STEM-oriented primary school children: participation in informal STEM programmes and career aspirations

ABSTRACT

It has been found that participation in informal, year-long STEM programmes for primary school children encourages further STEM learning and motivates STEM career aspirations. To date, no previous research has investigated such programmes. The study's goals were to collect and analyse demographic data about the 3rd-6th grade participants in such programmes (N = 276), to identify the reasons that led them to participate in the specific programmes (e.g. science, computers, robotics) and (3) to map their career aspirations. Qualitative data were obtained from two open-ended questions and analysed through summative content analysis. Key findings included: More boys participated in the programmes than girls; younger students participated more than older students; predominant motivations cited for participating in the programmes were interest and enjoyment as opposed to practical utility concerns; more 5th-6th graders cited utility concerns than 3rd-4th graders; about 1/3 of the participants had STEM career aspirations; more 5th-6th graders aspired to STEM careers than 3rd-4th graders; more 3rd-4th grade boys aspired to STEM careers than girls. We concluded that participation in these programmes signifies entry into STEM career pathways at a much earlier age than previously reported.

KEYWORDS:

• Informal education
• elementary/primary
• STEM/STEAM

Introduction

Most nations throughout the developed world have established educational goals and strategies to lead bright, motivated K-12 and post-secondary students to learn STEM disciplines (Science, Technology, Engineering and Math), aspire toward STEM careers and to choose one (National Science Foundation, 2016). Toward accomplishing these goals, a great deal of research has investigated the development of students’ career aspirations and their actual academic choices that lead to specific career choices, STEM or otherwise. One key finding has shown that STEM career aspirations are strongly and positively correlated with actual choices to learn STEM disciplines and topics over time; that is, youngsters who aspire to work in STEM careers choose to pursue further STEM learning throughout K-12 schooling and beyond (e.g. Chambers et al., 2018; DeWitt & Archer, 2015; Jiang et al., 2020; Lent et al., 1994; Lent et al., 2018).

Over the past decade, research has focused on the factors that influence STEM academic choices and career aspirations during ‘middle-childhood’ (ages: ≈6–12). Findings, cited below, reported that informal science education motivates youngsters to pursue STEM learning and to aspire toward STEM careers. Our current study focuses on one specific kind of informal science education, namely structured hands-on STEM programmes for science-oriented primary school children and the careers to which they aspired.

STEM career aspirations during middle-childhood

To begin, we note that some researchers contend that young children’s notions of career aspirations tend to be unrealistic and are unlikely to have any relation with what they might actually do in the future (Ginzberg, 1952; Gore et al., 2017). However, over the past decade, the importance of such career aspirations has been recognised and established. Findings have shown that around age nine (3rd graders), future career orientation is based in reality and that such aspirations shed light on children’s perceptions of and ideas about adult work (e.g. Archer et al., 2013; Archer et al., 2020; Chambers et al., 2018; Maltese & Cooper, 2017; Shin et al., 2019). Even more so, from research findings among children, it has been found that career aspirations that coalesce at this time are deeply ingrained and persist over time (e.g. Archer et al., 2013, 2020; Chambers et al., 2018; DeWitt & Archer, 2015; Shin et al., 2019; Vinni-Laakso et al., 2019). In addition, retrospective studies among adults corroborated findings that STEM career aspirations began at a young age (Dabney et al., 2013; Maltese & Tai, 2010; Venville et al., 2013). For example, Maltese and Tai (2010) interviewed 116 graduate students and scientists about when and how they initially became interested in science. They found that nearly two-thirds recounted that their interest in science began in primary school.

Informal STEM education

Allen and Peterman (2019) defined informal STEM education as ‘learning in science, technology, engineering and math that takes place across a multitude of designed settings and experiences outside of the formal classroom’ (p.18). As such, its domain is vast and includes different environments (e.g. botanical gardens, planetariums and science museums) and initiatives (e.g. camps, clubs and competitions). The informal STEM programmes for primary school students that we studied may be classified as ‘clubs’. In differentiating informal STEM clubs from formal classroom environments, Fitzgerald et al. (2020) noted different goals (affective goals often supersede cognitive ones), instructional strategies (active inquiry-based learning versus passive knowledge acquisition through books and lectures), learning outcomes (open-ended versus the need to meet narrow standards of knowledge and skills) and assessment procedures (ongoing and formative as opposed to summative).

Differences between the clubs themselves are also wide-ranging especially regarding duration (from several weeks to multi-year time frames) and target populations, programmes for gifted and science oriented students as well as for students from under-represented groups based on race, ethnicity, gender, and/or socio-economic status (Archer et al., 2021; Ayers et al., 2020; Hite & White, 2021). Although the differences among STEM clubs exacerbate research that attempts to evaluate their impact on participants, findings have shown that participation often contributes to student interest in STEM and to their ability related beliefs (self-efficacy) which, in turn, predict the pursuit of further STEM learning and career aspirations (e.g. Caspi et al., 2020; Fallik et al., 2013; Gecu-Parmaksiz et al., 2021; Kong et al., 2014; Newell et al., 2015; Rende et al., 2022; Sha et al., 2015; Young et al., 2016).

