Factors Associated with Technology Learning and STEM Vocations in High School-The Case of Technovation Chile

Abstract

This study explores key factors associated with the development of technological thinking and preferences for STEM-related occupations among high school students in Chile, within the context of the Technovation program. We focus on three central indicators, which reflect on the main goals of the program: conceptual understanding of technology, systems thinking (defined as the ability to approach problems through logic and structured reasoning), and occupational preferences in STEM fields. Using pre- and post-program survey data, we assess the evolution of these indicators: while gender gaps persist in STEM career preferences, the program contributes to narrowing conceptual and systems thinking gaps. Also, our results indicate that students with stronger academic performance and higher problem-solving disposition tend to perform better in both technological dimensions, according to the pre-program survey data. The same factors, plus "evaluation of the Teamwork experience", play a key role in the improvement of most of these indicators, comparing the trajectories between initial and closing performance.

Headnote

Abstract

This study explores key factors associated with the development of technological thinking and preferences for STEM-related occupations among high school students in Chile, within the context of the Technovation program. We focus on three central indicators, which reflect on the main goals of the program: conceptual understanding of technology, systems thinking (defined as the ability to approach problems through logic and structured reasoning), and occupational preferences in STEM fields. Using pre- and post-program survey data, we assess the evolution of these indicators: while gender gaps persist in STEM career preferences, the program contributes to narrowing conceptual and systems thinking gaps. Also, our results indicate that students with stronger academic performance and higher problem-solving disposition tend to perform better in both technological dimensions, according to the pre-program survey data. The same factors, plus "evaluation of the Teamwork experience", play a key role in the improvement of most of these indicators, comparing the trajectories between initial and closing performance.

Keywords

Technology learning; STEM vocation; Occupational preferences; Attitudes towards problem solving; Teamwork experience; Education; High school; Technology

1. Introduction

Since 2018, the NGO Technology with a Woman's Name (Tecnologia con Nombre de Mujer in Spanish, abbreviated TecMujer from now on) has been implementing the 21st Century Digital Skills Program based on Technova-tion's curriculum. In 2023, we conducted evaluations in both formats: institutional settings (as part of the school's syllabus) and after-school workshops where the program is implemented. The main variables of interest for the program's goals are:

- Work interests towards STEM occupational areas.

- "Technological thinking," defined as skills associated with technological projects and technology, which considers two dimensions:

o Understanding of basic technology concepts, or "conceptual understanding".

o The problem-solving of logical challenges through "systems thinking" (through the use of sequence logic, hierarchies, and selection of relevant variables).

We conducted a survey at the beginning and the end of the program, which included these variables and also technology-related attitudes, and perceived self-efficacy across various dimensions.

Here we explore which factor might explain the difference in performance of our three Key variables, and which ones might explain their evolution (higher or lower performance on the exit survey).

2. Methodology

2.1 Technovation Program in Chile (Context)

The Technovation program aims to spark interest in science and technology, especially among girls and adolescents, in order to increase female representation in STEM fields and reduce gender gaps in this aspect. This is achieved through:

- Hands-on experience, where participants define and design a tech project that addresses a real-world problem.

- Team collaboration, with groups of 2 to 5 students who assume different roles and share ideas, interests, and responsibilities.

- Community engagement, involving various organizations and educational institutions in the program's implementation.

The program is active in over 100 countries. Each country develops its implementation and assessment process independently, adapting the program's guidelines to better align with local contexts and needs. In Chile, we employ a "Project-Based Learning" model through two formats:

(1) Educational Institutions (public and private schools, lyceums, as well as charter schools) or School Program. These institutions sign agreements with TecMujer on a voluntary basis. Its teachers are trained in the program's curriculum, which covers technology, problem definition, and project methodologies. Students participate as part of the school's syllabus (usually as part of the "technology" course). This format is known as the "21st Century Digital Skills Development Program".

(2) Interschool Program: Based on the Technovation curriculum, these workshops are carried out directly by TecMujer instructors, in partnership with higher education institutions that provide infrastructure (classrooms and devices) across Chile. Students voluntarily enroll, and the activities take place out of school hours, often on Saturdays (the semester workshop format) or during summer breaks (in a bootcamp format).

