Introduction

STEM (science, technology, engineering and mathematics) domains play an important role in sustaining innovation capacity and promoting economic growth. Due to the growing demand for STEM professionals and the insufficient number of students willing to enter STEM fields, STEM education is gaining increasing attention worldwide, including in the United States (National Research Council, 2014), Australia (Education Council, 2016) and China (Hong Kong Education Bureau, 2016; National Institute of Educational Sciences, 2017).

It is expected that through STEM initiatives, more students will be interested in STEM careers. In addition, a scientifically literate citizen should understand how STEM and STEM professional’s work are related to the society through having the potential to benefit societal development and influence social issues directly and indirectly. In addition, knowledge of STEM careers, which could be students’ potential career pathways, can help students made informed career-related decisions in the future. Nevertheless, STEM careersFootnote 1 are often misunderstood by young students (Blažev et al., 2017; Carli et al., 2016; Cheryan et al., 2011; Hillman et al., 2014). Narrow perceptions and stereotypes could have a negative impact on students’ attitudes towards STEM learning and STEM careers (Luo et al., 2021). For example, according to the views of parents, some children reject a career in science because they think people doing science are “geeks” or “boffins” (Archer et al., 2013). In addition, high school students who had more negative stereotypes of STEM careers tended to have lower confidence in their ability to learn science, which in turn negatively influenced their interest in pursuing a STEM career (Luo et al., 2021; Garriott et al., 2017).

With the promotion of STEM education, it is necessary to carefully examine students’ perceptions of different STEM careers which are their potential career paths in the future, including the gender stereotypes, stereotypical views of STEM professionals’ work and perceptions regarding categories of STEM careers they may have. Perceptions refer to students’ understandings of concepts or objects, while stereotypes refer to the perceptions that are often negative or over-generalized (Matsumoto, 2009; VandenBos & American Psychological Association, 2015). Learning about students’ perceptions of STEM professionals may enable educators and researchers to better understand why some students lose interest in STEM careers and could enable educators to make better-informed decisions when developing interventions in STEM education.

Gender stereotypes of scientists and engineers

Having gender stereotypes related to STEM careers could be a barrier to students’ career development. It can be theoretically inferred from uncertainty-identity theory (Hogg, 2007, 2021), which claims that individuals tend to seek group identification to lower sense of uncertainty in their identities, that students tend to aspire careers that reinforce their identities, including gender identities. Likewise, psychologist Gottfredson (2002) argued that gender stereotypes may trigger children’s rejection of careers that conflict with their gender-related identities. Empirical evidence supports such claims by showing that holding gender stereotypes can have a detrimental effect on students’ self-concept of academic ability in general (Brown, 2019) and in STEM disciplines (Guimond & Roussel, 2001) and on students' STEM career aspirations (Makarova et al., 2019).

Studies have shown that when drawing a scientist or an engineer, more students (both boys and girls) drew males than females (Capobianco et al., 2011; Farland-Smith, 2009; Hansen et al., 2017; Medina-Jerez et al., 2011), which indicates that students may implicitly gender scientists and engineers as male. Moreover, researchers found that gender stereotypes regarding intelligence were observed on children as young as six (Bian et al., 2017) and that elementary students began to hold the implicit belief that “math is for boys” (Cvencek et al., 2011). A meta-analysis performed on 41 studies using the Draw-A-Scientist task spanning five decades showed that, on average, boys drew a higher percentage (96%) of male scientists than girls (58%) (Miller et al., 2018), showing a gender difference regarding gender stereotypes. According to the meta-analysis, students drew female scientists more often in later studies (Miller et al., 2018). It is worth noting that a recent study on Greek elementary students’ drawings of scientists showed that most girls had begun to draw female scientists (Emvalotis & Koutsianou, 2018).