The informal STEM programs for Israeli primary school students

For Israeli primary school children, the influence of participation in these programmes on the pursuit of further STEM learning has been reported. Caspi et al. (2020) found that participation in at least one such programme predicted the choices of 6^th grade primary-school graduates’ (≈12-year-olds) to enrol in advanced science programmes over the three-year duration of middle-school. Caspi et al. (2019) found that this academic choice, in turn, substantially increased the predictive outcome of middle-school graduates’ (≈15-year-olds) choice of a STEM major in high school (a combination of prescribed and elective courses in the chosen discipline) by an Odds Ratio of about 16:1. In other words, a meaningful academic choice made by primary school students may signify an even earlier initial entry than heretofore reported into what has been called STEM pathways (Cannady et al., 2014). Such pathways lead to a STEM career often via a benchmark high school STEM major.

The informal STEM programmes we investigated are aimed at very bright, highly motivated 3rd-6th grade primary school students. To find such students, homeroom teachers received intentionally fuzzy guidelines for recommending potential participants: ‘Recommend the best and brightest students in your class aiming at approximately the top 15%’. In other words, the measures defining best and brightest are determined by each homeroom teacher. The cutoff implicitly conveys that the programmes are selective, intended only for a minority of students. The guidelines however also emphasise that the number of students recommended is flexible, not a quota to be met; that is, if there are fewer such students, then teachers are guided to recommend only them; if there are more, teachers may recommend all who meet these criteria. From a macro perspective, based on Ministry of Education statistics, the programmes are aimed at about 100,000 primary school students nationwide.

The programmes have two main goals: cognitive, to develop students’ scientific reasoning skills through a series of guided, inquiry-based encounters; affective, to encourage positive attitudes toward and interest in STEM topics and careers. In the science programmes, instructional strategies enable students to understand natural and physical phenomena and the laws governing them through guided hands-on exploration. Students build models and test them, do experiments, observe and record data, and infer scientific principles, all beyond the formal school science curricula. Lessons generally open with a brief trigger activity (e.g. question, demonstration or video) to arouse interest and elicit students’ prior knowledge of the subject to be studied, followed by a hands-on activity done alone or in small groups. Ideally, the final 10 minutes are for sharing findings and conclusions. In the robotics programme, in addition to learning the underlying principles, scientific (e.g. energy, force, hydraulics, etc.) and engineering (e.g. algorithms, flow charts, etc.), students build and programme five robots which, at the programme's end, they can take home and operate from there. Various computer programmes include an introduction to programming as well as developing games, applications and websites. About 12–16 students of similar age and abilities meet once-a-week after school for 75 min. Instructors are usually university students majoring in STEM disciplines.

Research aims and the study's importance

Given the significant influence of participation in these programmes on the pursuit of further STEM learning (Caspi et al., 2019, 2020) and that no previous research has investigated them, the study's aims were to gain insights into: (1) the influence of age and gender on participation, (2) the reasons that led students to the specific programmes they chose (e.g. science, robotics), (3) to map their career aspirations, and (4) to determine if a relationship exists between participation in a specific programme and career aspirations; and, if so, is the relationship moderated by age or gender.

The study's importance lies precisely in gaining these initial insights. To date, from anecdotal sources only (our own experience and from colleagues who run such programmes at universities, science education centres and science museums), we know tentatively that participation in these kinds of informal programmes decreases with age and that more boys participate than girls. From the literature, however, we know that the under-representation of girls and young women in STEM studies and career aspirations has been reported repeatedly (e.g. Archer et al., 2020; Chambers et al., 2018; Clark et al., 2021; Holmes et al., 2018). Our current study will try to determine whether these two phenomena begin at these early ages.

Regarding the motivation to participate in such programmes, from the literature we know the key factors that motivate older students; these include interest and enjoyment (e.g. Alexander et al., 2019; Caspi et al., 2019, 2020; Moote et al., 2020; Shin et al., 2019), practical utility concerns for the near and more distant future (e.g. Caspi et al., 2019, 2020; Cheng et al., 2022; Shin et al., 2019), self concept of ability / self-efficacy (e.g. Abe & Chikoko, 2020; Caspi et al., 2020; DeWitt & Archer, 2015; Luo et al., 2021), peers and friends (e.g. Caspi et al., 2019, 2020; DeWitt & Archer, 2015, 2017; Nugent et al., 2015; Raabe et al., 2019; Raabe & Wölfer, 2019), parents (e.g. Archer et al., 2020; Caspi et al., 2019, 2020; Moote et al., 2020; Neitzel et al., 2019; Nugent et al., 2015; Šimunović et al., 2018) and previous formal and informal school experiences (e.g. Archer et al., 2020; Caspi et al., 2019, 2020; Moote et al., 2020). We expect that during middle-childhood, findings for the younger students may be different.

Methods

Research design

We used a descriptive research design to profile the participants through the use of frequency distributions. Two demographic items recorded age and gender; two open-ended questions elicited their reasons for enrolling in the specific STEM programmes and their career aspirations. These data will provide an overview of the programmes’ participants and enable us to determine if any statistically significant relationships exist between and among the distributions. Research questions and hypotheses follow.

Research questions and hypotheses

Where possible, hypotheses follow the questions. For question 4, neither theory nor prior evidence enables us to establish hypotheses.

Q1: Are there significant age and gender differences among the participants?

H1A: 3rd-4th graders will participate more than the 5th-6th graders (e.g. Archer et al., 2020; Blotnicky et al., 2018; Shaby et al., 2021).