Each technology project is submitted through the Technovation platform, where teams are guided by a teacher or mentor who supports them throughout the process and ensures compliance with participation requirements. This enables them to enter the Technovation Challenge (or International Challenge1). Per Technovation's guidelines, only women may present projects at the international level (although most of the school's teams are all-gendered).

Under this scenario, the measurement that accompanies the implementation consists of a questionnaire on attitudes, perceptions, as well as technological competence questions, implemented at the beginning and at the end of the program.

1 A competition that invites teams of girls ages 8 to 18 from around the world to learn and develop skills to solve world problems through technology.

2.2 This Survey Use as a Standardized Tool for Measuring Technological Capability

This survey, as a measurement tool, uses multiple-choice questions on "technological thinking" as a way to estimate basic levels of competence in this area, and thus accounts for the learning process that may have occurred during the program. In this sense, it is worth noting that this assessment is similar to standardized tests using multiple-choice questions with a limited number of responses. This implies some disadvantages and limitations, which are offset by some adjustments in the implementation as well as broader considerations:

a) It is a limited way of measuring a complex mental process, such as the mastery of types of knowledge and their learning. In that sense, this measurement is a supplementary way of measuring learning, since other important aspects (of learning) are the workshop outcomes themselves, in terms of the number of participants who manage to finish and present their projects, and among them those who manage to qualify for the international challenge.

b) We know that training or workshop orientations are complex processes, in which many factors have an influence throughout students' learning career, and this is an effort to measure contributions in trend changes, and by measuring such contributions, we hope to present bifurcations on this path. In addition, there are other initiatives that the NGO is working on, aiming to identify impacts on former participants of the program, some months or years after its completion.

c) The main objections to standardized tests lie in their use as ranking tools and high-stakes decisions (such as university selection or resources for an institution)2. In this sense, these questions should be seen as a reference and not as an absolute parameter, whereby participants are tested on what they know or how they reason at that moment, in a way that is as brief as possible; we put an emphasis on making clear to students that their answers should be as natural as possible, without performance or ranking/comparison anxieties; in this aspect, no performance pressure is imposed on either students or institutions. Likewise, the questions avoid equivocal phrases that include options such as "all of the above" (which is a way of punishing more risk-averse people) and seek to incorporate a certain degree of reading comprehension for their resolution (Reardon, 2018).

d) The implementation of the same instrument (designed according to the specific learning process in accordance with the program) allowed us to have a significant number of comparable cases, so that we can make inferences about statistically significant patterns.

2.3 Population

As mentioned above, the program is implemented in two modalities: school program and interschool workshops; within the interschool workshops, participants are divided into those who participate in semester modalities (12 sessions, 4 to 5 hours each class, mainly Saturdays) and those who take part in intensive summer programs ("bootcamps") in which the 12 sessions are concentrated in 2 weeks. During the year, we have the following cases:

o People who enroll and drop out of the program, mainly in the interschool workshops (at educational institutions, there are only dropouts when there is extended absenteeism, which is less frequent).

o People who participate during the entire period but do not manage to finish developing a project.

o People who complete a project, yet do not manage to upload it onto the international challenge platform (they do not meet all the requirements).

In this sense, the definition of "achievement" for each modality is also differentiated:

o For participants of the interschool program, the first definition of achievement (and the main impact indicator) is to get as many participants as possible to the stage of uploading their projects onto the international platform.

o For participants of the school program, the definition of achievement is defined as completing all the content and activities (which are graded by the teacher, as part of a course) during the entire period.

So, the definition of the research population consists of:

a) Participants from the school program, all those who participated in the program for the entire period (even if they didn't upload projects) and passed the course, with all the workshop's activities finished.

b) Participants of the interschool workshops, who participated in semester workshops and managed to upload their

2 Some examples of the unintended consequences of using a standardized assessment for high-stake decisions include the incentiviza-tion of adjusting teaching processes for testing purposes rather than for learning; the diversion of focus and time away from knowledge transmission toward test-taking strategies; and the presence of implicit gender biases in frequent standardized test implementations- for instance, discounting incorrect answers from the final score can discourage more risk-averse individuals, often women, from attempting all questions (Berwick, 2019).

projects onto the platform (which is the achievement criterion for this modality). Participants of the bootcamps were not included because the deadlines and dates of their development were not compatible with the logistics of the data collection.

