Comparison of Perceived Achievement of Complex Thinking Competency Among American, European, and Asian University Students

Abstract

Despite the growing focus of educational institutions on students’ practical abilities beyond theoretical knowledge, the perception that students have of their competencies is crucial for their effective application in professional contexts. Accordingly, this paper reports a study of 435 university students attending ten universities in eight countries in the Americas (Chile, Colombia, Mexico), Asia (Pakistan and the Philippines), and Europe (Spain, Finland, and Serbia). The goal was to measure their perceptions of their achievement of complex thinking competency and its sub-competencies. The intention was to identify how cultural, educational, and socioeconomic differences among countries account for the variances in the students’ self-assessment of competencies, impacting their professional preparedness. The study focused on the competency of complex thinking, considering its critical importance in solving current environmental problems. The analysis employed the non-parametric Brown–Forsythe statistical test and Bonferroni correction, given the non-normality and heteroscedasticity of the data. It was found that (i) there is no statistically significant difference by gender; (ii) there are statistically significant differences in all types of thinking per country, geographical area (continent), and Human Development Index (HDI).

1. Introduction

Undoubtedly, developing students’ skills and competencies is a primary goal of modern universities. Regardless of geographical location, higher education institutions have increased their focus on students’ practical abilities, extending beyond just the acquisition of theoretical knowledge (Baena et al. 2022). However, ensuring the development of these competencies is not enough; it is essential to foster a positive perception of these skills among future professionals, who must recognize and value their abilities to effectively apply them in real contexts. This self-awareness of their abilities directly influences their confidence and motivation, impacting their academic performance and ability to address complex challenges in their future professional lives (Vázquez-Parra et al. 2022).

Additionally, it is crucial to recognize that the perception of competencies and skills can vary considerably depending on the country where the universities are located. Cultural, educational, and socioeconomic variations among nations can impact how students value and perceive their capabilities. In particular cultural contexts, a collectivist orientation may predominate, promoting collaborative skills and teamwork, while in others, an individualistic approach may emphasize the importance of autonomy and competitiveness (Cole 2010). These divergences not only affect students’ self-assessment of their skills but also how they prepare for their professional careers, adapting their competencies to what they consider valued in their socio-cultural environments (Cheon et al. 2019).

This article presents the results of a study of 435 university students attending ten universities in eight countries in the Americas (Chile, Colombia, Mexico), Asia (Pakistan and the Philippines), and Europe (Spain, Finland, and Serbia). The goal was to measure their perception of achievement in complex thinking competency and its sub-competencies The choice of complex thinking competency was based on its relevance to solving environmental problems. The purpose of the study was to identify possible similarities and differences in university students’ perceptions of achievement per country, geographical area (continent), the country’s Human Development Index (HDI), and gender. The analysis employed the non-parametric Brown–Forsythe statistical test and Bonferroni correction, given the non-normality (validated by the multivariate Henze–Zirkler test and the univariate Shapiro–Wilk test) and heteroscedasticity of the data (verified with the Levene test). This analysis is fundamental to understanding how cultural, educational, and socioeconomic differences impact students’ professional preparation and their ability to address complex problems in their future careers.

2. Theoretical Framework

2.1. The Relevance of Complex Thinking in Professional Training

In the contemporary educational and professional landscape, complex thinking stands out as essential for equipping students to face the challenges and opportunities of today’s world (Tobón and Luna 2021). This cognitive paradigm, which promotes the integration of various perspectives and disciplines, focuses on the interrelation, diversity, and uncertainty that characterize real-world problems. Thus, complex thinking not only addresses elements in isolation but also examines systems, patterns, and contexts for a more comprehensive and profound understanding of phenomena (Silva and Iturra 2021).

In this sense, it is crucial to recognize complex thinking as a mega-competency that encapsulates four essential sub-competencies: critical, systemic, scientific, and innovative thinking. According to Edgar Morin (Morin 1990), complex thinking is distinguished by its focus on multidimensionality, meaning that problems must be approached considering multiple dimensions and factors simultaneously. Morin also emphasizes the importance of interdisciplinarity, arguing that deep understanding arises from the convergence of various disciplines and perspectives. Additionally, this approach underscores the notion of uncertainty, recognizing that knowledge is never complete or final, and there is always space for revision and constructive skepticism. Thus, complex thinking not only addresses how we think but also how that thinking can adapt to the inherent and changing complexity of today’s globalized and technologically advanced reality (Tobón 2010).

Delving into the sub-competencies, critical thinking is defined as the ability to analyze, evaluate, and synthesize information, discriminating facts and opinions to form grounded judgments (Cui et al. 2021). This type of thinking is indispensable for discerning truth from misleading information, a crucial skill in a world where misinformation predominates. Future leaders must possess sharp critical thinking to arrive at complex decisions and develop effective long-term strategies (Ma and Zhou 2022). Critical thinking facilitates a clear vision, allows for anticipating consequences, and fosters strategic decision-making, equipping individuals to lead responsibly and ethically in their professional areas (Hapsari 2016).

