Sage Journals: Discover world-class research

Abstract

This study examined characteristics of 458 Grade 10 students who earned a Bachelor of Science degree in computer science while completing their secondary school studies. Cluster analysis of SAT-M and Raven Progressive Matrices Test scores identified four distinct groups: G-EM: generally gifted, excelling in mathematics; AG-AEM: averagely gifted, average mathematical excellence; AG-NEM: averagely gifted, non-excelling in mathematics; NG-AEM: not generally gifted, average mathematical excellence. The G-EM group demonstrated significantly higher general and mathematical creativity compared to all other groups. Post-hoc analysis revealed cluster-dependent and task-dependent correlations between measured criteria. Results suggest creativity levels could serve as an effective criterion for program enrollment decisions.

Keywords

general giftedness mathematical competencies STEM computer science general creativity mathematical creativity

Background

Although the importance of STEM professions is growing exponentially, student interest in STEM professions remains less than optimal (Ardianto et al., 2023; Gallagher, 2000; Wai, 2015; Wai et al., 2010). Educational opportunities in these areas are essential not only for a prosperous society, but also for helping children transform their academic talents into professional achievements (Warne et al., 2019). Despite awareness of the personal and societal value of STEM professions, many students choose this career pathway late in life (Constan & Spicer, 2015; Miller et al., 2007; Sadler et al., 2012). Early high achievement in mathematics and science supports later success, and participation in high-quality STEM courses can strengthen students’ inclination to pursue these directions for their future (Young et al., 2017). Lichtenberger and George-Jackson (2013) noted that academic qualification and preparation for STEM are associated with the mathematics and science courses students take, as well as the grades they receive in these courses. This study addresses the initial phase of longitudinal research designed to track students throughout their academic journey, aiming to identify the characteristics that most effectively predict success in STEM education.

Educational programs especially designed for gifted students aim to provide opportunities that are not typically available in traditional classrooms (Young et al., 2017). University programs and courses for high school students designed for talented youth offer new challenges and support them in fulfilling their intellectual potential (Leikin, 2009a). Furthermore, these programs cultivate human assets to meet the needs of society and make higher education more accessible to a diverse range of gifted and talented students, regardless of their socioeconomic status (Weisblay, 2018). Engagement in out-of-school programs may play a critical role in helping students realize their intellectual potential (Greene et al., 2013). These programs are increasingly recognized as effective tools for promoting positive youth development (Eccles & Templeton, 2002; Goerge et al., 2007; Heath et al., 2022).

Advanced Placement (AP) courses are a well-established format that provide students with the opportunity to begin college-level academic studies while still in high school. The AP program offers college-level coursework to high school students, easing their transition to higher education (Dougherty et al., 2006; Hoepner, 2010; Warne et al., 2019). Research shows that students who take AP courses tend to perform better in both high school and college, particularly in the subjects in which they completed AP coursework (Ackerman et al., 2013). These college-level courses both ease the transition to postsecondary education and contribute to the development of human capital (Conger et al., 2021). Success in AP courses is associated with improved academic outcomes (Mo et al., 2011) and serves as an indicator of college-level readiness prior to high school graduation. Moreover, students who participated at the highest levels of AP coursework demonstrated lower failure rates in core university subjects, particularly in mathematics and science (Goerge et al., 2007). Participation in AP courses can serve to motivate students to pursue STEM careers by offering a college-level academic experience during high school (Warne et al., 2019). Research suggests that participation in AP STEM courses is associated with selecting a STEM major in college as well as career STEM achievement (Ackerman et al., 2013; Wai et al., 2010).

Motivation is a key predictor of academic performance (Dogan, 2015) and long-term academic achievement (Kori et al., 2016). Hence, to maximize educational success, young people must be motivated and willing to work hard (Jerrim et al., 2020). Motivation is often a result of strong implicit beliefs about one’s own abilities (Ommundsen et al., 2005; Wang & Eccles, 2013). Implicit beliefs about ability are defined as individuals’ perceptions of the differences they perceive between ability and effort; if an ability is seen as stable, it cannot be changed by effort, yet, when the ability has been acquired through training, then it can be improved by making greater effort (Li & Lee, 2004).

Adolescent motivation plays a critical role in shaping educational pathways and influencing the decision to pursue a STEM career (e.g., Musu-Gillette et al., 2015; Wang & Eccles, 2013). Therefore, increasing student participation in STEM fields requires intentional efforts to foster and sustain motivation (Rosenzweig & Wigfield, 2016). During adolescence motivation tends to decline more sharply in mathematics and science than in other academic domains (Wigfield et al., 2015). Research indicates that students’ educational environments can positively influence their motivational beliefs as well as their attitudes towards science (Fortus & Vedder-Weiss, 2014; Vedder-Weiss & Fortus, 2011). One effective way to enhance student motivation is through specially designed curricula that promote active engagement in the learning process (Chittum et al., 2017; Cutucache et al., 2016). Participation in an afterschool STEM program was associated with strengthened motivational beliefs and greater resilience against declines of motivation (Chittum et al., 2017).

It is widely believed that students who participate in AP and other out-of-school programs have high intellectual potential. Intellectual potential (Leikin, 2009a) comprises four elements: cognitive abilities (e.g., intelligence, domain-general and domain-specific skills, general and domain-specific creativity), affective characteristics (motivation, attitudes, beliefs, emotional responses), personality traits, and learning opportunities. The complex construct of intellectual potential framed our study design.

Research consistently highlights that student motivation is a key factor that influences academic success (Chowdhury et al., 2013). Highly motivated students tend to engage more actively in their learning and achieve higher academic outcomes (Chittum et al., 2017). Whereas previous studies have shown that STEM programs enhance engagement by fostering interest and motivation (Chittum et al., 2017; Ugras, 2018), the present study investigates the relationships between cognitive characteristics (intelligence and creativity) and student motivation.

