Several studies have been published about research productivity in educational psychology between 1991 and 2021 (Fong et al., 2022; Greenbaum et al., 2016; Hsieh et al., 2004; Jones et al., 2010; Smith et al., 1998, 2003). In these studies, a “standard” set of journals have been repeatedly used to investigate individual and institutional productivity (i.e., Contemporary Educational Psychology, Educational Psychologist, Educational Psychology Review, and the Journal of Educational Psychology). In addition to this standard set of journals, the previous ranking studies included several other journals to examine productivity (e.g., Cognition and Instruction; see Table 1). These studies have shown that some highly productive researchers in educational psychology are publishing at a high rate in top-tier journals, in some cases even over several decades (e.g., Patricia A. Alexander, Herbert W. Marsh, and Richard E. Mayer).

In addition to ranking studies that have examined researchers' productivity in educational psychology in general, there are also studies that have focused specifically on women's productivity in the field (Evans et al., 2005; Fong et al., 2009; Greenbaum et al., 2018; Griffin et al., 2023; Robinson et al., 1998). Furthermore, qualitative interviews have been conducted in follow-up studies with a small number of highly successful educational psychologists (Flanigan et al., 2018; Kiewra and Creswell, 2000; Kiewra et al., 2021; Patterson-Hazley and Kiewra, 2013) and specifically with selected highly productive female educational psychologists (Mayrath, 2008; Prinz et al., 2020). In these studies, researchers answered questions, for example, about important mentors and their strategies for successfully managing their time and research resources.

As a traditional continuation of previous rankings, the most recent study by Fong et al. (2022) identified the highly productive (early career) researchers and the top-producing institutions for the period from 2015 to 2021.

One major limitation of prior research on productivity in educational psychology is the unsystematic selection of journals used to determine individual productivity. An analysis of the journals included in previous productivity studies shows that most studies (i.e., Fong et al., 2022; Greenbaum et al., 2016; Hsieh et al., 2004; Jones et al., 2010; Smith et al., 1998, 2003) examined articles published primarily in the four “standard” journals of educational psychology (i.e., Educational Psychologist, the Journal of Educational Psychology, Educational Psychologist, and Educational Psychology Review) and additionally in Cognition and Instruction. So far, the same set of journals has been selected to assess the productivity of researchers over time, without a convincing rationale as to why these five journals in particular were selected and other top-tier journals were not. One of the earlier productivity studies (Smith et al., 1998) described its rationale as follows:

Fong et al. (2022) reported that these five journals were chosen as they have been selected in previous productivity studies (for a similar reasoning, see Greenbaum et al., 2018; Hsieh et al., 2004), and that they were “representative” (p. 2) journals within educational psychology (see also Griffin et al., 2023).

But what exactly do the authors mean by representative? In prior studies on productivity, no objective criteria were specified to identify representative journals in educational psychology. Smith et al. (1998) referred to Walberg and Haertel (1992), who regarded journals to be core journals when “identified in the 1988 Social Sciences Citation Index” as “those that are frequently cited by each other and by related journals from allied fields” (p. 13). Based on this criterion, Smith et al. (1998) selected the Journal of Educational Psychology, Educational Psychologist, and Contemporary Educational Psychology. Furthermore, they selected Educational Psychology Review and Cognition and Instruction (CI) because “CI has, arguably, become synonymous with work in the field of educational psychology” (p. 175) and “it seems reasonable to assume that many educational psychologists target their work for publication in these journals and read the contents of most of these journals” (p. 175).

This set of journals originally selected by Smith et al. (1998) became the “standard” set of journals for productivity studies, and has never been actively challenged or updated toward the ever-changing, increasingly international landscape of educational psychology journals. Thus, the journal selection criteria in previous ranking studies were unclear, and if available, outdated, failing to adequately reflect the selected journals' evolving status and quality. This begs the question of how these “Champions League Journals” (Alexander Renkl in Flanigan et al., 2018) should be characterized.

We argue that journals used to determine the productivity of researchers in educational psychology should be both high quality and broad in scope, and should be selected based on (relatively) objective and valid criteria for all journals in the field of educational psychology. In the present study, we defined top-tier journals as those journals that were among the top 10 journals in educational psychology based on their average rank in the Web of Science “Psychology, Educational” list from 2017 to 2022. This list ranks the impact and significance of journals within the field based on a variety of criteria, including editorial quality, international score, and citation counts. As the research landscape changes over time, not all of the previously included journals have maintained their status. For example, Cognition and Instruction has not been a part of the top 10 in the Web of Science “Psychology, Educational” list since 2019 and has an average rank of 17.17 between 2017 and 2022 (for a similar argument, see also Griffin et al., 2023). In contrast, Learning and Instruction and the Journal of the Learning Sciences have been top-tier journals for years, but have never been included in productivity rankings (e.g., Greenbaum et al., 2016; Hsieh et al., 2004; Jones et al., 2010; Smith et al., 1998, 2003). To date, the study by Fong et al. (2022) is the only study to include articles published in Learning and Instruction, which resulted in notable changes in the productivity ranking compared to the “standard” set of journals and Cognition and Instruction. However, no objective criterion was reported for the decision to include Learning and Instruction, and only one of two scoring methods was applied (i.e., the count-based method, see below). Thus, it is unclear why only this journal was added, and only one scoring method was used to rank the productivity with this new set of journals. In the present study, we defined broad-scope journals as those that cover a broad range of key topics within educational psychology and are not tied to specific subfields, populations, or applied contexts. For this reason, we excluded journals that target specific subfields within educational psychology (e.g., Child Development).

