Abstract
Background
While the importance of physician research has been underscored, a shortage of qualified physicians engaged in research persists. Early exposure to research could potentially ignite medical students’ interest in research, thereby motivating them to pursue research-related careers.
Objective
The study aims to examine early research engagement and medical graduates’ interest in incorporating research into their future career paths.
Methods
This was a national cross-sectional survey administered in 2020, with 152,624 medical students from 119 medical schools in China completing it. We selected and resampled the graduates’ data, and the final sample included 17,451 respondents graduating from 101 medical schools.
Results
For graduates engaged in research, 63.4% (3054) had the interest in integrating research into their future careers. Such interest in research did differ between medical graduates with and without research engagement by linear probability regression (β, 0.50; 95%CI, 0.48 to 0.52), but did not differ in instrumental variable regression analysis (β, 0.31; 95%CI, − 0.18 to 0.80). Furthermore, engaging in research significantly increased the top 50% of academically ranked graduates’ research interest in instrumental variable regression analysis (β, 0.44; 95%CI, 0.01 to 0.86).
Conclusions
Contrary to expectations, research engagement does not necessarily enhance medical graduates’ interest in integrating research into their future careers. However, graduates with strong academic performance are more inclined to develop this research interest. In light of these findings, we propose recommendations for nurturing research interest within medical education.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40670-024-02152-3.
Keywords: Research engagement, Research interest, Medical graduates, Instrumental variable analysis
Introduction
The value of physician research has been made clear; however, there exists a shortage of physician researchers, including physician-scientists and clinical researchers [1–6]. Physician-scientists are uniquely positioned to bridge the gap between laboratory-based research and clinical practice, translating scientific discoveries into innovative medical treatments [3–6], and clinical researchers’ primary focus is on conducting clinical studies that involve human participants, often evaluating the safety, efficacy, and effectiveness of medical treatments, interventions, or procedures in real-world clinical settings [1]. Their contributions have a profound impact on healthcare delivery, medical practice, and ultimately, patient health outcomes [1,2,6]. Despite this, there has been an alarm raised worldwide about the shortage of qualified physicians involved in research across various medical domains [7–12]. One of the issues contributing to the decline is the reduced number of individuals entering research careers in the first place [1].
To reverse such a trend, medical leaders and medical education researchers have called for engaging medical students in research activities [3,13–18]. Medical schools across countries have thus been integrating research training from research electives to projects into both undergraduate and postgraduate education [14,19–23]. The primary aim of early research engagement is to stimulate medical students’ interest in research, and consequently increase physician researchers’ headcount and diversity [1,20,24].
Through a theoretical lens, this interest can be triggered and developed over time through research engagement [25–27]. Defined by Hidi and Renninger (2006, p. 112) as “the psychological inclination to engage or reengage with specific classes of objects, events, or ideas over time,” interest is commonly categorized into situational interest, the affective reaction triggered by the environment; and individual interest, an individual predisposition to reengage a particular content [25]. Building on this classification, Hidi and Renninger (2006) elucidate a four-phase framework delineating the developmental trajectory of interest. Initially, interest is trigged by situational circumstances, evolving into sustained engagement if perpetuated. This sustained engagement may prompt the emergence of individual interest, ultimately leading to a well-established and enduring commitment. As regards medical students’ research interest, a situational research interest can be initiated by research opportunities presented within medical schools, and a temporary, short-term research engagement may cultivate sustained situational interest. Over time, individual interest in research may emerge and mature, signifying medical students’ enduring commitment to engaging in research activities [25].
To date, research has examined the effect of medical students’ research engagement on their situational research interest. Especially, most studies evaluated the short-term effect of research engagement on students’ interest in research, and found that medical students demonstrated increased interest in research after participating in specific research activities [6,28–30]. While fostering situational research interest serves as a crucial initial step, the ultimate goal is to transform this temporary engagement into lasting individual interest [25,26]. Such interest holds the potential to inspire medical graduates to incorporate research seamlessly into their future professional careers.