Creativity is widely regarded as one of the most essential skills for success in the 21st century (Henriksen et al., 2019). Creativity involves generating ideas that are both novel and useful, and it is widely considered an intrinsic component of human intelligence (Sternberg & Lubart, 2000). Creativity is a critical asset across nearly all modern fields and is defined as a complex cognitive process that results in the production of novel ideas (Simonton, 2000). It is a particularly important skill for STEM education (Ugras, 2018). Students who can apply their knowledge creatively are better equipped to solve complex, real-world problems (Leikin, 2018). Several studies over the past decade have demonstrated that mathematical creativity is closely linked to mathematical achievement (Haavold, 2020; Leikin & Lev, 2013; Schoevers et al., 2020). In many of these studies, higher levels of mathematical achievement were positively associated with greater mathematical creativity. To examine mathematical creativity, most of these studies employed Multiple Solution Tasks (MSTs; Leikin, 2009b), with the mathematical problems used by researchers differing in terms of topics, levels, and skills required from study participants.

The Study

Research Goal

The goal of the current study was to examine the characteristics of high school students who are motivated to pursue a Bachelor of Science (BSc) degree in computer science. Admission cut-off criteria for the program were determined by program leaders and based on assessments of intelligence and mathematical competence. The study investigated general and mathematical creativity, mathematical competence, general intelligence, and motivation among students before entering the BSc computer science program. Cluster analysis was conducted to explore differences in motivation and creativity among participants categorized by the existing cut-off criteria. We asked the following research questions: How do groups with different combinations of general intelligence and mathematical competencies differ in motivation and creativity? What correlations can be found between the cut-off criteria, motivation, and creativity components?

The Context

The Challenge Program is a 4-year program that offers gifted adolescents the unique opportunity to earn a BSc degree concurrently with their high school studies. This is a highly competitive program that offers challenging academic learning experiences for students seeking studies beyond the standard high school curriculum. Its goal is to fully cultivate students’ intellectual potential, allowing them to earn a prestigious degree in computer science at an early age and providing a strong foundation for future career success.

Study Participants

The participants comprised a group (N = 458) of high school students who applied to take the program’s admission test during the 2019–2020 school year. The participants (average age = 16 years) included students from both the Jewish and Arab sectors. All students were enrolled in high-level mathematics courses, with performance scores placing them within the top 1% of their age group, demonstrating exceptional achievement in school mathematics. As part of the application process, students were required to submit recommendations from their schools. The entrance examination was administered in Hebrew, per university admission requirements. Table 1 presents demographic information for the study’s participants.

Table 1.

Study Participants

	Male	Female	Total
Jewish sector	137	52	189
Arabic sector	147	122	269
Total	284	174	458

Tools and Data Collection

This study used data from admission tests administered following the initial screening phase in the application process. The following measures were used to address the goals of the current study.

The Raven Progressive Matrices Test (RPMT) was used to evaluate the level of students’ general intelligence (Raven et al., 2000). A shortened version of the test, adapted and validated by Zohar (1990), was used. This version consists of 30 items allotted with a 15-minute time limit. Each item included 3 × 3 matrices with a missing image, and participants were required to select the correct image from 6 to 8 alternatives. Successfully solving this test involves identifying the underlying rules that explain the progression of shapes. Black items are presented on a white background, with difficulty increasing throughout the test. The scores range from 0 to 30, representing the number of correct responses.

The math section on the SAT (SAT-M) was used to evaluate mathematical competencies. This standardized test is widely used for college admissions in the United States. A shortened version developed and validated by Zohar (1990), was used in this study. Participants were asked to answer 35 mathematical multiple-choice questions within a 30-minute time limit. The test contains questions from various mathematical topics and is designed as a multi-component assessment. Scores range from 0 to 35, representing the number of correct responses.

Multiple Solution Tasks were used to assess both general and mathematical creativity. A scoring model and scoring scheme validated by R. Leikin (2009b, 2013), was applied. The tasks explicitly required participants to generate multiple solution strategies. Although the model was initially developed to assess mathematical creativity, research demonstrates its utility for evaluating general creativity as well (R. Leikin, 2013). Each solution strategy was scored across four domains: fluency, flexibility, originality, and creativity. Fluency was measured by the number of appropriate solution strategies generated. Flexibility reflected the variety of strategies used and the participant’s ability to “be different from themselves.” Originality of the solution strategies was based on the rarity of a solution strategies within the sample, capturing the participant’s ability to “be different from others.” The Creativity score for each solution was calculated by multiplying its flexibility and originality ratings. The full model and the scoring scheme are presented in the Appendix.

The Pictorial MST was used to assess general creativity (Leikin, 2009b, 2013, M. Leikin, 2013). In this test, participants are presented with a black-and-white image depicting a small, child-like raccoon attempting to retrieve its hat from a high hanger but unable to do so (see Figure 1). Participants were instructed to “help the raccoon to get its hat in as many ways as possible.” Students were given 10 minutes to complete the test.

Figure 1.

MST-1 Pictorial MST.

The Mathematical MST Test was used to assess students’ mathematical creativity (Leikin, 2009b, 2013; see Figure 2). The test included four problems: an arithmetic calculation, a system of equations, a word problem, and a geometry problem. These problems, seemingly simple for high-achieving mathematics students, were made challenging by requiring participants to create as many different solution strategies as possible. Students were given 40 minutes to complete all four problems. The tasks were adapted from R. Leikin (2013) and Leikin and Guberman (2023).

Figure 2.

Mathematical Problems for the Evaluation of Mathematical Creativity.

The Learning Motivations Questionnaire was used to assess the students’ motivation. The questionnaire was adapted from Midgley et al. (2000). This 20-item questionnaire uses a five-point Likert-type scale ranging from 1 (Not at all true) to 5 (Very true). It evaluates five motivational components: (a) enjoyment of learning (e.g., “I am intrigued by learning new educational material”), (b) overcoming challenges (e.g., “I despair of learning when the material is too difficult”), (c) learning initiative (e.g., “I read educational material out of personal desire”), (d) independence (e.g., “When I don’t know the answer to a question, I copy from others”), and (e) learning goals for future success (e.g., “Studies have no impact on my future”). Responses were averaged to compute scores of each scale, and a reliability analysis was conducted. Table 2 presents the internal consistency of each scale, with Cronbach’s α values exceeding 0.60, indicating satisfactory reliability.

