Education for Health

ORIGINAL RESEARCH PAPER
Year
: 2012  |  Volume : 25  |  Issue : 1  |  Page : 55--63

A Social and Academic Enrichment Program Promotes Medical School Matriculation and Graduation for Disadvantaged Students


L Keith1, D Hollar2,  
1 Deceased, Assistant Dean, UNC School of Medicine, Chapel Hill, North Carolina, USA
2 Assistant Professor, UNC School of Medicine, Chapel Hill, North Carolina, USA

Correspondence Address:
D Hollar
Assistant Professor, UNC School of Medicine, 410 Berryhill Building, CB 7530, Chapel Hill, NC - 27599
USA

Abstract

Introduction: This study assessed the impact of a pre-medical pipeline program on successful completion of medical school and the capacity of this program to address achievement gaps experienced by disadvantaged students. The University of North Carolina (USA) Medical Education Development (MED) program provides intensive academic and test skills preparation for admission to medical, dental, and other allied health professions schools. Methods: This retrospective study evaluated the academic progress of a longitudinal sample of 1738 disadvantaged college students who completed MED between 1974 and 2001. Data sources included MED participant data, medical school admissions data for the host school, aggregate data from the Association of American Medical Colleges (AAMC), and individual MED participant data from AAMC. Methods of analysis utilized Chi-square, independent samples t test, and logistic regression to examine associations between factors. Results: Of the 935 students in MED from 1974 to 2001, who had indicated an interest in medical school, 887 (94.9%) successfully matriculated and 801 (85.7%) successfully earned the MD degree. Using logistic regression, factors that were significantly correlated with earning the medical degree included the student«SQ»s race, college undergraduate total and science grade point averages, with Hispanic, African American, and Native American participants earning the medical degree at rates comparable to Caucasian participants. MED students successfully earned the MD degree despite having significantly lower Medical College Admissions Test (MCAT) scores and undergraduate grade point averages compared to all United States medical school applicants: MCAT scores had little relationship with student«SQ»s success. Conclusions: These findings suggest that an intensive, nine-week, pre-medical academic enrichment program that incorporates confidence-building and small-group tutoring and peer support activities can build a foundation on which disadvantaged students can successfully earn matriculation to and graduation from medical school.



How to cite this article:
Keith L, Hollar D. A Social and Academic Enrichment Program Promotes Medical School Matriculation and Graduation for Disadvantaged Students.Educ Health 2012;25:55-63


How to cite this URL:
Keith L, Hollar D. A Social and Academic Enrichment Program Promotes Medical School Matriculation and Graduation for Disadvantaged Students. Educ Health [serial online] 2012 [cited 2019 Dec 7 ];25:55-63
Available from: http://www.educationforhealth.net/text.asp?2012/25/1/55/99208


Full Text

 Introduction



Since the late 1980s, many educational enhancement programs have been offered to improve the percentage of disadvantaged students who pursue science careers. [1],[2],[3] Nevertheless, overall entrance rates into graduate school remains low for low-income/first-generation college students (34.5%), African-American (32.9%) and Hispanic/Latino students (30.1%), compared to Caucasian graduates (48.0%) who participate in these programs. [3]

Disadvantaged students, who include students of all races and ethnicities, typically report a lack of social support as a primary reason for lack of persistence in pursuing graduate programs. [3] Lee [4] demonstrated that satisfying relationships in summer science programs contributed to students' persistence in science activities. Grandy [5] showed that support and social development in high school are strong predictors for successful science career values, a finding echoed by other studies. [6],[7] Jacobs et al. [8] found that among 12 th grade girls, friends' support for science significantly correlated with girls' persistence with science careers. Small peer workgroup coursework modules have yielded comparable academic performance for disadvantaged and non-disadvantaged students. [9] Taylor and Antony [10] found in a cross-sectional study of six research universities that stereotype threat reduction with respect to academic science programs can be achieved through successful role models, academic advising, and faculty and peer support. These studies suggest that social support mechanisms are crucial for stability in continued mathematics and science achievement by women, minorities, and disadvantaged students. Based upon these findings, it appears that self-concept reinforcements could be combined with social support interventions to improve science and mathematics achievement. [11],[12],[13]