H1B: More boys will participate than girls (e.g. Archer et al., 2020; Clark et al., 2021; Kang et al., 2019; Stoet & Geary, 2018; Vrieler et al., 2020).

Q2: What motivated students to select the specific programmes they chose and are their motivations moderated by age and gender?

H2: Interest and enjoyment will be cited as the predominant motivation factors for all programs regardless of age and gender (e.g. Archer et al., 2020; Caspi et al., 2019, 2020).

Q3: Are STEM career aspirations moderated by age and gender?

H3A: Career aspirations during middle-childhood are relatively stable and will not be modified by age (Archer et al., 2020).

H3B: More boys than girls will aspire to STEM careers (e.g. Archer et al., 2020; Chambers et al., 2018; Clark et al., 2021; Holmes et al., 2018).

Q4: Are STEM career aspirations and the choice of a particular STEM programme linked? If so, do age and gender moderate this link?

Procedure

The study was approved by an Institutional Review Board contingent on getting signed parental consent for the children to fill out the survey. Parental consent was obtained during the registration procedure. In each STEM programme, the children completed anonymous, paper-and-pencil questionnaires at the start of the first lesson, preceding classroom activities of any kind. The rationale was explained by research assistants who had no prior association with the participants while the one page form was being distributed face down on the tables. Based on observations of the participants made by the assistants and a review of the forms, all the filled out questionnaires were deemed acceptable.

Participants

Participants included 276 students (3rd-6th graders; 186 boys / 90 girls), predominantly Jewish, enrolled in diverse primary schools located throughout a mid-sized Israeli city. All participants are fluent Hebrew speakers and there is little or no extreme wealth or poverty or affiliation with extreme religious groups. We further categorised participants into two groups based on distinctions made by the Israeli Ministry of Education which defines three primary school grade levels: lower (1st-2nd), middle (3rd-4th) and upper (5th-6th). Therefore, participants included 173 3rd-4th graders (103 boys; 70 girls) and 103 5th-6th graders (83 boys; 20 girls).

Participation in the programmes is limited to highly qualified and motivated primary school students who were recommended by their homeroom teachers. Enrolment was generally done by parents accompanied by their child who, in the vast majority of cases, knew beforehand which specific programme he or she wanted. The children were often adamant about their choices.

Instrumentation and data analysis

Two demographic items yielded unambiguous binary data. Age was recorded as ‘grade level’ while gender was either ‘boy’ or ‘girl’ (in Israel today, it is very uncommon to use other gender categories for primary school children). To test for statistically significant differences, we used the Chi square test. To test for effect size, we used Cramer's V where values range between 0 and +1; a relatively strong effect is present for values greater than 0.25. Two open-ended questions followed the demographic items:

1. Why did you choose this programme? List all the reasons you can think of, important and seemingly unimportant.

2. What do you want to work at when you grow up?

These questions allowed participants to freely express the reasons underlying their academic choices and career aspirations while avoiding any bias or influence that could have resulted from the use of closed-ended items that may suggest unintended or spurious responses (Callegaro et al., 2015).

The qualitative data from these questions were coded into predetermined categories. For question 1, we included all the key known predictors for academic choice reported in the literature cited above, namely interest and enjoyment, practical utility concerns, self concept of ability/self-efficacy, peers and friends, parents, and previous formal and informal school experiences.

For question 2, we defined three categories: STEM careers (e.g. science, mathematics and engineering); STEM-related careers (e.g. medicine, STEM teacher); all others (e.g. sports, business).

Summative content analysis was used to categorise responses. Two independent raters, experienced science educators and researchers, categorised all student responses; that is, both scored all the responses and then compared their results. Interrater reliability was 93.20% (Cohen's κ = .89). Disagreement was resolved through discussion and consensus. We also noted how many reasons were cited for enrolling in the STEM programme (representativeness) and the number of reasons per respondent (willingness to cooperate).

Results

Demographics (Q1)

Participants in our study included 276 3rd-6th grade students.

• In support of hypothesis H1A, more 3rd-4th graders (62.68%; 173/276) participated in the programmes than 5th-6th graders (37.32%; 103/276) [χ²(1) = 17.754, p < .009, Cramer's V = 0.25].

• In support of hypothesis H1B, more boys (67.39%; 186/276) participated than girls (32.61%; 90/276) [χ²(1) = 33.391, p < .001, Cramer's V = 0.35].

• We found that gender disparity increases with age: 40.46% (70/173) of the 3rd-4th graders were girls while 19.42% (20/103) of the 5th-6th graders were girls [χ²(1) = 13.01, p < .001, Cramer's V = 0.22].

Table 1 shows gender and age frequencies for the participants (N = 276) in the specific programmes.

Table 1. Age and gender frequencies per programmes.