Table 2.

Cronbach’s Alpha Distribution for the Learning Motivations Questionnaire

Scale	Cronbach’s α
Enjoyment from learning	0.626
Overcoming challenge	0.623
Learning initiative	0.836
Learning goals for future success	0.773

Data Analysis

A cluster analysis was conducted to identify common participant profiles based on general giftedness and mathematical competencies. The cluster analysis was performed using z-scores from the SAT-M and RPMT. Silhouette values, which range from −1 to +1, were used to evaluate cluster quality. Higher values indicate stronger cohesion within clusters and weaker similarity to neighboring clusters. A silhouette score of 0.5 is generally considered indicative of fair cohesion and separation. Predictor importance was also assessed, standardized on a scale from 0 to 1, with the most influential predictor set to 1. Both variables were found to be predictors, with the SAT-M contributing the most significantly to cluster formation.

A multivariate analysis of variance (MANOVA) was conducted to examine between-group differences in creativity components for each of the MSTs. Post-hoc pairwise comparisons were performed using Bonferroni adjustment. Only statistically significant findings are reported. Additionally, Pearson correlation coefficients were calculated to explore the relationships between the clustering criteria and creativity levels exhibited by students in the different clusters.

Findings

Four Clusters of Participants

Cluster analysis revealed four distinct participant groups based on their levels of general giftedness and mathematical competencies. Descriptive statistics and z-scores for the four clusters are presented in Table 3 and illustrated in Figure 3. A z-score represents the number of standard deviations a data point is from the mean of its distribution. By transforming values from any normal distribution into a standardized scale (mean = 0, SD = 1), z-scores allow for meaningful comparisons across datasets with differing units or scales.

Table 3.

Descriptive Statistics for the Four Clusters of Participants

Four clusters	G-EM	AG-AEM	AG-NEM	NG-AEM
n	94	179	106	79
% of total	20.52%	39.09%	23.14%	17.25%
Mean (SD)
RPMT	27.45 (1.89)	24.78 (2.11)	23.08 (2.10)	16.99 (3.07)
SAT-M	26.29 (3.21)	18.76 (2.98)	9.72 (2.75)	15.37 (5.36)

Cluster quality: Silhouette measure of cohesion and separation is 0.5; predictor importance: RPMT = 0.73 and SAT-M = 1.

Note. G-EM = generally gifted and excelling in mathematics; AG-AEM = averagely gifted and averagely excelling in mathematics; AG-NEM = averagely gifted and non-excelling in mathematics; NG-AEM = not generally gifted and averagely excelling in mathematics.

Figure 3.

Z-Scores in the Four Clusters of Participants

A cluster analysis of 458 participants, based on their SAT-M and RPMT scores, identified four distinct groups (see Figure 3 and Table 3):

• The G-EM group included students who were generally gifted and excelled in mathematics, as reflected in their high RPMT and SAT-M scores. A total of 94 participants (20.52%) were classified into this group.

• The AG-AEM group included students who were averagely gifted and averagely excelling in mathematics, with mid-level scores on both the RPMT and SAT-M. A total of 179 participants (39.09%) were classified into this group.

• The AG-NEM group consisted of averagely gifted non-excelling in mathematics students who attained mid-level RPTM scores and low SAT-M scores (despite excelling in high school mathematics). A total of 106 participants (23.14%) were classified into this group.

• The NG-AEM group included students who exhibited low RPMT scores and mid-level AST-M scores and thus were identified as averagely excelling in mathematics but not being generally gifted. A total of 79 participants (17.25%) were classified into this group.

Because all applicants to the Challenge Program had excelled in high school mathematics, the cluster analysis did not generate a not generally gifted and not excelling in mathematics (NG-NEM) profile. Interestingly, the analysis did not identify an averagely gifted and excelling in mathematics (AG-EM) cluster. We assume that the general intelligence (G) factor plays a significant role in the self-esteem of excelling mathematically (EM) students, which may have discouraged AG-EM students from applying to the program. Because the participants come from a wide variety of high schools with diverse requirements and teaching levels, the group was quite heterogeneous, and thus, the AG-AEM group included the largest amount of students (39.09%).

Motivation of the Participants in the Four Clusters

Motivation levels were examined for each participant group. Descriptive statistics for the four groups are presented in Table 4. All study participants demonstrated high levels of motivation, with mean scores exceeding 4 out of 5, as anticipated. Surprisingly, no significant between-group differences in motivation were found between the four groups. These findings suggest that students who applied to the Challenge Program were equally motivated to pursue a BSc in computer science while also completing their high school requirements, regardless of differences in general abilities and mathematical competencies.

Table 4.

Means and Standard Deviations of the Motivation Scales in the Four Clusters of Participants

Four clusters	G-EM	AG-AEM	AG-NEM	NG-AEM
n	94	179	106	79
Learning motivations questionnaire	Mean (SD)
Enjoyment from learning	4.44 (0.50)	4.26 (0.61)	4.29 (0.56)	4.30 (0.55)
Overcoming challenge	4.05 (0.62)	4.03 (0.52)	4.03 (0.54)	4.06 (0.52)
Learning initiative	4.37 (0.65)	4.37 (0.63)	4.39 (0.67)	4.50 (0.54)
Learning goals for future success	4.46 (0.62)	4.35 (0.73)	4.25 (0.60)	4.25 (0.82)

Creativity of the Participants in the Four Clusters

Creativity performance was analyzed for each participant group across the five creativity tasks (see Figures 1 and 2). Descriptive statistics for the four research groups are presented in Table 5, and significant between-group differences in creativity components are reported in Table 6. Students in the G-EM group demonstrated higher levels of creativity on both the general and mathematical MSTs compared to students in the other groups. Specifically, in MST-1 and MST-3, G-EM students achieved higher scores across all creativity components. All between-group differences in these tasks were found to be statistically significant.

Table 5.

