Int. J. Sci. Educ., 35(17), 2999–3025.
33. Charles M. and Bradley K., (2009), Indulging Our Gendered Selves? Sex Segregation by Field of Study in 44 Countries, Am. J. Sociol., 114(4), 924–976.
34. Chen, F. F., (2007), Sensitivity of goodness of fit indexes to lack of measurement invariance, Struct. Equ. Modeling, 14(3), 464–504.
35. Cheryan S., Ziegler S. A., Montoya A. K. and Jiang L., (2017), Why are some STEM fields more gender balanced than others? Psychol. Bull., 143(1), 1–35.
36. Cooper C. I. and Pearson P. T., (2012), A Genetically Optimized Predictive System for Success in General Chemistry Using a Diagnostic Algebra Test, J. Sci. Educ. Technol., 21(1), 197–205.
37. Cousins A. and Mills M., (2015), Gender and high school chemistry: student perceptions on achievement in a selective setting, Camb. J. Educ., 45(2), 187–204.
38. Dabney K. P., Tai R. H., Almarode J. T., Miller-Friedmann J. L., Sonnert G., Sadler P. M. and Hazari Z., (2012), Out-of-School Time Science Activities and Their Association with Career Interest in STEM, Int. J. Sci. Educ., Part B, 2(1), 63–79.
39. Delaney J. and Devereux P. J., (2019), Understanding gender differences in STEM: Evidence from college applications, Econ. Educ. Rev., 72, 219–238.
40. Devine A., Fawcett K., Szűcs D. and Dowker A., (2012), Gender differences in mathematics anxiety and the relation to mathematics performance while controlling for test anxiety, Behav. Brain Func., 8, 33.
41. Dillon J., (2009), On Scientific Literacy and Curriculum Reform, Int. J. Environ. Sci. Educ., 4(3), 201–213.
42. Eccles J., (2011), Gendered educational and occupational choices: applying the Eccles et al. model of achievement-related choices, Int. J. Behav. Dev., 35(3), 195–201.
43. Eccles J. S. and Wang M.-T., (2016), What motivates females and males to pursue careers in mathematics and science? Int. J. Behav. Dev., 40(2), 100–106.
44. Eddy S. L. and Brownell S. E., (2016), Beneath the numbers: a review of gender disparities in undergraduate education across science, technology, engineering, and math disciplines, Phys. Rev. Phys. Educ. Res., 12(2), 020106.
45. Ellis J., Fosdick B. K. and Rasmussen C., (2016), Women 1.5 Times More Likely to Leave STEM Pipeline after Calculus Compared to Men: Lack of Mathematical Confidence a Potential Culprit, PLoS One, 11(7), e0157447.
46. England B. J., Brigati J. R., Schussler E. E. and Chen M. M., (2019), Student Anxiety and Perception of Difficulty Impact Performance and Persistence in Introductory Biology Courses, CBE Life Sci. Educ., 18(2), ar21.
47. Ertl B., Luttenberger S. and Paechter M., (2017), The Impact of Gender Stereotypes on the Self-Concept of Female Students in STEM Subjects with an Under-Representation of Females, Front. Psychol, 8, 703.
48. European Commission, (2004), Increasing human resources for science and technology in Europe: report of the High Level Group on Human Resources for Science and Technology in Europe, in Gago J. M. (ed.) Luxembourg: Office for Official Publications of the European Communities.
49. Farenga S. J. and Joyce B. A., (1999), Intentions of young students to enroll in science courses in the future: An examination of gender differences, Sci. Ed., 83(1), 55–75.
50. Field, A., (2013), Discovering statistics using IBM SPSS statistics, London: Sage.
51. Fink A., Frey R. F. and Solomon E. D., (2020), Belonging in general chemistry predicts first-year undergraduates’ performance and attrition, Chem. Educ. Res. Pract., 21(4), 1042–1062.
52. Fischer F., Schult J. and Hell B., (2013), Sex differences in secondary school success: why female students perform better, Eur. J. Psychol. Educ., 28(2), 529–543.