Programmes	3rd-4th grades (N = 173)				5th-6th grades (N = 103)				Totals (N = 276)
	Boys		Girls		Boys		Girls		Boys		Girls
	N	%	N	%	N	%	N	%	N	%	N	%
Computers	63	61.17%	39	55.71%	68	81.93%	13	65.00%	131	70.43%	52	57.78%
Robotics	Not applicable				10	12.05%	2	10.00%	10	5.38%	2	2.22%
Science	28	27.18%	25	35.71%	Not applicable				28	15.05%	25	27.78%
Math	12	11.65%	6	8.57%	5	6.02%	5	25.00%	17	9.14%	11	12.22%
Totals	103	100%	70	100%	83	100%	20	100%	186	100%	90	100%

We next tested for gender disparities among the 3rd-4th graders regarding their choice to participate in the specific programmes – science, technology or math. No significant differences were found:

• Science: 27.18% (28/103) of the boys made this choice versus 35.71% (25/70) of the girls.

[χ² (1) = 1.43, p = .23, Cramer's V = 0.09].

• Technology: 61.17% (63/103) of the boys chose computers versus 55.71% (39/70) of the girls.

[χ² (1) = 0.51, p = .48, Cramer's V = 0.05].

• Math: 11.65% (12/103) of the boys versus 8.57% (6/70) of the girls.

[χ² (1) = 0.42, p = .52, Cramer's V = 0.04].Regarding gender disparities among the 5th-6th graders, given that only 20 girls participated in the programmes, the subsequent sample sizes were too small for statistical testing.

Student motivations (Q2)

To begin, we tested responses for representativeness and willingness to cooperate which establish the data's reliability (Krosnick & Presser, 2010). Table 2 shows the number of reasons the participants cited.

Table 2. Number of reasons cited by participants for choosing the specific informal programmes.

# ofreasons cited	Grades 3–4 (N = 173)				Grades 5–6 (N = 103)				Totals (N = 276)
	Boys		Girls		Boys		Girls		Boys		Girls
	N	%	N	%	N	%	N	%	N	%	N	%
0	10	9.71%	3	4.29%	4	4.82%	0	0%	14	7.53%	3	3.33%
1	46	44.66%	25	35.71%	21	25.30%	6	30%	67	36.02%	31	34.44%
2	37	35.92%	29	41.43%	31	37.35%	8	40%	68	36.56%	37	41.11%
3	10	9.71%	13	18.57%	27	32.53%	6	30%	37	19.89%	19	21.11%
Tot.	103	100%	70	100%	83	100%	20	100%	186	100%	90	100%

Total responses for all students (N = 276) showed high rates of representativeness: 93.84% (259/276) cited at least 1 reason and willingness to cooperate: 58.33% (161/276) cited 2 or more reasons. There were no significant gender differences; in other words, boys and girls were equally representative and willing to cooperate. Table 3 lists the reasons for participating in the programmes along with illustrative examples; Table 4 shows the reasons for choosing the specific programmes per grade levels and gender.

Table 3. Typical examples per category.

Categories	Examples
Interest/enjoyment	I love math; I like to do experiments; Science is interesting; I’m interested in learning about robots.
Utility	I want to be smarter; I want to earn a lot of money.
Self-efficacy	I'm good at programming; I'm really, really good at building robots.
Friends	I want to be with my best friend; to be with friends.
Parents	My parents really want me to enrol; Enrolling will make my parents happy.

Table 4. Reasons for choosing the programmes per grade levels and gender.

Categories	Grades 3–4 (268)				Grades 5–6 (N = 197)				Totals (N = 465)
	Boys		Girls		Boys		Girls		Boys		Girls
	N	%	N	%	N	%	N	%	N	%	N	%
Interest/enjoyment	117	79.59%	101	83.47%	111	70.70%	32	80%	228	75%	133	82.61%
Utility	16	10.88%	11	9.09%	32	20.38%	6	15%	48	15.79%	17	10.56%
Self-efficacy	3	2.04%	1	0.83%	3	1.91%	0	0%	6	1.97%	1	0.62%
Friends	0	0%	2	1.65%	3	1.91%	0	0%	3	0.99%	2	1.24%
Parents	11	7.48%	6	4.96%	8	5.10%	2	5%	19	6.25%	8	4.97%
Totals	147	100%	121	100%	157	100%	40	100%	304	100%	161	100%

In support of hypothesis H2, interest/enjoyment was by far the predominant factor. Furthermore, regarding the moderating effects of age on motivation, no significant difference was found between the two grade levels [χ²(5) = 9.072, p = .110, Cramer's V = 0.06]. However, although not statistically significant, we note that the percentage of 5th-6th grade students citing utility was almost twice that of their younger counterparts 19.29%, (38/197) vs. 10.07% (27/268). Regarding gender difference in motivation, no significant difference was found between boys and girls [χ²(5) = 4.488, p = .481, Cramer's V = 0.04].

Career aspirations (Q3)

Career aspirations were noted by 76.45% (211/276) of the participants; the remaining 23.55% (65/276) either did not answer or wrote ‘don't know’. Regarding ‘no answer’ or ‘don't know’, age and gender parity were found across both grade levels. Career aspirations are shown in Table 5.

Table 5. Career aspirations by grade and gender.