A second limitation of previous productivity studies is the way the number of published articles is scored, which also impacts the measurement of researchers' productivity. Early studies used different scoring methods to operationalize research productivity, resulting in partially divergent results (Hsieh et al., 2004). For example, Smith et al. (1998, 2003) selectively used a point-based score to measure research productivity, while Hsieh et al. (2004) used a count-based score. The count-based score is derived by simply counting the number of articles published, regardless of the author position (Hsieh et al., 2004). The point-based score takes author position into account, with higher positions resulting in more points awarded (Smith et al., 1998, 2003; for details, see Section 2). The two scores can be inconsistent, especially when researchers differ in the number of articles and/or their position in the author line. The point score penalizes researchers who work in large teams with many co-authors or with early career researchers who typically serve as the first author in collaborations. In contrast, the count score favors researchers with middle author positions and many collaborations, in which they might not have contributed as much. Recent ranking studies (e.g., Fong et al., 2022; Greenbaum et al., 2016; Jones et al., 2010) used both scoring methods and found differences in the ranking results for research productivity (e.g., the ranking order changed and new researchers entered the top 30 list of highly productive researchers, see Fong et al., 2022).

A third limitation is that prior research has provided little information about the background and diversity of highly productive researchers. This information would help the field better understand highly productive researchers and what strategies, work habits, and/or working conditions may have helped them to reach the top of the productivity rankings. At the same time, such insights could also stimulate discussions about which social groups (e.g., racial/ethnic, socioeconomic, gender) are (or are not) represented at the top of the productivity rankings and why this may be the case.

Fourth, prior research has typically identified a rather small sample of 20–30 productive researchers, which is unlikely to contribute to a diverse representation of individuals. A larger sample of highly productive researchers (a) is likely to provide a more holistic view of the field, (b) provides broader recognition of a greater number of individuals, and (c) is more likely to include and thus encourage emerging and early career researchers.

In this productivity study, we aimed to identify highly productive researchers based on their articles published in top-tier broad-scope journals in educational psychology between 2017 and 2022. To do so, we applied objective inclusion criteria (i.e., average ranking position within the top 10 in the Web of Science “Psychology, Educational” list from 2017 to 2022; see Section 2 for details). This 6-year window aligns with the time frames used in prior ranking studies and reflected the most recent complete publication period available when coding occurred in 2023. By targeting the top 50 most productive researchers, we aimed to achieve a more diverse sample of researchers. To acknowledge the controversy in measuring productivity, we determined the most productive researchers based on both scoring methods. Furthermore, we proposed and calculated a combined score based on the averaged point- and count-based scores to (partially) compensate for their individual shortcomings.

One important goal of this study is also to gain further insight into the characteristics of highly productive researchers in educational psychology. Indeed, the characteristics of the most productive researchers remain largely unexplored, because prior studies have, at most, assessed only singular background information (e.g., the year of PhD and their gender). To this end, we conducted an additional survey study to assess background characteristics (e.g., gender, ethnicity, and first-generation student status), working habits, and care responsibilities of the selected 50 most productive researchers. Women often carry a disproportionate share of domestic and caregiving responsibilities, including childcare, elder care, and other forms of family-related care work (e.g., Horne et al., 2018). Because balancing these responsibilities with the demands of a scientific career can be challenging and may affect women's productivity (e.g., Collins et al., 2021; Jolly et al., 2014), we paid particular attention to gender differences with respect to childbearing, division of care work, and work habits.

An additional goal was to explore general trends and developments in educational psychology, such as the degree to which researchers collaborate (as measured by the average number of co-authors) and the types of studies that are published (e.g., original studies, research syntheses). Furthermore, we analyzed the research topics of highly productive researchers, the number of studies per publication, and the adoption of open science practices.

To analyze researchers' productivity in educational psychology, we examined individuals' number of published articles in all issues of the following top-tier, broad-scope educational psychology journals between 2017 and 2022: Educational Psychology Review, Educational Psychologist, Journal of Educational Psychology, Learning and Instruction, Journal of the Learning Sciences, and Contemporary Educational Psychology. By March 22, 2023, we arrived at this modified list of journals in educational psychology and identified the top 50 most productive researchers from 2017 to 2022 in the three steps detailed below.