It is important to provide support at crucial transitional moments in students' careers. For both female and disadvantaged students, several studies have shown comparable math performance with male and non-disadvantaged peers up to middle and high school, [14],[15] at which point interests in science and mathematics decline. Zuckerman and Cole [16] as well as Long and Fox [17] emphasized that at each stage of educational development in the sciences, women and minorities experience "successive filtering," [16] a phenomenon that occurs due to environmental contexts of mentorship, advising, teamwork, and institutional facilities.

Longitudinal, multivariate studies of adolescent student achievement and development of positive self-concept have shown a "Big Fish, Little Pond" effect, [18],[19] in which students excel when they are provided with challenging material and a supportive, nonthreatening (i.e., not hypercompetitive) environment. Summers and Hrabowski, [20] p. 1870 cited "cultural isolation, motivation and performance vulnerability in the face of low expectations, peers who are not supportive of academic success, and discrimination, whether perceived or actual" as major barriers for disadvantaged students to continue in science and engineering careers. However, a few studies of academic achievement for disadvantaged students in the medical and other health professions' have addressed these pertinent, contextual issues.

The purpose of this study was to assess the impact of academic and socioeconomic factors on successful completion of medical school, using longitudinal data from 1738 undergraduate disadvantaged students in the United States who completed the University of North Carolina at Chapel Hill (UNC-CH) School of Medicine's Medical Education Development (MED) program between 1974 and 2001. In light of research supporting the need for social support, peer mentoring, and immersion in subject areas with adequate support, the academic social support networks that the MED program provides are particularly interesting. It provides disadvantaged students powerful evidence of self-efficacy in science courses within the context of a successful medical school and university collaboration.

The need for greater numbers of professionals from disadvantaged backgrounds in the sciences and health professions was noted by Flexner [21] in 1910, although there has been little increase in the proportion of physicians from disadvantaged backgrounds in the intervening century. [22] Therefore, academic preparation and enrichment programs such as MED continue to be needed in the United States as well as in all countries because of the pervasiveness of disadvantaged populations across societies. [23],[24] Furthermore, the problem of medical migration to the United States adds a further threat to physician availability in developing nations, as fewer physicians in these nations strain already weakened health systems and lower representation from disadvantaged populations at home. [25],[26]

 The Medical Education Development Program



The School of Medicine at UNC-CH initiated the MED program to address the U.S. Health Resources Services Administration's (HRSA) Health Careers Opportunity Program (HCOP) goal of preparing individuals from disadvantaged backgrounds to pursue careers in the medical and allied health professions. To carry out this goal, UNC-CH established a collaborative partnership involving the following institutions:



UNC School of Dentistry and Gillings School of Global Public Health;North Carolina Health Careers Access Program (NC-HCAP), operated by partnership institutions and which provides outreach workshops conducted on-site at North Carolina rural and inner city elementary, middle, and high schools;The five campuses of the UNC statewide system of universities with large enrolments of disadvantaged students-Elizabeth City State University (ECSU), Fayetteville State University (FSU), North Carolina Agricultural and Technical University (NC A&T), North Carolina Central University (NCCU), and UNC-Pembroke (UNC-P);St. Augustine's College, a private institution in Raleigh, NC, with a high proportion of disadvantaged students; andAll nine regional Area Health Education Centers (AHEC) across North Carolina, operated by the University of North Carolina Schools of Medicine, Dentistry, Nursing, Pharmacy, and Public Health in coordination with the state's three other university medical schools (Duke, Wake Forest, and East Carolina), ten additional state universities, six additional nursing schools, and 24 community colleges for medical student clerkship training, residency training, and health careers/science enrichment courses for high school and college students (www.med.unc.edu/ahec).