It is important to note that critical thinking should not be considered solely as a set of epistemological skills, such as analyzing assertions, forming beliefs, or reaching logical conclusions based on premises. From the perspective of educational philosophers such as John Dewey, critical thinking is intrinsically linked to intellectual virtues that go beyond its technical dimension. Virtues such as truth-seeking, intellectual humility, and commitment to ethical reasoning are essential to ensure that critical thinking is not used as a neutral and potentially dangerous tool, for example, to justify political propaganda. These virtues underscore the role of critical thinking in the formation of autonomous and responsible individuals capable of exercising judgment for the common good (Bailin and Siegel 2003; Lipman 2003).

Likewise, the development of critical thinking in higher education should not be limited to preparing students for the labor market, as emphasized by international organizations such as the OECD (OECD 2018b). Instead, it should be aligned with a Humboldtian vision of education, which prioritizes the strengthening of the individual’s rational autonomy. This approach considers the university as a space for promoting self-reflection and the capacity for self-determination. Thus, critical thinking becomes not only a cognitive tool but also a means to achieve a state of rational autonomy, where individuals can make informed and ethically justified decisions in a complex and interconnected world (Siegel 1988; Hitchcock 2018).

Scientific thinking, on the other hand, means applying the scientific method and empirical and evidence-based reasoning, even in everyday contexts. This methodological approach requires skills in observation, hypothesis formulation, experimentation, and data analysis, where students tackle problems structurally and methodically (Suryansyah et al. 2021). An empirical approach significantly improves accuracy and efficacy in problem-solving, not only in sciences but in any professional field, endowing professionals with a set of critical skills applicable to a broad spectrum of situations and disciplines (Koerber and Osterhaus 2019).

Systemic thinking focuses on understanding how parts of a system interact with each other and how changes in one part can impact the entire system (Abuabara et al. 2023). This perspective is vital in contexts where decisions or actions in one area can have significant and often unexpected repercussions in others. Systemic thinking is crucial for understanding the structure and dynamics of complex systems; it is fundamental for sustainable innovation and informed decision-making. It facilitates an integrated and strategic vision necessary for leading and contributing effectively in a work environment that demands interconnected understanding and long-term planning (Khammarnia et al. 2017).

Finally, innovative thinking generates novel ideas and creative solutions to complex problems. This thinking is essential in a labor world characterized by rapid technological and market changes, where the ability to innovate not only distinguishes individuals and organizations but is also crucial for their survival and success (Qosimova 2022). It fosters the development of new products, services, and processes and encourages a proactive approach to challenges and opportunities. In the 21st century, the ability to think outside conventional schemes and find unconventional solutions is invaluable, especially in fields such as technology, business management, health, and sustainable development, where creativity and innovation are key drivers of progress and efficiency (Saienko et al. 2021).

The development of this macro competency and its sub-competencies in university students is essential to adequately prepare them for the challenges of the contemporary professional and academic world (Brown 2019). By fostering these transversal competencies, universities equip students with the necessary tools to lead, innovate, and contribute significantly to their respective fields, thus promoting a positive and lasting impact on problems in their environments (Ramirez-Montoya et al. 2022).

2.2. Perceived Competency as a Key Factor in Professional Problem Solving

The perceived achievement of competency or skill is fundamental for university students to apply these abilities in professional contexts effectively. A positive and accurate perception of their skills not only increases their confidence in their capabilities but also enhances their willingness to tackle and solve complex and challenging problems (Alkhatib 2019). This is crucial in the professional realm, where initiative and autonomy are often required to navigate uncertain or evolving situations. Students who recognize and value their competencies are better prepared to apply them effectively, optimizing outcomes and facilitating innovative solutions (Cruz-Sandoval et al. 2023).

Moreover, an adequate perception of one’s skills allows future professionals to address their career development better. They are aware of their strengths and areas for improvement and can direct their continuous learning and professional development toward maximizing their effectiveness at work (Hiver et al. 2021). Ultimately, this enables them to meet and exceed workplace expectations, positioning them as valuable and competent professionals (Rios and Suarez 2017).

On the other hand, a good perception of competencies fosters resilience by enabling individuals to handle setbacks and challenges better, as they possess a solid foundation of confidence in their abilities to think critically and find solutions. This resilience translates into greater adaptability and an ability to handle stress and pressure effectively, indispensable abilities in many of today’s dynamic and demanding professional environments (Lising et al. 2004). Finally, a realistic and positive perception of one’s skills helps university students communicate their value in the labor market effectively. By understanding and articulating what they can do well, students are better equipped to negotiate roles, responsibilities, and remunerations in their future jobs (Cruz-Sandoval et al. 2022).