Nevertheless, relatively little evidence exists demonstrating the impact of medical students’ research engagement on individual research interest, that is, medical students sustained and enduring interest in research [6,28–30]. Besides, prior investigations have predominantly concentrated on the impact of biomedical research [6,28–30], affording limited consideration to alternative forms of research engagement, such as clinical and social science research. Therefore, our study employed samples of medical graduates in China to investigate their interest in incorporating research into future career paths subsequent to their participation in various research activities.
Methods
Data and Sample
In this cross-sectional study, we employed survey data from the 2020 China Medical Students Survey (CMSS). The CMSS aimed to map the landscape of medical students’ overall learning experience and further provide evidence for stakeholders, such as medical schools, and teachers, to improve the quality of medical education in China. The CMSS questionnaire was designed by experts from education, medicine, and management based on the American Medical School Graduate Questionnaire and the American National Survey of Student Engagement [31,32]. It had six parts (e.g., demographic characteristics, learning experiences, and career plans) and had good reliability and validity [33].
The survey has been administered annually each summer using a web-based mode by the National Centre for Health Professions Education Development (NCHPED, link: https://medudata.bjmu.edu.cn/) since 2019. Of the 165 medical schools in China, 119 (72.1%) medical schools volunteered to participate in this survey. The online anonymous questionnaire was administered to 180,348 medical students and 152,645 questionnaires were completed, yielding an 84.6% response rate [34]. This study used the graduates’ data for two reasons: graduates could reflect deeply on their overall research experience and graduates’ interest in research was regarded as an individual interest, rather than situational interest compared with medical students from other grades. We further resampled the graduates’ data according to gender, home location, university location, and self-reported academic GPA to make the sample nationally representative. The final resampled data consisted of 17,451 medical graduates from 101 medical schools. Peking University’s Institutional Review Board granted ethical approval (IRB00001052-20069).
Variables
The CMSS assessed student research engagement with the question, “How many research projects have you engaged in?” Answers included three options: none, one, two or more. Those who responded “one” or “two or more” to this question could respond to another item, “What type(s) of research was/were it/them?” The three responses were biomedical research, clinical research, and humanities and social science research. Medical students’ research interest was evaluated by the question “Would you be interested in incorporating research into your future career?” Replies were “No,” “Neutral,” and “Yes.” To explore whether research engagement promoted medical students’ interest in future research, we dichotomized the responses of research engagement into two categories: non-engagement (none) and engagement (one, two or more), while the responses of medical students’ research interest were divided into a binary variable: yes and no (no or neutral). Those who replied “Neutral” were classified as the “no” group as they did not report interest.
Demographic variables included gender (male vs female), home location (rural vs urban), and parental educational levels. Parental educational levels were measured by years of education the respondent’s father and mother received, respectively. Academic variables included university geographic location (located in the east, middle, or west of China), length of the degree program (5-year program, 8-year program, “5 + 3” program), and GPA (top 50%, bottom 50%). Our study included three types of degree programs in China’s medical education system according to the duration of studies: a 5-year program leading to a Bachelor of Medicine degree, a “5 + 3” program leading to a Master of Medicine degree, and an 8-year program leading to a Doctor of Medicine degree [35,36]. We considered the 8-year program and “5 + 3” program as long-term degree programs which were offered by leading and research-oriented medical schools in China, and the 5-year program was provided by all medical schools.
Statistical Analyses
We calculated descriptive statistics for the demographic and academic variables using proportions because all the variables were categorical except parents’ educational attainments reported as mean. Next, the linear probability regression models were used to explore the association between research engagement and medical graduates’ research interest. The logistical model is commonly used if the dependent variable is binary, but we chose linear probability regression models given the large sample [37]. Those demographic and academic variables, treated as control variables, were controlled for by entering them into the regression models. Finally, we employed the accreditation status of the medical institution as the instrumental variable to address the endogeneity, deeply exploring the relationship.