The MED program focuses on (1) disadvantaged students already admitted and now preparing to enter a United States medical school, and (2) disadvantaged rising college seniors and postgraduates who plan to apply to medical or dental school. The program uses the HRSA operational definition for disadvantaged student as "one who comes from (1) an environment that has inhibited the individual from obtaining the knowledge, skills, and abilities required to enroll in and graduate from a health professions school, or from a program providing education or training in an allied health profession; or (2) a family with an annual income below a level based on low-income thresholds according to family size, published by the United States Bureau of the Census, adjusted annually for changes in the Consumer Price Index, and adjusted by the Secretary for use in all health and allied health professions programs." [27] The program is advertised via mailings and site visits to disadvantaged college students at the partnering institutions (listed above) as well as at colleges and universities across the United States that enrol high percentages of disadvantaged students. For each year's summer program, students must apply by February of that year, providing transcripts, faculty letters of recommendation, and personal essays. From hundreds of applications, a committee of UNC medical and dental faculty plus Special Programs Office personnel of the university selects candidates for interviews. The committee selects among MED applicants based upon a combination of students' demonstrated academic performance, personal essays, letters of recommendation and interview.

Each year's cohort of students (originally 30-40 students per summer; currently approximately 80-100 students per summer) participate in intensive, nine weeks of pre-medical and/or pre-dental school coursework and preparation for standardized admission tests (e.g., Medical College Admissions Test, Dental Admissions Test). Students participate in 41 class hours of gross anatomy, 37 hours of histology, 25 hours of physiology, and 56 hours of clinical biochemistry instruction, plus the option for 56 hours of either immunology and microbiology or dental technique. Students live in a residence hall on the UNC campus. Program staff members support students with evening and weekend counseling and tutoring sessions. Student confidence is built through group teamwork activities, social events, and motivational speakers from UNC faculty and past graduates of the program who are in medical and dental school or who have graduated and are currently in practice.

 Methods



Study Sample

In this retrospective analysis of secondary data, historical data was collected on 1738 disadvantaged undergraduate students who participated in MED from 1974 through 2001. These MED participants should have had ample time to enter and complete a Doctor of Medicine (MD) degree program by 2009. Student MD completion was corroborated via a data matching agreement with the Association of American Medical Colleges (AAMCs), which maintains records of all students who matriculate into and graduate from United States medical schools. The MED and AAMC data did not distinguish between Doctor of Medicine (MD) and Doctor of Osteopathic Medicine (DO), the latter degree unique to the United States but similar to the MD degree other than providing stronger training in musculoskeletal medicine. [28]

We compared the MED students on undergraduate grade point averages and United States Medical College Admissions Test (MCAT) scores (if applicable) to a large cohort (n = 30,402) of American students who applied to the medical school that is home to MED during this period. We also compared aggregate MED MCAT scores by year to aggregate national MCAT scores made available by AAMC. [29],[30] For the 1738 students, we selected 935 students who specifically expressed their interest in medical careers when they started the program. The remaining 803 students pursued and are continuing to pursue other advanced degrees, including 147 Doctors of Dental Science (DDS), Masters of Science (MS) and Doctor of Philosophy (PhD) degrees, and other health professions degrees. We compared successful to unsuccessful completion of the MD degree from this primary sample of n = 935 students on various educational and social independent variables, as described below.

Variables Tested

The primary dependent or outcome variable was whether or not the individual completed the medical degree, based on AAMC data. The study and data sharing agreement were approved by the University of North Carolina Behavioral Internal Review Board (IRB), and all data were de-identified prior to analyses.

Academic independent variables included total college undergraduate grade point average (GPA), college undergraduate science GPA, and Medical College Admission Test (MCAT) scores. The MCAT is required of applicants to most United States medical schools; it consists of subtests that measure the applicant's proficiencies at understanding biological sciences, physical sciences, and English verbal ability/comprehension. We used aggregate data for 1994-2001 available from AAMC to compare MED participant mean annual scores to national means.