Given the importance of competency perception in the development and professional performance of students, universities must pay attention to how they foster and assess this perception in their educational programs, ensuring that students not only acquire technical and theoretical skills but also develop a deep understanding and appropriate valuation of them (Vázquez-Parra et al. 2023). This not only improves academic outcomes but also prepares students to face real-world challenges with confidence and competency, thus strengthening their impact on society and their future work environments (Ramírez et al. 2021)

Despite this, it is critical to consider how cognitive biases may influence self-perception of complex thinking competencies. The Dunning–Kruger effect (Kruger and Dunning 1999) describes a phenomenon in which individuals with limited skills in an area tend to overestimate their competence, while those with superior skills may underestimate their performance. In the context of complex thinking, this implies that some students may overestimate their abilities without being aware of their limitations, whereas more competent students may evaluate themselves more critically and modestly.

This effect has significant implications for studies that rely on self-reports, as students’ perception of their own competencies might not accurately reflect their actual ability. As Dunning (Dunning 2011) suggests, the lack of metacognitive awareness in less competent individuals prevents them from recognizing their deficiencies, leading to inflated self-assessment. On the other hand, those with higher competence tend to have better metacognition and thus are more aware of the complexities and challenges involved in complex thinking, which may lead them to be humbler in their self-assessment.

Considering the Dunning–Kruger effect is essential for interpreting the results of studies on self-perceived competence. Recognizing this limitation, the importance of complementing self-reports with objective assessments and mixed methodologies to obtain a more accurate understanding of the development of complex thinking in university students is reinforced.

3. Methodology

3.1. Participants

The sample comprised 435 higher education students with an average age of 21. These students attended ten universities in eight countries on three continents: the Americas, Europe, and Asia. Students from six different disciplines of study participated: Humanities and Education, Social Sciences, Health Sciences, Business, Engineering and Architecture, and Art. For clarity about the sample,

Table 1. Study population.

Country	Men	Women	Total
Chile	18	28	46
Colombia	11	31	42
Mexico	29	30	59
Philippines	29	29	58
Pakistan	29	4	33
Spain	31	67	98
Finland	23	26	49
Serbia	24	26	50

provides data on origin and gender.

Notably, although a balance between men and women was sought, the sample was generated proportionally to the number of male and female students enrolled. Thus, although the sample was not even in numbers between genders, it was proportional to the enrollments of the participating institutions. In the case of Spain, note that two institutions participated, one in the North and the other in the South, to have a broader and more objective perception of the reality of this Iberian country. The sample of students from the North of Spain comprised 45 students (30 women and 15 men), and the sample from the South had 53 students (37 women and 16 men). Figure 1 shows the means for competency and the sub-competencies (types of thinking) per country.

The multivariate Henze–Zirkler test (HZ = 1.740; p-value < 0.001) and the univariate Shapiro–Wilk test established that the data did not follow a normal distribution (

Table 2. Shapiro–Wilk univariate normality test.

Variable	Statistic	p-Value	Normality
HDI 2022	0.8823	<0.001	No
Gender	0.6323	<0.001	No
Complex thinking	0.9734	<0.001	No
Systemic thinking	0.9648	<0.001	No
Scientific thinking	0.9692	<0.001	No
Critical thinking	0.9343	<0.001	No
Innovative thinking	0.9334	<0.001	No

Source: own elaboration using the MVN package of R (version 5.9 from 2021) (Korkmaz et al. 2022).

), nor were they homoscedastic according to Levene’s test (for example, for segmentation by gender, p-value < 0.046894, leading to rejecting the null hypothesis of homogeneity of variances).

3.2. Sample Limitations

Although the sample included 435 university students from ten universities in eight countries in the Americas, Europe, and Asia, there are limitations that could affect the generalizability of the results. First, the sample selection was based on students’ availability and willingness to participate, which introduces a self-selection bias. It is possible that students who agreed to participate have a particular interest in complex thinking or feel more confident in their abilities, which may not represent the general student population. In addition, the distribution of the sample across countries and genders was not uniform. For example, in Pakistan, there was a higher representation of males (29 males and 4 females), which could influence the results and limit the comparability between genders in that country. Also, Spain contributed the largest number of participants (98 students), while other countries had smaller samples, which could affect the weighting of results and statistical power in comparative analyses.

Another limitation is the lack of representativeness of the participating institutions within each country. The universities selected may have particular characteristics in terms of resources, pedagogical approaches, or student profiles that do not reflect the full educational landscape of the country or region. This may introduce an institutional bias that influences students’ perceptions of their competencies. These limitations could generate biases in the results, affecting the external validity of the study. Self-selection bias and lack of randomization in the selection of participants may limit the generalizability of the findings to all student populations in the included countries. Differences in sample sizes by country may also affect the sensitivity of statistical tests, giving more weight to data from countries with larger samples.

In addition, cultural differences in how students perceive and report their competencies may influence the results. For example, in some cultures, individuals may be more likely to underestimate their skills due to norms of modesty, while in others they may overestimate their skills. This phenomenon, known as cultural response bias, may affect the comparability of perceptions across countries. To address these potential biases, the Brown–Forsythe nonparametric statistical test and the Bonferroni correction were used, adapting the analysis to the non-normality and heteroscedasticity of the data. However, it is important to recognize that these statistical measures do not completely eliminate the biases inherent in the sample limitations. Therefore, the results should be interpreted with caution, considering these factors.