As a first step, we determined top-tier journals in educational psychology by selecting the journals with an average rank of ≤ 10 on the Web of Science “Psychology, Educational” list between 2017 and 2022. This resulted in nine journals with an average ranking position of ≤ 10: Educational Psychology Review (average ranking position 1.33), Educational Psychologist (average ranking position 2.00), Journal of Educational Psychology (average ranking position 3.50), Child Development (average ranking position 4.33), Learning and Instruction (average ranking position 5.33), Journal of Counseling Psychology (average ranking position 7.50), Journal of the Learning Sciences (average ranking position 7.83), Contemporary Educational Psychology (ranking position 9.33), and Journal of School Psychology (average ranking position 9.83).

As a second step, we assessed the scope and aims of the journals to determine whether they could be considered broad in scope, that is, journals that cover all topics in educational psychology and that are not tied to specific thematic areas or populations. Based on this criterion of topic generality, three of these journals were excluded: Child Development, which focuses on research on child development; the Journal of Counseling Psychology, which publishes research in counseling psychology for client and non-client populations with a specific focus, for example, on optimizing their potential and enhancing their wellbeing; and the Journal of School Psychology, which centers on psychological and behavioral processes in school settings. The remaining six top-tier, broad-scope journals were Educational Psychology Review, Educational Psychologist, the Journal of Educational Psychology, Learning and Instruction, the Journal of the Learning Sciences, and Contemporary Educational Psychology.

As a third step, we coded all articles from 2017 to 2022 from the selected journals. In total, our database included 7170 articles and 4428 authors. Similar to prior research (e.g., Fong et al., 2022), we calculated productivity scores for each researcher using the count-based and point-based scoring method, resulting in partially overlapping but not identical lists. For the count-based method, we counted the number of articles published independently of the author's position (Hsieh et al., 2004). For example, Researcher A published one article as a first author (with one co-author), two articles as a second author (with two co-authors), and one article as a last author (with three co-authors). Based on the count system, Researcher A receives a score of 4 (i.e., 4 published articles) in total. Based on the point-system scoring, we calculate the credit of Researcher A with the formula of Howard et al. (1987): credit=1.5n-i∑i=1n1.5i-1, with n as the number of authors and i as the ordinal author position. For the first author publication (with one co-author), Researcher A receives a score of 1.52-11.51+1.50 = 0.6 credits; as a second author (with two co-authors), Researcher A earns 1.53-21.52+1.51+1.50 = 0.316 credits for each of the two published articles; for the last author publication (with three co-authors), Researcher A earns 1.54-41.53+1.52+1.51+1.50 = 0.123 credits. In total, Researcher A receives 1.355 credits (1 × 0.6 + 2 × 0.316 + 1 × 0.123) using the point-based scoring method. We then identified the 50 most productive researchers based on the point-based and count-based scoring method¹. Additionally, we created a third ranking list by averaging the ranks of the researchers obtained by the count- and point-based scoring methods. For example, Researcher A has a rank of 1 with the count-based scoring method and a rank of 5 based on the point-based scoring method. Based on the combined method, we averaged both ranks, resulting in a score of 3 (i.e., (1 + 5)/2). We then identified the 50 most productive researchers based on the combined method (i.e., the 50 researchers with the highest combined ranks).

For all 7,170 identified articles, we extracted and coded the following information: publication year, author names, author position, institutional affiliations, and article type. Similar to prior research (e.g., Fong et al., 2022; Greenbaum et al., 2016; Jones et al., 2010), we included editorials (i.e., introductions to special issues), reviews, essays, and interviews, but not errata and in memoria articles. We coded different article/study types using the following a priori categories:

Illustrative example publications for each study type are provided in Supplementary Table S1.1 of the online supplement (OS).

Furthermore, we coded information about the number of first-, last-, and sole-authored articles, number of studies per publication (for original studies), and the degree to which open science practices were used by counting the number of preregistrations and registered reports for all researchers in the rankings.

Based on the frequency of the articles' keywords, we identified up to three descriptors per researcher that best represented the primary topic on which they had published (e.g., self-regulated learning, example-based learning, and multimedia learning). For each researcher, we then calculated the relative percentage of each descriptor using the formula:

For example, for Researcher A, who published four articles, we calculated the relative proportion of research topic from two keywords related to self-regulated learning (2/4 × 100= 50%), three keywords related to example-based learning (3/4 × 100= 75%), and one keyword related to multimedia learning (1/4 × 100= 25%). To derive these descriptors, semantically related keywords were grouped into broader topic clusters when such grouping was necessary. For instance, keywords such as immersion, immersive virtual reality, virtual field trip, and virtual reality were consolidated under the broader descriptor “virtual reality.” One author conducted the initial semantic grouping, and a second author reviewed the proposed clusters for conceptual consistency. Any discrepancies were discussed until agreement was reached.

