An Integrated Analysis of School Students’ Aspirations for STEM Careers: Which Student and School Factors Are Most Predictive?

Internationally, there is concern over the low number of people studying and working in the science, technology, engineering and mathematics (STEM) areas (Office of the Chief Scientist, 2013; Wang & Degol, 2013; Wilson & Mack, 2014). Declining enrolments and participation in STEM disciplines is a significant issue because building capacity in the STEM fields is pivotal to maintaining/increasing productivity and international competiveness (Marginson, Tytler, Freeman & Roberts, 2013; Office of the Chief Scientist, 2013). In Australia, student participation in year 12 mathematics and science subjects is declining and for science is at its lowest point in 20 years (Kennedy, Lyons & Quinn, 2014). Furthermore, comparative international tests of mathematical and scientific literacy reveal that Australian students’ performance has consistently dropped since 2003 (Office of the Chief Scientist, 2014; Thomson, Wennert, O’Grady & Rodrigues, 2016). At the same time, over the coming decades, demand in some STEM fields is predicted to grow substantially with an 11.7% increase in engineers, a 9% increase in business analysts and programmers and an 8% increase in natural and physical scientists (PwC, 2015).

Focus on the STEM disciplines has arisen not only out of a perceived lack of skilled workers in new highly technological fields of employment but also in relation to concerns about STEM being taught as discrete subjects in schools rather than as part of an integrated curriculum (Office of the Chief Scientist, 2014). Within the Australian context, this decline in STEM participation in both the education sector and the workforce has prompted the Chief Scientist to call for school curricula to reflect the evolving nature of STEM, provide better STEM teaching, address the under-representation of females and disadvantaged students in STEM and, at all levels of education, highlight the future needs of a well-qualified STEM workforce to help guide students’ study decisions (Office of the Chief Scientist, 2013, 2014).

In this paper, we analyse the aspirations for STEM careers of a sample of Australian school students in relation to student background and school-related factors in order to shed light on the relative influence of these factors. At a time when shortfalls in STEM participation are becoming critical, a better understanding of who is and is not interested in STEM will provide timely insights for teachers, school careers advisors and tertiary educators, concerning fertile avenues for redressing the apparent decline in student interest.

The relationship between the career choices of adults and their earlier aspirations have been studied extensively over many decades (Hartung, Porfeli & Vondracek, 2005; Howard, Carlstrom, Katz, Chew, Ray, Laine & Calum, 2011). While there is evidence to suggest that young people with high aspirations are more likely to enter high-status occupations (Schoon & Parsons, 2002), there is also evidence that other factors such as gender, prior achievement, educational aspiration and SES can be highly influential in the formation and actualisation of youth aspirations (Gutman & Akerman, 2008; Gutman & Schoon, 2012). In relation to the investigation of aspirations for STEM careers, many different research perspectives have been employed, including, for example, sociological explorations of career choice (Archer, DeWitt & Wong, 2014; Beal & Crockett, 2013), psychological examinations of motivation and interest (Robnett & Leaper, 2013; Watt, Shapka, Morris, Durik, Keating & Eccles, 2012), and educational studies focussed on curriculum, teachers and career guidance (Schmitt-Wilson & Welsh, 2012). While most of these studies examine the aspirations of students in the upper years of secondary school, our study is unique in examining the aspirations of students from years 3 to 12, thus providing new insights into the development of STEM aspirations across the school years.