53. Fisher C. R., Thompson C. D. and Brookes R. H., (2020a), ‘95% of the time things have been okay’: the experience of undergraduate students in science disciplines with higher female representation, Int. J. Sci. Educ., 42(9), 1430–1446.
54. Fisher C. R., Thompson C. D. and Brookes R. H., (2020b), Gender differences in the Australian undergraduate STEM student experience: a systematic review, High. Educ. Res. Dev., 39(6), 1155–1168.
55. Fisher C. R., Brookes R. H. and Thompson C. D., (2022), ‘I don’t Study Physics Anymore’: a Cross-Institutional Australian Study on Factors Impacting the Persistence of Undergraduate Science Students, Res. Sci. Educ., 52, 1565–1581.
56. Gil-Pérez D. and Vilches A., (2005), The contribution of science and technological education to citizens’ culture, Can. J. Sci. Math. Technol. Educ., 5(2), 253–263.
57. Glynn S. M., Brickman P., Armstrong N. and Taasoobshirazi G., (2011), Science motivation questionnaire II: validation with science majors and nonscience majors, J. Res. Sci. Teach., 48(10), 1159–1176.
58. Good C., Rattan A. and Dweck C. S., (2012), Why do women opt out? Sense of belonging and women's representation in mathematics, J. Pers. Soc. Psychol., 102(4), 700–717.
59. Griffith A. L., (2010), Persistence of women and minorities in STEM field majors: Is it the school that matters? Econ. Educ. Rev., 29(6), 911–922.
60. Gunderson E. A., Ramirez G., Levine S. C. and Beilock S. L., (2012), The Role of Parents and Teachers in the Development of Gender-Related Math Attitudes, Sex Roles, 66(3–4), 153–166.
61. Guo J., Parker P. D., Marsh H. W. and Morin A. J. S., (2015), Achievement, motivation, and educational choices: a longitudinal study of expectancy and value using a multiplicative perspective, Dev. Psychol., 51(8), 1163–1176.
62. Halim L., Abd Rahman N., Zamri R. and Mohtar L., (2018), The roles of parents in cultivating children's interest towards science learning and careers, Kasetsart J. Soc. Sci., 39(2), 190–196.
63. Harackiewicz J. M., Rozek C. S., Hulleman C. S. and Hyde J. S., (2012), Helping Parents to Motivate Adolescents in Mathematics and Science: An Experimental Test of a Utility-Value Intervention, Psychol. Sci., 23(8), 899–906.
64. Hazari Z., Sonnert G., Sadler P. M. and Shanahan M.-C., (2010), Connecting high school physics experiences, outcome expectations, physics identity, and physics career choice: a gender study, J. Res. Sci. Teach., 47(8), 978–1003.
65. Holman L., Stuart-Fox D. and Hauser C. E., (2018), The gender gap in science: How long until women are equally represented? PLoS Biol., 16(4), e2004956.
66. House J. D. and Telese J. A., (2017), Confidence in Science and Achievement Outcomes of Fourth-Grade Students in Korea: Results from the TIMSS 2011 Assessment. Education, 137(4), 389–392.
67. Huang C., (2013), Gender differences in academic self-efficacy: a meta-analysis, Eur. J. Psychol. Educ., 28(1), 1–35.
68. Huang X., Zhang J. and Hudson L., (2019), Impact of math self-efficacy, math anxiety, and growth mindset on math and science career interest for middle school students: the gender moderating effect, Eur. J. Psychol. Educ., 34(3), 621–640.
69. Hudson P. and Matthews K., (2012), Identities and Transformational Experiences for Quantitative Problem Solving: Gender Comparisons of First-Year University Science Students, J. Sci. Math. Educ. Southeast Asia, 35(1), 22–42.
70. Ing M., (2014), Can Parents Influence Children's Mathematics Achievement and Persistence in STEM Careers? J. Career Dev., 41(2), 87–103.
71. Isaacs B., (2001), Mystery of the Missing Women Engineers: A Solution, J. Prof. Issues. Eng. Educ. Pract., 127(2), 85–91.