Lastly, as an indicator of the degree of collaboration, we calculated the average number of authors for each article.

To describe the sample with regards to background characteristics (i.e., diversity), working habits, and working conditions, we contacted individuals who appeared in any of the three lists and sent them an individual link to an online survey using the Qualtrics software (Qualtrics, 2023; a copy of the online survey is available in the OSF repository: https://doi.org/10.17605/OSF.IO/DWQFX). Because these lists were not identical and partially overlapped, they included N = 74 researchers, of whom 71 responded (however, the number of responses varied by question).

After providing informed consent, participants were asked to provide background information (e.g., age, gender, nationality, and ethnicity), academic background information (e.g., year of PhD, years in science, main PhD supervisor(s), academic position, first-generation student status, number of work hours per week, number of PhD students and PostDocs in total), and information about family and household commitment (e.g., number of hours for house work and/or care, nursing or custodial work; for all measured variables, see Supplementary Table S1.2).

The study was approved by the ethics committee of Ruhr University Bochum (EPE-2022–037) and informed consent was obtained from all participants. Participants had the opportunity to complete the survey from February 20 to June 26, 2023 and received up to three reminders, after which we stopped data collection. Because three researchers did not participate in the survey, we obtained basic information for them from Google Scholar, ORCID, ResearchGate, and/or university websites.

Tables 2–4 display the most productive researchers from 2017 to 2022 using the count-based, point-based, and combined method, respectively. The point-based method identified 50 highly productive researchers, the count-based method resulted in a list of 57 highly productive researchers (because 13 researchers shared rank 45), and the combined method resulted in a list of 50 highly productive researchers. The individual productivity results show that mostly senior researchers were ranked highest. In particular, three researchers appeared in the top 5 in all three rankings (i.e., Richard E. Mayer, Reinhard Pekrun, and Herbert W. Marsh) and one additional researcher appeared in the top 5 on only two of the three lists (i.e., Andrew J. Martin). These four researchers were also listed in prior ranking studies and thereby have proven to be highly productive researchers in the field of educational psychology over the last decades. Notably, a substantial number of early career researchers (here defined as those who received their doctorate in 2013 or later) have emerged as a new generation of top-performing early career researchers who were not included in prior rankings (except for the recent ranking by Fong et al., 2022). Among the most productive researchers, 40% were early career researchers in the point-based ranking, 25% in the count-based ranking, and 37% in the combined ranking. For example, Rebecca J. Collie, Logan Fiorella, Hanna Gaspard, and Vincent Hoogerheide appeared on all three lists of the present ranking study.

In general, the three scoring methods resulted in overlap in the researchers listed, such that 33 researchers (45%) were listed in all three rankings and 17 researchers (23%) were listed in two of the three rankings. However, there were 24 researchers (32%) who appeared in only one of the rankings. Overall, the rankings based on the different scoring methods deviated from each other, suggesting that researchers may employ different publication strategies.

Highly productive researchers (n = 70) reported to have supervised, on average, 17.37 PhD students (SD = 14.45, range: 0–60), and 5.60 PostDocs (SD = 7.32, range: 0 and 40) during their research career. Information regarding the main supervisor(s) and inspiring role model(s) of the highly productive researchers are provided in the OS (Supplementary Tables S2.2, S2.3).

Open science practices were rarely used by the most productive researchers. None of their studies were registered reports and only 1.38% (n = 8) were preregistered. Among the most productive researchers, nine had authored preregistered articles. The first preregistered study in our sample was published in 2020. There were four preregistered studies published in 2021 and three in 2022. Two preregistered studies were published in each of the journals Contemporary Educational Psychology, Educational Psychology Review, Journal of Educational Psychology, and Learning and Instruction. However, in Educational Psychologist and Journal of the Learning Sciences, there were no preregistered studies published.

The average weekly hours spent in a typical term on work, teaching, housework, as well as care, nursing, or custodial work are shown in Table 10.

In terms of the distribution of working hours, highly productive researchers dedicated 40.73% (SD = 17.68) of their work hours to research (n = 71), 20.86% (SD = 11.69) to teaching (n = 62), 19.16% (SD = 9.47) to supervision (n = 66), and 17.89% (SD = 14.16) to administrative tasks (n = 62) during a typical term. The rest of their working hours (9.49%, SD = 7.55) was spent on other tasks.

Furthermore, among highly productive researchers, 70% confirmed their involvement in some form of care work at various points in their careers, while 28% reported no such involvement (n = 71). At the time the survey was administered, 39% of highly productive researchers are engaged in care work, while 31% have done so in the past. In this group of scholars, 68% reported having at least one child (n = 69) and providing an average of 12.92 years (SD = 7.37) of care, nursing, or custodial work during their academic careers (n = 45). Highly productive researchers managed, on average, 47.92% (SD = 19.66) of their care responsibilities, indicating that the remainder was shared with others (n = 50).