A comprehensive synthesis of research focussed on aspirations for STEM careers, conducted by van Tuijl and van der Molen (2015), points to three important interrelated factors: students’ knowledge of STEM fields, the affective value placed on STEM careers and students’ ability beliefs. The authors suggest that it is important to acknowledge the role of students’, parents’ and teachers’ knowledge (or lack thereof) of STEM fields, children’s knowledge of the self in STEM activities, and parents’ and teachers’ knowledge of the early circumscription processes that can occur between the ages of 8 and 16. They suggest that more awareness of the range of STEM careers from a young age is crucial for STEM career development. Also, their overview of the literature reveals the persistent negative affective view that STEM fields “are predominately ‘things- and male-oriented’ and form a threat to more feminine lifestyles or to a good family-work balance” (p.16) leading them to reinforce the importance of male and female STEM role models for children (van Tuijl & van der Molen, 2015). Lastly, their research synthesis reveals the central role of students’ ability beliefs and the prevalent view that STEM subjects are difficult and complex, requiring a great deal of effort. They suggest that students with an entity view of intelligence, rather than those with a growth mindset (Dweck, 2006) may be unwilling to take on the perceived challenge of STEM subjects at school. Each of these factors (knowledge, affective value and ability beliefs) identified by van Tuijl and van der Molen (2015) are themselves multi-faceted and intersect with other influential factors identified in the literature such as gender, prior achievement, socioeconomic status and notions of social and cultural capital.

Gender has been a key feature of many studies examining aspirations for STEM careers (Archer et al., 2014; Eccles, 1994; Hernandez-Martinez, Black, Williams, Davis, Pampaka & Wake, 2008; Novakovic & Fouad, 2013; Packard & Nguyen, 2003; Sadler, Sonnert, Hazari & Tai, 2012; Shapka, Domene & Keating, 2006; Watt et al., 2012). In general, this body of research points to a persistent gap in the proportions of boys and girls aspiring to STEM careers, with boys more likely to do so from as early as the middle school years (Archer et al., 2014; Kanny, Sax & Riggers-Piehl, 2014). Indeed, there is evidence of a subsequent gender gap in those taking up STEM careers (Beede, Julian, Langdon, McKittrick, Khan & Doms, 2011; Broadley, 2015). The reasons for this disparity have been identified as complex and interrelated, ranging from issues related to a lack of female role models, irrelevant science curricula, lack of mathematics preparation, cultural stereotypes, field-specific ability beliefs and teacher predispositions (Blickenstaff, 2005; Cheryan, Master & Meltzoff, 2015; Leslie, Cimpian, Meyer & Freeland, 2015).

It may be the case that girls are not completely free to pursue STEM as they are constrained by gender stereotypes within academic cultures about who is suited to, and likely to succeed in, these disciplines (Stout, Dasgupta, Hunsinger & McManus, 2011). These stereotypes can affect girls’ beliefs about their STEM competence and the value of STEM careers (Blickenstaff, 2005; Broadley, 2015). There is also a widespread belief that the study of STEM subjects requires in-born brilliance, a trait more often believed to be held by males (Cheryan et al., 2015; Leslie et al., 2015). Furthermore, females of equal ability to males often report that they perceive themselves as being less capable (MacPhee, Farro & Canetto, 2013; Mann, Legewie & DiPrete, 2015; Sax, Kanny, Riggers-Piehl, Whang & Paulson, 2015; Shumow & Schmidt, 2013). In this study, we test the notion that female students are not as interested in STEM careers as their male counterparts and examine the intersection of gender with other factors such as prior achievement.

Alongside gender issues, researchers have also focussed on the critical role of prior achievement in the formation and maintenance of STEM career aspirations over time, finding that secondary school achievement levels, particularly in mathematics, can act as a critical filter for future career choices (Gemici, Bednarz, Karmel & Lim, 2014; Gore, Holmes, Smith, Southgate & Albright, 2015) and for STEM careers in particular (Shapka et al., 2006). Drawing on data from the Longitudinal Study of Australian Youth (LSAY), Anlezark, Lim, Semo and Nguyen (2008) found that half of those from the highest mathematics achievement quartile ended up working in a STEM career, and that the students in this quartile were twice as likely to choose a STEM career as those in the lowest achievement quartile.

Differences in prior achievement, however, do not explain the persistent gender gaps in STEM career aspirations (Riegle-Crumb, King, Grodsky & Muller, 2012) with some pointing to individual persistence and motivation as being more important explanatory factors for the observed lack of female participation (Shumow & Schmidt, 2013). Others have suggested that female lack of interest in STEM subjects at school is not solely related to prior achievement but rather is dependent on the value that they place on STEM tasks (Shumow & Schmidt, 2013). The implication here is that changes in instructional approaches to mathematics and science, in ways that make the social benefits explicit, have some potential to increase female interest and participation in STEM.