72. Jacob M., Iannelli C., Duta A. and Smyth E., (2020), Secondary school subjects and gendered STEM enrollment in higher education in Germany, Ireland, and Scotland, Int. J. Comp. Sociol., 61(1), 59–78.
73. Jacobs J. E., (2005), Twenty-five years of research on gender and ethnic differences in math and science career choices: What have we learned? New Dir. Child Adolesc. Dev., 2005(110), 85–94.
74. Kang J., Hense J., Scheersoi A. and Keinonen T., (2019), Gender study on the relationships between science interest and future career perspectives. Int. J. Sci. Educ., 41(1), 80–101.
75. Kelly A. M., (2016), Social cognitive perspective of gender disparities in undergraduate physics, Phys. Rev. Phys. Educ. Res., 12(2), 020116.
76. Kimmel L. G., Miller J. D. and Eccles J. S., (2012), Do the Paths to STEM Professions Differ by Gender? Peabody J. Educ., 87(1), 92–113.
77. Lane K. A., Goh J. X. and Driver-Linn E., (2012), Implicit Science Stereotypes Mediate the Relationship between Gender and Academic Participation. Sex Roles, 66(3–4), 220–234.
78. Legewie J. and DiPrete T. A., (2014), The High School Environment and the Gender Gap in Science and Engineering, Sociol. Educ., 87(4), 259–280.
79. Lewis K. L., Stout J. G., Pollock S. J., Finkelstein N. D. and Ito T. A., (2016), Fitting in or opting out: A review of key social-psychological factors influencing a sense of belonging for women in physics, Phys. Rev. Phys. Educ. Res., 12(2), 020110.
80. Lewis S. E. and Lewis J. E., (2007), Predicting at-risk students in general chemistry: comparing formal thought to a general achievement measure, Chem. Educ. Res. Pract., 8(1), 32–51.
81. López-Martín E. and Ardura D., (2023), The effect size in the scientific publication, Educación XX1, 26(1), 9–17.
82. Lv B., Wang J., Zheng Y., Peng X. and Ping X., (2022), Gender differences in high school students’ STEM career expectations: an analysis based on multi-group structural equation model, J. Res. Sci. Teach., 59(10), 1739–1764.
83. Lyons T., (2006a), Different Countries, Same Science Classes: Students’ Experiences of School Science in Their Own Words, Int. J. Sci. Educ., 28(6), 591–613.
84. Lyons T., (2006b), The Puzzle of Falling Enrolments in Physics and Chemistry Courses: Putting Some Pieces Together, Res. Sci. Educ., 36(3), 285–311.
85. Makarova E., Aeschlimann B. and Herzog W., (2019), The Gender Gap in STEM Fields: The Impact of the Gender Stereotype of Math and Science on Secondary Students’ Career Aspirations, Front. Educ., 4, 60.
86. Maltese A. V. and Tai R. H., (2011), Pipeline persistence: Examining the association of educational experiences with earned degrees in STEM among U.S. students, Sci. Educ., 95(5), 877–907.
87. Mann A. and DiPrete T. A., (2013), Trends in gender segregation in the choice of science and engineering majors, Soc. Sci. Res., 42(6), 1519–1541.
88. Mau W.-C., (2003), Factors That Influence Persistence in Science and Engineering Career Aspirations, Career Dev. Q., 51(3), 234–243.
89. McDonald C. V., (2016), STEM Education: A review of the contribution of the disciplines of science, technology, engineering and mathematics, Sci. Educ. Int., 27(4), 530–569.
90. Mejía-Rodríguez A. M., Luyten H. and Meelissen M. R. M., (2021), Gender Differences in Mathematics Self-concept Across the World: an Exploration of Student and Parent Data of TIMSS 2015, Int. J. Sci. Math. Educ., 19, 1229–1250.
91. Michell D., Szorenyi A., Falkner K. and Szabo C., (2017), Broadening participation not border protection: how universities can support women in computer science, J. High. Educ. Policy Manag., 39(4), 406–422.