Socioeconomic independent variables included in analyses were father's education level, mother's education level, and family income. These data were obtained from a consolidated database of all student participants in the MED program and from data provided by the AAMC. Demographic independent variables included student race (self-reported) and gender.

Methodology

Analyses involved non-parametric statistics (i.e., Chi-square for dichotomous comparisons, Mann-Whitney U test for polychotomous comparisons) and parametric/multivariate statistics (i.e., independent samples t-tests, Analysis of Variance - ANOVA, and logistic regression). Statistical analyses utilized Statistical Package for the Social Sciences (SPSS), Version 16.0 (IBM SPSS, Chicago, Illinois). [31]

 Results



Demographically, the 1738 student matriculants of the MED program from 1974 to 2001 were 57.2% female, with 78.7% African-American, 10.1% Caucasian, 5.2% Native American, 4.0% Asian, and 1.9% Hispanic. By 2009, 924 (53.2%) had earned an MD or other advanced degree.

Of the 935 MED students who reported at the start of the program an aspiration to enter medical school, 55.7% were female, 82.4% were African-American, 6.2% were Native American, and 8.0% were Caucasian. Twelve students (1.3%) were of Hispanic/Latino/Latina ancestry. [Table 1] provides MED student gender and race based upon matriculation to medical school and earning the MD degree. Of the MED matriculants to medical school, African-American males had the highest proportional non-matriculation to medical school (X42 = 22.1, P ≤ 0.001) compared to all other groups based upon gender and race, with 19 of 284 participants not matriculating [Table 1]. Five of nine Asian-American male participants matriculated [Table 2], although the small sample size for this group makes these findings difficult to interpret. Overall, male and female participants of all races had proportionally equivalent success at earning the MD degree [Table 1].{Figure 1}{Table 2}

Compared to available 1994-2001 AAMC data on all United States medical school applicants, the 935 MED students had significantly lower final undergraduate grade point averages [t14 = -3.52, P = 0.003; [Figure 1]], significantly lower undergraduate science grade point averages [t14 = -3.77, P = 0.002; [Figure 2]], and significantly lower Medical College Admissions Test (MCAT) scores [t 14 = -9.85, P < 0.001; [Figure 3]]. The MED participant college undergraduate grade point averages [Figure 1] ranged from 3.08 to 3.36 on a 0-4.0 scale, compared to a range of 3.28 to 3.45 for all United States medical school applicants, with all medical school applicants being higher each year. The MED participant college undergraduate science course grade point averages [Figure 2] ranged from 2.99 to 3.31, compared to a range of 3.18 to 3.36 for all United States medical school applicants, with all medical school applicants again being higher each year. The MED participant MCAT score averages [Figure 3] ranged from 17.8 to 21.8, compared to a range of 25.1 to 27.0 for all United States medical school applicants, with all medical school applicants being higher each year.{Figure 2}{Figure 3}

Within the MCAT scores, MED participants had significantly lower MCAT verbal scores (range 6-7.29) compared to all United States medical school applicants (range 8.3-8.7) (t 14 = -10.9, P < 0.001), significantly lower MCAT Physical Science scores (range 5.76-7.03 for MED versus range 8.3-9.0 for all United States applicants; t 14 = -11.4, P < 0.001), and significantly lower MCAT Biological Science scores (range 5.87-7.58 for MED versus range 8.5-9.3 for all United States applicants; t 14 = -7.04, P < 0.001). Note that this comparison involved MED participants versus all United States medical school applicants. The aggregate comparisons [Figure 1], [Figure 2] and [Figure 3] are further supported by statistical comparisons between the same MED students and n = 30,402 applicants to the program's host medical school during the same period [Table 3]. MED students had significantly lower MCAT scores (mean 20.5 overall) and undergraduate total/science grade point averages (3.19 and 3.05) compared to same institution medical school applicants (MCAT 26.4; undergraduate GPA 3.36; science GPA 3.28), respectively) who did not participate in a pre-medical education program.{Table 3}