The accreditation status of medical schools in China served as an instrumental variable in our study. The accreditation activities of undergraduate medical education programs have been steadily conducted by the Chinese Ministry of Education’s Working Committee for the Accreditation of Medical Education since the Ministry of Education of the People’s Republic of China and the National Health Commission of the People’s Republic of China jointly issued the Standards for Basic Medical Education in 2008 [38]. Among the medical schools, 65 were accredited, including 11,811(68.7%) students, whereas 36 schools, with 5633 students (32.3%), were not accredited. Since the external quality evaluation mechanism was established relatively later, the Chinese accreditation procedure was different from the Western where medical schools were qualified to provide education programs after accreditation associations accredit degree programs. The accreditation status of medical schools was suitable for an instrumental variable because: (1) It was strongly associated with medical students’ research interest, irrespective of other confounding variables, as research engagement was listed as an important part of the Accreditation Standards for Basic Medical Education in China [38]. (2) It was not itself associated with medical students’ research interest through any mechanism except research engagement, and it was not related to other demographic and academic confounding factors included in the regression models. The under-identification test and weak identification test, as well as the over-identification test, all indicated that the accreditation status of the medical institution was an appropriate instrumental variable (Supplementary 1). The statistical analysis was performed using Stata MP version 14.0 software (Stata Inc., Chicago, IL, USA). Only a two-paired p-value < 0.05 was considered to have statistical significance.
Results
Descriptive Characteristics of the Sample
The final sampled data we used in this study were nationally representative in terms of medical graduates’ demographic and academic characteristics, involving 17,451 medical graduates from 101 medical schools. As for research engagement (RE), 12,717 (72.9%) of those surveyed had no research experience, while only 4734 (27.1%) medical graduates participated in research projects, with 2760 (15.8%) in biomedical research, 1528 (8.8%) in clinical research, and 446 (2.6%) in humanities and social science research. Merely 4820 (27.6%) participants had research interest (RI); the others had either no interest (4109, 23.6%) or neutral interest attitude towards research (8522, 48.9%). However, for those with research experience, almost 63.4% were interested in incorporating research into future career, while only 19.5% expressed no interest. The demographic and academic characteristics of respondents are shown in Table 1.
Table 1.
Characteristics of medical graduates from 101 medical schools (N = 17,451)
| Characteristics | Study groupa | Nationally[[34]] | RE of study group | ||
|---|---|---|---|---|---|
| Engagement | Non-engagement | pb | |||
| Gender | |||||
| Female | 9613 (55.0) | 55.0 | 2597 (27.0) | 7016 (73.0) | 0.713 |
| Male | 7838 (45.0) | 45.0 | 2137 (27.3) | 5701 (72.7) | |
| Parental educational attainments | |||||
| Father’s educational level | 10.2 (± 3.8) | 10.5 | 9.9 (± 3.7) | 10.9 (± 4.0) | p < 0.001 |
| Mother’s educational level | 8.9 (± 4.3) | 9.4 | 8.6 (± 4.2) | 9.7 (± 4.4) | p < 0.001 |
| Home location | |||||
| Rural | 8638 (49.5) | 46.0 | 1956 (22.6) | 6682 (77.4) | p < 0.001 |
| Urban | 8813 (50.5) | 54.0 | 2778 (31.5) | 6035 (68.5) | |
| GPA | |||||
| Bottom 50% | 8749 (50.1) | 2059 (23.5) | 6690 (76.5) | p < 0.001 | |
| Top 50% | 8693 (49.9) | 2675 (30.7) | 6027 (69.3) | ||
| Types of the degree program | |||||
| 5-year program | 16,018 (91.8) | 92.4 | 3894 (24.3) | 12,124 (75.7) | p < 0.001 |
| Long-term program | 1433 (8.2) | 7.6 | 840 (58.6) | 593 (41.4) | |
| University location | |||||