Means and Standard Deviations of the Creativity Components in the Four Clusters of Participants

Four clusters n		G-EM	AG-AEM	AG-NEM	NG-AEM
Four clusters n		94	179	106	79
Mean (SD)
MST-1 Pictorial	Fluency	8.47 (4.19)	6.05 (3.65)	4.83 (3.41)	5.48 (3.17)
	Flexibility	56.77 (28.03)	41.44 (23.18)	34.96 (22.23)	37.14 (21.03)
	Originality	18.85 (16.91)	11.38 (14.65)	7.75 (12.11)	7.91 (10.06)
	Creativity	141.65 (144.58)	98.11 (120.32)	72.32 (122.69)	68.95 (97.25)
MST-2 Calculation	Fluency	1.95 (1.19)	1.40 (1.02)	1.53 (0.92)	1.29 (0.70)
	Flexibility	13.20 (7.69)	11.37 (7.41)	14.13 (8.30)	12.17 (6.15)
	Originality	4.34 (6.63)	2.91 (5.30)	3.62 (6.40)	2.72 (4.84)
	Creativity	34.89 (49.88)	24.67 (47.44)	33.46 (60.10)	24.71 (43.94)
MST-3 System of equations	Fluency	2.62 (1.15)	1.72 (1.01)	1.61 (0.92)	1.47 (0.86)
	Flexibility	14.78 (5.90)	11.76 (5.91)	13.30 (7.51)	12.25 (6.34)
	Originality	2.96 (5.07)	0.99 (2.46)	1.68 (3.30)	1.06 (2.28)
	Creativity	28.26 (51.80)	8.62 (23.55)	15.38 (31.98)	9.57 (22.32)
MST-4 Word problem	Fluency	0.89 (0.91)	0.42 (0.58)	0.52 (0.71)	0.53 (0.71)
	Flexibility	6.84 (6.22)	3.93 (5.26)	4.93 (6.55)	5.08 (6.80)
	Originality	2.07 (4.09)	0.87 (2.65)	1.18 (2.78)	1.18 (3.33)
	Creativity	20.50 (43.59)	8.51 (26.59)	11.56 (27.76)	11.68 (32.98)
MST-5 Geometry	Fluency	1.11 (0.99)	0.68 (0.73)	1.00 (1.13)	0.89 (0.91)
	Flexibility	8.72 (6.58)	6.15 (6.28)	8.70 (8.73)	8.05 (7.67)
	Originality	1.31 (3.20)	1.02 (2.87)	1.88 (4.91)	1.09 (3.06)
	Creativity	11.39 (30.45)	9.07 (27.00)	15.93 (40.28)	9.59 (28.79)

Table 6.

Between-Group Differences in Creativity Components in the Four Clusters of Participants

		Cluster effect F (3, 270)	Pairwise differences
MST-1 Pictorial	Fluency	11.413*** $η_{p}^{2} = . 113$	C1>C2* C1>C3* C1>C4***
	Flexibility	10.086*** $η_{p}^{2} = . 101$	C1>C2* C1>C3* C1>C4***
	Originality	8.407*** $η_{p}^{2} = . 085$	C1>C2* C1>C3* C1>C4*
	Creativity	4.406** $η_{p}^{2} = . 047$	C1>C3* C1>C4*
MST-2 Calculation	Fluency	8.172*** $η_{p}^{2} = 0.51$	C1>C2*** C1>C3* C1>C4***
MST-2 Calculation	Flexibility	3.355* $η_{p}^{2} = 0.22$	C3>C2*
MST-3 System of equations	Fluency	25.384*** $η_{p}^{2} = . 144$	C1>C2* C1>C3* C1>C4***
	Flexibility	5.008** $η_{p}^{2} = . 032$	C1>C2***
	Originality	7.966*** $η_{p}^{2} = . 050$	C1>C2*** C1>C3* C1>C4***
	Creativity	7.951*** $η_{p}^{2} = . 050$	C1>C2*** C1>C3* C1>C4***
MST-4 Word problem	Fluency	9.150*** $η_{p}^{2} = . 057$	C1>C2* C1>C3* C1>C4**
	Flexibility	4.664** $η_{p}^{2} = . 030$	C1>C2***
	Originality	3.003* $η_{p}^{2} = . 019$	C1>C2*
	Creativity	2.897* $η_{p}^{2} = . 019$	C1>C2*
MST-5 Geometry	Fluency	5.432*** $η_{p}^{2} = . 035$	C1>C2**
MST-5 Geometry	Flexibility	4.084** $η_{p}^{2} = . 026$	C1>C2* C3>C2*

*p < .05 **p < .01 ***p < .001.

MST-2, a calculation-based problem drawn from the elementary school curriculum, required basic formal mathematical knowledge. In this task, the G-EM group outperformed all other groups significantly, but only in the fluency component. Surprisingly, no significant differences were observed between the G-EM group and other groups in the remaining creativity components for this problem. In the word problem (MST-4), G-EM participants differed significantly in the fluency component compared to all other groups. However, for the remaining three creativity components, the G-EM participants scored significantly higher only in comparison to the AG-AEM group. Interestingly, the students in the AG-NEM group demonstrated higher mathematical flexibility than those in the NG-AEM group when solving calculation and geometry problems. This finding suggests that general intelligence may play a more significant role than mathematical competencies in supporting flexibility when solving relatively simple mathematical problems.

In MST-5, significant between-group differences were found only in fluency and flexibility, indicating that geometry problems are generally more challenging for high school students. Additionally, given that the participants in this study are oriented toward computer science, they may prefer word problems and calculation-based problems.

Correlations in the Four Clusters

Pearson correlation coefficients were calculated for each participant group to examine the relationships between the clustering criteria and the creativity components across the five MST tasks. Table 5 presents the creativity components for which significant correlations were identified.