Given program goals, the gender dimension is central to the analysis and review of the trends in all the parameters and variables measured in these studies. However, it should also be considered that the two formats in which the program is implemented (school and interschool programs) present significant differences in the composition of participants and the dynamics they show during the workshops: in the case of the interschool program, they are volunteers, who dedicate their weekends and free time to the development of activities; as opposed to the people from educational institutions, who do not usually have the same degree of voluntariness and for whom the workshop implies a grade that affects their school grades.

For all the above reasons, the variable "segment" is created, which in the case of institutions differentiates self-identification by gender, while in the interschool program, all identities are grouped together, considering that by definition of this program, only people with a female sex (assigned at birth) can enroll.

In Table 3, we see that the participants' gender distribution in educational institutions is balanced, and in Table 4 (which shows data on respondents), we also see that a similar female/male proportion (similar response rate).

2.4 Variables Related to Participation in the Survey

The recording of variables from the initial survey allows us to explore the profiles associated with the higher or lower participation rate in the closing measurement/survey in greater depth, thus identifying possible biases in the analysis of the results of the respondents who participated in both measurements (compared to those who participated only in the first measurement).

We conducted an analysis of the variables related to participation in the closing survey.

As can be seen in Table 5, this data analysis shows that there was a significant difference in participation based on (in order of importance):

o The previous year's grade point average (note that in Chile, grades go from 1 to 7, being grades lower than 4 a failing grade, grades between 4 and 4.9 are "barely passable", grades between 5 and 5.9 are "good", and grades above 6 are "very good").

o Performance conceptual understanding, as the average percentage of questions answered correctly.

o Performance in systems thinking, as the average percentage of questions answered correctly.

We can observe a correlation between the closing measurement participation rate and the average grades of the previous year, in which grades lower than 5 had a 41% participation rate and grades higher than 6 had a 59% participation rate in the closing survey. In other words, there was a consistent profile of students who were discouraged from continuing to participate in the measurements.

However, the fact that the level of technological thinking (both in conceptual understanding and systems thinking dimensions) is related to participation may imply that the impacts detected (an increase between initial and closing values) in these variables are being underestimated, since those who respond both times have less cases with greater potential for improvement and, on the other hand, increases the average of these variables in the initial measurement.

For the school program participants, we can observe the following:

- Conceptual understanding: For female participants of the schools' program, we see an increase of 6% (percentage of correct answers), closing an initial 7% gap between them and their male classmates, with a gender gap of 1% in the closing survey (at the end of the workshop). Non-binary/other gender also closes their initial gap with their male classmates, from an initial 3% gap to a 0% gap at the end of the workshop.

- Systems thinking: In the initial survey, we see no major differences in the performance of all genders, and even progress at the end of the workshop, between 4% and 5% for all segments.

- STEM-related occupational preference: In this indicator, we can see a significant gender gap between males and females (11% gap) and other genders (15%) classmates. The evolution of this indicator does not show a reduction of this gap, only a small increase for all school program participants.

4 The closing participation rate considers the number of respondents in the closure measurement divided by the number of respondents in the initial measurement. In 2023, this rate was 51%.

For the Interschool program participants, we can see a higher initial level on all indicators, especially on conceptual understanding, with an average of 72% of the answers correct. This implies that there is less chance of improving an already high level of this indicator. For systems thinking, we can see a similar 5% improvement in the closing survey. Finally, for this format, we can see the greatest increase in STEM occupational preferences, with a 7% leap from the initial 48%.

3. Analysis of Variables Correlated with the Impact Indicators

We conducted an exploratory analysis of the 2023 data that allowed us to make some hypotheses on how to achieve greater impact on the variables of interest. The exploration sought to identify:

- The variables most correlated with the indicators of interest (particularly in the initial measurements).

- The variables most correlated with the variation of these indicators between the initial and closure measurements. Two variables were not explored because some of their categories had few cases:

- The variable "region" was not shown, given the uneven distribution of the program at the national level, considering that in some regions there was only one place where the program was carried out, which contrasts with other regions with a large number of schools and interschool programs.

- The exploration of correlations by the school's type of administration (dependency- ownership, ration and funding) was not included in all crosstab analysis since, analogously to what happens with regions, some categories (such as municipal departments of education or municipal corporations) contain one or two schools, so the correlations could refer more to the behavior of a particular school than to the relevance of the variable itself.