3.3. Instrument

To assess the students’ perceived achievement in complex thinking competency and its sub-competencies, we applied the eComplexity instrument (Vázquez-Parra et al. 2024), which consists of 25 items grouped into four dimensions that measure the sub-competencies of systemic, scientific, critical, and innovative thinking (see

Table 3. E-Complexity instrument reliability per item, dimension, and overall.

Items	McDonald’s ω	Cronbach’s α
Systemic thinking	0.80 [0.74–0.84]	0.79 [0.74–0.85]
1. I have the ability to find associations between variables, conditions and constraints in a project.	0.79	0.78
2. I identify data from my discipline and from other areas that contribute to solve problems.	0.76	0.76
3. I participate in projects that have to be solved using inter/multidisciplinary perspectives.	0.78	0.78
4. I organize information to solve problems.	0.77	0.76
5. I enjoy learning about different perspectives on a problem.	0.79	0.78
6. I lean toward strategies for understanding the parts and whole of a problem.	0.79	0.79
7. I have the ability to Identify the essential components of a problem in order to formulate a research question.	0.78	0.77
8. I know the structure and formats for writing research reports used in my area or discipline.	0.79	0.79
Scientific thinking	0.76 [0.71–0.82]	0.76 [0.70–0.81]
9. I identify the structure of a research article that is used in my area or discipline.	0.74	0.74
10. I identify the elements to formulate a research question.	0.71	0.71
11. I design research instruments coherent with the research method used.	0.73	0.72
12. I formulate and test research hypotheses.	0.73	0.73
13. I am inclined to use scientific data to analyze research problems.	0.73	0.73
14. I have the ability to critically analyze problems from different perspectives.	0.75	0.75
Critical thinking	0.80 [0.75–0.84]	0.80 [0.74–0.84]
15. I identify the basis of my own and others’ judgments in order to recognize false arguments.	0.78	0.78
16. I self-evaluate the level of progress and achievement of my goals to make the necessary adjustments.	0.76	0.76
17. I use reasoning based on scientific knowledge to make judgments in the face of a problem.	0.78	0.78
18. I make sure to review the practical guidelines of the projects in which I participate.	0.79	0.79
19. I appreciate criticism in the development of projects in order to improve them.	0.77	0.77
20. I know the criteria for determining a problem.	0.77	0.77
21. I have the ability to identify variables, from various disciplines, that can help answer questions.	0.78	0.78
Innovative thinking	0.77 [0.72–0.82]	0.77 [0.71–0.82]
22. I apply innovative solutions to diverse problems.	0.73	0.73
23. I solve problems by interpreting data from different disciplines.	0.72	0.71
24. I analyze research problems contemplating the context to create solutions.	0.73	0.73
25. I tend to critically evaluate and innovate solutions derived from a problem.	0.74	0.72
Complex thinking	0.93 [0.91–0.94]	0.92 [0.91–0.94]

). This instrument underwent an initial validation process in two parts: theoretical validation and content validation with experts. The theoretical validation, based on the analysis of instruments that measure complex reasoning and its sub-competencies, revealed the lack of an integrating instrument, which led to the design of one that incorporated these dimensions (Castillo-Martínez et al. 2022). As for the content validation, experts evaluated the instrument items for clarity, coherence, and relevance using an online questionnaire. The results were high scores in all categories, with considerable independence between them.

In a pilot study with 999 participants, the instrument demonstrated validity and internal consistency, with a KMO index above 0.80, a p-value less than 0.05, and a Cronbach’s Alpha of 0.93. The scale’s reliability was calculated by item, by dimension, and overall using McDonald’s Omega and Cronbach’s Alpha indices. In each of the various reliability calculations, the values obtained exceeded “acceptable”, ensuring that the instrument employed in the study had adequate internal consistency (see Table 3).

This pilot study established that the instrument had high validity, reliability, and internal consistency, highlighting the independence of its four sub-competencies: systemic, scientific, critical, and inventive thinking (Castillo-Martínez et al. 2024).

3.4. Hypothesis

The study was based on the following hypotheses, formulated from the literature and theoretical framework:

Hypothesis 1 (H1).

There is a significant difference in the perceived achievement of complex thinking competency and its subcompetencies among university students from different countries.

Hypothesis 2 (H2).

Perceived achievement of complex thinking competence and its subcompetencies varies significantly by students’ geographic area (continent).

Hypothesis 3 (H3).

The Human Development Index (HDI) of the country significantly influences the perceived achievement of the complex thinking competency and its subcompetencies.

Hypothesis 4 (H4).

There is a significant difference in the perception of achievement of the complex thinking competency and its subcompetencies between males and females.

These hypotheses are relevant because they address how cultural, educational, and socioeconomic factors may influence self-perception of critical competencies for professional performance in the 21st century as follows:

-

H1 and H2: Based on studies suggesting that educational practices and cultural values affect the development and perception of cognitive skills (Nisbett 2003; Hofstede 2011). Exploring differences between countries and continents allows us to understand how different contexts shape these perceptions.