As shown in Table 7, the G-EM group exhibited the highest number of significant correlations between creativity components and SAT-M scores. Specifically, significant correlations were found for fluency and flexibility in MST-1 and MST-4, as well as originality and creativity in MST-2 and MST-3. In the AG-AEM group, all creativity components in MST-5 were significantly correlated with SAT-M scores. Surprisingly, fewer correlations emerged between creativity components and RPMT scores. No significant correlations were observed between RPMT and any creativity component in MST-4. Only two significant correlations with RPMT were found in MST-2, MST-3, and MST-5. The fewest correlations were found for the NG-AEM group. Among the five MST tasks, MST-1 revealed the highest number of significant correlations with RPMT. These findings highlight the importance of mathematical creativity in supporting creative problem-solving, even beyond strictly mathematical tasks. Overall, there were more significant correlations found for the SAT-M test than for RPMT, which is logical given that all the problems, apart from MST-1, are mathematical in nature.

Table 7.

Significant Correlations Between Creativity Components and Clustering Criteria

Four clusters n		G-EM		AG-AEM		AG-NEM		NG-AEM
Four clusters n		94		179		106		79
Criteria		RPMT	SAT-M	RPMT	SAT-M	RPMT	SAT-M	RPMT	SAT-M
MST-1 Pictorial	Fluency	.318*	.349*	.235*	−.026	.012	.242*	.081	.039
	Flexibility	.387**	.370*	.291**	−.047	.115	.313**	.026	−.100
	Creativity	.243	.136	.163	.046	−.097	.229*	.054	.034
MST-2 Calculation	Fluency	−.027	.304**	−.019	.126	.246*	.056	−.117	.077
	Flexibility	−.096	.121	−.059	.137	.227*	.075	−.147	.102
	Originality	.143	.220*	−.093	.104	.110	.073	.045	.126
	Creativity	.158	.290**	−.124	.097	.138	.098	.041	.087
MST-3 System of equations	Fluency	.194	.282**	.021	.140	.198*	.276**	.031	.293**
	Flexibility	.092	.148	−.106	.231**	.160	.101	−.023	.157
	Originality	.161	.394***	−.042	.102	.050	.195*	.226*	.170
MST-4 Word problem	Creativity	.171	.368***	−.041	.103	.041	.169	.210	.144
	Fluency	.153	.375***	−.055	.141	−.040	.052	−.009	.069
	Flexibility	.095	.234*	−.048	.157*	−.010	.038	−.032	.038
MST-5 Geometry	Fluency	.204*	.241*	−.113	.196**	.052	.031	.059	−.110
	Flexibility	.083	.170	−.074	.210**	.074	.025	.000	−.132
	Originality	.203*	−.027	.004	.198**	.003	.026	.022	−.064
	Creativity	.149	.016	.000	.194**	.039	.041	−.008	−.089

*p < .05 **p < .01 ***p < .001.

Summary and Discussion

The goal of the study described in this paper was to examine characteristics of high-achieving high school students in mathematics who were motivated to participate in a BSc in computer science program. Specifically, the students’ general and mathematical creativity, mathematical competencies, and general intelligence were examined. Students (N = 458) applied to the program and completed the program’s admission assessments, which included the RPMT, which was used for examining general giftedness; the SAT-M, which was used to evaluate mathematical competencies; and MSTs, which were used to assess both general and mathematical creativity (Leikin, 2009b, R. Leikin, 2013). Cluster analysis was conducted using participants’ RPMT and SAT-M scores, resulting in the identification of four distinct groups based on different combinations of general giftedness (G-factor) and mathematical competence (EM-factor; e.g., Leikin et al., 2014, 2017).

To the best of our knowledge, this study is the first to apply this type of cluster analysis among students pursing computer science degrees. The findings revealed that students of varying levels of abilities, both in general and in mathematics, exhibited similarly high levels of motivation to enroll in a STEM acceleration program. However, despite comparable motivational levels, these students differed in both general and mathematical creativity. The four groups that were identified using clustering procedure were: (a) G-EM group, which included generally gifted (high RPTM scores) and excelling in mathematics (high SAT-M scores) students; (b) AG-AEM group, which included students who were averagely gifted (with mid-level RPTM scores) and averagely excelling in mathematics (with mid-level SAT-M scores); (c) AG-NEM group, which consisted of averagely gifted non-excelling in mathematics students who attained mid-level RPTM scores and low SAT-M scores (despite their excellence in school mathematics); and (d) NG-AEM group, which included nongifted students averagely excelling in mathematics (i.e., students who attained low RPTM scores and mid-level SAT-M scores).

Students who applied to the Challenge Program demonstrated high levels of motivation and self-confidence in their ability to succeed in computer science studies while continuing their high school education. No significant differences in motivation levels were found among the four participant groups. These consistently high levels of motivation align with previous research indicating a strong association between motivational beliefs and active participation in academic challenging STEM programs.

Participants’ creativity was assessed using five different creativity problems. Overall, students in the G-EM group outperformed all other groups, demonstrating significantly higher levels of creativity when solving both general and mathematical problems. In most cases, the differences between the groups were significantly greater in favor of the G-EM group. In the case of MST-1 and MST-3, students from the G-EM group scored higher across all creativity components, with all between-group differences reaching statistical significance. Interestingly, students from the AG-NEM group, although not identified as gifted, demonstrated significantly higher levels of flexibility in the calculation problem (MST-2) than other groups. In the word problem (MST-4) the G-EM group showed higher scores in all creativity components except fluency, but these differences were only significant when compared to the AG-AEM group. In the geometry problem (MST-5), significant differences between groups were found only in fluency and flexibility. In both cases, the G-EM group outperformed the AG-AEM group. Surprisingly, in terms of flexibility, the AG-NEM group also performed better than the AG-AEM group, suggesting that factors beyond general giftedness may influence creative performance in certain types of mathematical tasks.

Furthermore, the G-EM group exhibited the highest number of significant correlations between creativity components and both clustering criteria: RPMT and SAT-M. The strongest associations with RPMT were observed in MST-1, with a smaller number of significant correlations also found with SAT-M in this problem. In contrast, MST-3 revealed the highest number of correlations with SAT-M. Overall, more significant correlations were found with SAT-M than with RPMT, which was expected given that all problems, apart from MST-1, are mathematical in nature. Notably, with MST-4, no significant correlation was found with RPMT, an unexpected finding considering that word problems often require reading comprehension in addition to mathematical reasoning. In general, RPMT was linked to levels of general creativity among students in the two higher performing groups (G-EM and AG-AEM) and showed no significant relationship to either mathematical or general creativity among students in the two lower performing groups (AG-NEM and NG-AEM). These findings suggest that the relationship between cognitive ability and creativity is both cluster-dependent and task-dependent.