3.1 Variables Correlated with Occupational Preferences in the STEM Area

To review this phenomenon, we analyzed both program formats separately since, as we have shown, they presented different dynamics due to their composition, the implementation process, and the initial levels of motivation and technological thinking (conceptual understanding and systems thinking).

3.1.1 School Program

We should keep in mind that gender was a main factor correlated with a career choice in the STEM area, but beyond this variable, Table 7 shows other significant correlations.

- The correlation observed in 2022 between the previous year's grade point average and the preference for STEM are-as (a higher grade would imply a greater disposition towards that area) was ratified; however, while in 2022, people with lower averages lost motivation for the STEM area, their number increased slightly in the closure measurement in 2023.

- The type of administration also had an impact on the initial level of motivation, highlighting delegated administration institutions, which showed the lowest preference for the STEM area (31%) and at the same time represented 50% of the respondents of this segment (14 educational institutions), being mostly technical/vocational schools. It is worth comparing this with private charter schools, since the latter category includes 9 schools (3 of which have a technical/professional orientation) and represents 26% of the respondents, with a 39% preference for STEM careers. The other categories that showed higher percentages in this parameter included only 1 or 2 schools for each category, so the contrast might have corresponded more to the particularities of these schools than to some common structural factors.

- Disposition towards problem-solving (or to solve problems) is one of the variables that is most correlated with initial inclination toward STEM (13% difference in STEM orientation between low and high disposition), and at the same time is associated with increased preferences in the closure measurement.

- The level of conceptual understanding had a significant impact on the initial disposition towards the STEM area (12% difference), but did not seem to have much impact on the evolution of this item.

- The level of systems thinking also affects the initial disposition towards STEM areas (6% difference between the low and high levels), and the high level of systems thinking was also associated with an increase in this preference (by 4% in the closure measurement).

3.1.2 Interschool Program

In the interschool program, which overall had a 7% increase in inclination toward STEM areas with respect to schools, we can observe the following patterns in Table 8:

- The variable with the strongest association with preference for STEM careers was the disposition towards problem-solving, with a 31% difference between those with low and high dispositions. However, this variable did not make much difference in the evolution with the closure measurement (all levels of disposition increased similarly by between 6% and 8%).

- The variable "average grade from the previous year", on the other hand, had a 26% difference in preferences for STEM occupations between those who had averages below 5 and those who had an average of 6 or higher. At the same time, those with an average grade of 6 increased their preference for STEM from 51% to 60% in the closure measurement, while those with lower average grades did not increase (or even decreased) their preference for STEM in the closure measurement.

- The level of the interschool program (Junior or Senior) also showed an initial difference in this dimension, with seniors having 15% more inclination towards STEM areas. However, junior participants showed a more significant increase in the closure measurement (12%), closing the gap with the senior level by 10 points.

- Table 8. also shows that those who had a lower level of conceptual understanding showed a greater initial tendency towards STEM areas, which is curious, since, in the case of the school program this correlation goes in the opposite direction (keep in mind that in that case a higher level of conceptual understanding is associated with a greater willingness to pursue a STEM occupation, as shown in Table 7). - The correlation between the level of systems thinking and inclination towards STEM careers is linear, showing a 20% difference in this preference between those with low and high levels of this variable.

3.2 Variables Correlated with Conceptual Understanding

In an exploratory analysis, we identified the variables that correlated most strongly with the level of conceptual understanding in technology issues and their evolution, broken down by type of program.

3.2.1 School Program

In the program implemented at schools, we identified some variables related to the performance and evolution of conceptual understanding:

- The average grade from the previous year was associated with a 15% variation in the performance of conceptual understanding (47% for those with an average of less than 5 and 62% for those with an average of 6 or higher). This variable was also associated with a greater increase in this index, with those with lower averages improving more in the closure measurement (4%) than those with higher averages (1% increase), i.e., an initial gap based on previous academic performance was narrowed.

- Disposition towards problem-solving also experienced a 15% difference between those with a low disposition (53% in conceptual understanding score) and those with a high disposition (68% in conceptual understanding score).

- Conceptual understanding was also closely related to the other aspect of technological thinking, "systems thinking", i.e., there are aspects that complement each other, with a 13% difference between those who had a low level of systems thinking (51% in conceptual understanding score) and a high level (64%).