| East | 5345 (30.6) | 31.2 | 1848 (34.6) | 3497 (65.4) | p < 0.001 |
| Middle | 5543 (32.4) | 32.9 | 1275 (23.0) | 4268 (77.0) | |
| West | 6563 (37.6) | 35.9 | 1611 (24.6) | 4952 (75.4) | |
| Research engagement | 4734 (27.1) | 23.1 | |||
| Biomedical | 2760 (15.8) | 13.7 | p < 0.001 | ||
| Clinical | 1528 (8.8) | 7.1 | |||
| Social science | 446 (2.6) | 2.4 | |||
| Research interest (binary) | |||||
| Yes | 4820 (27.6) | 3054 (63.4) | 1766 (36.6) | p < 0.001 | |
| Neutral | 8522 (48.9) | 877 (10.3) | 7645 (89.7) | ||
| No | 4109 (23.6) | 803 (19.5) | 3306 (80.5) | ||
| Research interest (trichotomous) | |||||
| Yes | 4820 (27.6) | 3054 (63.4) | 1766 (36.6) | p < 0.001 | |
| No (no or neutral) | 12,631 (72.4) | 1680 (13.3) | 10,951 (86.7) | ||
Note:
aExcept for parental education attainment, data are expressed as the number (%) of students. Percentages have been rounded and may not total 100
bSignificance tests were conducted to examine the differences between engagement and non-engagement for each variable. For continuous variables, t-test was used, and for categorical variables, chi-square analysis was used
Association Between Research Engagement and Research Interest
As shown in Table 2, medical graduates’ research interest did differ between engagement and non-engagement groups by linear probability regression (β, 0.50; 95%CI, 0.48 to 0.52), but no significant difference was observed on IV regression analysis (β, 0.31; 95%CI, − 0.18 to 0.80). Furthermore, there was no significant difference between the three types of research engagement and research interest (Table 3). However, compared with graduates engaged in clinical research and biomedical research, respondents who participated in humanities and social science research had a higher likelihood of showing interest in incorporating research into future career (β, 0.80; 95%CI, − 2.50 to 4.09).
Table 2.
Association between research engagement and research interest: the LPM model and the IV-LPM model
| Variables | LPM (β, 95%CI)a | p | IV-LPM (β, 95%CI)b | p |
|---|---|---|---|---|
| RE | ||||
| Non-engagement | 0 [reference] | 0 [reference] | ||
| Engagement | 0.50 (0.48 to 0.52) | p < 0.001 | 0.31 (− 0.18 to 0.80) | 0.230 |
| Gender | ||||
| Female | 0 [reference] | 0 [reference] | ||
| Male | 0.01 (0.00 to 0.03) | 0.035 | 0.01 (0.00 to 0.03) | 0.040 |
| GPA | ||||
| Bottom 50% | 0 [reference] | 0 [reference] | ||
| Top 50% | 0.06 (0.04 to 0.07) | p < 0.001 | 0.07 (0.03 to 0.12) | 0.001 |
| Constant | 0.11 (− 0.09 to 0.14) | p < 0.001 | 0.14 (0. 07 to 0.23) | 0.001 |
aLine probability model estimated the coefficient for medical student RI among students engaged in research with those who did not have research experience as the reference group. Covariates included gender, parental educational attainments, home location, university location, types of a degree program, and GPA. The regression coefficients of parental educational attainments, home location, university location, and types of a degree program were not statistically significant and were not reported. Standard errors are clustered at the medical school level
bInstrumental variable regression models were estimated as linear probability model using the accreditation status of the medical institution as an IV for the presence of RE. The same covariates used in LPM were included. Coefficients represented the difference in probability of RI. Standard errors are clustered at the medical school level
Table 3.
Association between research engagement and research interest: IV-LPM regression modelsa
| Variables | Research interest | |||||
|---|---|---|---|---|---|---|
| β, 95%CI | p | β, 95%CI | p | β, 95%CI | p | |
| Biomedical research engagement | ||||||
| No | 0 [reference] | |||||
| Yes | 0.28 (− 0.37 to 0.94) | 0.401 | ||||
| Clinical research engagement | ||||||
| No | 0 [reference] | |||||
| Yes | 0.17 (− 0.80 to 1.15) | 0.73 | ||||
| Social science research engagement | ||||||
| No | 0 [reference] | |||||
| Yes | 0.80 (− 2.50 to 4.09) | 0.63 | ||||
| Gender | ||||||