-

H3: Supported by research linking socioeconomic development to the quality and focus of education (OECD 2018a). Higher HDI could be associated with better educational resources and thus higher perceptions of competencies.

-

H4: Although some studies indicate gender differences in certain cognitive skills (Halpern 2012), others suggest that these differences are diminishing or nonexistent (Hyde 2005). Exploring this hypothesis contributes to the debate and understanding of gender equity in education.

By substantiating these hypotheses, the study seeks to provide empirical evidence that contributes to the understanding of how various factors influence the perception of complex thinking, thus informing more inclusive and effective educational policies and practices.

3.5. Procedure

To collect the data, we contacted participating institutions and teachers interested in implementing the instrument with their students. The application was conducted through a digitalized Microsoft Form instrument, which included an option for each participant to agree to participate voluntarily and acknowledge that they understood how their responses would be used for research purposes. The implementation was carried out in both Spanish and English.

The interdisciplinary research group R4C at the Institute for the Future of Education at Tecnologico de Monterrey supported this implementation. The institutional ethics committee rated the implementation (ID: IFE-2024-01) as low risk and approved it to proceed. The study observed the Terms and Conditions of the Research for Challenges Privacy Notice (https://tec.mx/es/aviso-privacidad-research-challenges, accessed on 30 November 2024).

3.6. Data Analysis

R software (version 3.0 from 2023) was used for data analysis. Initially, data normality was validated using the multivariate Henze–Zirkler test and the univariate Shapiro–Wilk test with the MVN package (Korkmaz et al. 2022). Subsequently, homoscedasticity was checked using Levene’s test (Katsileros et al. 2024).

The assessment of differences in the students’ self-perceived achievement of complex, systemic, scientific, critical, and innovative thinking was analyzed by country. A summary for each country can be found in

Table A1. Information for each country.

Country	Muestra	ST	ScT	CT	IT	CT	HDI 2022	Continent
Chile	46	4.08	4.07	4.13	4.00	4.08	Very high	Latin America
Colombia	42	4.24	4.17	4.24	4.26	4.23	High	Latin America
Spain	98	3.57	3.39	3.31	3.34	3.42	Very High	Europe
Finland	49	3.51	3.40	3.29	3.59	3.44	Very High	Europe
Mexico	59	4.18	4.10	4.18	4.17	4.16	High	Latin America
Pakistan	33	3.90	3.84	3.92	3.86	3.88	Low	Asia
Philippines	58	3.80	3.60	3.84	3.78	3.76	High	Asia
Serbia	50	3.58	3.08	3.47	3.51	3.42	Very High	Europe

(Chile, Colombia, Spain, Finland, Mexico, Pakistan, the Philippines, and Serbia), geographic area (Asia, Europe, and Latin America), gender (male, female), and the 2022 Human Development Index range (very high, high, low). The middle range was not considered because none of the countries in the sample were in that stage. The criteria were Low (<0.550), Medium (0.550–0.699), High (0.700–0.799), and Very high (≥0.800) (UNDP 2024). We employed the non-parametric Brown–Forsythe statistical test (Elamir 2023; Brown and Forsythe 1974) due to the lack of normality and homoscedasticity in the data. For this purpose, we used the onewaytest package in R (Dag et al. 2023).

The Brown–Forsythe mean difference test is based on transforming the data to their trimmed mean or using group medians. The transformation makes the test more sensitive to violations of normality and heteroscedasticity.

The null and alternative hypotheses were as follows:

-

H 0: all group means are equal (no variation in group means).

H 0: μ 1 = μ 2 = … = μ k (where k is the number of groups).

-

H a: At least, the mean of one group is different from that of other groups.

H a: The μ are not equal.

The Brown–Forsythe test statistic for equality of means (De Almeida et al. 2008) is

$$F={\displaystyle \frac{N-k}{k-1}}\times {\displaystyle \frac{\sum n_{1 }{({\overline{z}}_{i.}-{\overline{z}}_{..})}^2}{\sum \sum {(z_{ij}-{\overline{z}}_{i.} ))}^2}}$$

where

$$z_{ij}=\left|y_{ij}-{\overline{y}}_i\right|$$

$${\overline{y}}_i=median of group i$$

The test statistics had approximately an F-distribution with degrees of freedom of $k-1$ and $N-k$.

Finally, in cases where significant differences were identified, a group comparison was performed using the Bonferroni correction (Armstrong 2014).

4. Results

Since there was neither normality nor homoscedasticity, the Brown–Forsythe test for difference in means was applied. The results are synthesized in

Table 4. Results of the Brown–Forsythe mean difference test (alpha = 0.05).