Perceptions of scientists and engineers at work

Many elementary and middle school students from different cultural contexts (such as the United States, Australia, and Turkey) have been found to hold inadequate or even stereotypical understandings of scientists and engineers (Capobianco et al., 2011; Fralick et al., 2009; Karaçam, 2016; Laubach et al., 2012). A considerable number of studies on students’ perceptions of STEM careers have used drawing tasks (Capobianco et al., 2011; Chambers, 1983; Hansen et al., 2017). According to these studies, scientists are commonly believed to work indoors and conduct experiments (Finson, 2002; Fralick et al., 2009). In addition, some studies have focused on the work of engineers and found that elementary students’ perceptions of engineers’ work are also stereotypical and naïve; for example, they perceive them as usually performing manual labor and focusing on repairing, installing, fixing, building or using vehicles (Capobianco et al., 2011; Fralick et al., 2009; Lachapelle et al., 2012).

Some researchers have claimed that drawings of scientists, engineers and other STEM professionals could be better interpreted by comparing the similarities and differences between a student’s drawings of these different careers. Fralick et al. (2009) compared 1600 middle school students’ presentations of scientists and engineers using drawing tasks. The results revealed that the students’ perceptions of scientists and engineers were different with regard to the content of their work. However, their study required students to draw either a scientist or an engineer. Therefore, the data failed to address the differences between students’ perceptions of scientists and engineers at the individual level.

Categories of careers in students’ drawings of scientists and engineers

Only a few previous studies have analyzed what categories of scientists or engineers are present in students’ drawings. Buldu (2006) analyzed 30 5- to 8-year-old children’s drawings of scientists and classified them into two categories: people doing research and people inventing new materials. However, because of the study’s small sample size, this finding needs verification from further studies with larger samples. Capobianco et al. (2011) collected approximately 400 first- to fourth-grade children’s drawings of engineers and interviewed some of them regarding their drawings. The researchers categorized students’ understandings of engineers into four types: a mechanic, a laborer, a technician, and a designer. This indicated that many elementary students have a rather superficial understanding of engineers’ work as handy work.

Research gaps and research questions

To date, no existing study has examined students’ perceptions of technologists as STEM professionals. Technologists here refer to people who work with specific technological products and/or perform technological processes, who comprise a substantial number of and an important part of the STEM workforce. In addition, there is few study that systematically compares students’ perceptions of scientists, engineers, and technologists as STEM professionals at the same time while using unified checklists. It would be meaningful to know if the gender and other stereotypes students hold regarding technologists are similar to those they hold about scientists and engineers. It is also worth exploring how students differentiate among scientists, engineers, and technologists and how they relate their concepts of different careers (for example, inventors) with scientists, engineers, or technologists.

To address these research gaps and to provide a broader understanding of students’ perceptions of STEM careers, there is a need to explore students’ perceptions of technologists as STEM professionals and to compare the similarities and distinctions among students’ perceptions of scientists, engineers, and technologists. The three careers are included mainly because (1) completion time is limited due to elementary students’ attention span and (2) these career types constitute a large percentage of all STEM careers.

The study is focused on upper elementary (fourth to sixth grade) students due to the following considerations. First, previous studies on perceptions of scientists or engineers have indicated that stereotypes (including gender stereotypes) are noticeable in the upper-elementary age period (Capobianco et al., 2011; Farland-Smith, 2009; Hansen et al., 2017; Miller et al., 2018). Second, some researchers argued that age 10–14 is a critical period in which students’ narrow perceptions of scientists may deter them from aspiring to science careers ( DeWitt et al., 2013). Furthermore, interventions have been proven effective in terms of increasing elementary students’ perceptions of (Colston et al., 2017) or attitudes towards STEM careers (Tyler-Wood et al., 2012). Given that Hong Kong special administrative region has initiated STEM programs at all elementary schools (Hong Kong Education Bureau, 2016), upper elementary years can be a target age range for this study and for future interventions.