CAREERS	Grades 3–4 (N = 173)				Grades 5–6 (N = 103)				Total (N = 276)
	Boys (N = 103)		Girls (N = 70)		Boys (N = 83)		Girls (N = 20)		Boys (N = 186)		Girls (N = 90)
	N	%	N	%	N	%	N	%	N	%	N	%
STEM
1. Computers	15	14.56%	4	5.71%	19	22.89%	7	35%	34	18.28%	11	12.22%
2. Hi tech	6	5.83%	1	1.43%	8	9.64%	3	15%	14	7.53%	4	4.44%
3. Scientist	8	7.77%	6	8.57%	2	2.41%	1	5%	10	5.38%	7	7.78%
4. Engineering	2	1.94%	1	1.43%	12	14.46%	0	0%	14	7.53%	1	1.11%
5. Mathematics	2	1.94%	0	0%	0	0%	0	0%	2	1.08%	0	0%
Sub-Totals	33	32.04%	12	17.14%	41	49.40%	11	55%	74	39.78%	23	25.56%
STEM-related
1. M.D.	5	4.85%	8	11.43%	0	0%	3	15%	5	2.69%	11	12.22%
2. Astronaut	1	0.97%	3	4.29%	3	3.61%	0	0%	4	2.15%	3	3.33%
3. Inventor	2	1.94%	0	0%	1	1.20%	1	5%	3	1.61%	1	1.11%
4. Veterinarian	0	0%	2	2.86%	0	0%	0	0%	0	0%	2	2.22%
5. STEM teacher	1	0.97%	2	2.86%	1	1.20%	0	0%	2	1.08%	2	2.22%
Sub-Totals	9	8.74%	15	21.43%	5	6.02%	4	20%	14	7.53%	19	21.11%
Non-STEM
1. Soccer	12	11.65%	0	0%	5	6.02%	0	0%	17	9.14%	0	0%
2. Sports (other)	2	1.94%	3	4.29%	1	1.20%	0	0%	3	1.61%	3	3.33%
3. Business	6	5.83%	0	0%	2	2.41%	0	0%	8	4.30%	0	0%
4. Pilot or military	2	1.94%	0	0%	4	4.82%	0	0%	6	3.23%	0	0%
5. Music	2	1.94%	3	4.29%	0	0%	0	0%	2	1.08%	3	3.33%
6. Fashion design	0	0%	5	7.14%	0	0%	0	0%	0	0%	5	5.56%
7. Teacher	0	0%	5	7.14%	0	0%	0	0%	0	0%	5	5.56%
8. Law / banking	2	1.94%	0	0%	2	2.41%	0	0%	4	2.15%	0	0%
9. Police	2	1.94%	1	1.43%	1	1.20%	0	0%	3	1.61%	1	1.11%
10. Artist	2	1.94%	1	1.43%	0	0%	0	0%	2	1.08%	1	1.11%
11. Actor/actress	0	0%	1	1.43%	2	2.41%	0	0%	2	1.08%	1	1.11%
12. Dance	0	0%	3	4.29%	0	0%	0	0%	0	0%	3	3.33%
13. Paramedic	2	1.94%	0	0%	0	0%	0	0%	2	1.08%	0	0%
14. Film	0	0%	0	0%	1	1.20%	0	0%	1	0.54%	0	0%
15. Misc.	4	3.88%	4	5.71%	1	1.20%	0	0%	5	2.69%	4	4.44%
Sub-Totals	36	34.95%	26	37.14%	19	22.89%	0	0%	55	29.57%	26	28.89%
No answer or ‘Don't know’	25	24.27%	17	24.29%	18	21.69%	5	25%	43	23.12%	22	24.44%
TOTALS	103	100%	70	100%	83	100%	20	100%	186	100%	90	100%

Contrary to H3A that hypothesised the stability of career aspirations during middle-childhood, we found that significantly more 5th-6th graders (50.49%; 52/103) aspired to STEM careers than their younger 3rd-4th grade counterparts (26.01%; 45/173) [χ²(1) = 16.97, p < .001, Cramer's V = 0.28]. Furthermore, among those aspiring to STEM careers, further investigation showed the effect of age on technology career aspirations (careers 1, 2 and 4 in Table 5): Nearly all such 5th-6th graders 94.23% (49/52) aspired to technology careers versus 64.44% (29/45) of the younger 3rd-4th graders [χ²(1) = 13.59, p < .001, Cramer's V = 0.37].

In support of hypothesis H3B, we found that significantly more boys (39.78%; 74/186) aspired to STEM careers than girls (25.56%; 23/90) [χ²(1) = 5.39, p = .020, Cramer's V = 0.14]. However, we found no significant gender effects vis-à-vis technology career aspirations: 83.78% (62/74) of the boys who cited STEM careers indicated technology fields vs. 69.57% (16/23) of the girls [χ ²(1) = 2.25, p = 0.13, Cramer's V = 0.15].

Regarding STEM-related careers, 21.11% (19/90) of the girls vs. 7.53% (14/186) of the boys aspired to work in such fields. Regarding non-STEM career choices, gender parity was found: 29.57% (55/186) of the boys cited such aspirations vs. 28.89% (26/90) of the girls. Specifically, 36.36% (20/55) of the boys’ cited sports, especially soccer, while 11.54% (3/26) of the girls cited a sport as a career aspiration; 50% (13/26) of the girls aspired to work in the arts or performing arts (careers 5, 6, 10, 11, 12 and 14 in Table 1) while only 12.73% (7/55) of the boys cited such occupations.

The relationship between STEM career aspirations and the choice of a STEM programme (Q4)

We found significant differences for the effects of age and gender. Table 6 shows the data.

Table 6. Relationship between the choice of a specific STEM programme and STEM career aspirations (The % columns show the percentage of participants in each programme who cited a STEM career).