As can be seen in Table 10, there were no gender differences in the group of highly productive researchers regarding the number of working hours, the amount of teaching, and housework. Furthermore, both genders allocated similar amounts of time to research, teaching, supervision, administrative, or other activities, and appear to be comparable in terms of child rearing and care during their careers [women: 82%; men: 66%; χ²₍₁₎ = 1.471, p = 0.225]⁴. Additionally, women were as likely as men to have children (b = −0.02; 95% CI [−0.48, 0.43]). Among those highly productive researchers who have provided care work in the past, women (14.00 years, SD = 7.68) reported a similar amount of caregiving as men (17.85 years, SD = 8.32; 95% CI [−3.45, 11.14]). However, women overall reported to take a significantly greater share of care responsibilities per week (M = 58.48%, SD = 20.70) than men (M = 38.93%, SD = 13.52%; b = −14.48, 95% CI [−26.75, −2.21]), and reported significantly higher hourly commitments to care, nursing, or custodial work during their academic careers (M = 35.00 h, SD = 27.14) than men did (M = 20.52 h, SD = 12.42; b = −14.48, 95% CI [−26.75, −2.21]).

Overall, the count-based, point-based, and combined methods resulted in different lists, with a substantial overlap in the researchers listed (i.e., 45% of the researchers were identified by all three scoring methods, 23% were identified by only two scoring methods, and 32% were identified by only one scoring method). Variations between these lists might reflect that highly productive researchers have different publication strategies and/or different roles in collaborative projects. One such strategy involves taking on more supervisory roles in research teams and thus occupying the last author position, which is penalized in the point-based scoring method but not in the count-based method. For example, based on the point-based (but not count-based) scoring method, Patricia A. Alexander (#2) and Logan Fiorella (#7) were ranked among the 10 most productive researchers, suggesting they are less likely to occupy the last author position in their publications. In contrast, based on the count-based (but not the point-based) scoring method, Ulrich Trautwein (#2), Oliver Luüdtke (#6), and Tamara van Gog (#10) appeared among the top 10 (for the full lists of highly productive researchers based on each scoring method, see Tables 2–4).

As in prior ranking studies, senior researchers achieved the highest ranks. Specifically, from 2017 to 2022, three researchers (i.e., Richard E. Mayer, Reinhard Pekrun, and Herbert W. Marsh) appeared in the top 5 across all three rankings. Notably, Richard E. Mayer and Herbert W. Marsh have consistently been listed among the top-performing researchers since 1991 (Fong et al., 2022; Greenbaum et al., 2016; Hsieh et al., 2004; Jones et al., 2010; Smith et al., 1998, 2003). Using the point-based ranking method, Patricia A. Alexander has additionally been listed among the top-performing researchers since 1991, and for the count-based method John Sweller has additionally been included. Alongside established researchers, the rankings also shed light on the rise of early career researchers. Notably, 40% of the top 50 in the point-based ranking were early career researchers, whereas 25% of the individuals in the count-based ranking were early career researchers, and 37% in the combined ranking (e.g., Rebecca J. Collie, Logan Fiorella, Hanna Gaspard, and Vincent Hoogerheide appeared in all three rankings). These statistics provide a glimpse into the diversity of contributors to the field, including both seasoned and emerging researchers.

We observed several important publication trends. First, multiple-study publications are still rare, even in the selected top-tier journals in educational psychology. In this analysis, an average of only 1.24 studies was reported per article. Future studies are encouraged to include more than one study per publication to strengthen the evidence and replicability of findings (for a similar argument, see Plucker and Makel, 2021).

Second, most of the studies published in the selected journals by the most productive researchers were quantitative in nature. Most articles had a correlational study design (50%) or an experimental study design (27%), while less than 1% of the studies involved qualitative research (e.g., interview study designs or mixed methods study designs). This finding speaks to the trend in educational psychology toward a greater percentage of observational and correlational studies (Brady et al., 2023; Hsieh et al., 2004), with a decrease in the percentage of experimental studies (Brady et al., 2023).

Third, we observed a persistent trend toward increased collaboration. In the present ranking study, most productive researchers published only 5% of their articles as single authors, while a substantial majority of 95% were collaborations, with an average of 4.20 authors per publication. In prior ranking studies, publications involved fewer authors (see e.g., Fong et al., 2022; Jones et al., 2010). These results also show that, overall, the most productive researchers are “collaborative researchers.”

Fourth, our ranking study is the first to shed light on the adoption of open science practices by assessing the number of preregistered articles and registered reports published by the most productive researchers. We found that only 1.38% of the articles were preregistered, while none of the studies qualified as a registered report. We expect to see more preregistrations in the future, as indicated by the slight increase in preregistrations between 2017 and 2022—likely also due to the increased adoption of open science policies by many journals in recent years (e.g., in the Journal of Educational Psychology, Learning and Instruction). To help our research to become more open, transparent, reproducible, and replicable, future ranking studies are encouraged to continue to assess the extent to which the most productive researchers in educational psychology preregister their hypotheses, sample sizes, and analysis plans. Moreover, we encourage future ranking studies to extend our efforts by considering a wider range of open access practices, such as whether researchers openly share their data, materials, and/or (reproducible) code.