Of the 935 MED students interested in medical school, 887 students (94.9%) successfully matriculated to medical school by 2009. The average time differential from the undergraduate MED program participation to medical school matriculation was a mean of 1.45 +/- 1.87 years. A total of 177 (19.3%) of these students matriculated in less than one year, 163 of which were senior undergraduate students taking MED during the summer preceding medical school after already being accepted to medical school. Another 56.2% matriculated in one year, 11.9% matriculated between one and two years, and the remaining 12.6% of MED students matriculated into medical school in 3 or more years, which ranged up to 22 years. In all, 98% of MED matriculants were accepted into medical school in five years or less from their undergraduate completion.

For the primary dependent variable, 801 of the 887 matriculating students (90.3%) successfully earned the MD degree by 2009. This statistic represents 85.7% of all 935 MED students who indicated an interest in attending medical school. There were no significant differences between successful and unsuccessful MD students at matriculating to medical school based upon gender or race.

Because 163 of the 935 MED participants in this study had already been accepted to medical school prior to MED participation, it should be stated that 720 of the remaining 772 true applicants (93.3%) successfully matriculated into medical school. Of these 720 true MED-associated matriculants, 651 students (90.4%) successfully earned the MD degree, compared to 150 (92.0%) of the 163 pre-accepted MED participants (X12 = 0.239, P = 0.320).

Nine of ten (90.0%) Hispanic/Latino/Latina students, 51 of 54 (94.4%) Native American students, 641 of 713 (89.9%) African-American students, and 65 of 70 (92.9%) of Caucasian students earned the MD degree [Table 1]. For this cohort, those who went on to graduate medical school had lower undergraduate total grade point averages (3.10 +/- 0.45) than non-graduates (3.19 +/- 0.46) [F1,868 = 4.40, P = 0.036; [Table 2]]. Those who successfully graduated also had lower science grade point averages [2.97 +/- 0.53 versus 3.08 +/- 0.51; F1,868 = 4.31, P = 0.038; [Table 2]]. MCAT scores did not correlate with earning an MD degree [B = 0.002, P = 0.969, [Table 4]; Mean 22.2 for MD graduates versus 22.9 for non-graduates, [Table 2]].{Table 4}

Logistic regression log-likelihood ratios identified three significant factors (P < 0.05) associated with graduation with a medical degree among those who had matriculated in medical school [Table 4]: Race (B = 0.698, P = 0.037), Undergraduate Total GPA (B = 2.33, P = 0.028), and Undergraduate Science GPA (B = 1.78, P = 0.049), with the model having a weak Nagelkerke R2 = 0.100).

 Discussion



The findings demonstrate the success of a MED nine-week intensive summer program in preparing disadvantaged undergraduate students, most of whom are minority students, for a medical science curriculum. Between 1974 and 2001, 94.9% of MED students at the University of North Carolina who stated that their goal was to go to medical school had succeeded in matriculating into medical school by 2009. Furthermore, 85.7% of these students (90.3% of matriculating students) earned the MD degree by 2009.

There were no significant differences between successful and unsuccessful MD students at matriculating to medical school based upon gender or race. Furthermore, the graduation rates of the MED participants was slightly higher than rates for disadvantaged medical student generally in 1995, as reported by the AAMC (e.g., 61% graduating within four years, 90% within ten years for dual degree students). [29]

This assessment demonstrates the effectiveness of the MED program at producing a high proportion of medical school matriculants and graduates from disadvantaged student participants who initially indicated an intention to pursue a career in medicine. It identifies special subpopulations (e.g., African-American males) requiring additional support mechanisms in such pipeline programs. And it finds consistently higher success rates over time.