	Country	Continent	Sex	HDI 2022
Complex thinking	statistics: 18.82565 p-value: 1.908064 × 10−21 **	statistics: 62.37788 p-value: 1.60335 × 10−23 **	statistics: 0.6814961 p-value: 0.4095335	statistics: 29.3456 p-value: 2.335169 × 10−11 **
Systemic thinking	statistics: 12.28141 p-value: 5.6588 × 10−14 **	statistics: 37.86528 p-value: 3.280106 × 10−15 **	statistics: 0.07240087 p-value: 0.7880085	statistics: 17.01148 p-value: 3.204551 × 10−7 **
Scientific thinking	statistics: 17.25935 p-value: 1.448874 × 10−19 **	statistics: 56.38872 p-value: 6.624039 × 10−22 **	statistic: 2.334048 p-value: 0.1273108	statistics: 23.55255 p-value: 1.007007 × 10−9 **
Critical thinking	statistics: 17.41693 p-value: 1.065009 × 10−19 **	statistics: 61.71231 p-value: 1.371212 × 10−23 **	statistics: 1.375459 p-value: 0.2415289	statistics: 30.51421 p-value: 1.422345 × 10−11 **
Innovative thinking	statistics: 10.88496 p-value: 1.920951 × 10−12 **	statistics: 36.6638 p-value: 2.750721 × 10−15 **	statistics: 0.2397674 p-value: 0.6246235	statistics: 20.46455 p-value: 1.646816 × 10−8 **

Source: own elaboration using the onewaytest package of R (version 3.0 from 2023) (Dag et al. 2023). Note: ** Statistically significant difference at 5%.

and

Table 5. Interpretation of the Brown–Forsythe mean difference test.

	Country	Continent	Sex	HDI 2022
Complex thinking	Yes	Yes	No	Yes
Systemic thinking	Yes	Yes	No	Yes
Scientific thinking	Yes	Yes	No	Yes
Critical thinking	Yes	Yes	No	Yes
Innovative thinking	Yes	Yes	No	Yes

Source: own elaboration using the onewaytest package of R (Dag et al. 2023).

using an alpha of 0.05.

The interpretation of Table 5 indicates the following:

(i)

There is no statistically significant difference per gender.

(ii)

There are statistically significant differences in all types of thinking per country, geographic area (continent), and HDI level.

In cases where statistically significant differences were identified, a group comparison was carried out using the Bonferroni correction (Armstrong 2014) (

Table 6. Comparison of groups by Bonferroni correction (Alpha = 0.05) per country.

Country 1	Country 2	p-Value CoT	Ho CoT	p-Value ST	Ho ST	p-Value ScT	Ho ScT	p-Value CT	Ho CT	p-Value IT	Ho IT	Interpretation
Chile	Colombia	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Chile	Mexico	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There are no differences in any type of thinking
Chile	Pakistan	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Chile	Philippines	8.6686 × 104	Not reject	3.8241 × 106	Not reject	3.2520 × 103	Not reject	3.1867 × 105	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Chile	Finland	8.1202	Not reject	1.1139 × 102	Not reject	6.6731 × 102	Not reject	2.8464	Not reject	2.3520 × 105	Not reject	There is no difference in any type of thinking
Chile	Serbia	1.0437 × 10−1	Not reject	4.8492 × 10−1	Not reject	3.7662 × 10−3	Reject	2.9566 × 102	Not reject	9.2208 × 104	Not reject	There is no difference, except scientific thinking
Chile	Spain	8.3006 × 10−4	Reject	7.0545 × 10−1	Not reject	9.0058 × 10−3	Reject	5.9033 × 10−4	Reject	1.2300 × 101	Not reject	There is difference in all types of thinking
Colombia	Mexico	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Colombia	Pakistan	4.3336 × 105	Not reject	7.1149 × 105	Not reject	9.0988 × 105	Not reject	1.0000 × 106	Not reject	5.0765 × 105	Not reject	There is no difference in any type of thinking
Colombia	Finland	7.3759 × 10−1	Not reject	1.5027 × 101	Not reject	2.2681 × 102	Not reject	4.4281 × 10−1	Not reject	2.1344 × 103	Not reject	There is a difference in all types of thinking
Colombia	Philippines	5.9546 × 103	Not reject	3.5118 × 104	Not reject	1.3764 × 103	Not reject	4.8587 × 104	Not reject	1.1291 × 104	Not reject	There is difference in all types of thinking
Colombia	Serbia	2.2210 × 10−2	Reject	1.2630	Not reject	3.0878 × 10−3	Reject	4.5542 × 101	Not reject	9.0469 × 102	Not reject	There is difference in all types of thinking
Colombia	Spain	1.4586 × 10−3	Reject	1.4888	Not reject	1.0956 × 10−1	Not reject	3.1070 × 10−4	Reject	1.4223 × 10−2	Reject	There is difference in all types of thinking
Finland	Pakistan	8.3424 × 104	Not reject	4.8183 × 105	Not reject	3.5815 × 105	Not reject	2.2310 × 104	Not reject	1.0000 × 106	Not reject	There is difference in critical thinking
Finland	Philippines	4.4405 × 105	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.6717 × 104	Not reject	1.0000 × 106	Not reject	There is difference in critical thinking
Finland	Serbia	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Finland	Spain	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Finland	Mexico	6.7505 × 10−1	Not reject	1.7825 × 101	Not reject	4.3456 × 102	Not reject	5.1851 × 10−1	Not Reject	7.2094 × 103	Not Reject	There is difference in all types of thinking
Mexico	Pakistan	9.5524 × 105	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Mexico	Philippines	1.0369 × 104	Reject	7.2216 × 104	Not reject	2.1615 × 103	Not reject	7.4429 × 104	Not reject	4.0273 × 104	Not reject	There is difference in complex, scientific, and innovative thinking
Mexico	Serbia	5.7012 × 10−3	Reject	1.9652 × 10−1	Not reject	2.4020 × 10−3	Not reject	6.0831 × 101	Not reject	2.9980 × 103	Not reject	There is difference in all types of thinking
Mexico	Spain	2.8940 × 10−5	Reject	2.7762 × 10−1	Not reject	1.0275 × 10−2	Reject	1.7664 × 10−5	Reject	1.7292 × 10−2	Reject	There is difference in all types of thinking
Pakistan	Philippines	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking
Pakistan	Serbia	2.7204 × 104	Not reject	6.3945 × 105	Not reject	2.7842 × 102	Not Reject	3.7076 × 105	Not reject	1.0000 × 106	Not reject	There are differences in complex and scientific thinking
Pakistan	Spain	1.6464 × 104	Not reject	6.4721 × 105	Not reject	3.8392 × 104	Not reject	7.7721 × 103	Not reject	6.8595 × 104	Not reject	There are differences in complex, critical, and scientific thinking
Philippines	Serbia	1.4312 × 105	Not reject	1.0000 × 106	Not reject	1.8351 × 104	Not reject	4.6375 × 105	Not reject	1.0000 × 106	Not reject	There is a difference in scientific thinking
Philippines	Spain	7.9306 × 10+04	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	1.7144 × 103	Not reject	2.3290 × 104	Not Reject	There is a difference in innovative and critical thinking
Serbia	Spain	1.0000 × 106	Not reject	1.0000 × 106	Not reject	6.3372 × 106	Not reject	1.0000 × 106	Not reject	1.0000 × 106	Not reject	There is no difference in any type of thinking