Based on the results of our study, we argue that effectively teaching diverse groups of students requires integration of varied task types tailored to their cognitive profiles. The results clearly demonstrate that the development of mathematical creativity is essential for success in STEM studies. Although prior research has consistently highlighted the relationship between mathematical creativity and mathematical ability, our findings suggest that general creativity is equally important. Therefore, incorporating activities that foster general creativity is just as critical as those that develop mathematical skills in preparing students for success in STEM education.

Our findings align with previous research demonstrating a positive relationship between intelligence and creativity (Cramond et al., 2010). Although most prior studies that examined relationships between creativity and STEM programs focused on how such programs cultivate creativity, the present study took a different approach. The current study sought to characterize students who are highly motivated to enroll in a STEM acceleration program by examining their general and mathematical creativity profiles.

The findings of this study demonstrate that students with high levels of both general giftedness and mathematical competencies perform better than all other groups in creativity tests, exhibiting higher levels of creativity. This finding is consistent with previous studies (e.g., Leikin et al., 2014, 2017). Prior research focusing on the cognitive characteristics of generally gifted (G) students has shown that they consistently outperform their nongifted (NG) peers in applying diverse problem-solving strategies (Steiner & Carr, 2003). In this study, we obtained unanticipated findings for the AG-NEM group, which demonstrated a significantly higher level of flexibility in two mathematical problems, suggesting that, to some extent, giftedness compensates for a lower level of mathematical competence. We agree with Leikin et al. (2017), who argued that the combination of G and EM is a necessary condition for success in solving insight-based mathematical problems. This combination of abilities has been linked to neural efficiency in learning-based problem-solving tasks (Waisman et al., 2016).

Based on the findings from this study, we suggest that both general and mathematical creativity can serve as criteria for students’ enrollment in STEM programs. Fostering creativity in school settings may increase students’ likelihood of both enrolling in and succeeding within such programs. To better identify students most suited for accelerated STEM tracks, we recommend incorporating MSTs into the admissions process. This study represents the initial stage of a longitudinal investigation that follows students from their application to the computer science acceleration program through graduation. The goal of the follow-up study is to identify the components of STEM-related intellectual potential that most accurately predict long-term success in the program.

Footnotes

Acknowledgments

This study was made possible through the support of a grant by the Israel Science Foundation grant 1475/18, Grant No. 2018141 from the United States-Israel Binational Science Foundation (BSF) and the Aaron Gutwirth Foundation. This publication reflects only the views of its author(s) and not necessarily those of the Foundations. Our sincere thanks go to the University of Haifa for its generous support of this study.

ORCID iD

Roza Leikin

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was made possible through the support of a grant by the Israel Science Foundation grant 1475/18, Grant No. 2018141 from the United States-Israel Binational Science Foundation (BSF) and the Aaron Gutwirth Foundation.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author Biographies

Lior Miller Markovitz, PhD, is an educational consultant at the University of Haifa for youth. She holds an MA in educational counseling and a teaching certificate in special education. Her PhD dissertation at the Faculty of Education in the University of Haifa examined high school students who succeed in academic studies in computer science.

Gad M. Landau, PhD, is a Research Professor at NYU and a Professor Emeritus at the University of Haifa. Professor Landau’s research interests focus on string algorithms, data structures, computational biology, and parallel computation. Landau has coauthored more than 140 peer-reviewed scientific papers. He has fathered several academic projects at the University of Haifa, most notably the Etgar undergraduate program for highly talented high school students throughout the north of Israel.

Roza Leikin, PhD, is a Full Professor of Mathematics Education and Gifted Education at the Faculty of Education of the University of Haifa. Since October 2020 she has served as the Faculty Dean. Her research and practice focuses task design, creativity and expertise in students and teachers. From 2012–2017 she served as the President of the International Group for Mathematical Creativity and Giftedness. She has edited 17 volumes related to research in mathematics and gifted education and has published more than 190 peer-reviewed papers and book chapters.

Appendix

References

Ackerman

P. L.

Kanfer

Calderwood

(2013). High school advanced placement and student performance in college: STEM majors, non-STEM majors, and gender differences. Teachers College Record, 115(10), 1–43. https://doi.org/10.1177/016146811311501003

Ardianto

Rubini

Pursitasari

I. D.

(2023). Assessing STEM career interest among secondary students: A Rasch model measurement analysis. Eurasia Journal of Mathematics, Science and Technology Education, 19(1), em2213. https://doi.org/10.29333/ejmste/12796

Chittum

J. R.

Jones

B. D.

Akalin

Schram

Á. B.

(2017). The effects of an after-school STEM program on students’ motivation and engagement. International Journal of STEM Education, 4(1), 11–16. https://doi.org/10.1186/s40594-017-0065-4

Chowdhury

A. M. R.

Bhuiya

Chowdhury

M. E.

Rasheed

Hussain

Chen

L. C.

(2013). The Bangladesh paradox: Exceptional health achievement despite economic poverty. The Lancet, 382(9906), 1734–1745. https://doi.org/10.1016/S0140-6736(13)62148-0

Conger

Kennedy

A. I.

Long

M. C.

McGhee

(2021). The effect of advanced placement science on students’ skills, confidence, and stress. Journal of Human Resources, 56(1), 93–124. https://doi.org/10.3368/jhr.56.1.0118-9298R3

Constan

Spicer

J. J.

(2015). Maximizing future potential in physics and STEM: Evaluating a summer program through a partnership between science outreach and education research. Journal of Higher Education Outreach and Engagement, 19(2), 117–136.