- Finally, it can be observed that one of the emotions (sadness) had a negative correlation with conceptual understanding, varying by 8% between those who reported a feeling of sadness (51%) and those who did not have this emotion (59%) in the initial measurement. This gap continued to exist in the closure measurement. In another section, we will check whether any other emotion had a significant impact on any of the key indicators.

3.2.2 Interschool Programs

The same variables identified in the program of educational institutions also had a significant degree of correlation with the understanding of technology concepts in the interschool program. However, we cannot lose sight of the fact that in this format there was no relevant increase in the closure measurement in 2023, considering that in the initial measurement the level of conceptual understanding was already significantly high (73%), so there was not so much room for improvement and a risk of worsening over time. It is for this reason that the analysis of the evolution of parameters is focused on factors associated with "performance decline."

- The variable most related to conceptual understanding was the average grade from the previous year, with a 10% variation between those with an average grade below 5 (63%) and those with an average grade above 6 (73%). Likewise, those with a lower average grade had a worse performance in this variable in the closure measurement, dropping 13% in that measurement, while those with an average of 5 to 5.9 decreased by 5% in this parameter. In other words, those with lower academic performance would not have as consolidated "systems thinking," which implies that their performance would be less consistent.

- Systems thinking also had a relevant relationship with conceptual understanding, with a variation of 9% between those with a low level (68% of conceptual understanding score) and a high level (77%). Those with a low level showed a 4% drop in the closure measurement.

- The disposition towards problem-solving in the initial measurement showed an unexpected correlation: those with low disposition had 5% more in conceptual understanding score (78%) than those with high disposition (73%). However, the most important difference occurred in evolution since the number of those with low disposition decreased by 14% in conceptual understanding, while the number of those with high disposition increased by 2%.

- Finally, the correlation of the emotion "sadness" was reiterated: those who did not have it showed a 9% better understanding of concepts than those who had it (in the initial measurement).

3.3 Variables Correlated with Systems Thinking

Unlike the exploratory analysis carried out on factors associated with STEM career preferences and level of conceptual understanding, both program modalities (educational institutions and interschool programs) shared several correlation factors. However, in this case, we observed that the variables associated with the greatest differences were distinct.

3.3.1 School Program

- The variable that correlated the most with the level of systems thinking was the level of concepts, showing a 12% variation between low and high levels of conceptual understanding.

- Disposition towards problem-solving is also relevant, with a 10% difference between those with low disposition and those with high disposition.

- Academic performance was associated with an 8% variance in the average number of correct systems thinking responses.

- The presence of optimism (in the initial measurement) was also associated with an 8% average higher performance.

3.3.2 Interschool Programs

- In the interschool program, the variable most associated with systems thinking performance was the level of conceptual understanding, with a difference of 11% between those with low and high levels of conceptual understanding.

- Preference for STEM careers was also associated with a higher level of systems thinking, with an average of 64% of correct responses from those who considered STEM careers, in contrast to 56% of those who did not. On the other hand, the number of those who did not consider a STEM career increased their performance in the closure measurement by 7% (in contrast to those who did choose a STEM career and who increased their performance in the closure measurement).

- The previous year's average grade was associated with only a 5% difference between the lowest and highest averages. However, it was one of the variables most associated with the increase in systems thinking performance, observing that those with an average of 6 or higher increased by an average of 6%, while those with an average below 5 did not improve in the closure measurement.

- Finally, there were almost no differences in the level of systems thinking according to the level of the interschool workshop (Junior or Senior), but it was observed that junior participants improved by 7% in the closure measurement, in contrast to the senior participants' improvement, which increased by 4%.

4. Conclusions

The following are the main conclusions on impacts and factors associated with them:

o In terms of the gender gap, considering the evolution in schools (of the School Program), there is evidence that the technological thinking gap is narrowing in Conceptual Understanding, and with consistent advances in systems thinking. In terms of the gap in STEM occupations, a more modest progress is achieved.

o The interschool program in general registered higher initial levels of performance in the technological thinking dimensions as well as greater motivation towards STEM careers (in addition to a significant advance in this aspect towards the closure measurement, with 7%).

o Regarding the disposition towards future STEM careers in the School Program, we see that female and male segments of schools registered an average of 1% increase, maintaining a gender gap of 11% in 2023.

o The conceptual understanding level reached in the interschool program (72%) makes it difficult to observe improvements in the closure measurement. On the other hand, the schools showed progress, with greater progress being observed in 2023 in the female segment (narrowing the initial gender gap in this aspect).

o In systems thinking, significant progress was observed in both programs. The gender gap in the closing measurement of this parameter was only 2%.