92. Miller-Friedmann J., Childs A. and Hillier J., (2018), Approaching gender equity in academic chemistry: lessons learned from successful female chemists in the UK, Chem. Educ. Res. Pract., 19(1), 24–41.
93. Miner K. N., Walker J. M., Bergman M. E., Jean V. A., Carter-Sowell A., January S. C. and Kaunas C., (2018), From “Her” Problem to “Our” Problem: Using an Individual Lens Versus a Social-Structural Lens to Understand Gender Inequity in STEM, Ind. Organ. Psychol., 11(2), 267–290.
94. Moakler Jr. M. W. and Kim M. M., (2014), College Major Choice in STEM: Revisiting Confidence and Demographic Factors, Career Dev. Q., 62(2), 128–142.
95. Moore R. and Burrus J., (2019), Predicting STEM Major and Career Intentions with the Theory of Planned Behavior, Career Dev. Q., 67(2), 139–155.
96. Mujtaba T. and Reiss M. J., (2014), A Survey of Psyschological, Motivational, Family and Perceptions of Physics Education Factors that Explain 15 Year-Old Student's Aspirations to Study Physics in Post-Compulsory English Schools, Int. J. Sci. Math. Educ., 12(2), 371–393.
97. Nugent G., Barker B., Welch G., Grandgenett N., Wu C. and Nelson C., (2015), A Model of Factors Contributing to STEM Learning and Career Orientation, Int. J. Sci. Educ., 37(7), 1067–1088.
98. OECD, (2016), PISA 2015 Assessment and Analytical Framework, OECD Publishing.
99. Ogunde J. C., Overton T. L., Thompson C. D., Mewis R. and Boniface S., (2017), Beyond graduation: motivations and career aspirations of undergraduate chemistry students, Chem. Educ. Res. Pract., 18(3), 457–471.
100. Oon P.-T. and Subramaniam R., (2010), Views of physics teachers on how to address the declining enrolment in physics at the university level, Res. Sci. Technol. Educ., 28(3), 277–289.
101. Osborne J., Simon S. and Collins S., (2003), Attitudes towards science: a review of the literature and its implications, Int. J. Sci. Educ., 25(9), 1049–1079.
102. Ost B., (2010), The role of peers and grades in determining major persistence in the sciences, Econ. Educ. Rev., 29(6), 923–934.
103. Palmer T.-A., Burke P. F. and Aubusson P., (2017), Why school students choose and reject science: a study of the factors that students consider when selecting subjects, Int. J. Sci. Educ., 39(6), 645–662.
104. Petrucci C. J., (2009), A Primer for Social Worker Researchers on How to Conduct a Multinomial Logistic Regression, J. Soc. Serv. Res., 35(2), 193–205.
105. Potvin P. and Hasni A., (2014), Interest, motivation and attitude towards science and technology at K-12 levels: a systematic review of 12 years of educational research, Stud. Sci. Educ., 50(1), 85–129.
106. Raabe I. J., Boda Z. and Stadtfeld C., (2019), The Social Pipeline: How Friend Influence and Peer Exposure Widen the STEM Gender Gap, Sociol. Educ., 92(2), 105–123.
107. Rainey K., Dancy M., Mickelson R., Stearns E. and Moller S., (2018), Race and gender differences in how sense of belonging influences decisions to major in STEM, Int. J. STEM Educ., 5(1), 10.
108. Reinhold S., Holzberger D. and Seidel T., (2018), Encouraging a career in science: a research review of secondary schools’ effects on students’ STEM orientation, Stud. Sci. Educ., 54(1), 69–103.
109. Robnett R. D. and Leaper C., (2013), Friendship Groups, Personal Motivation, and Gender in Relation to High School Students’ STEM Career Interest, J. Res. Adolesc., 23(4), 652–664.
110. Rocabado G. A., Komperda R., Lewis J. E. and Barbera J., (2020), Addressing diversity and inclusion through group comparisons: a primer on measurement invariance testing, Chem. Educ. Res. Pract., 21(3), 969–988.