Source: own elaboration using the onewaytest package of R (Dag et al. 2023).

,

Table 7. Comparison of groups by Bonferroni correction (Alpha = 0.05) per geographic area.

Area 1	Asia	Asia	Europe
Area 2	Europe	Latin America	Latin America
p-value CoT	2.1766 × 106	5.7083 × 101	6.6036 × 10−21
Ho CoT	Not Reject	Not Reject	Reject
p-value ST	7.9039 × 103	6.7097 × 102	6.7839 × 10−14
Ho ST	Not Reject	Not Reject	Reject
p-value ScT	2.3999 × 102	9.0883	2.6378 × 10−17
Ho ScT	Not Reject	Not Reject	Reject
p-value CT	5.6882 × 10−1	9.8983 × 102	4.0215 × 10−18
Ho CT	Not Reject	Not Reject	Reject
p-value IT	1.0381 × 103	6.9971 × 102	7.9461 × 10−9
Ho IT	Not Reject	Not Reject	Reject
Interpretation	No significant difference in types of thinking	No significant difference in types of thinking	There is a difference in all types of thinking

Source: own elaboration using the onewaytest package of R (Dag et al. 2023).

and

Table 8. Comparison of groups using the Bonferroni correction (Alpha = 0.05) per area classification in the HDI.

IDH1	High	High	Low
IDH2	Low	Very High	Very High
p-value CoT	6.75 × 105	1.63 × 10−6	1.72 × 104
Ho CoT	Not Reject	Reject	Not Reject
p-value ST	7.18 × 105	1.87 × 10−2	2.03 × 105
Ho ST	Not Reject	Reject	Not Reject
p-value ScT	1.00 × 106	1.68 × 106	1.02 × 104
Ho ScT	Not Reject	Not Reject	Not Reject
p-value CT	8.64 × 105	1.51 × 10−7	1.27 × 104
Ho CT	Not Reject	Reject	Not Reject
p-value IT	5.46 × 105	1.83 × 10−3	1.25 × 105
Ho IT	Not Reject	Reject	Not Reject
Interpretation	No significant difference in types of thinking	Significant differences exist in some types of thinking	No significant difference in types of thinking

Source: own elaboration using the onewaytest package of R (Dag et al. 2023).

). This test is known for being extremely conservative, as it adjusts the significance level by dividing it by the number of tests (α/n), thereby significantly reducing the risk of Type I errors.

5. Discussion

The results of this study reveal significant differences in the perceived achievement of the complex thinking competency and its subcompetencies among university students from different countries, geographic areas, and Human Development Index (HDI) levels. However, no significant gender differences were found.