Finally, the research topics of highly productive researchers were very diverse. Our analysis of the keywords used in the publications revealed that the 10 most frequently listed topics (among the top 3 keywords used for publication) were: motivation, quantitative methods, multimedia learning, achievement, self-concept, self-regulated learning, cognitive load, emotions, classroom processes, and engagement. All of these can be considered classic educational psychology topics. While contemporary topics such as digital learning technology and virtual reality appeared once, other emerging trends such as artificial intelligence and personalized learning have not yet become prominent keywords among the most productive researchers in publications from 2017 to 2022. However, we expect that these emerging topics are likely to gain prominence alongside established classical topics in future publications in top-tier, broad-scope educational psychology journals.

On average, highly productive researchers reported working 48.60 h and teaching 6.63 h per week in a typical term. It is noteworthy that they reported spending the largest proportion of their working time (about 41%) on research-related activities, followed at some distance by teaching-related activities, supervision, administrative activities, and other activities. The time spent focusing on research activities is certainly a pivotal factor for individual productivity.

A key contribution of our study is that it sheds light on the diversity and international involvement of highly productive researchers, as these factors are beneficial to the scientific advancement of educational psychology. Our results showed that most of the highly productive researchers self-identified as White (86%) and male (59%), did not have a first-generation student status (60%) but did have a first-generation faculty status (86%), did not speak English as their first language (59%), and have moved within countries for their career (79%) but not across countries (72%). Overall, diversity was limited, particularly with regard to race/ethnicity but also gender.

With respect to gender, women and men within the group of highly productive researchers reported broadly similar behaviors in terms of housework and/or care work, childbearing, and work habits. However, somewhat fewer women than men reach this level of productivity, and highly productive female researchers report more hours engaged in care work per week and a higher share of care work within their family. The underrepresentation of women among highly productive researchers may be linked to the attrition of women at the postdoctoral stage—a period that often coincides with the onset of family responsibilities, making it challenging to manage career uncertainties alongside caring responsibilities. In addition, women's greater caring responsibilities, also observed in the present study, may constrain available research time and thereby reduce productivity (e.g., Ysseldyk et al., 2019).

The underrepresentation of certain racial and ethnic groups among the most productive researchers likely reflects multiple structural factors, including systemic barriers such as historical and ongoing institutional bias and discrimination that create barriers to education and career opportunities, but also a lack of representation and role models in academia (e.g., Dupree and Boykin, 2021; Dupree and Kraus, 2022; Fox Tree and Vaid, 2022). In addition, research in the US context indicates that members of minority groups in academia often carry a heavier service load, including committee work, mentoring, and community engagement (Fox Tree and Vaid, 2022; Syed, 2017), which can reduce time available for research and publication.

Our results also showed substantial global imbalances. Most highly productive researchers had European (45%), North American (39%), or Australian (7%) nationalities, whereas only a minority had Asian nationalities (6%), and none were African or South American citizens. Similarly, most respondents were employed in Northwestern Europe (45%), the United States (34%), Australia (13%), or Canada (6%). These disparities are likely linked to structural inequalities across countries and continents. High-income countries typically invest substantially more in research and development, providing better funding structures, infrastructures, and opportunities for scientific careers. For example, high-income countries spend 2.76% of their GDP on research and development, compared to 1.25% in middle-income and 0.21% in low-income countries [United Nations Educational, Scientific and Cultural Organization (UNESCO), 2023]. Researchers from some countries also face visa restrictions for international conferences, research stays, or relocation to other countries, which can hinder career progression. In addition, locally concentrated clusters of highly productive researchers may reflect training lineages: Highly productive researchers may pass on effective research practices to their doctoral students and postdoctoral researchers, increasing the likelihood that these individuals also become highly productive. We saw this in the advisor/advisee relationships in the top 50 most productive researchers (see below and Supplementary Table S.2.2 of the OS), and this assumption is also supported by expertise research (e.g., Ericsson, 2009).

A central question arising from these findings is why diversity within the research community matters in the first place. Prior research shows that diverse research communities generate more innovative and higher-impact science. Diverse teams produce publications with more citations and higher peer-rated quality (AlShebli et al., 2018; Bosetti et al., 2012; Campbell et al., 2013), and researchers from underrepresented groups contribute disproportionately to scientific novelty (Hofstra et al., 2020). Diversity also improves the epistemic process by broadening research questions, revealing biases in existing models, and enabling more accurate data collection—particularly in studies involving human participants (Intemann, 2009). Beyond demographic aspects, global diversity similarly broadens the epistemic scope of psychological research and strengthens generalizability across sociocultural contexts (Lin and Li, 2023b). Thus, the demographic patterns observed in our sample may indicate missed epistemic opportunities. Promoting diversity within educational psychology is therefore not only a matter of representation but directly relevant to scientific quality and innovation.