Cramond

Kim

K. H.

VanTassel-Baska

(2010). The relationship between creativity and intelligence. In Kaufman

J. C.

Sternberg

R. J.

(Eds.), The Cambridge handbook of creativity (pp. 395–412). Cambridge University Press. https://doi.org/10.1017/CBO9780511763205.025

Cutucache

C. E.

Luhr

J. L.

Nelson

K. L.

Grandgenett

N. F.

Tapprich

W. E.

(2016). NE STEM 4U: An out-of-school time academic program to improve achievement of socioeconomically disadvantaged youth in STEM areas. International Journal of STEM Education, 3(1), 6. https://doi.org/10.1186/s40594-016-0037-0

Dogan

(2015). Student engagement, academic self-efficacy, and academic motivation as predictors of academic performance. The Anthropologist, 20(3), 553–561. https://doi.org/10.1080/09720073.2015.11891759

10.

Dougherty

Mellor

Jian

(2006). The relationship between advanced placement and college graduation. 2005 AP study series, report 1. National Center for Educational Accountability.

11.

Eccles

J. S.

Templeton

(2002). Chapter 4: Extracurricular and other after-school activities for youth. Review of Research in Education, 26(1), 113–180. https://doi.org/10.3102/0091732X026001113

12.

Fortus

Vedder‐Weiss

(2014). Measuring students’ continuing motivation for science learning. Journal of Research in Science Teaching, 51(4), 497–522. https://doi.org/10.1002/tea.21136

13.

Gallagher

J. J.

(2000). Unthinkable thoughts: Education of gifted students. Gifted Child Quarterly, 44(1), 5–12. https://doi.org/10.1177/001698620004400102

14.

Goerge

Cusick

G. R.

Wasserman

Gladden

(2007). After school programs and academic impact: A study of Chicago’s after school matters. Chapin Hall Center for Children.

15.

Greene

K. M.

Lee

Constance

Hynes

(2013). Examining youth and program predictors of engagement in out-of-school time programs. Journal of Youth and Adolescence, 42(10), 1557–1572. https://doi.org/10.1007/s10964-012-9814-3

16.

Haavold

P. Ø.

(2020). An investigation of the relationship between age, achievement, and creativity in mathematics. The Journal of Creative Behavior, 54(3), 555–566. https://doi.org/10.1002/jocb.390

17.

Heath

R. D.

Anderson

Turner

A. C.

Payne

C. M.

(2022). Extracurricular activities and disadvantaged youth: A complicated—but promising—story. Urban Education, 57(8), 1415–1449. https://doi.org/10.1177/0042085918805797

18.

Henriksen

Creely

Henderson

(2019). Failing in creativity: The problem of policy and practice in Australia and the United States. Kappa Delta Pi Record, 55(1), 4–10. https://doi.org/10.1080/00228958.2019.1549429

19.

Hoepner

C. C.

(2010). Advanced placement math and science courses: Influential factors and predictors for success in college STEM majors. University of California.

20.

Jerrim

Shure

Wyness

(2020). Driven to succeed? Teenagers’ drive, ambition and performance on high-stakes examinations. Institute of Labor Economics (IZA). (IZA Discussion Paper No. 13525). https://doi.org/10.2139/ssrn.3660272

21.

Kori

Pedaste

Leijen

Ä.

Tõnisson

(2016). The role of programming experience in ICT students’ learning motivation and academic achievement. International Journal of Information and Education Technology, 6(5), 331–337. https://doi.org/10.7763/IJIET.2016.V6.709

22.

Leikin

(2013). The effect of bilingualism on creativity: Developmental and educational perspectives. International Journal of Bilingualism, 17(4), 431–447. https://doi.org/10.1177/1367006912438300

23.

Leikin

(2009a). Bridging research and theory in mathematics education with research and theory in creativity and giftedness. In Leikin

Berman

Koichu

(Eds.), Creativity in Mathematics and the Education of Gifted Students (pp. 385–411). Sense.

24.

Leikin

(2009b). Exploring mathematical creativity using multiple solution tasks. In Leikin

Berman

Koichu

(Eds.), Creativity in Mathematics and the Education of Gifted Students (pp. 129–145). Sense.

25.

Leikin

(2013). Evaluating mathematical creativity: The interplay between multiplicity and insight. Psychological Test and Assessment Modeling, 55(4), 385–400.

26.

Leikin

(2018). Openness and constraints associated with creativity-directed activities in mathematics for all students. In Amado

Carreira

Jones

(Eds.), Broadening the Scope of Research on Mathematical Problem Solving: A Focus on Technology, Creativity and Affect (pp. 387–397). Springer.

27.

Leikin

Guberman

(2023). Creativity and challenge: Task complexity as a function of insight and multiplicity of solutions. In Leikin

(Ed.), Mathematical Challenges for All (pp. 325–342). Springer International.

28.

Leikin

Paz-Baruch

Waisman

Lev

(2017). On the four types of characteristics of super mathematically gifted students. High Ability Studies, 28(1), 107–125. https://doi.org/10.1080/13598139.2017.1305330

29.

Leikin

Lev

(2013). Mathematical creativity in generally gifted and mathematically excelling adolescents: What makes the difference? ZDM—The International Journal on Mathematics Education, 45(2), 183–197. https://doi.org/10.1007/s11858-012-0460-8

30.

Leikin

Paz-Baruch

Leikin

(2014). Cognitive characteristics of students with superior performance in mathematics. Journal of Individual Differences, 35(3), 119–129. https://doi.org/10.1027/1614-0001/a000140

31.

Lee

(2004). A review of conceptions of ability and related motivational constructs in achievement motivation. Quest, 56(4), 439–461. https://doi.org/10.1080/00336297.2004.10491836

32.

Lichtenberger

George-Jackson

(2013). Predicting high school students’ interest in majoring in a STEM field: Insight into high school students’ postsecondary plans. Journal of Career and Technical Education, 28(1), 19–38. https://doi.org/10.21061/jcte.v28i1.571

33.

Midgley

Maehr

M. L.

Hruda

L. Z.

Anderman

Freeman

K. E.

Urdan

(2000). Manual for the patterns of adaptive learning scales. University of Michigan.

34.