The absence of significant gender differences in the perception of complex thinking is consistent with the gender similarity hypothesis proposed by Hyde (Hyde 2005), who argues that men and women are more similar than different in most cognitive skills. This finding suggests that, in the current context, educational opportunities and pedagogical practices may be promoting gender equity in the development of advanced cognitive skills. Moreover, it contradicts previous studies that indicated gender differences in certain cognitive skills (Halpern 2012), which could be due to changes in educational policies and greater awareness of the importance of gender equality in education (UNESCO 2019).

On the other hand, the significant differences found between countries and continents support the idea that the cultural and educational context influences the perception and development of complex thinking. Nisbett (Nisbett 2003) posits that Western cultures tend toward analytical and critical thinking, while Eastern cultures favor holistic and systems thinking. This aligns with our findings, where students from different regions showed variations in their self-perception of their complex thinking subcompetencies. For example, students from Asian countries may perceive greater skills in systems thinking, influenced by cultural values that emphasize harmony and interpersonal relationships (Markus and Kitayama 1991).

The influence of HDI on the perception of competencies suggests that socioeconomic and human development factors affect the self-assessment of cognitive skills. Zhao (Zhao 2010) argues that, in countries with higher human development, education systems tend to have more resources and policies that encourage innovation and critical thinking, which could explain the observed differences. Likewise, students in countries with lower HDI may have less access to quality educational resources, affecting their confidence and perception of competencies (OECD 2018a).

It is important to consider that self-perceived competence may be subject to cultural response biases. Heine et al. (Heine et al. 2002) note that individuals from different cultures may vary in their tendency to evaluate themselves positively or negatively, influenced by cultural norms about modesty or self-esteem. This could have affected participants’ responses, especially in cultures where modesty is valued and self-criticism is common.

From the results and the methodology employed in the study, we identified the following significant findings:

• Absence of Normality and Homoscedasticity: The Henze–Zirkler and Shapiro–Wilk tests confirmed that the data do not follow a normal distribution, and the Levene test indicated the lack of homoscedasticity. This justified the choice of the Brown–Forsythe test for the analysis, which is suitable for data that do not meet these statistical assumptions.
• The results of the Brown–Forsythe test are as follows:

  -

  By Gender: No statistically significant differences in self-perceived achievement in the different types of thinking were found between males and females. This result suggests that gender does not significantly influence how graduates perceive their ability in complex systems, scientific thinking, critical thinking, and innovative thinking.

  -

  By Country and Geographic Area: There were significant differences in self-perceived achievement among different countries and geographic areas (continents). This indicates that cultural and regional contexts could have a substantial impact on how graduates value their thinking skills.

  -

  By HDI Level: Significant differences were also observed according to the 2022 Human Development Index (HDI) rank. This could reflect how socioeconomic development conditions influence education and self-assessment of cognitive and thinking skills.

• Post Hoc analysis with Bonferroni correction: For variables where significant differences were found (country, geographic area, and HDI), a Post Hoc analysis compared means between specific groups using Bonferroni correction, which helps to control Type I errors in multiple comparisons.

The study’s findings contribute significantly to multicultural learning theory and the analysis of cognitive development in various contexts. From a theoretical perspective, the results underscore the importance of considering socioeconomic and cultural contexts as a determining factor in the self-perception of complex thinking skills. This challenges existing theories that assume uniformity in the acquisition and assessment of cognitive skills across different contexts, suggesting instead that cognitive development and its perception may be deeply contextual. Furthermore, the absence of significant differences in perception between genders may call into question theories that postulate innate or socially constructed differences in cognitive abilities between men and women, indicating that perceptions of ability may be more influenced by external factors than by gender per se.

In practical terms, these findings have direct implications for the design and implementation of educational policies and development programs. Educators and policymakers should consider adapting curricula and teaching methods to reflect and respect students’ cultural and socioeconomic diversity. This could involve integrating culturally relevant teaching materials and pedagogical approaches that promote inclusion and equity in education. Additionally, the understanding that the level of human development (HDI) influences self-perceived thinking skills suggests that educational interventions should mainly target improving education in low HDI regions, potentially through policies that encourage more investment in educational resources and teacher training in these areas.

6. Conclusions

The objective of this study was to evaluate students’ perceived achievement of complex thinking skills and their sub-competencies in ten universities in eight countries, seeking to identify statistically significant differences. The Brown–Forsythe test, selected due to the lack of normality and homoscedasticity in the data, allowed us to reject the null hypothesis that there are no differences in group means by country, geographic area, and HDI level, thus highlighting the influence of socioeconomic and cultural environments on self-perceived cognitive abilities. However, the same hypothesis was not rejected with respect to gender, indicating the absence of significant differences in the perception of thinking skills between men and women.

Despite these findings, the study has limitations, such as reliance on self-perceptions, which may be influenced by social desirability biases. The sample distribution may not be representative of all geographic areas. In addition, other variables of the educational environment that could affect the perception of skills were not controlled. Future research could explore the impact of educational interventions in low HDI contexts, longitudinal studies to follow the evolution of these perceptions, and the effect of more specific educational factors on the self-perception of thinking skills. These directions would advance our understanding of how academic and cultural conditions affect cognitive development and guide the development of more inclusive and effective educational policies worldwide.