The effectiveness of the MED program is also seen by comparing its graduates who applied to the UNC School of Medicine to non-disadvantaged applicants who applied to the school. Between 1992 and 2001, MED graduates had substantially lower GPAs and MCAT scores than their non-disadvantaged counterparts but the percentage of MED graduates who graduated or were making academic progress at the school remained comparable to non-MED participants during the same time period. The differences in admissions scores (e.g., MCAT, grade point averages) between MED participants and all United States medical school applicants would be greater if the comparison involved United States medical school matriculants, who had mean MCAT scores of 29.7 (subscale means: 8.7 verbal, 9.0 physical sciences, 9.3 biological sciences) during this time period. [30] These data indicate that students with substantially lower MCAT and GPAs who successfully complete the MED program completed medical school at about the same rate as students with better grades.

Similar outcomes have been documented elsewhere among re-applicants to medical school. Blakely and Broussard [32] studied a similar MED program at the University of California, Davis (UCD), where 95 of 115 students who had been rejected by medical schools (unlike the UNC MED program) were successful at medical school matriculation following a 10-week summer program. Similarly, Grumbach and Chen [33] demonstrated that a University of California, San Francisco (UCSF) undergraduate program could boost medical school matriculation for disadvantaged students from 22% to 67%. The UNC MED program demonstrates success with students who have not previously applied to medical school because 93.3% of these true applicants at the time of MED participation successfully matriculated to medical school. Of these 720 true MED-associated matriculants, 651 students (90.4%) successfully earned the MD degree.

These results also support previous educational research indicating that disadvantaged populations studying in the fields of mathematics, science, and health professions face barriers to academic achievement including family socioeconomic history and the limitations of standardized tests. The issue of social support measures was not adequately represented in this model, given that most programs and their funding sources do not allow extensive immersion in academic programs with simultaneous intensive peer mentoring support throughout the academic year. Nevertheless, the ongoing refinement of the MED curriculum will soon include longitudinal peer support and broadened content to encompass the entire undergraduate research curriculum. In the future, it will be important to include measures of self-concept and educational/psychological development [11],[12],[13] among outcomes measured.

Three factors-race, college/undergraduate total and science grade point averages-contribute significantly to the likelihood of earning a medical degree for these program graduates. These factors have been previously identified in the educational research literature. However, these results indirectly implicate social support issues that have been documented in the educational research literature for science achievement. Therefore, we recommend that pre-medical enrichment programs should carefully integrate coursework within each student's social support needs. The MED program emphasizes tutor-supported small-group activities that promote student self-confidence, teamwork, and academic achievement using models of self concept. [11],[12],[13] Reinforcement of self-confidence and teamwork encourages students to work with others for improved learning and test performance.

The continuation of MED-style programs in the United States can serve to increase medical, dental, and other health professions opportunities for disadvantaged individuals and potentially can return many of these professionals to underserved rural and urban communities. [34] Several studies indicate that disadvantaged health professionals tend to practice in these areas. [35],[36] Finally, the educational research literature suggests that MED-type programs can promote early and prolonged interest in science careers when they are provided to middle- and high-school age adolescents of different cultures. [14],[15],[16],[17],[18],[19],[20] We advocate expansion of MED-type programs to school-age youth.

Evaluating the MED program has shown that academic preparation and social supports, even for a brief, nine-week period, can substantially improve opportunities for increasing the diversity of physicians in our societies. Future studies should assess whether involving disadvantaged students in a culture of medical education, respect, and confidence-building is a substantial factor in promoting student self concepts, [11],[12],[13],[23],[24] thus inspiring them to learn and pursue medical careers.

Programs such as MED exist at medical, dental, nursing, and public health schools across the United States to increase career opportunities for disadvantaged students. The continued underrepresentation of disadvantaged students in health careers within the United States [22] and many other countries [23],[24],[25],[26] justifies these programs, although evaluation of programs within the United States show varying levels of success. [3] The success of the MED program indicates that academic and social support mechanisms might be curriculum strategies that can be replicated in other programs globally. International studies have identified social support, promotion, and infrastructure development for increasing and retaining disadvantaged and indigenous health professionals. [37] Finally, social and behavioral sciences as well as web-based technological approaches to medicine suggest new models for medical education where medical students in different countries interact and utilize similar curricula, cooperative efforts that could positively promote the training of disadvantaged health professionals internationally. [38]

Limitations of the study include the absence of external comparison groups (e.g., medical education preparation program for disadvantaged students at another university, multiple socioeconomic levels of students) and time effects due to historical changes in individuals, culture and variables (e.g., MCAT scores and assessments). Additionally, the study did not explicitly test which of the MED program's several components (e.g., coursework, mentoring, test preparation, tutoring) directly contributed to the program's success.

