2.1 Data processing and analysis

The processing and normalization of the data, as well as the execution of the analyses, were carried out using Excel software and the R programming language. The following actions were taken regarding processing and normalization:

a) The degrees reported in the Lattes CV were grouped. Initially, the terms “degree," "graduate," "baccalaureate and "bachelor" were eliminated from the degree descriptions, as their relevance was deemed negligible.

b) A subsequent investigation revealed that 0.3% (six cases) of the total number of researchers possess two master's degrees, while 0.2% (five cases) hold two doctorates. It was decided to maintain the area and period of the last master's and doctorate. The areas of the degrees (undergraduate, master's, and doctorate) were grouped based on the radicals or common terminations of the words, and the set of letters "engi" was always linked to engineering courses. Consequently, all terms containing "engi" in the engineering field were consolidated into a single group. This process was repeated multiple times for each degree level, with the CAPES⁴ areas and sub-areas serving as a reference. For instance, master's and doctoral degrees belonging to the fields of "mathematics education," "science education," "science and mathematics education," "science teaching," "science and mathematics teaching," and "science education," among other combinations, were classified under sub-area 90201000 Science and Mathematics Teaching. During this process, numerous typing errors were identified, as some CVs listed the degrees in the following spellings: "pedgogia," "pegagogia," "bilogia," among others. In such cases, the data has been standardized.

It was determined that 28 researchers (representing 1.4% of the total sample) did not report their graduate coursework in their CVs, despite having reported their master's and doctoral degrees. In these cases, the degrees were not considered; only the other degrees were reported.

e) The multivariate linear regression method (^{AGRESTI and FINLAY, 2012}) was used to assess the relationship between the number of articles in journals by RGs and the proportion of researchers with doctoral degrees controlled by the areas of interest of education and psychology (see Graph 6).

3.1 Profile of the researchers who constitute the Research Groups

A total of 96 RGs were analyzed, 64 of whom were registered in the field of Education and 32 in the field of psychology. This yielded a total of 1,963 member researchers, of whom 1,263 were members of the Education area and 674 of the Psychology area. The analysis revealed that 26 researchers are affiliated with RGs from both domains, with an average of 20.4 researchers per RG.

Graphs 1 (A-C) illustrate the academic backgrounds of the researchers affiliated with Education and Psychology RGs, categorized by degree level (undergraduate, master's, and doctoral). The analysis identified a total of 145 distinct graduate courses, which suggests that the RGs possess a multidisciplinary background (see Graph 1A).The three graduate courses that were most frequently identified were Pedagogy (n = 735), representing 31.2% of the total, followed by Psychology (n = 648), accounting for 27.5%, and Mathematics (n = 176), constituting 7.5%.

Source: Prepared by the authors, based on survey data

Graph 1 Degree areas of the researchers who are members of the CNPq Research Groups in the areas of Education and Psychology: (A) Bachelor's, (B) Master's and (C) Doctorate with the highest frequency 

A comparison of the data (Graph 1A) reveals a discrepancy between the number of RGs in the Education area and the number of RGs in the Psychology area. While the former has twice the number of members as the latter, this ratio is not proportionally reflected in the areas of graduation. Consequently, RGs in Education are more diverse in their graduate course areas and exhibit greater heterogeneity.

In the context of postgraduate studies (see Graphs 1B-C), a discernible gap emerges between the Education and Psychology domains. This disparity persists across both masters and doctoral programs, suggesting that researchers specializing in RGs may predominantly employ a cultural-historical theoretical framework within the educational domain, with challenges arising predominantly from the Education sector rather than the Psychology domain.

This observation is further substantiated by the predominance of researchers holding postgraduate degrees in Science and Mathematics Teaching across both master's and doctoral programs. The substantial presence of researchers with postgraduate qualifications in this field aligns with the findings reported by ^{Camillo and Mattos (2019}). The findings of this study corroborate the results obtained by ^{Bonfim, Solino, and Gehlen (2019}) on the use of Vygotsky as a theoretical reference in dissertations and theses in the area of Science Education in Brazil. According to these authors, there has been a significant increase in theses and dissertations defended between 1991 and 2016 that use Vygotskian concepts. In this study, the majority of researchers obtained their doctoral degrees in Science and Mathematics Teaching from 2016 onwards.