Therefore, the aim of this study is to exploratorily examine fourth- to sixth-grade students’ perceptions of STEM professionals using the newly developed checklists. The main research question is as follows: What are elementary students’ perceptions of STEM professionals (i.e., scientists, engineers, and technologists) according to their presentations (visual depictions and written descriptions)?

There are three sub-questions as follows:

  1. (1)

    How do elementary students characterize scientists, engineers, and technologists’ gender in their presentations of these professionals?

  2. (2)

    How do elementary students characterize scientists, engineers, and technologists’ work and STEM professionals themselves?

  3. (3)

    What are the categories of careers emerging from the students’ presentations of scientists, engineers, and technologists?

Method

This research adopted a qualitative survey investigation method involving drawing tasks and follow-up questions. This design was based on the consideration that drawing is age-appropriate for fourth to sixth graders, while follow-up questions can clarify the meaning of the drawings and provide supplementary information.

Instrument development

A series of tools have been used to examine students’ perceptions of scientists or engineers. Chambers (1983) developed the Draw-A-Scientist Test (DAST) to gather K-5 students’ drawings of scientists. Many subsequent studies on perceptions of scientists or engineers have adopted the drawing approach (Farland-Smith, 2009, 2012; Fralick et al., 2009; Subramaniam et al., 2013). Schibeci and Sorensen (1983) administered the DAST to Australian elementary students and found that it is a reliable tool for examining students’ perceptions of scientists.

Some later versions of the DAST included the Draw-A-Scientist-At-Work Test (Flick, 1990) and the Draw-An-Engineer-At-Work Task (Fralick et al., 2009). The instructions given for such tasks have evolved from “draw a scientist/engineer” to “draw a scientist at work” or “draw an engineer at work/doing engineering work” (Ambusaidi et al., 2015; Hillman et al., 2014). Researchers have argued that these instructions elicit more information regarding students’ perceptions of the work of the professionals (Capobianco et al., 2011; Farland-Smith & Tiarani, 2016; Flick, 1990; Huber & Burton, 1995).

The instrument in this study was developed in several rounds of discussion and revisions with reference to the Draw-A-Scientist Test (Chambers, 1983) and its later versions (Farland-Smith & Tiarani, 2016; Flick, 1990; Fralick et al., 2009). The instrument combined drawing tasks with follow-up questions regarding the drawings to address the concern that analysis of students’ perceptions based on drawings alone can be misleading (Bielenberg, 1997; Hillman, et al., 2014).

The instrument comprises three parts, one each for scientists, engineers, and technologists. Each drawing task states, “Please draw a … (STEM professional) at work” (with “STEM professional” being a scientist, engineer, or technologist depending on the part of the task at hand). The task was followed by several questions regarding the drawing itself, including the drawn STEM professional’s gender, what the STEM professional is doing and what the most important task in the STEM professional’s work is. Students’ responses to these last two questions were taken as descriptions of the work of the STEM professionals they had drawn.

The draft instrument was reviewed and revised by four other experts in science education (two university researchers and two Ph.D. graduates). According to the feedback, the revised instrument showed good face validity. In addition, an elementary school teacher was invited to review and revise the items to make the instrument more understandable for fourth- to sixth-grade students. In addition, the instrument was piloted on nine students (six fifth graders and three sixth graders) from a public elementary school and the students were also interviewed to ensure that they could fully understand the survey.

Participants

The participants were 564 elementary students from Hong Kong, of whom 260 were fourth- to sixth-grade students from classes in three public elementary schools and 304 were students (fourth- to sixth-grade) participating in an annual STEM event, in which the participants voluntarily engaged in STEM project-based learning during their out-of-school time for weeks. The three schools are all government-aided and co-ed schools (the school types that constitutes 81% and 94% of all elementary schools in Hong Kong respectively) from three different districts. The sampling strategy ensured the heterogeneity in both school locations and students’ degrees of STEM participation, which varies from STEM participation in school only to both formal and intensive informal STEM participation.