- It is also noteworthy that some values of the "emotional state" of the initial measurement (in particular, optimism and sadness) would be associated with the level and evolution of systems thinking and conceptual understanding, respectively.

- Regarding technological thinking, the data indicated that there would be a complementarity between conceptual understanding and systems thinking since a good foundation in one dimension favored the performance and improvement of the other.

- The incorporation of the variable "disposition towards problem-solving" seems to be relevant, because it appeared as a relevant factor associated both with the initial difference of the indicators and with the more favorable evolution of the inclination towards STEM areas of conceptual understanding and systems thinking. Likewise, this indicator was also associated with the previous year's academic performance, which was also relevant when analyzing orientation to the STEM area and systems thinking.

- In this sense, this correlation makes us notice that motivating inclination towards STEM areas is a particular challenge, as it seeks to attract adolescents towards an activity that requires patience and a certain degree of frustration tolerance, since addressing motivation towards occupations in the STEM area does not seem to have an immediate solution, especially when it comes to such aspects as self-image, self-confidence, attitude toward risks, academic performance, stereotypes / social imaginary, feedback from teachers and family members shaped by cultural factors, etc. that have shaped preferences for years before starting the workshop.

- Regarding the learning process, a keynote should be taken of what was revealed in the qualitative stage and the figures of this study on the importance of teamwork experience. There are quantitative indicators that show that this experience can have an impact on key dimensions such as conceptual understanding or the disposition towards problem-solving (which, as we have already indicated, is relevant to explain several aspects of the difference in initial level and the evolution of key indicators).

o As a complement to the analyses carried out and to illustrate this point, Table 13 shows an association between teamwork experience and the evolution of technological concepts, where a good experience is associated with an increase in their performance in this parameter.

- In the case of Table 14, an association was observed between the teamwork experience and the evolution of the disposition to solve problems, which showed that a bad experience would be associated with a significant decrease in the score of this dimension (it would be a hygienic or "discouraging" factor of these attitudes). In this sense, this correlation should be explored since the management of a better teamwork experience (of which some interviews provide guidance) can favor the achievement of impacts, both in STEM and in technological thinking.

- To schematize the observations about the pathway to move program participants toward STEM areas, the data seems to indicate that:

o Obtaining technological thinking tools, in both dimensions, would make the inclination toward STEM areas more likely. In that sense, there are indications that these aspects can be, to some extent, "leveled" by the program if it starts below a certain threshold. This is in line with studies on motivations to study STEM specializations (Wang, 2017), in which performance in the associated fields is one of the determinants of the choice to pursue a career or major in STEM.

o Motivation towards STEM areas - at the time workshops are implemented- already carries a backpack of previous factors, among which is the decision the type of school where the students are enrolled - which has implications on a future occupational orientation - and previous academic performance, which is associated with possibilities of academic and occupational orientations as well as a certain level of confidence in one's own abilities.

o Teamwork experience is a relevant factor to consider at the planning stage, since aspects of coexistence, emotional management, and perceptions about the distribution of workload often affect motivation in relevant points such as disposition to solve problems. In contrast to what is usually the experience of group work in schools, whereby several specific projects are delivered with different groups during the year, the Digital Skills and Technovation Program considers the development of the same project with the same team throughout the year. The qualitative survey provides some clues in this regard: that the establishment of teams is voluntary but guided, that the evaluation (at educational institutions where a grade is given to the course) incorporates a group grade and a personal grade (based on whether a participant can respond to own and teammates' components), the generation of instances and tools for the resolution of conflicts or disagreements.