Graph 2illustrates the temporal progression of the primary domains in which doctoral dissertations were submitted for consideration. Among the 1,963 researchers from the RGs who were examined, 683 (34.8%) have successfully completed their doctorates, which were conferred between 1979 and 2020.

Source: Prepared by the authors, based on survey data

Graph 2 Number of researchers who are members of CNPq Research Groups in the areas of Education and Psychology with a doctorate by area and year of graduation 

Accordingly, the first doctorate was conferred in 1979 in the field of psychology, followed by the conferral of the first doctorate in education in 1984. Although a PhD defense in mathematics science teaching was conducted in 1983, this was an exception. In the area of science and mathematics teaching, the majority of doctorates were awarded after 2013. It is noteworthy that the first doctoral defense in a field apart from psychology, education, science and mathematics teaching, and medicine occurred in 1989. However, it was not until 1995 that other fields began to emerge more frequently. Given the diversity of doctoral programs within the RGs, it is essential to examine the trajectory of those who have completed or are currently completing a doctorate. Graph 3illustrates the movement of the areas in which all researchers with a doctorate have completed or are in progress, with n = 893 researchers constituting 45.5% of the total.

Caption: The colors used refer to the area of the doctorate. Red indicates a doctorate in education, while blue indicates a doctorate in psychology. Source: Prepared by the authors, based on survey data.

Graph 3 Movements in the areas of the 893 researchers who are members of the CNPq Research Groups in the areas of Education and Psychology with completed or ongoing doctorates 

As illustrated in Graph 3, two distinct pathways were identified: 1) the attainment of a master's degree in education and a doctorate in education was indicative of graduation in pedagogy, which corresponded to 193 researchers (21.6%), and 2) the possession of a master's degree in psychology and a doctorate in psychology was indicative of a degree in psychology, which corresponded to 168 researchers (18.8%). Notably, the transition between these two areas corresponded to more than half of the researchers with a completed doctorate or a doctorate in progress (59.0%), indicating a significant shift in academic trajectories. For instance, some researchers initially pursued a degree in psychology at the undergraduate level but subsequently obtained a doctorate or are currently enrolled in a doctorate program in Postgraduate Programs in Education. It is also noteworthy to examine the academic background of individuals who hold masters and doctoral degrees in Science and Mathematics Teaching. A significant proportion of these individuals (11 researchers) possess a degree in physics, with most concentrating in a single RG. This observation underscores the prevalence of diverse pathways leading to doctoral degrees. Graph 4 provides a visual representation of the graduation dates of the researchers comprising the 96 RGs.

Source: Prepared by the authors, based on survey data.

Graph 4 Year of master's and doctoral degrees awarded to researchers in the 96 CNPq Research Groups in the areas of Education and Psychology 

One interpretation of the decline in master's degrees from 2014 and doctorates from 2017 may be associated with the delay in updating the CVs of its members on the RG portal, as well as the restriction of funds from primary scientific research funding agencies, as Silva Júnior and Fargoni (2021) have noted.

3.2 Scientific production

A comprehensive analysis of the scientific output of the 1,963 researchers affiliated with the RGs from 2000 to 2020 revealed a total of 43,876 presentations at scientific events, 2,851 books, 9,410 book chapters, 13,956 full papers in conference proceedings, and 10,974 full articles in journals. On average, researchers present 2,000 papers at national and international scientific events per year. It is important to note that these figures represent the total scientific output per researcher. However, some publications have multiple authors, and both authors are members of the RGs analyzed. Therefore, if each author included the publication in question in their respective Lattes CV, then it was counted more than once when surveying with Script Lattes. Consequently, a comprehensive analysis of individual scientific production is being developed for future studies.