111. Rocard M., Csermely P., Jorde D., Lenzen D., Walberg-Henriksson H. and Hemmo V., (2007), Science Education Now: A Renewed Pedagogy for the Future of Europe, Brussels, Directorate-General for Research, European Commission.
112. Roth W.-M. and Lee S., (2004), Science Education as/for Participation in the Community, Sci. Educ., 88(2), 263–291.
113. Rüschenpöhler, L. and Markic, S. (2020). Secondary school students’ chemistry self-concepts: Gender and culture, and the impact of chemistry self-concept on learning behaviour, Chem. Educ. Res. Pract., 21(1), 209–219.
114. Sachdev A., (2018), Gender Disparity in STEM Across Cultures, Ind. Organ. Psychol., 11(2), 309–313.
115. Sadler P. M., Sonnert G., Hazari Z. and Tai R., (2012), Stability and volatility of STEM career interest in high school: a gender study, Sci. Educ., 96(3), 411–427.
116. Sahin A. and Waxman H. C., (2020), Characteristics of Secondary Students who have Intentions to Choose a STEM Major in College: Findings from a Three-Year Study, EURASIA J. Math. Sci. Technol. Educ., 16(12), em1922.
117. Salta K. and Koulougliotis D., (2015), Assessing motivation to learn chemistry: adaptation and validation of Science Motivation Questionnaire II with Greek secondary school students, Chem. Educ. Res. Pract., 16(2), 237–250.
118. Salta K. and Koulougliotis D., (2020), Domain specificity of motivation: chemistry and physics learning among undergraduate students of three academic majors, Int. J. Sci. Educ., 42(2), 253–270.
119. Sánchez-Aguilar M., Rosas A., Molina-Zavaleta J. G. M. and Romo-Vázquez A., (2016), Exploring High-achieving Students’ Images of Mathematicians, Int. J. Sci. Math. Educ., 14(3), 527–548.
120. Sax L. J., Kanny M. A., Riggers-Piehl T. A., Whang H. and Paulson L. N., (2015), “But I’m Not Good at Math”: The Changing Salience of Mathematical Self-Concept in Shaping Women's and Men's STEM Aspirations, Res. High. Educ., 56(8), 813–842.
121. Schumm M. F. and Bogner F. X., (2016), Measuring adolescent science motivation, Int. J. Sci. Educ., 38(3), 434–449.
122. Scott F. J., (2012), Is mathematics to blame? An investigation into high school students’ difficulty in performing calculations in chemistry, Chem. Educ. Res. Pract., 13(3), 330–336.
123. Seo E., Shen Y. and Alfaro E. C., (2019), Adolescents’ Beliefs about Math Ability and Their Relations to STEM Career Attainment: Joint Consideration of Race/ethnicity and Gender, J. Youth Adolesc., 48(2), 306–325.
124. Sha L., Schunn C. and Bathgate M., (2015), Measuring Choice to Participate in Optional Science Learning Experiences During Early Adolescence, J. Res. Sci. Teach., 52(5), 686–709.
125. Sha L., Schunn C., Bathgate M. and Ben-Eliyahu A., (2016), Families Support Their Children's Success in Science Learning by Influencing Interest and Self-Efficacy, J. Res. Sci. Teach., 53(3), 450–472.
126. Sheldrake R., Mujtaba T. and Reiss M. J., (2017), Science teaching and students’ attitudes and aspirations: the importance of conveying the applications and relevance of science, Int. J. Educ. Res., 85, 167–183.
127. Shirazi S., (2017), Student experience of school science, Int. J. Sci. Educ., 39(14), 1891–1912.
128. Shwartz G., Shav-Artza O. and Dori Y. J., (2021), Choosing Chemistry at Different Education and Career Stages: Chemists, Chemical Engineers, and Teachers, J. Sci. Educ. Technol., 30(5), 692–705.
129. Simon R. M., Wagner A. and Killion B., (2017), Gender and choosing a STEM major in college: femininity, masculinity, chilly climate, and occupational values, J. Res. Sci. Teach., 54(3), 299–323.