| Female | 0 [reference] | 0 [reference] | 0 [reference] | |||
| Male | 0.02 (0.01 to 0.04) | 0.004 | 0.02 (0.00 to 0.40) | 0.02 | 0.02 (0.00 to 0.05) | 0.10 |
| GPA | ||||||
| Bottom 50% | 0 [reference] | 0 [reference] | 0 [reference] | |||
| Top 50% | 0.07 (0.02 to 0.18) | 0.007 | 0.06 (0.02 to 0.10) | 0.001 | 0.06 (0.02 to 0.10) | 0.007 |
| Constant | 0.13 (0.07 to 0.20) | p < 0.001 | 0.13 (0.06 to 0.20) | p < 0.001 | 0.13 (0.07 to 0.20) | p < 0.001 |
aInstrumental variable regression models were estimated as linear probability models using the accreditation status of the medical institution as an IV for the presence of RE. Covariates included gender, parental educational attainments, home location, university location, types of a degree program, and GPA. The regression coefficients of parental educational attainments, home location, university location, and types of a degree program were not statistically significant and were not reported. Coefficients represented the difference in probability of RI. Standard errors are clustered at the medical school level
Both Table 2 and Table 3 illustrated there were significant differences between males and females in the linear probability model (β, 0.01; 95%CI, 0.00 to 0.03) as well as in IV analysis (β, 0.01; 95%CI, 0.00 to 0.03). Males tended to show interest compared with females. Concerning GPA, compared with respondents ranking below the average, the top 50% of academically ranked students were more likely to express research interest (β, 0.06; 95%CI, 0.04 to 0.07). After addressing potential endogeneity with instrumental variable analysis, significant differences remained (β, 0.07; 95%CI, 0.03 to 0.12). Similar features were also found between three different types of research engagement and research interest (see Table 3).
Heterogeneity Effects
Interestingly, obvious heterogeneity was observed among sub-groups of academic performance (Table 4). For graduates with high GPAs, those who experienced research were significantly more likely to be interested in incorporating research into future career (β, 0.44; 95%CI, 0.01 to 0.86). Table 5 displays the associations between RE and RI by comparing the respondents’ gender. We found that both female and male respondents with the top 50% GPA reported higher interest in future research (male students ranked top 50%, β, 0.09; 95%CI, 0.02 to 0.16; female students ranked top 50%, β, 0.06; 95%CI, 0.03 to 0.09).
Table 4.
Association between research engagement and research interest by GPA: IV-LPM regression modelsa
| Variables | Top 50% | p | Bottom 50% | p |
|---|---|---|---|---|
| RE | ||||
| Non-engagement | 0 [reference] | 0 [reference] | ||
| Engagement | 0.44 (0.01 to 0.86) | 0.044 | 0.04 (− 0.84 to 0.93) | 0.923 |
| Gender | ||||
| Female | 0 [reference] | 0 [reference] | ||
| Male | 0.02 (0.001 to 0.042) | 0.038 | 0.003 (− 0.016 to 0.022) | 0.743 |
| Constant | 0.18 (0.07 to 0.30) | 0.001 | 0.19 (0.04 to 0.33) | 0.013 |
aSame as in Table 3
Table 5.
Association between research engagement and research interest by gender: IV-LPM regression modelsa
| Variables | Male | p | Female | p |
|---|---|---|---|---|
| RE | ||||
| Non-engagement | 1 [reference] | 1 [reference] | ||
| Engagement | 0.21 (− 0.48 to 0.90) | 0.546 | 0.35 (− 0.10 to 0.80) | 0.123 |
| GPA | ||||
| Bottom 50% | 1 [reference] | 1 [reference] | ||
| Top 50% | 0.09 (0.02 to 0.16) | 0.013 | 0.0 6(0.03 to 0.09) | 0.001 |
| Constant | 0.18 (0.05 to 0.31) | 0.006 | 0.13 (0.05 to 0.20) | 0.001 |
aSame as in Table 3
Discussion
To our knowledge, this is the first in-depth study that explores the relationship between research engagement and medical students’ individual research interest. For graduates engaged in research activities, 63.4% reported being interested in incorporating research into future career paths. Our findings indicated that research engagement might not significantly promote research interest among all medical graduates. However, it did benefit the research interest of participants who were academically ranked in the top 50%.