Students who agreed to participate in the study and whose parents gave informed consent completed the printed, anonymous instruments in their classrooms under the supervision of their teachers in approximately 15 min. A standardized instruction sheet was provided for teachers for the administration of the survey completion. Both the instrument and the students’ responses were in Chinese, and the inputted data contained no identifiable information. In total, there were 564 valid responses collected, among which 40.8% were from girls and 57.8% were from boys. The students were in the fourth grade (27.8%), fifth grade (45.8%), and sixth grade (26.4%). A review of the existing literature examining perceptions of scientists and engineers using a similar qualitative methodology suggested that the sample size in this study was adequate (Finson et al., 1995; Subramaniam et al., 2013). Along with the instrument, items asking the students whether they knew scientists/engineers/technologists in their lives were included. Only 4.1%, 18.3% and 7.3% of participating students indicated that they knew at least one scientist, an engineer or a technologist, respectively, including parents, relatives, and neighbors.

Data analysis

Drawing data were scanned, and students’ writing was input and coded for simple descriptive statistical analysis, which was conducted using SPSS 24.0. Moreover, content analysis using checklists and categorizing was adopted to analyze students’ perceptions of scientists, engineers, and technologists as the method is appropriate for inductively reveal the presence of themes in children’s drawings and writings.

Analysis of the gender of the drawn STEM professionals

In addition to descriptive statistical analysis, chi-square statistics were adopted to analyze whether boys and girls differed in terms of the gender of the scientists, engineers, and technologists they had drawn.

Analysis of drawings and writings regarding STEM professionals at work

To characterize students’ drawings and written descriptions of STEM professionals at work, two checklists were developed and used. The widely used Draw-A-Scientist Test Checklist (DAST-C) developed by Finson et al. (1995) was not used in this study because it is not designed for drawings of engineers or technologists. Moreover, DAST-C does not fully capture the features of students’ drawings of scientists, according to a systematic review by Ferguson and Lezotte (2020). Therefore, two checklists were newly developed and applied in this study, with one analyzing the visual presentations and the other analyzing the written descriptions. The drawing and writing are usually complementary, exhibiting different aspects of students’ perceptions. The drawings, in some cases, showed stereotypical features of the visual presentation of the STEM professionals at work, including the appearance, tools, and surroundings, while the writings described the practices of the STEM professionals at work. Interrater reliability (IRR) was calculated for the coding using the checklists.

Development of the checklist for the work of STEM professionals

To better summarize the students’ written descriptions of the STEM professionals’ work (i.e., what the STEM professional is doing and what the most important task in the STEM professional’s work is), a checklist for the STEM professionals’ work in students’ written descriptions of their drawings was developed. The researcher first reviewed hundreds of students’ responses and generated an initial list of items. The list was then simplified until the items in the checklist achieved the minimum number that still allowed for all the aspects the students mentioned to be labelled. Approximately 30% of the surveys were coded by the researcher and a trained coder. The IRR was calculated using the following formula:

$${\text{IRR}}\, = \,\left( {{1} - {\text{ differently coded items}}/{\text{total number of items coded}}} \right)*{1}00\%$$

The final IRR for each item of the 22 items in the checklists ranged from 85 to 100%, which showed that the coding process was trustworthy. Thus, the remaining data were coded only by the trained coder.

Development of the checklist for the drawings of STEM professionals

Another checklist was developed to extract features of students’ image presentation of the three STEM professionals at work. First, one researcher browsed 150 students’ drawings (approximately 450 pictures in total) and wrote down the features. Another coder did the same for another 150 students’ drawings. The features generated by the researcher and the coder were then compared, combined, and organized to form a checklist. The items (39 items in total) in the checklist could be grouped into appearance and clothing, emotions and thoughts, objects, and other. Then, each picture was coded by the coder with the guidance of the checklist.

To avoid subjectivity, approximately 27% of the drawings were coded by another coder who was trained by the researcher. The IRR for each item in this checklist ranged from 86 to 100%, and the remaining drawings were coded by the coder who was involved in the checklist development.