 Acknowledgements



Funding for this study was provided by the School of Medicine, University of North Carolina at Chapel Hill and the U.S. Health Resources Services Administration (Grant Number D18HP13633 - Health Careers Opportunity Program. We thank Karen Stone, MBA, Cedric Bright, MD, Warren Newton, MD, Nerissa Price, MD, Claudis Polk, MA, Lisa Long, MS and Michel Branch, MA, Sandra Void, BA at UNC-Chapel Hill and Gwen E. Garrison, PhD, of the American Association for Medical Colleges (AAMC). Larry Keith is especially thankful for the support of his family: Wilma, Channte and Quinton.

Ethical Approval : The study was approved by the University of North Carolina School of Medicine Behavioral IRB for all years.

References

1Hexter H. A description of federal information and outreach programs and selected state, institutional, and community models. Washington, D.C: Advisory Committee on Student Financial Assistance symposium, Information Resources Services and Programs; 1990.
2McElroy EJ, Armesto M. TRIO and upward bound: History, programs, and issues-past, present, and future. The Journal of Negro Education. 1998; 67(4):373-380.
3U.S. Department of Education. A profile of the Ronald E. McNair postbaccalaureate achievement program: 1997-1998 through 2001-2002. Washington, D.C, U.S.: Department of Education, Office of Postsecondary Education; 2005.
4Lee JD. More than ability: Gender and personal relationships influence science and technology involvement. Sociology Education. 2002; 75(4):349-373.
5Grandy J. Persistence in science of high-ability minority students: Results of a longitudinal study. Journal of Higher Education. 1998; 69(6):589-620.
6Kahle JB. Measuring progress toward equity in science and mathematics education. National Institute for Science Education NISE Brief. Vol. 2. Madison: University of Wisconsin-Madison; 1998. p. 2-11.
7Phinney JS, Alipuria LL. Ethnic identity in college students from four ethnic groups. Journal of Adolescence. 1990; 13(2):171-183.
8Jacobs JE, Finken LL, Griffin NL, Wright JD. The career plans of science-talented rural adolescent girls. American Education Research Journal. 1998; 35(4):681-704.
9Fullilove RE, Treisman PU. Mathematics achievement among African American undergraduates at the University of California, Berkeley: An evaluation of the mathematics workshop program. The Journal of Negro Education. 1990; 59(3):463-478.
10Taylor E, Antony JS. Stereotype threat reduction and wise schooling: Towards the successful socialization of African American doctoral students in education. The Journal of Negro Education. 2000; 69(3):184-198.
11Hattie J. Self concept. Hillsdale. NJ: Lawrence Erlbaum; 1992.
12Byrne BM. Measuring self-concept across the lifespan: Issues and instrumentation. Washington, DC: American Psychological Association; 1996.
13Marsh HW, Yeung AS. Longitudinal structural equation models of academic self-concept and achievement: Gender differences in the development of math and English constructs. American Education Research Journal. 1998; 35(4):705-738.
14Gross S. Early mathematics performance and achievement: Results of a study within a large suburban school system. The Journal of Negro Education. 1993; 62(3):269-287.
15Park HS, Bauer S. Computational mathematical abilities of African American girls. Journal of Black Studies. 1999; 30(2):204-215.
16Zuckerman H, Cole JR. Women in American science. Minerva. 1975; 13:82-102.
17Long JS, Fox MF. Scientific careers: Universalism and particularism. Annual Review of Sociology. 1995; 21:45-71.
18Marsh HW, O'Mara A. Reciprocal effects between academic self-concept, self-esteem, achievement, and attainment over seven adolescent years: Unidimensional and multidimensional perspectives of self-concept. Personality and Social Psychology Bulletin. 2008; 34(4):542-552.