Programmes	3rd-4th grades
	Boys (N = 103)		Girls (N = 70)		Totals (N = 173)
	N	%	N	%	N	%
Computers	25/63	39.68%	14/39	35.90%	39/102	38.24%
Science	11/28	39.29%	11/25	44.00%	22/53	41.51%
Math	6/12	50.00%	2/6	33.33%	8/18	44.44%
Totals	42/103	40.78%	27/70	38.57%	69/173	39.88%
	5th-6th grades
	Boys (N = 83)		Girls (N = 20)		Totals (N = 103)
Programmes	N	%	N	%	N	%
Computers	41/68	60.29%	12/13	92.31%	53/81	65.43%
Robotics	4/10	40%	2/2	100%	6/12	50%
Math	1/5	20%	1/5	20%	2/10	20%
Totals	46/83	55.42%	15/20	75.00%	61/103	59.22%

A between-group analysis showed that 65.43% (53/81) of the older 5th-6th graders who participated in a computer programme aspired to STEM careers vs. 38.24% (39/102) of their younger counterparts. This finding is statistically significant [χ²(1) = 13.36, p < .001, Cramer's V = 0.27]. Small sample sizes precluded making inferences regarding the math programmes.

A within-group analysis showed that for the younger 3rd-4th graders who participated in the computer and science programmes, there were no significant gender effects regarding STEM career aspirations. For the older 5th-6th graders who participated in the technology programmes (computers and robotics), 93.33% (14/15) of the girls aspired to STEM careers while only 57.69% (45/78) of the boys had such aspirations [χ ²(1) = 6.89, p < .01, Cramer's V = 0.27].

Summary of key results

Participants in the informal programmes, their reasons for participation and the specific programmes chosen

More boys than girls participated in the programmes; overall participation declined with age, especially among girls. Interest and enjoyment were by far the key reasons for participating, followed by practical utility concerns that were cited more often by the older students. Among 3rd-4th graders, we found no gender disparity regarding the choice of a specific programme; that is, boys and girls participated equally in the science, technology and math programmes.

Career aspirations

Of the participants who cited career aspirations (>75%), about 1/3 named STEM occupations. Boys aspired to STEM careers more than girls; girls aspired to STEM-related careers more than boys. The older 5th-6th grade students aspired to STEM careers more than their younger counterparts. Among such 5th-6th students, the overwhelming majority (94%) aspired to technology careers. Although the younger boys aspired to STEM careers more than girls, among the older students, gender parity was found.

Career aspirations and academic choice

The older students who participated in a technology programme aspired to STEM careers more than their younger counterparts. Regarding gender, the older 5th-6th grade girls who participated in the technology programmes aspired to STEM careers much more than their male counterparts.

Discussion

In this section, we discuss (1) the participants’ age and gender, (2) their reasons for participation, (3) their STEM career aspirations, and lastly (4) the relationship between the choice of specific programmes and STEM career aspirations. Throughout this section, we will note how the study's context – the participants were bright and highly motivated – may have influenced the findings.

Participants

Age

Participation in the informal STEM programmes declined with age; more 3rd-4th graders participated than 5th-6th graders. This finding confirms anecdotal data from our own experience and from colleagues who run such programmes at universities, science education centres and science museums. Although this finding for middle-childhood students parallels the well-documented decline in adolescents’ motivation to learn science (e.g. Archer et al., 2020; Blotnicky et al., 2018; Shaby et al., 2021), it does not necessarily reflect less motivation to learn STEM. The older 5th-6th students may simply be allocating precious after school free time to explore other interests (e.g. artistic and sport activities) without lessening their interest in STEM which may be satisfied in other ways including formal schooling.

Gender

Overall (3rd-6th graders), we found that more than twice as many boys as girls participated in the programmes.

However, at a higher resolution, most striking was the more than 4:1 ratio of 5th-6th grade boys to girls (83/20) versus the almost 1.5:1 ratio of 3rd-4th grade boys to girls (103/70). For the 5th-6th grade girls, this may indeed represent the age at which even very bright and STEM capable girls begin to disengage from technology oriented disciplines. We note that the programmes offered were technology oriented (computers and robotics), not science. Had science programmes been offered, perhaps more girls would have enrolled.

These findings reflect the widely reported under-representation of women who study computer sciences and engineering (e.g. Archer et al., 2020; Clark et al., 2021; Kang et al., 2019; Stoet & Geary, 2018; Vrieler et al., 2020) and show that these biases are already evident at these very young ages, not just in middle-school, high school or university; even more so, it reflects disinterest and unwillingness to participate in informal technology programmes at these young ages.

A similar gender disparity was found in the math programmes where two-thirds of the 3rd-4th graders were boys. Again, older girls' and young women's aversion to math is widely reported in more representative contexts (e.g. Else-Quest et al., 2013; Stoet & Geary, 2018; Su & Rounds, 2015; Wang et al., 2015); here, however, we have shown its manifestation at young ages amongst bright and capable girls.