130. Simpkins S. D., Price C. D. and Garcia K., (2015), Parental support and high school students’ motivation in biology, chemistry, and physics: Understanding differences among latino and caucasian boys and girls, J. Res. Sci. Teach., 52(10), 1386–1407.
131. Sjøberg S., Schreiner C. and Schreiner C., (2010), The ROSE project An overview and key findings. Retrieved from: http://ntsnet.dk/sites/default/files/Rose20project20-overview-2010.pdf.
132. Smyth E. and Hannan C., (2006), School Effects and Subject Choice: The uptake of scientific subjects in Ireland, Sch. Eff. Sch. Improv., 17(3), 303–327.
133. Sokolowski H. M., Hawes Z. and Lyons I. M., (2019), What explains sex differences in math anxiety? A closer look at the role of spatial processing, Cognition, 182, 193–212.
134. Stake J. E., (2006), The Critical Mediating Role of Social Encouragement for Science Motivation and Confidence Among High School Girls and Boys, J. Appl. Soc. Psychol., 36(4), 1017–1045.
135. Stoet G. and Geary D. C., (2018), The Gender-Equality Paradox in Science, Technology, Engineering, and Mathematics Education, Psychol. Sci., 29(4), 581–593.
136. Stokking K. M., (2000), Predicting the choice of physics in secondary education, Int. J. Sci. Educ., 22(12), 1261–1283.
137. Su R. and Rounds J., (2015), All STEM fields are not created equal: People and things interests explain gender disparities across STEM fields, Front. Psychol., 6, 189.
138. Taskinen P. H., Dietrich J. and Kracke B., (2016), The Role of Parental Values and Child-specific Expectations in the Science Motivation and Achievement of Adolescent Girls and Boys, Int. J. Gend. Sci. Technol., 8(1), 103–123.
139. Toma, R. B., Greca, I. M. and Orozco Gómez, M. L., (2019), Attitudes towards science and views of nature of science among elementary school students in terms of gender, cultural background and grade level variables. Res. Sci. Technol. Educ., 37(4), 492–515.
140. Turner R. C. and Lindsay H. A., (2003), Gender Differences in Cognitive and Noncognitive Factors Related to Achievement in Organic Chemistry, J. Chem. Educ., 80(5), 563–568.
141. Ulriksen L., Madsen L. M. and Holmegaard H. T., (2010), What do we know about explanations for drop out/opt out among young people from STM higher education programmes? Stud. Sci. Educ., 46(2), 209–244.
142. Vedder-Weiss D. and Fortus D., (2013), School, teacher, peers, and parents’ goals emphases and adolescents’ motivation to learn science in and out of school, J. Res. Sci. Teach., 50(8), 952–988.
143. Vincent-Ruz P. and Schunn C. D., (2018), The nature of science identity and its role as the driver of student choices, Int. J. STEM Educ., 5(1), 48.
144. Wagner E. P., Sasser H. and DiBiase W. J., (2002), Predicting Students at Risk in General Chemistry Using Pre-semester Assessments and Demographic Information, J. Chem. Educ., 79(6), 749–755.
145. Wang M.-T. and Degol J. L., (2017), Gender Gap in Science, Technology, Engineering, and Mathematics (STEM): Current Knowledge, Implications for Practice, Policy, and Future Directions, Educ. Psychol. Rev., 29(1), 119–140.
146. Wang J. and Staver J. R., (2001), Examining Relationships Between Factors of Science Education and Student Career Aspiration, J. Educ. Res., 94(5), 312–319.
147. Xu X., Villafane S. M. and Lewis J. E., (2013), College students’ attitudes toward chemistry, conceptual knowledge and achievement: structural equation model analysis, Chem. Educ. Res. Pract., 14(2), 188–200.

---

Footnote

† Current address: Fachbereich Biologie, Chemie, Pharmazie, Institut für Chemie und Biochemie – Anorganische Chemie, Freie Universität Berlin, Fabeckstraße 34/36, DE-14195 Berlin, Germany.

---