19Marsh HW, Trautwein U, Ludtke O, Koller O. Social comparison and big-fish-little-pond effects on self-concept and other self-belief constructs: Role of generalized and specific others. Journal of Educational Psychology. 2008; 100(3):510-524.
20Summers MF, Hrabowski FA. Preparing minority scientists and engineers. Science. 2006; 311(5769):1870-1871.
21Flexner A. Medical education in the United States and Canada: A report to the Carnegie Foundation for the Advancement of Teaching. New York: The Carnegie Foundation for the Advancement of Teaching; 1910.
22Sullivan LW, Mittman IS. The state of diversity in the health professions a century after Flexner. Academic Medicine. 2010; 85(2):246-253.
23Mda T. The structure and entrenchment of disadvantage in South Africa. In: Snyder I, Nieuwenhuysen, editors. Closing the gap in education? Improving outcomes in Southern World societies. Clayton, Victoria, Australia: Monash University Publishing; 2010. Available from: http://books.publishing.monash.edu. Retrieved 2 March, 2012.
24Perez LM, Martinez J. Community health workers: Social justice and policy advocates for community health and well-being. American Journal of Public Health. 2008; 98(1):11-14.
25Bundred PE, Levitt C. Medical migration: Who are the real losers? The Lancet. 2000; 356(9225):245-246.
26Hagopian A, Thompson MJ, Fordyce M, Johnson KE, Hart LG. The migration of physicians from sub-Saharan Africa to the United States of America: Measures of the African brain drain. Human Resources for Health. 2004; 21:17.
27Health Resources Services Administration. Health careers opportunity program (HCOP) self governance inventory. Available from: http://aspe.hhs.gov/SelfGovernance/inventory/Hrsa/822.htm. Retrieved 18 October, 2011.
28Gimpel JR. Getting "beyond the barriers" in reforming osteopathic medical education. The Journal of the American Osteopathic Association. 2007; 107(7):270-275.
29Garrison G, Mikesell C, Matthew D. Medical school graduation and attrition rates. AAMC Analysis in Brief. 2007; 7(2):1-2.
30Association of American Medical Colleges. Applicant matriculant data file Table 17: MCAT scores and GPAs for applicants and matriculants to U.S. medical schools, 1999-2010. Washington, DC: Author ; 2009.
31IBM SPSS. Statistical package for the social sciences. Chicago, IL: IBM SPSS Inc.; 2005.
32Blakely AW, Broussard LG. Blueprint for establishing an effective postbaccalaureate medical school pre-entry program for educationally disadvantaged students. Academic Medicine. 2003; 78(5):437-447.
33Grumbach K, Chen E. Effectiveness of University of California postbaccalaureate premedical programs in increasing medical school matriculation for minority and disadvantaged students. Journal of the American Medical Association. 2006; 296(9):1079-1085.
34Department of Health and Human Services. Public Health Service Act: Rural physician training grant program, definition of "Underserved Rural Community." Federal Register. 2010; 75(101):29447-29451.
35Kindig DA, Yan G. Physician supply in rural areas with large minority populations. Health Affairs. 1993; 12(12):177-184.
36U.S. Department of Health and Human Services. Location patterns of minority and other health professionals (DHHS Publication number HRS-P-OD-85-2). Washington, DC: Author ; 1985.
37Felton-Busch CM, Solomon SD, McBain KE, De La Rue S. Barriers to advanced education for indigenous Australian health workers: An exploratory study. Education for Health. 2009; 22(2):187.
38Harden RM. International medical education and future directions: A global perspective. Academic Medicine. 2006; 81(12):S22-S29.