After considering the endogeneity, no significant relationship was found between research engagement and graduates’ interest in incorporating research into future career paths, although a large number of students expressed a positive interest attitude after research activities. Earlier studies found that students enhanced their situational research interest [24,28,39–41]. One plausible explanation is that medical students do develop situational interest during short-term research engagements; however, this situational interest may not evolve into enduring individual interest upon graduation, potentially due to irregular or inconsistent research engagement [25]. Another potential reason is that certain participants engage in research solely for pragmatic motives, such as enhancing their chances of securing desirable residency programs or advancing their post-graduate education, instead of being driven by an intrinsic appreciation for the value of research itself [6,18,42,43]. One systematic review study indicated that almost half of the medical students performed research to build a strong resume rather than out of interest [6]. This finding indicates a pause occurs during the transition from medical students’ unstable situational research interest to the establishment of stable individual research interest during their bachelor’s studies.
Remarkably, graduates’ involvement in humanities and social science research demonstrates a greater propensity for fostering research interest compared to biomedical and clinical research, albeit no statistically significant differences were observed among these three types of research engagement. Given the scarcity of studies exploring the correlation between humanities and social science research engagement and medical students’ interest in research, further investigation is warranted to elucidate the underlying reasons behind this finding.
Students with good academic performance tended to report interest in research. These findings are also in alignment with the results in a Saudi Arabian study [44]. One possible explanation is that high-GPA students are more inclined to actively participate in research activities, thereby acquiring medical knowledge and cultivating skills pertinent to research, teamwork, and communication [13,44]. These acquired knowledge and abilities not only instilled them with confidence but also significantly bolstered their interest in research [44]. In other words, students with lower GPAs might feel greater pressure to focus on essential aspects of the curriculum, leaving them with less time to devote to research activities. This limited engagement in research could lead to diminished readiness and interest in participating in future research activities.
To cultivate students’ interest in engaging in research activities throughout their future careers, we propose several suggestions based on our findings. Firstly, medical schools should ensure the provision of consistent research opportunities for medical students, prioritizing long-term engagement. This approach is anticipated to nurture students’ individual research interest, thereby enhancing their likelihood of pursuing research-related careers in the future. Secondly, recognizing the positive impact of research engagement in humanities and social science on medical students’ research interest, medical schools lacking adequate equipment to support biomedical and clinical research could also consider providing or expanding opportunities in these areas for students. Thirdly, in countries with constrained resources yet aspiring to cultivate physician-scientists, it is advisable to identify talented students, such as those highly engaged in research, with an inclination towards research and furnish them with adequate support [45].
Our study is limited in several regards. Although the instrumental variable technique approximates causality, a causal association between research engagement and medical graduates’ interest in incorporating research in future career paths cannot be definitively ascertained. Moreover, potential self-reported bias may exist, particularly among medical students intending to pursue further studies. While they may express an interest in research, their actual inclination towards research might differ. Furthermore, our understanding of the development of individual research interest from situational research interest among students remains limited, necessitating additional qualitative research to explore this aspect further. Despite these limitations, our study first employed the instrument variable technique and explored the relationship between research engagement and medical graduates’ interest in integrating research into future career paths. Future studies should consider assessing whether practicing physicians with prior research experience persist in research endeavors or incorporate research in their practice compared to those without.
Conclusions
To the best of our knowledge, this is the first Chinese study to explore the relationship between research engagement and medical graduates’ interest in integrating research into future career paths. This study shows that research does not promote graduates in integrating research into future career paths. However, graduates with good academic performance are more likely to report interest in future research engagement.
Supplementary Information
Below is the link to the electronic supplementary material.
Author Contribution
HBW made substantial contributions to the study conception and design. HBW and GYZ conducted the data analyses and drafted the manuscript. HBW interpreted the results with the support from GYZ, LL, and XXM. XXM made editing contributions. All authors reviewed the final manuscript and have approved the final version.
Funding
The research project was funded by the National Natural Science Foundation of China (Grant No. 72174013) and the China Scholarship Council (File No. 202208310033). The funders had no role in the design and development of the study protocol or the decision to publish.
Data Availability
Data will be made available on request.
Declarations
Conflict of Interest
The authors report there are no competing interests to declare.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Cianciolo AT, et al. Physician-scientist or basic scientist? Exploring the nature of clinicians’ research engagement. Adv Health Sci Educ Theory Pract. 2021;26(2):353–67. [DOI] [PubMed] [Google Scholar]
- 2.Mokresh ME, et al. Knowledge, attitudes, and barriers toward research among medical students: a cross-sectional study in Turkey. Med Sci Educ. 2024;34(2):387–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Association of American Medical Colleges. Physician-scientists. 2023; https://www.aamc.org/what-we-do/mission-areas/medical-research/physician-scientist.