On a more positive note, the gender parity we found in the 3rd-4th grade science programmes has also been reported – not in actual academic choice as in our case – but in closely related young students’ positive attitudes toward science (DeWitt et al., 2013; OECD, 2007) and interest in science (DeWitt et al., 2013; OECD, 2007). This gender parity in pursuing science studies in general does not decline over time (Caspi et al., 2020; Cheryan et al., 2017); it emerges however for older girls and young women in their choices of specific science disciplines. On the one hand, the disengagement of girls from physics and computer science starts at adolescence and increases with age (e.g. Archer et al., 2020; Blotnicky et al., 2018; Cheryan et al., 2017; Shaby et al., 2021). On the other, Cheryan et al. (2017) reported that women obtain more than half of U.S. undergraduate degrees in biology and chemistry. Thus, changes along STEM career paths may often reflect gender-related issues.

Reasons for participation in the STEM programs

Interest/enjoyment and practical utility concerns

Interest and enjoyment were the predominant reasons for participation in the programmes among our selected 3rd-6th graders. This finding is not unique to bright and highly motivated students, it is also in accord with those from similar age groups and older students (e.g. Alexander et al., 2019; Caspi et al., 2019, 2020; DeWitt & Archer, 2015; Moote et al., 2020; Sha et al., 2015; Shin et al., 2019; Vinni-Laakso et al., 2019). For example, DeWitt and Archer (2015) found that 6th grade students whose ‘attitudes to school science’ (a construct very similar to interest and enjoyment) were more positive had higher ‘science aspirations’ (i.e. the pursuit of further STEM learning). This predominance of interest and enjoyment spans the K-12 continuum. In Finland, Vinni-Laakso et al. (2019) found that 1st-2nd graders reported that these factors motivated their future pursuit of further STEM learning and STEM career aspirations.

The increased importance of utility for the older 5th-6th grade students is especially noteworthy. From a developmental perspective, we may be looking at an earlier loss of naivety or purity of intent based on interest and enjoyment only. Shin et al. (2019) implemented a science utility intervention to increase the motivation (i.e. interest in science and the intent to engage in science-related activities) of representative Korean 5th-6th graders. Their study showed the increased role played by practical utility concerns for younger students from a more representative sample.

Expectancies for success (self-efficacy)

Less than 1% of the participants cited self-efficacy as a reason for participating; in other words, their beliefs about whether they can successfully meet the perceived demands of the STEM programmes were not cited. On the one hand, this finding is in accord with Vinni-Laakso et al. (2019) who found that the ‘self-concept of ability’ of the Finnish 1st-2nd graders did not predict their pursuit of further STEM learning or STEM career aspirations. On the other hand, DeWitt and Archer (2015) found that 6th grade students’ pursuit of further STEM learning and STEM career aspirations was predicted by self-concept in science. Beyond inconclusiveness regarding the motivational influence of self-efficacy during middle-childhood, we know that for adolescents and older students it predicts their pursuit of further STEM learning and STEM career aspirations (e.g. Abe & Chikoko, 2020; Adedokun et al., 2013; Badri et al., 2016; Caspi et al., 2020; Roberts et al., 2018). We therefore suggest that self-efficacy vis-à-vis STEM activities is either below an awareness threshold or not yet developed at this young age perhaps due to lack of opportunity.

Parents and peers

Despite voluminous research regarding the influence of parents on the pursuit of further STEM learning, (e.g. Archer et al., 2020; Caspi et al., 2019, 2020; Moote et al., 2020; Neitzel et al., 2019; Šimunović et al., 2018), less than 5% of the participants cited them as reasons for enrolling. In the same vein, although the influence of peers is also widely reported (e.g. DeWitt & Archer, 2015; DeWitt & Archer, 2017; Nugent et al., 2015; Raabe et al., 2019; Raabe & Wölfer, 2019), only 1% cited them as a reason for participating. We cannot infer from these findings, however, that parents and peers had no influence on the participants; rather, ‘self-concept of independence’, the capacity to decide autonomously that is developed around age three (Helwig, 2006) may explain the responses. Here, despite student limitations (they cannot enrol or pay for participation on their own), they still may have a sense of choice and perceive the decision to participate as made solely on their own.

STEM career aspirations

About 1/3 of the participants aspired to STEM occupations. On a nationwide scale, this provides further evidence that for tens of thousands of Israeli primary school students, pathways to STEM career aspirations begin at a much earlier age than previous recorded, namely 3rd-6th graders. We also found that more 5th-6th graders aspired to STEM careers than their 3rd-4th grade counterparts (59% vs. 30% respectively); this finding points to an age demarcation that may indicate a fuzzy boundary where STEM learning preferences coalesce into a career inclination or mindset or disposition.

Gender related findings about STEM career aspirations showed that overall more boys aspired to STEM careers than girls and that more girls aspired to STEM-related careers than boys. Similar findings have been reported in the literature not just for bright students in selected contexts (e.g. Chambers et al., 2018; Cheryan et al., 2017; Stoet & Geary, 2022). We also found that while more 3rd-4th grade boys aspired to STEM careers than girls, among 5th-6th graders, gender parity was found. These findings further strengthen the contention that entry into STEM career pathways begins at younger ages. They also show the positive relation between participation in informal STEM programmes and STEM career aspirations during middle-childhood (e.g. Bindis, 2020; Caspi et al., 2020; Habig et al., 2020; Maiorca et al., 2021; Vrieler et al., 2020). These relations do not demonstrate causality. It is plausible that participation in such programmes influenced STEM career aspirations. It is equally plausible that STEM career aspirations led to participation in the programmes.