The performance of one’s peers has also been linked to aspirations for STEM careers with highly competitive school environments having a negative impact on these aspirations, particularly for girls (Mann et al., 2015). However, this aspiration gender gap lessens for high-performing students in strong performance environments (Mann et al., 2015).

While students from low SES backgrounds are under-represented in higher education, scholars have questioned the view that this phenomenon stems from a lack of individual aspiration, recognising the complex nature of social disadvantage (Gale, Parker, Rodd, Stratton & Sealey, 2013; MacPhee et al., 2013; St Clair & Benjamin, 2011; Zipin, Sellar, Brennan & Gale, 2015). With reference to STEM aspirations, Archer, Osborne, DeWitt, Dillon, Wong and Willis (2013) found in the ASPIRES study that less than 15% of middle school students aspire to become scientists and that these preferences remain stable over time. The ASPIRES study determined that familial ‘science capital’, having family members with science qualifications or science careers, was a key factor in increasing the likelihood of students’ science aspirations. However, it should also be noted that these higher levels of ‘science capital’ tend to be associated with middle to high SES families.

The connection between SES and prior achievement has also been researched, with acknowledgement that students’ achievement in school can mediate the impact of SES on aspirations (Gemici et al., 2014). Similarly, SES has been found to intersect with gender and race, indicating that caution should be taken when considering students from low SES backgrounds as a distinct group (Kanny et al., 2014). All of these factors are examined concurrently in our analysis, resulting in new understandings of their relative weight in explaining students’ aspirations for STEM careers.

Family cultural capital plays a role in supporting student aspirations and achievement through exposure to abstract ideas via art or culture, through provision of resources in the home to support learning and through acquaintance with knowledge about further education and careers (Khattab, 2015). Family characteristics are also linked to students’ educational aspirations which, in turn, have a major influence on career trajectories. The children of parents with higher education qualifications are more likely to aspire to achieving at least the same level of education as their parents, in comparison to those students whose parents do not have post-secondary qualifications (Gil-Flores, Padilla-Carmona & Suárez-Ortega, 2011). Also, there is longitudinal evidence that those who do aspire to complete university are more likely to actually do so, reinforcing the importance of aspiration as a first step toward achievement (Homel & Ryan, 2014).

Factors related to student background and the school setting, for example, gender, SES, cultural capital, indigenous status, language background, location, year level, prior achievement and students’ self-perceptions of their ability, all interact with student aspirations and with each other to some extent. This ‘fuzzy’ complexity complicates the interpretation of related research but some common ground is emerging: gender gaps in STEM education and its associated careers are consistently prevalent along with stereotypical gendered views of many STEM careers, students’ ability beliefs and prior achievement impact on their choices to study STEM subjects in school, the key influence of family factors including parents’ knowledge of or experience with STEM careers is prevalent and the pivotal role of teachers and schools as potential enablers for STEM aspirations remains a key site for intervention. While the vast majority of this empirical research into career development and occupational aspirations have been conducted with adolescents (Watson, Nota & McMahon, 2015), few studies have focussed on children. Our study responds to calls to ‘focus more systematic research attention on the childhood antecedents and dimensions of occupational choice and vocational development across the life span’ (Hartung et al., 2005, p.412). We employ an accelerated longitudinal design to examine the development of aspirations for STEM careers from years 3 to 12, providing key insights into changes in aspirations from childhood through to adolescence, in conjunction with other key variables such as gender, prior achievement and SES.

Using a unique dataset, involving students from 8 to 18 years of age, we examine the formation of aspirations for STEM careers and make recommendations for STEM career education across the school years.