- Although the correlations are not entirely consistent in this aspect, it is worth noting the emotional aspects that may be involved in the development of the program, particularly at the stage in which the workshop takes place (in adolescence, self-image aspects undergo a period of intense redefinition and the so-called "math anxiety"). These emotions can favor (Usan, 2022) or hinder the performance or acquisition of these skills (Abi, 2020). From a program perspective, an implementation design that successively delivers skills (technological thinking) and tests them while providing teamwork experience that supports and fosters a willingness to face certain types of challenges (requiring patience, trying various paths) can lead participants to consider STEM career options.

References

References

Abi, A., Garcia, J., & Lopez, M. (2020). Predicting mathematics achievement in secondary education: The role of cognitive, motivational, and emotional variables. Frontiers in Psychology, 11, Article 528. https://doi.org/10.3389/fpsyg.2020.00528

Berwick, C. (2019, October 25). What does the research say about testing? Edutopia. https://www.edutopia.org/article/what-does-research-say-about-testing

Brockhoff, S., & Weber, M. (2015). Towards empirically measuring patience. UniversalJournal of Management, 5(12), 501-508. https://doi.org/10.13189/ujm.2015.031207

Chin, C, & Brown, D. (2000). Learning in science: A comparison of deep and surface approaches. Journal of Research in Science Teaching, 37(2), 109-138. https://doi.org/10.1002/(SICI)1098-2736(200002)37:2<109:AID-TEA3>3.0.CO;2-7

Demirel, M., Demirel, D., & Keles, E. (2015). A study on the relationship between reflective thinking skills towards problem solving and attitudes towards mathematics. In the 7th World Conference on Educational Sciences (WCES-2015), 05-07 February 2015, Athens, Greece.

Erdemir, N. (2009). Determining students' attitude towards physics through problem-solving strategy. Asia-Pacific Forum on Science Learning and Teaching, 10(2). https://www.eduhk.hk/apfslt/vl0_issue2/erdemir/

Granato, S. (2023). Early influences and the choice of college major: Can policies reduce the gender gap in scientific curricula (STEM)? Journal of Policy Modeling, 45(3), 494-521. https://doi.Org/10.1016/j.jpolmod.2023.03.002

Granese, L., Herrera, C, & Jimenez, R. (2023). Effects of the pandemic on the socioemotional well-being of children and adolescents in Chile and the world. Centro de Estudios Publicos (CEP), Chile, https://www.cepchile.cl/

Ibrahim, N., Rahman, M., & Halim, M. (2023). Post-pandemic blues: Collaborative learning and communicative competence in English as a second language (ESL). International Journal of Academic Research in Business and Social Sciences, 13(4), 89-104. https://doi.org/10.6007/IJARBSS/vl3-i4/16828

Mohd, A., Noraini, I., & Mohamed, S. (2011). The effects of attitude towards problem solving in mathematics achievements. Australian Journal of Basic and Applied Sciences, 5(12), 1857-1864.

Prince, M. J., & Felder, R. M. (2006). The many faces of inductive teaching and learning. Journal of College Science Teaching, 35(5), 14-20.

Reardon, S. F., Fahle, E. M., Kalogrides, D., Podolsky, A., & Zarate, R. (2018). The relationship between test item format and gender achievement gaps on math and ELA tests in fourth and eighth grades. Educational Researcher, 47(5), 284-294. https://doi.org/10.3102/0013189X18762105

Usan, A., Salavera, C, & Teruel, P. (2022). Self-efficacy, optimism, and academic performance as psychoeducational variables: Mediation approach in students. Children, 9(3), 420. https://doi.org/10.3390/children9030420

Wang, X. (2013). Why students choose STEM majors: Motivation, high school learning, and postsecondary context of support. American Educational Research Journal, 50(5), 1081-1121. https://doi.org/10.3102/0002831213488622

Wickstrom, A., & Zeiler, K. (2021). The performativity of surveys: Teenagers' meaning-making of the 'Health Behavior in School-aged Children Survey' in Sweden. Children & Society, 35(3), 428-444. https://doi.Org/10.l 111/chso. 12416

World Economic Forum. (2020, October). These are the top 10 job skills of tomorrow - and how long it takes to learn them. https://www.weforum.org/agenda/2020/10/top-10-work-skills-of-tomorrow-how-long-it-takes-to-learn-them/

Zakaria, E., Chin, C. L., & Daud, Y. (2004). The reliability and construct validity of scores on the attitudes toward problem solving scale. Journal of Science and Mathematics Education in Southeast Asia, 27(2), 123-139.