- 4.Schafer AI. The vanishing physician-scientist? Transl Res. 2010;155(1):1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Paracha M, et al. Scholarly impact of student participation in radiation oncology research. Int J Radiat Oncol Biol Phys. 2018;101(4):779–83. [DOI] [PubMed] [Google Scholar]
- 6.Amgad M, et al. Medical student research: an integrated mixed-methods systematic review and meta-analysis. PLoS One. 2015;10(6). 10.1371/journal.pone.0127470 [DOI] [PMC free article] [PubMed]
- 7.Anderson DJ. Population and the environment: time for another contraception revolution. N Engl J Med. 2019;381(5):397–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Milewicz DM, et al. Rescuing the physician-scientist workforce: the time for action is now. J Clin Invest. 2015;125(10):3742–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Schroeder AR, et al. Outcomes of follow-up visits after bronchiolitis hospitalizations. JAMA Pediatr. 2018;172(3):296–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ommering BWC, et al. Future physician-scientists: could we catch them young? Factors influencing intrinsic and extrinsic motivation for research among first-year medical students. Perspect Med Educ. 2018;7(4):248–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Goldacre M, et al. Junior doctors’ views about careers in academic medicine. Med Educ. 1999;33(5):318–26. [DOI] [PubMed] [Google Scholar]
- 12.Rosenzweig EQ, et al Inside the STEM pipeline: changes in students’ biomedical career plans across the college years. Sci Adv 2021:7(18). 10.1126/sciadv.abe0985 [DOI] [PMC free article] [PubMed]
- 13.Burgoyne LN, O’Flynn S, Boylan GB. Undergraduate medical research: the student perspective. Med Educ Online. 2010;15. 10.3402/meo.v15i0.5212 [DOI] [PMC free article] [PubMed]
- 14.DiBiase RM, et al. A medical student scholarly concentrations program: scholarly self-efficacy and impact on future research activities. Med Educ Online. 2020;25(1):1786210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Funston G, et al. Medical student perceptions of research and research-orientated careers: an international questionnaire study. Med Teach. 2016;38(10):1041–8. [DOI] [PubMed] [Google Scholar]
- 16.Greenberg RB, et al. Medical student interest in academic medical careers: a multi-institutional study. Perspect Med Educ. 2013;2(5–6):298–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ommering BWC, et al. Promoting positive perceptions of and motivation for research among undergraduate medical students to stimulate future research involvement: a grounded theory study. BMC Med Educ. 2020;20(1):204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Adeboye W, et al. Predictors of self-reported research engagement and academic-career interest amongst medical students in the United Kingdom: a national cross-sectional survey. Postgrad Med J. 2023;99(1177):1189–96. [DOI] [PubMed] [Google Scholar]
- 19.Colmenares C, Bierer SB, Graham LM. Impact of a 5-year research-oriented medical school curriculum on medical student research interest, scholarly output, and career intentions. Med Sci Educ. 2013;23(1):88–91. [Google Scholar]
- 20.Laskowitz DT, et al. Engaging students in dedicated research and scholarship during medical school: the long-term experiences at Duke and Stanford. Acad Med. 2010;85(3):419–28. [DOI] [PubMed] [Google Scholar]
- 21.Stevenson MD, et al. Increasing scholarly activity productivity during residency: a systematic review. Acad Med. 2017;92(2):250–66. [DOI] [PubMed] [Google Scholar]
- 22.Yuan HF, Xu WD, Hu HY. Young Chinese doctors and the pressure of publication. Lancet. 2013;381(9864). 10.1016/S0140-6736(13)60174-9 [DOI] [PubMed]
- 23.Philibert I, et al. Scholarly activity in the next accreditation system: moving from structure and process to outcomes. J Grad Med Educ. 2013;5(4):714–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Stone C, et al. Contemporary global perspectives of medical students on research during undergraduate medical education: a systematic literature review. Med Educ Online. 2018;23(1):1537430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hidi S, Renninger KA. The four-phase model of interest development. Educ Psychol. 2006;41(2):111–27. [Google Scholar]