Male students were more likely to aspire to a STEM career than females (OR = 3.18) and students in the older cohorts were also more likely (OR = 1.35 for the year 5–8 cohort; OR = 1.58 for the year 7–10 cohort; OR = 1.54 for the year 9–12 cohort) in comparison to students in the year 3–6 cohort. With respect to cultural capital, students in the highest two quartiles were more likely to aspire to a STEM career than those in the lowest quartile (OR = 1.27 Q3; OR = 1.47 Q4). Students’ prior achievement in reading and numeracy, as measured by NAPLAN, was also a significant variable. In relation to reading achievement students in quartiles other than the lowest quartile were increasingly likely to aspire to a STEM career (OR = 2.13 Q4; OR = 1.74 Q3; OR = 1.66 Q2). Achievement in numeracy was also related to the choice of a STEM career with those in the top numeracy quartile more likely to express interest in these careers than those in the lowest quartile (OR = 2.71). Those in the second quartile for numeracy were also more likely to aspire to a STEM career (OR = 2.02), as were those in the third quartile (OR = 1.52) than those in the bottom quartile for numeracy.

The addition of school-related variables in model 2 mediates the effects of some of the student background factors which were found to be significant in model 1. While we found that students with an Indigenous background were significantly less likely to express interest in a STEM career than their non-indigenous peers (5% Indigenous; 11% non-Indigenous), and that the proportion of students choosing STEM significantly increased across the SES quartiles (quartile 1, 8%; quartile 2, 9%; quartile 3, 11%; quartile 4, 14%), both of these factors were not significant in the presence of school-related factors measuring students’ prior achievement in reading and numeracy. In other words, Indigenous and low SES students with relatively high levels of prior achievement are just as likely to aspire to a STEM career as their non-Indigenous or high SES peers. In contrast, prior achievement does not mediate the effect of gender, with males significantly more likely to aspire to a STEM career than females even when these other factors are taken into account.

Two other factors related to student background were found to be significant predictors of the aspiration for a STEM career: exposure to a parent in a STEM career (OR = 1.42) and being in the higher two quartiles for cultural capital (OR = 1.47 Q4; OR = 1.27 Q3). Other factors related to student background were found not to be significant in explaining aspiration for a STEM career in the presence of the variables in model 2 representing school-related factors.

This paper has provided a detailed examination of primary and secondary school student aspirations for STEM careers. In general, more students are interested in science careers, followed by engineering, technology and mathematics careers, respectively. We have examined a combination of student background and school-related factors to determine those which are significant predictors of school students’ likelihood of expressing interest in a post-school STEM career. In terms of student background, the analysis revealed that students were more likely to aspire to a STEM career as they get older, more so for males than females. Cultural capital was found to be a salient factor related to students’ home life, as was their exposure to a parent working in a STEM career. In relation to the school-related factors tested, students’ prior achievement in both reading and numeracy were found to be significant predictors of aspiration for a STEM career. Other factors, seemingly related to the choice of STEM when considered independently, for example, SES, indigenous status, language background and students’ self-rated ability, were found not to be significant in explaining aspirations for STEM in the presence of the other variables.

At the individual level, consideration of a STEM career remains a highly gendered concern but the reasons remain unclear. Students’ choices in this study aligned with the stereotypical notion that STEM careers are largely ‘male oriented’ and are therefore, not desirable for females, particularly in the fields of technology and engineering. Previous research has concluded that females’ lack of aspiration in STEM may be linked to poor academic self-concept even when their achievement levels are comparable to males (Sax et al., 2015). This study finds that actual academic achievement rather than students’ perception of their achievement is more predictive of aspirations for a STEM career. However, it could also be that academic self-concept is a precursor to academic achievement (MacPhee et al., 2013) and so its importance as a factor for indirectly enabling aspirations for STEM careers remains. The findings highlight the need for teachers to be aware of the importance of maintaining student interest and achievement levels in mathematics in the earlier years of schooling so that students are able to sustain their belief that a STEM career is achievable.