To increase diversity in educational psychology, Kucirkova (2023) recommends providing diverse mentors who can help students and early career researchers from marginalized groups navigate the implicit structures and routines of academia (i.e., the “hidden curriculum”). Moreover, strengthening dispositions such as confidence and self-efficacy—often lower in women yet crucial for leadership positions—in professional development is important (Howe-Walsh and Turnbull, 2016). Addressing the extreme work culture in academia, which disproportionately affects women and minorities, could additionally be effective, for example, through training in workload management and wellbeing strategies (Kucirkova, 2023). Universities should cultivate environments that support individuals, especially those from marginalized groups, through failures and setbacks and recognize the diverse experiences and challenges one faces in academic careers (Kucirkova, 2023). Further recommendations include increasing diversity among editors, reviewers, authors, and participants, and introducing recognition systems (e.g., badges) for publications that involve research teams or samples extending beyond White, Educated, Industrialized, Rich, and Democratic (WEIRD) populations (Lin and Li, 2023a,b; Roberts et al., 2020). Several initiatives within educational psychology indicate growing attention to diversity—for example reduced conference fees for researchers from low-GDP countries, childcare services at major conferences, awards recognizing minority scholars, lower open-access fees for scholars from low-income countries, and recent racial reckoning efforts across professional organizations [e.g., American Educational Research Association (AERA), 2020; American Psychological Association (APA), 2021]. These measures represent important first steps, yet more sustained and comprehensive efforts will be necessary to address ongoing structural inequities in the field.

Relative to prior research, we have identified a more extensive list of the most productive researchers (i.e., the top 50) and applied more objective criteria for selecting the top-tier, broad-scope educational psychology journals, increasing the number of target journals from five to six. We focused on publications in these journals because publication volume in top-tier outlets remains a common proxy for research productivity (e.g., Smith et al., 1998) and captures aspects of both quantity and quality (Prinz et al., 2020). Acknowledging that this proxy is reductionistic and does not fully capture the breadth of scientific contributions (Griffin et al., 2023; Schönbrodt et al., 2025), we used both count- and point-based scoring methods and a combined scoring method, and assessed additional indicators of research quality (e.g., number of studies per article, adoption of open science practices). While these enhancements strengthen the present study, several limitations remain and point to directions for future research.

First, although we have extended the list of most productive researchers to 50, future research may benefit from identifying an even higher number of highly productive researchers (e.g., top 500). This would also allow for more diversity in the sample and thereby more fine-grained intersectional analyses (e.g., how different intersectional groups, such as women of color, are represented among highly productive researchers) that may also consider the role of highly productive researchers' socioeconomic status (e.g., parental educational level, income) and its association with researchers' academic career.

Second, while we have considered the authorship position of the most productive researchers in educational psychology, future research may examine more explicitly the role and distributed influence of the collaborative network of co-authors and specify their contributor roles using the metadata of the CRediT taxonomy, at least for more recent publications.

Third, in contrast with prior ranking studies, we focused on top-tier, broad-scope journals within educational psychology, using an objective inclusion criterion of an average rank of ≤ 10 on the Web of Science “Psychology, Educational” list (2017–2022). This threshold expanded the set of journals included in earlier studies, while still restricting the selection to outlets that can reasonably be considered top-tier. Future research may conduct sensitivity analyses to examine potential benchmark effects by varying this threshold (e.g., ≤ 5 or ≤ 15) or the publication window (e.g., 5 vs. 30 years). Furthermore, we excluded both specialized high-impact journals within educational psychology (e.g., Child Development, Journal of School Psychology, Journal of Counseling Psychology) and broader high-impact journals outside the scope of educational psychology (e.g., Educational Research Review, Psychological Bulletin). This decision aligns with the rationale underpinning prior ranking studies: journals that cover the breadth of educational psychology and are not tied to specific subfields or populations offer a more balanced representation of the field than specialized outlets that may disproportionately reflect larger or more active subdomains. However, limiting the analysis to top-tier, broad-scope educational psychology journals may overemphasize productivity within this specific set while overlooking interdisciplinary work published in broader psychology or education journals, as well as high-quality research on specialized topics typically appearing in niche outlets. To capture a fuller range of productivity profiles within and across educational psychology, future research could incorporate both broad-scope journals beyond the educational psychology category and high-impact specialized outlets. Such an expanded scope would not only increase the likelihood of identifying highly productive researchers who publish across diverse contexts but also enable examinations of the external validity of prior ranking results—specifically, the extent to which observed productivity patterns generalize to educational psychology in its broadest sense.