STEM career aspirations: our findings versus others cited in the literature

We found no research into the career aspirations of young students who resembled our sample population, namely very bright and highly motivated 3rd-6th graders who participated in selective science clubs. We will however compare our findings with others for illustrative purposes which contrast the orders of magnitude vis-à-vis STEM career aspirations between the dissimilar groups being compared.

In our study, 34% of the participants expressed STEM career aspirations. This finding contrasts with those from two large-scale representative studies carried out in the United Kingdom. First, DeWitt and Archer (2015) investigated the career aspirations of more than 9,000 6th grade students aged 10-11. They found that 7.8% held high science aspirations, a construct that included both the pursuit of further science learning and a science career. Also in the UK, Chambers et al. (2018) investigated the career aspirations of more than 13,000 primary school children aged 7-11; their findings that 6.7% aspired to careers in science or engineering are similar to those of DeWitt and Archer (2015).

We reiterate that we investigated a select, non-representative group and that the purpose of this comparison is to emphasise a second key factor that discriminates between the studies, namely participation in the informal STEM programmes. We further dampen the comparison by noting the possibility that participation in such programmes may emerge from already existing STEM career aspirations, or any other predisposed variables that define the current population. In other words, based on the current qualitative data, we cannot conclude that participation in informal STEM programmes positively influences (i.e. increases) STEM career aspirations.

STEM career aspirations and the choice of specific programs

For the younger students, we found no relationship between STEM career aspirations and the programmes they chose, science or technology. However, the older students aspired to STEM careers more than their younger counterparts. This reinforces the possibility that this is indeed, as noted above, an age demarcation where some students start to chart their way toward what may become a long-life identity. This does not negate the later entrance of others into STEM career pathways. Future studies may test for differences between those who felt at a very early age their aspiration to be a scientist or engineer versus others who did so later or even very much later.

The older girls who participated in the male dominated technology programmes aspired to STEM careers much more than their younger counterparts, both boys and girls, and, in particular, more than their age equivalent male counterparts. This finding not only supports our previous speculation that we are looking at an age demarcation for developing career aspirations, but also shows that these girls broke entrenched gender stereotypes at a very early age. We also reiterate that these girls were a small minority.

The sources of such pervasive stereotypes are well documented and include ever-present socialising agents such as teachers, parents and peers. For example, some K-8 teachers exhibit implied gender stereotypes when evaluating the performance of male students, favour male students and think that girls are less skilled at math than boys (Copur-Gencturk et al., 2021). Parents too may hold similar biases (Kim et al., 2018) and may encourage their sons to take part in STEM-related activities more than their daughters (Eccles, 2005). Peer biases, especially from boys, are also prevalent. Girls who do well in STEM domains may threaten an implied gender status hierarchy and cause boys to react in a negative manner (Kuchynka et al., 2022).

Given this subtle and pervasive ‘pro-boys against-girl’ climate, the choice of these girls to commit to informal STEM studies is noteworthy and exceptional. Future studies should investigate the factors that led such girls to enrol in the technology programmes and aspire to STEM career aspirations at such an early age.

Limitations

The findings from this study must be viewed within their limitations both broad design-related and specific procedural/methodological ones.

• This study surveyed only participants in the STEM programmes; it investigated their motivations to enrol in the specific programmes and their career aspirations. A broader study would include (1) students who were invited to participate in the programmes but did not, (2) students that participate in non-STEM programmes, and (3) students who do not participate because they were not invited to.

• Data in this study were based only on students’ self-reports, obtained through written questionnaires. A broader study should include other sources (e.g. teachers, parents and peers) and collection methods, especially interviews and quantitative closed-ended questions which can indicate significant correlations among and between the different factors at play and determine how much variance (vis-à-vis academic choice) may be attributed to each factor.

Conclusions and implications

Our major conclusion is that findings from this study add further support to previous ones showing that participation in these kinds of programmes – at least for very bright and highly motivated youngsters – marks entry into STEM pathways that lead to the pursuit of further STEM learning and STEM career aspirations at a much earlier age than previously recognised (Caspi et al., 2019, 2020). These findings mark the importance of informal STEM studies, not only as a venue to strengthen science, technology and mathematics knowledge, but also as an opportunity for science educators to sustain already existent student interest in these domains.

Although we are hesitant to jump the gun and speculate beyond the limits of our data, we do so briefly in order to point out a hypothetical ‘what if’ situation that deals with a key policy issue regarding STEM education: What if the participation of selected primary school children in such informal programmes (‘ability-grouping with curriculum differentiation’) does work well at enhancing their scientific reasoning skills and fostering motivation to pursue future STEM studies and pursue STEM careers? To what extent, if at all, should such a ‘segregation’ policy be implemented? We emphasise that any decision should take into account the social implications of any action(s) undertaken. To summarise these implications and point out their inherent dilemma, we cite the name of an article by DeWitt and Archer (2017): Participation in informal science learning experiences: The rich get richer?

Ethical statement

1. This study was approved by the Ethics committee at the University where the first author is a faculty member and the second a Research Associate.

2. Signed parental consent was obtained for all participants.

Ethics approval and consent to participate

This research study was approved by the Ethics committee of the Dept. of Education and Psychology, The Open University of Israel (#3290). Signed parental consent was obtained for all participants.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.