Another gender difference in STEM disciplines is the relationship between student interest and skill. It has previously been found to be bi-directional for males but uni-directional for females (Lee, Lawson & McHale, 2015). In other words, male interest in STEM can lead to increased achievement and conversely high achievement can increase males’ interest in STEM careers. However, for females, while achievement in STEM disciplines can lead to increased interest, the reverse does not generally occur. Given that interest in STEM in early high school is a key predictor of interest at the end of formal schooling (Sadler et al., 2012), this finding reinforces the importance for females of supporting the development of sound STEM skills from an early age. Whereas males seem to maintain interest in STEM without necessarily having strong underlying skills, females are less likely to do so (Lee et al., 2015). Maintaining and supporting females’ academic achievement in STEM disciplines is vital for sustaining their interest as they progress through school. However, there is evidence that female students’ achievement in STEM can be undermined through teachers’ unconscious bias towards males in the classroom (Blickenstaff, 2005) leading them to spend more time interacting with male students in comparison to females (Shumow & Schmidt, 2013). It is important that teachers are aware of their potential unconscious bias in STEM classrooms which could impact negatively on female students’ achievement and aspirations.

Students cannot aspire to particular careers, including those in the STEM fields, if they lack knowledge of those careers (van Tuijl & van der Molen, 2015). This knowledge can be gained through careers education, generally in secondary school, or through exposure to adult family members or acquaintances outside of school. Our study finds that students are more likely to aspire to a STEM career if one or more of their parents/carers work in a STEM occupation, thereby giving students detailed knowledge of careers that they might otherwise not gain. Students without access to parents working in STEM fields are less likely to aspire to a STEM career, even if they have comparable levels of prior achievement. This finding reveals a significant gap in students’ knowledge about STEM careers, possibly leading some students to compromise their aspirations due to a lack of knowledge of the breadth of possibilities. The implication for careers education in STEM is that it is currently not filling this gap, leaving some students who may potentially have the ‘skill and will’ to pursue STEM without the necessary familiarity with these careers.

The influence that a parent’s STEM occupation can have on students’ aspirations indicates the importance of role models for career development. Several studies point to the lack of female STEM role models as a reason for the persistent lack of female interest in STEM (Beede et al., 2011; Broadley, 2015; Wang & Degol, 2013). However, the impact of role models, particularly for female students, is contentious. While the stereotypical view of women in STEM is that they are unfeminine, which may be a disincentive for some females considering a STEM career, the use of overtly feminine STEM role models has also been found to have a negative impact on the STEM aspirations of school-aged girls (Betz & Sekaquaptewa, 2012). This counterintuitive finding could indicate that the use of role models to entice females into STEM is a somewhat shallow approach, particularly if stereotypical views of the STEM disciplines remain ‘male’ and ‘thing’ oriented (van Tuijl & van der Molen, 2015). Approaches that reveal non-gendered but multi-faceted views of STEM careers, challenging the dominant stereotypes, might be more successful in attracting females.

The need to increase numbers of suitably qualified STEM professionals has prompted interest in the school sector and its role in developing the future STEM workforce. Our findings reveal a dire lack of interest in, or possible lack of knowledge of, careers that focussed on mathematics, with only eight out of 6492 students expressing such interest during the four years of the study. While this is concerning, the role of mathematics as a foundational component necessary for any STEM career is important. The way students perceive themselves as students learning mathematics can have a vital impact on aspirations for related STEM careers, particularly for females. To increase aspirations for STEM careers, schools need to plan instruction that increases students’ self-concept and achievement in mathematics in particular.

This study has revealed a series of interrelated key factors associated with school students’ aspirations for STEM careers. These are gender, cultural capital, age, parent STEM occupation and prior achievement in reading and numeracy. Schools clearly have a role to play in improving the achievement levels of all students in STEM disciplines. However, for females the importance of school achievement appears to be even more crucial than it is for males because it can drive interest in STEM. However, females with achievement comparable to males are considerably less likely to aspire to STEM careers, indicating that prior achievement is only one part of the puzzle in determining the relative lack of female interest in STEM.

The likelihood of student interest in STEM increases for those students who have a parent working in a STEM occupation, emphasising that insider knowledge of STEM careers can lead to increased awareness and interest. Providing more students with this kind of knowledge is a potentially fruitful approach for careers education in schools. However, in doing so, care must be taken to avoid the reinforcement of STEM gender stereotypes which are proving very resistant to change. Future research using the qualitative data from the Aspirations Longitudinal Study will provide further insights into the ways in which students think about STEM careers and their motivations for aspiring to them, shedding light on the seemingly intransigent patterns of participation reported in this paper.