Categorizing the students’ drawings of STEM professionals

Students’ presentations (drawings and written descriptions) of STEM professionals were categorized to reveal the major categories of career perceptions. An initial review of the students’ drawings and related responses indicated that the students’ drawings could be categorized into several different categories. The researcher summarized the initial categories and started categorizing the drawings with reference to the students’ own descriptions of the drawings. The categorization process in turn helped to revise the categories. This iterative process continued until 30% of the data had been categorized by the researcher; all the categories were mutually exclusive and exhaustive, as suggested by Stemler (2001). This 30% of the data was also coded by a trained coder, and the IRR of the two coders was 87%. The coding process was then continued by the trained coder. An example of the data analysis using the two checklists and the categorizing process is listed in the appendix.

Results

Gender of the STEM professionals drawn by the students

The results (Table 1) show the percentage of males and females in students’ drawings of scientists, engineers, and technologists. In the boys’ and girls’ combined data, the ratio of male scientists to female scientists in all drawings of scientists is approximately 6:1. In addition, the ratio for male engineers to female engineers is approximately 8:1, and for technologists, the ratio is approximately 4:1. Among all valid responses, 63.5% of boys drew the scientists, engineers, and technologists all as males, while only 0.3% of boys drew all of them as females. For girls, only 0.9% of the girls drew them all as females, compared to approximately 38.7% of the girls who drew all three professionals as males.

Table 1 Summary of the gender of students’ drawn STEM professionals
Full size table

Chi-square tests of independence were performed to examine the relationship between students’ gender and their judgments of the gender of the STEM professionals that they had drawn. The relationship between these variables is significant for scientists’ gender, with Χ2 (2, 547) = 76.457, p < 0.001, Cramer's V = 0.374, and for engineers’ gender, Χ2 (2, 544) = 21.624, p < 0.001, Cramer's V = 0.199, indicating that the boys and girls differed in their opinions regarding the scientists’ and engineers’ gender. In other words, boys drew significantly more male scientists (93.5% and 68.1%, respectively) and male engineers (92.2% and 81.3%, respectively) than girls. On the other hand, there is almost no difference in the boys’ and girls’ depictions of the gender of the drawn technologists, with Χ2 (2, 521) = 2.021, p > 0.05, Cramer's V = 0.062.

Characterizing STEM professionals at work

The most common themes mentioned for the work of scientists (Fig. 1) are researching (39.9%), experimenting (34.4%) and inventing (21.1%) followed by biology/medicine (21.1%) and chemistry (8.5%), that is, the students mentioned that the drawn scientist was working in the biology/medicine or chemistry field.

Fig. 1
figure 1

Features with high occurrence (> 10.0%) in students’ written descriptions regarding scientists’ work

Full size image

In the context of engineers’ work (Fig. 2), the most mentioned theme is that of buildings (50.9%), showing that about half of the students mentioned that the engineer was working on buildings. The other themes included constructing buildings (33%), designing (including redesigning/ refining design/ examining design/ testing design) (25.4%), fixing (9.6%), supervising others/giving orders (9.2%), and other specific actions (7.1%), such as moving or mixing things.

Fig. 2
figure 2

Features with high occurrence (> 10.0%) in students’ written descriptions regarding engineers’ work

Full size image

For the work of technologists (Fig. 3), the students primarily indicated the theme of electric/electronic devices (28.0%), which means that students indicated that the technologist was doing work related to electric/electronic devices, followed by coding (16.3%), fixing (16.1%), researching (12.2%), and other specific actions (10.6%), such as moving or mixing things.

Fig. 3
figure 3

Features with high occurrence (> 10.0%) in students’ written descriptions regarding technologists’ work

Full size image

The features with occurrences greater than 10% are presented in Figs. 4, 5 and 6. The results show that the students tended to draw scientists as people working with test tubes/flasks (76.7%) or at a lab desk (61%) and often with smiling faces (40.8%), glasses/goggles (27.3%) or with an explosive hairstyle (11.3%). Among them, the presence of tubes/flasks (76.7%) or a lab desk (61%) depicted the scientists to be engaged in their laboratory work.