A close examination of the distribution of the number of articles published in journals over the years reveals a substantial increase, with two periods of particular note: i) from 2008 to 2012, there was a continuous growth. The number of published articles doubled during this period compared to previous years. The second significant increase in publications occurred from 2019 to 2020, with a 25.9% growth rate, the highest percentage growth observed during the entire study period.

However, the findings do not allow for the establishment of a causal relationship with the points made in the literature, as previously noted by Salém (2012), ^{Silva and Hai (2016}), and ^{Bonfim, Solino, and Gehlen (2019}).The authors of the study attribute this increase to the growing use of a Vygotskian theoretical framework in recent decades. While the results of this study demonstrate an increase in the number of dissertations and theses defended, it cannot be inferred that this is correlated with the number of articles published in journals by the same researchers. In other words, it cannot be guaranteed that the continuous growth over the years in the number of articles published in journals is the result of an increase in the number of dissertations and theses defended.

Source: Prepared by the authors, based on survey data.

Graph 5 Distribution of the number of articles in journals by researchers who are members of CNPq Research Groups in the areas of Education and Psychology by year (2000-2020) 

Finally, about academic production, Graph 6presents three variables: 1) the number of articles in journals per RG, distinguishing between Education and Psychology; 2) the proportion of researchers with completed doctorates in RG; and 3) the number of researchers per RG.

Caption: Education Research Groups: 64; Psychology Research Groups: 32. The count of articles for each RG is per researcher, with 26 researchers belonging to two or more RGs classified in both areas, Education and Psychology; 53 researchers belonging to two or more RGs classified in the area of Education and 33 researchers belonging to two or more RGs classified in the area of Psychology. The filled line indicates the linear regression, and the gray shading indicates the standard deviation of the sample. Source: Prepared by the authors, based on survey data.

Graph 6 Distribution of the number of articles in journals by CNPq Research Group in the areas of Education and Psychology, 2000-2020 

Graph 6 shows two line segments representing the average number of journal articles per RG. Both segments have a positive slope, indicating that, on average, as the proportion of Ph.D. researchers per RG increases, so does the number of articles published by RGs. The size of the circle indicates the number of researchers per RG. The smallest group has two members and the largest 110.

The statistical treatment of the data presented in Figure 6 allows us to conclude that

i) There are no significant differences in the production of articles between the different fields of education or psychology. Both fields contribute equally to research based on cultural-historical theory.

ii) There are no significant differences in the production of articles by the different RGs. Only three of the 96 RGs produced significantly more than the average.

iii) The number of articles in journals is statistically significant in relation to the proportion of Ph.D. researchers in the RGs and in relation to the number of researchers per RGs⁵.

Therefore, in general, research in cultural-historical theory, represented by the RGs of education and psychology found by ^{Asbahr and Oliveira (2021}), and has developed symmetrically between the fields.

3.3 Training human resources

Given the significant contribution of the members of the RGs to the production of scientific knowledge, it is necessary to determine the number of supervisions completed or in progress by the researcher-supervisors.

Of the 1,963 researchers who form the 96 RGs, 315 (16.0%), researchers were responsible for 3,782 master's degrees, and 147 (7.5%) researchers were responsible for 1,016 doctoral degrees between 2000 and 2020, for a total of 4,798 masters and doctoral degrees. A smaller group of researchers, 60 (3.1%), were responsible for supervising 178 postdoctoral students. The growth in the number of M.Sc. and Ph.D. graduates per year was linear, as shown in Graph 7.

Graph 7 Number of master's and doctoral degrees awarded by researcher-supervisors who are members of CNPq Research Groups in the areas of Education and Psychology between 2000 and 2020 

The decrease in the number of master's and doctoral degrees awarded by researcher-supervisors who are members of RGs in 2020 may be related to the coronavirus pandemic, which has changed the dynamics of teaching and research institutions. In this scenario, some postgraduate programs, such as the Inter-University Postgraduate Program in Science Teaching (USP), exceptionally allowed an extension of the defense deadlines.