- 26.Krapp A. An educational-psychological theory of interest and its relation to SDT. In: Deci EL, Ryan RM, editors. Handbook of self-determination research. Rochester (NY): University of Rochester Press; 2002. p. 405–27. [Google Scholar]
- 27.Schiefele U, Krapp A, Winteler A. Interest as a predictor of academic achievement: a meta-analysis of research. In: Renninger KA, Hidi S, Krapp A, editors. The role of interest in learning and development. Hillsdale (NJ): Lawrence Erlbaum Associates; 1992. p. 183–212. [Google Scholar]
- 28.Houlden RL, et al. Medical students’ perceptions of an undergraduate research elective. Med Teach. 2004;26(7):659–61. [DOI] [PubMed] [Google Scholar]
- 29.Burge SK, Hill JH. The medical student summer research program in family medicine. Fam Med. 2014;46(1):45–8. [PubMed] [Google Scholar]
- 30.Cain L, Kramer G, Ferguson M. The Medical Student Summer Research Program at the University of Texas Medical Branch at Galveston: building research foundations. Med Educ Online. 2019;24(1):1581523. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Association of American Medical Colleges. Medical School Graduation Questionnaire: 2020 All Schools Summary Report. 2021. Available from: https://www.aamc.org/media/46851/download.
- 32.Center for Postsecondary Research in the Indiana University School of Education. NSSE Survey instruments. 2013. Available from: https://nsse.indiana.edu/nsse/survey-instruments/index.html.
- 33.Zhang G, et al. The association between medical student research engagement with learning outcomes. Med Educ Online. 2022;27(1):2100039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.National Centre for Health Professions Education Development. China Medical Students Survey 2020 All Schools Summary Report. 2020. Available from: https://medu.bjmu.edu.cn/cms/show.action?code=publish_4028801e6bb6cf11016be526c0dc0014&siteid=100000&newsid=50947b6c789248b39fca817dfb9ba3ee&channelid=0000000008.
- 35.Schwarz MR, Wojtczak A, Zhou T. Medical education in China’s leading medical schools. Med Teach. 2004;26(3):215–22. [DOI] [PubMed] [Google Scholar]
- 36.Zhu J, Li W, Chen L. Doctors in China: improving quality through modernisation of residency education. Lancet. 2016;388(10054):1922–9. [DOI] [PubMed] [Google Scholar]
- 37.Horrace WC, Oaxaca RL. Results on the bias and inconsistency of ordinary least squares for the linear probability model. Econ Lett. 2006;90(3):321–7. [Google Scholar]
- 38.Ministry of Education. Accreditation Standards for Basic Medical Education in China. 2016. Available from: http://ime.bjmu.edu.cn/cgzs/197708.htm.
- 39.de Oliveira NA, et al. Student views of research training programmes in medical schools. Med Educ. 2011;45(7):748–55. [DOI] [PubMed] [Google Scholar]
- 40.Wolfson RK, et al. The impact of a scholarly concentration program on student interest in career-long research: a longitudinal study. Acad Med. 2017;92(8):1196–203. [DOI] [PubMed] [Google Scholar]
- 41.O’Sullivan PS, et al. Becoming an academic doctor: perceptions of scholarly careers. Med Educ. 2009;43(4):335–41. [DOI] [PubMed] [Google Scholar]
- 42.Baig SA, et al. Reasons behind the increase in research activities among medical students of Karachi, Pakistan, a low-income country. Educ Health (Abingdon). 2013;26(2):117. [DOI] [PubMed] [Google Scholar]
- 43.Siemens DR, et al. A survey on the attitudes towards research in medical school. BMC Med Educ. 2010;10(1):1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Mina S, et al. Perceived influential factors toward participation in undergraduate research activities among medical students at Alfaisal University—College of Medicine: a Saudi Arabian perspective. Med Teach. 2016;38(sup1). 10.3109/0142159X.2016.1142508 [DOI] [PubMed]
- 45.Ommering BWC, Dekker FW. Medical students’ intrinsic versus extrinsic motivation to engage in research as preparation for residency. Perspect Med Educ. 2017;6(6):366–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data will be made available on request.