Fig. 4
figure 4

Features with high occurrence (> 10.0%) in the drawings of scientists

Full size image
Fig. 5
figure 5

Features with high occurrence (> 10.0%) in the drawings of engineers

Full size image
Fig. 6
figure 6

Features with high occurrence (> 10.0%) in the drawings of technologists

Full size image

The students drew engineers working with safety helmets (43.5%), buildings/scaffolds (35.3%), design drawings (23.6%), tools such as spanners/hammers (23.2%), cranes/derricks/ tractors/excavators (19.2%), building materials (14%), notebooks (not computers)/papers (12.2%) and pens (10.8%) (Fig. 5). It should be noted that 10.8% of the students drew engineers working with multiple people, despite the requirement to “draw an engineer at work,” compared to 0.9% of their presentations of scientists and 3% of their presentations of technologists that included multiple people.

For technologists, the students tended to draw them working with computers (52.2%) and/or with tools such as spanners/hammers (11.4%). The number of occurrences of smiling faces was similar in the students’ drawings of scientists (40.8%), engineers (40.6%), and technologists (40.4%).

Categories of career perceptions in students’ presentations of STEM professionals

After reviewing all the drawings and written descriptions, it was found that they could be characterized into the following 10 categories of careers: researchers, inventors, aerospace engineers, construction workers, building designers, construction site coordinators, programmers, and engineers that solve problems (Table 2). The seven major categories out of the 10 identified categories were shown in Fig. 7. For scientists, as shown in Fig. 7, the students tended to draw either a researcher (73.1%) or an inventor (23.8%).

Table 2 Definitions and occurrences of the categories of students’ presentations of STEM professionals (only occurrences > 1% are shown)
Full size table
Fig. 7
figure 7

Typical presentations of seven major categories of STEM professionals (percentages over 10% are shown; the “students’ description” below each drawing was written by the student regarding the drawing. The first description is the student’s response to the question regarding what the drawn professional is doing, and the second is the student’s description of what the most important task in the STEM professional’s work is.)

Full size image

As shown in Fig. 7, students also tended to draw four categories of engineers, with many drawing construction workers (category 3, constituting 43.3% of drawn engineers) or building designers (category 4, constituting 26.0% of drawn engineers). A smaller portion drew construction site coordinators that supervise workers/monitor projects (category 5, constituting 10.0% of the drawn engineers) or mechanical/electric/electronic technologists (category 6, covering 10.0% of the drawn engineers). The remaining drawings included engineers who solve problems (3.6%).

The students tended to draw three categories of technologists. The most common category is mechanic/electric/electronic technologists (category 6, constituting 38.5% of the drawn technologists), who are usually drawn with computers and in nature. The second category is programmers, whose work is coding, developing websites/ apps, or hacking (23.5%). In addition, a small number of students drew technologists as researchers (13.5%), which was the most common category for scientists, followed by inventors (8.4%) and clerks (5.3%).

Discussion

The findings extend previous conclusions by showing that both boys and girls exhibited gender stereotypes, not only for scientists and engineers, as studies involving DAST and DAET shows (e.g., Emvalotis & Koutsianou, 2018; Fralick et al., 2009), but also for technologists. In addition, the students tended to think of scientists as people working in laboratories doing research or inventing things, which is generally consistent with existing studies using DAST and DAST-C (e.g., Emvalotis & Koutsianou, 2018). Unlike in previous studies, the students had several presentations of engineers, mostly related to buildings. The study also supplemented previous research by showing that many students tended to think of technologists as people who work with computers (or other technological devices). The results indicated that students view scientists, engineers and technologists as different types of professionals while holding stereotypes and misperceptions regarding these careers.