Miller

K. C.

Carrick

Martínez-Sussmann

Levine

Andronicos

C. L.

Langford

R. P.

(2007). Effectiveness of a summer experience for inspiring interest in geoscience among Hispanic-American high school students. Journal of Geoscience Education, 55(6), 596–603. https://doi.org/10.5408/1089-9995-55.6.596

35.

Yang

Calaway

Nickey

(2011). ACT test performance by advanced placement students in Memphis city schools. The Journal of Educational Research, 104(5), 354–359. https://doi.org/10.1080/00220671.2010.486810

36.

Musu-Gillette

L. E.

Wigfield

Harring

Eccles

J. S.

(2015). Trajectories of change in student’s self-concepts of ability and values in math and college major choice. Educational Research and Evaluation, 21(4), 343–370. https://doi.org/10.1080/13803611.2015.1057161

37.

Ommundsen

Haugen

Lund

(2005). Academic self‐concept, implicit theories of ability, and self‐regulation strategies. Scandinavian Journal of Educational Research, 49(5), 461–474. https://doi.org/10.1080/00313830500267838

38.

Raven

J. C.

Court

J. H.

(2000). Manual for Raven’s progressive matrices and vocabulary scales. Oxford Psychologists.

39.

Rosenzweig

E. Q.

Wigfield

(2016). STEM motivation interventions for adolescents: A promising start, but further to go. Educational Psychologist, 51(2), 146–163. https://doi.org/10.1080/00461520.2016.1154792

40.

Sadler

P. M.

Sonnert

Hazari

Tai

(2012). Stability and volatility of STEM career interest in high school: A gender study. Science Education, 96(3), 411–427. https://doi.org/10.1002/sce.21007

41.

Schoevers

E. M.

Kroesbergen

E. H.

Kattou

(2020). Mathematical creativity: A combination of domain‐general creative and domain‐specific mathematical skills. The Journal of Creative Behavior, 54(2), 242–252. https://doi.org/10.1002/jocb.361

42.

Simonton

D. K.

(2000). Creativity: Cognitive, personal, developmental, and social aspects. American Psychologist, 55(1), 151–158. https://doi.org/10.1037/0003-066X.55.1.151

43.

Steiner

H. H.

Carr

(2003). Cognitive development in gifted children: Toward a more precise understanding of emerging differences in intelligence. Educational Psychology Review, 15(3), 215–246. https://doi.org/10.1023/A:1024636317011

44.

Sternberg

R. J.

Lubart

T. I.

(2000). The concept of creativity: Prospects and paradigms. In Sternberg

R. J.

(Ed.), Handbook of creativity (pp. 93–115). Cambridge University Press.

45.

Ugras

(2018). The effect of STEM activities on STEM attitudes, scientific creativity and motivation beliefs of the students and their views on STEM education. International Online Journal of Educational Sciences, 10(5), 165–182. https://doi.org/10.15345/iojes.2018.05.012

46.

Vedder‐Weiss

Fortus

(2011). Adolescents’ declining motivation to learn science: Inevitable or not? Journal of Research in Science Teaching, 48(2), 199–216. https://doi.org/10.1002/tea.20398

47.

Wai

(2015). Long-term effects of educational acceleration. In Assouline

S. G.

Colangelo

VanTassel-Baska

Lupkowski-Shoplik

A. E.

(Eds.), A nation empowered: Evidence trumps the excuses holding back America’s brightest students (Vol. 2, pp. 73–83). The Belin-Blank Center for Gifted and Talented Education.

48.

Wai

Lubinski

Benbow

C. P.

Steiger

J. H.

(2010). Accomplishment in science, technology, engineering, and mathematics (STEM) and its relation to STEM educational dose: A 25-year longitudinal study. Journal of Educational Psychology, 102(4), 860–871. https://doi.org/10.1037/a0019454

49.

Waisman

Leikin

(2016). Brain activity associated with logical inferences in geometry: Focusing on students with different levels of ability. ZDM – International Journal on Mathematics Education, 48(3), 321–335. https://doi.org/10.1007/s11858-016-0760-5

50.

Wang

M. T.

Eccles

J. S.

(2013). School context, achievement motivation, and academic engagement: A longitudinal study of school engagement using a multidimensional perspective. Learning and Instruction, 28, 12–23. https://doi.org/10.1016/j.learninstruc.2013.04.002

51.

Warne

R. T.

Sonnert

Sadler

P. M.

(2019). The relationship between advanced placement mathematics courses and students’ STEM career interest. Educational Researcher, 48(2), 101–111. https://doi.org/10.3102/0013189X19825811

52.

Weisblay

(2018). Integration of high school students from different population groups in studies. The Knesset Research and Information Center. https://www.knesset.gov.il/mmm/data/pdf/m04202.pdf

53.

Wigfield

Eccles

J. S.

Fredricks

J. A.

Simpkins

Roeser

R. W.

Schiefele

(2015). Development of achievement motivation and engagement. In Lamb

M. E.

Lerner

R. M.

(Eds.), Handbook of child psychology and developmental science: Socioemotional processes (pp. 657–700). John Wiley & Sons. https://doi.org/10.1002/9781118963418.childpsy316

54.

Young

J. L.

Young

J. R.

Ford

D. Y.

(2017). Standing in the gaps: Examining the effects of early gifted education on Black girl achievement in STEM. Journal of Advanced Academics, 28(4), 290–312. https://doi.org/10.1177/1932202X17730549

55.

Zohar

(1990). Mathematical reasoning ability: Its structure, and some aspects of its genetic transmission. Hebrew University. [Unpublished doctoral dissertation].

Characterization of STEM-Motivated High School Students: Focusing on Intelligence,Mathematical Competencies,and Creativity

Abstract

Keywords

Background

The Study

Research Goal

The Context

Study Participants

Tools and Data Collection

Data Analysis

Findings

Four Clusters of Participants

Motivation of the Participants in the Four Clusters

Creativity of the Participants in the Four Clusters

Correlations in the Four Clusters

Summary and Discussion

Footnotes

Acknowledgments

ORCID iD

Funding

Declaration of Conflicting Interests

Author Biographies

Appendix

References