Fourth, future ranking studies should continue to incorporate and emphasize indicators of research quality (e.g., Vazire and Holcombe, 2022). We recommend expanding the evaluation of high-quality science to include the rigorous application of appropriate research methods (e.g., state-of-the art statistical techniques, representative sampling, design-informed power analyses) and open science practices (e.g., preregistration, registered reports, embedded replication studies, and the sharing of data, materials, and reproducible analysis scripts). These indicators could be assessed using an explicit and transparent scoring scheme for research quality (Leising et al., 2022), thereby helping to recognize and incentivize high-quality, impactful, and reliable research. In addition, preregistering future ranking studies themselves—including the operationalization of productivity and the planned scoring procedures—would further strengthen transparency and methodological rigor.

Fifth, it is debatable whether existing ranking studies—including the present one—adequately capture the full breadth of what a “productive researcher” should be and accomplish. Work to date has relied on publication volume in selected journals as a proxy for productivity. We recommend expanding the conceptualization of research productivity to reflect not only the frequency of publication but also the quality, influence, and broader contributions of researchers. Ideally, a productive researcher would regularly publish high-quality work that also has a meaningful and lasting impact. Such impact may be reflected in high citation counts within and outside the field, which has been recently operationalized by field-specific and field-across impact indices of researchers by factoring in the number of citations within and/or across fields (see Wu et al., 2026). Beyond publication count and impact, productivity may also involve translating research into authentic educational settings, generating influential theoretical contributions, and contributing to the scientific ecosystem—for example, by securing external funding, mentoring junior scholars, and building research infrastructures. Active involvement in editorial and administrative roles (e.g., serving as an associate or chief editor, reviewing for journals, organizing conferences) further demonstrates engagement in advancing the field. Incorporating these additional factors would allow future evaluations to more comprehensively capture the multifaceted nature of scientific productivity, though an operational and transparent scoring framework to assess these contributions has yet to be developed. Broadening the definition of productivity may ultimately help incentivize and reward high-quality, high-impact, and high-engagement scholarship.

Sixth, the most productive researchers in our sample were predominantly based in Northwestern Europe and the United States, likely because universities in these regions provide particularly favorable conditions for research (e.g., strong funding structures, high-quality education, and advanced research infrastructures). Understanding who becomes highly productive—and which groups remain underrepresented—can help the field reflect on potential structural barriers and consider how greater diversity may strengthen the epistemic breadth and societal relevance of educational psychology. Although educational psychology operates within a global scientific community, research systems, resources, and publication practices vary substantially across regions. To examine whether the conclusions drawn in the present study also hold in other contexts, future work could conduct analogous ranking studies and surveys within specific continents, countries, or regions with lower publication density, such as Latin America or Africa. Such locally grounded analyses would help determine the extent to which our findings generalize beyond high-income research environments, identify visibility or language-related biases, and shed light on region-specific structural barriers. They may also inform targeted capacity-building efforts for local researchers and institutions and contribute to a more globally inclusive understanding of research productivity in educational psychology. Follow-up interview studies could further identify factors and strategies that enable researchers in diverse global contexts to become highly productive relative to their peers, and shed light on conditions that promote productivity among specific social groups (e.g., by race/ethnicity, socioeconomic status, gender) that are currently underrepresented in productivity rankings (see also Kiewra and Creswell, 2000; Prinz et al., 2020).

Addressing these research avenues would further enhance and complement the assessment of researchers' productivity within educational psychology.

Identifying highly productive researchers and understanding the factors associated with high productivity is valuable to current and future generations of researchers. Research productivity continues to be a key criterion for tenure, merit awards, and research funding (Győrffy et al., 2020). Although the present study cannot make causal claims about why the top 50 researchers are highly productive, students and early career researchers can learn from the environments in which these researchers work as well as from their characteristics, work habits, and publication strategies. Insights from ranking studies may therefore serve as one criterion when choosing a supervisor or institution and when planning an academic career.

One plausible hypothesis is that early career researchers may be more likely to become highly productive when mentored by highly productive scholars. This possibly is reflected in several mentor–mentee pairs in our sample [e.g., Richard Mayer (mentor)—Logan Fiorella (mentee); Tamara van Gog (mentor)—Vincent Hoogerheide (mentee); Herbert W. Marsh (mentor)—Jiesi Guo (mentee)]. As discussed earlier, research productivity encompasses not only publication volume and impact but also engagement within the educational psychology community. Adopting this broader understanding may help encourage early career researchers to pursue high-quality, high-impact, and high-engagement scholarship.

Many top-performing researchers emphasize that influence ultimately matters more than sheer output. As Patricia Alexander put it, “Do not aim to be a prolific scholar; aim to be the best and most influential scholar you can be” (p. 39; Patterson-Hazley and Kiewra, 2013). Accordingly, junior researchers should strive to conduct high-quality, impactful research and contribute to the educational psychology community (for further advice, see also Kiewra et al., 2021; Kiewra, 2025; Prinz et al., 2020).