Gender stereotypes of STEM professionals

This study supplemented previous literature by showing that, in addition to scientists and engineers, technologists were also more likely to be perceived as males. Compared to that for scientists (approximately 6:1) and technologists (approximately 4:1), the ratio of drawings of male engineers to female engineers was high (approximately 8:1), indicating that for students, engineers could be more gender stereotyped than scientists and technologists. The extremely low (less than 1%) percentage of both boys and girls who drew scientists, engineers, and technologists all as females reflects the difficulty of undoing gender role stereotyping among elementary students. It is thus assumed that elementary students may pervasively associate STEM careers with masculinity.

In contrast to a recent study performed by Emvalotis and Koutsianou (2018) in which most fourth- to sixth-grade female students from Greece drew female scientists, most fourth- to sixth-grade female students (68.1%) in this study still drew male scientists. The percentage of both boys and girls who drew male and female scientists is consistent with the results for 225 fourth- and fifth-grade students in South China in a study by Farland-Smith (2009). The similarities in percentages indicated that Chinese elementary students’ gender stereotypes of scientists may not have changed noticeably in the past decade. However, this conclusion needs further verification.

For the gender stereotypes in perceptions of engineers, this study indicates a higher level of gender stereotyping than was found in previous studies. The ratio of the male engineers to female engineers drawn in this study (approximately 8:1) is much higher than the ratio in a similar study involving seventh- to ninth-grade students (Liu & Chiang, 2019), which was close to 4:1. The ratio is also much higher than in Knight and Cunningham’s (2004) study on grades 3–12 in the United States.

The underlying mechanism explaining why upper elementary students hold gender stereotypes for scientists, engineers, and technologists is still unclear. It could be related to the stereotype that males are associated with “being smart”, as found by Bian et al. (2017), and thus are believed to be more suited to STEM careers, which demand intelligence. It could also be that students are exposed to more male STEM professionals in the mass media or in their real lives. This study indicates that efforts to eliminate gender stereotypes may start earlier and target STEM careers as a general category rather than focusing on scientists only.

Characteristics of STEM professionals at work

This study confirmed previous findings with DAST and DAST-C that most elementary students perceive scientists as researchers doing laboratory work surrounded by experimental instruments (Chambers, 1983; Emvalotis & Koutsianou, 2018; Losh et al., 2008) and some adopted less-common perceptions of scientists as inventors (Buldu, 2006).

In this study, engineers were mostly doing work related to constructing buildings, while in many previous studies involving DAET, engineers were believed to fix, repair and build things (e.g., Capobianco et al., 2011; Fralick et al., 2009; Karatas et al., 2011; Lachapelle et al., 2012). In this study, the most common features used to depict engineers’ work, namely the construction sites or tools related to construction (e.g., safety helmets), similar to findings from another study on Chinese high school students using DAET (Liu & Chiang, 2019), had much lower occurrences in most previous studies using DAET involving western samples (e.g., Capobianco et al., 2011; Lachapelle et al., 2012). Educators should be aware that students from different cultural backgrounds may hold different stereotypes of engineers.

Engineers drawn as scientists and technologists

In contrast to some previous findings with the DAET instrument that students lack perceptions of engineers (Fralick et al., 2009), this study adopts additional categorizing process and implies that many students had formed perceptions of engineering careers but did not see them as engineers. Among the seven major categories of careers presented in the students’ drawings, inventors, building designers and programmers were three actual engineering careers, constituting as much as 30% of all valid presentations. However, the students tended to perceive inventors as scientists and programmers as technologists rather than as engineers. In addition, apart from designing buildings, other engineering practices, including coding and inventing, were seldom mentioned as part of engineers’ work, showing detachment of these engineering careers (along with engineering practices) from the concept of engineers.

Because of students’ misperceptions of STEM professionals, it is reasonable to infer that some students’ attitudes, such as interest, are also “misplaced.” For example, the fact that one student was interested in being a scientist but not in being an engineer may be because he/she was interested in becoming a scientist who invents and would not like to be a worker at a construction site. If this student had informed perceptions of scientists’ and engineers’ work, he/she may had positive attitudes towards both careers.

The reason why inventors and programmers were not associated with engineers may be complex. Fundamentally, it is most likely that students do not have an adequate understanding of engineers’ work or engineering. Instead, they harbor a misunderstanding that engineers are involved in civil construction, which will be discussed in detail later. Second, in science courses, textbooks and popular science, inventors are often introduced as scientists only rather than as engineers. In addition, perhaps students perceive programmers as technologists because they perceive people who handle technological products as technologists.

Engineers are mostly limited to civil construction

Students in general tended to include various occupations related to building construction in their perceptions of engineers: building designers, construction workers, and construction site coordinators constituted approximately 79% of all presentations of engineers. Among the major categories of engineers that the students presented, only building designers are engineering professionals. Moreover, many major categories presented as engineers (construction workers, construction site coordinators and mechanic/electric/electronic technologists) are in fact technologists/technicians.

The reason why the students overwhelmingly believed that engineers are people doing work related to civic construction could well be related to the Chinese language. In Chinese, the word “engineering” has an alternative meaning, “project,” which often refers to civil engineering projects. Students may thus form the naïve understanding that people working on civil construction are engineers. The vast majority of students may integrate what they see in life related to civil construction into their perceptions of engineers and gradually form biased or stereotypical perceptions.

Limitations and future research

One limitation of this study is that the students’ drawings may not only reflect their “private” understanding of scientists, engineers, and technologists (Laubach et al., 2012; Palmer, 1997), but may have partly shown their understanding of other’s recognition of who a scientist, engineer, or technologist is, as Losh et al. (2008) pointed out. However, such limitation underneath many instruments derived from DAST also has its theoretical merit, because students may construct their STEM identities not only based upon their private perceptions of STEM careers but also how they think other people view these careers.

More types of careers can be covered in future research in addition to scientists, engineers, and technologists that are included in this study. Moreover, the external factors that may influence students’ perceptions of STEM professionals were not explored in this study. How students’ family background, including their parents’ occupations and socioeconomic status, and their school context influence students’ perceptions of STEM professionals and the association between students’ biased perceptions and their attitudes towards these careers are also worth exploring.

Implications

This study provides valuable information regarding the mechanism which explains why some young children and teenagers lose interest in STEM careers. For example, it can be reasonably hypothesized that, a student who is interested in engineering tasks such as designing and tinkering may think that being an engineer is not a desirable career because he/she not like the idea of engineers mostly doing work related to civil construction on a on a construction site. In addition, the gender stereotypes that associate STEM careers with masculinity, implicit as it may be, could still deter girls from pursuing STEM career paths.

This study can serve as a valuable reference for curriculum standards and textbooks at the primary level. There is a lack of content on engineers’ work and the implications of engineering for society in Hong Kong curriculum standards, which could be one of the reasons why students’ perceptions of engineers seem more biased than their perceptions of scientists and engineers. The findings of this study also suggest that the nature of engineers’ and technologists’ work should be introduced to students. If this is done, students’ naïve beliefs that engineers are mostly involved in civil construction and technologists are mostly involved in technologies may be transformed, giving students a deeper, more realistic, and more diversified understanding of their work.

The findings of this study could serve as a reference for educators and for curriculum developers when designing interventions or assessments in STEM education. STEM careers are often underrepresented in the mass media (Jennings et al., 2015) and in young children’s lives. Intervention is therefore needed to provide students with exposure to information regarding STEM careers (Grier & Johnston, 2012; Wyss, 2013), with special attention to the stereotypes revealed in this study, the misplaced career information, and the existence of gender stereotypes regarding STEM careers in general. The instruments and checklists developed in the study can also be used in future studies or in classroom assessments.