Advertisement

Educational debt and intended employment choice among dental school seniors

  • Author Footnotes
    1 Dr. Wanchek is an assistant professor, Department of Public Health Science, University of Virginia, P.O. Box 400206, Charlottesville, Va. 22904
    Tanya Wanchek
    Correspondence
    Address correspondence to Dr. Wanchek
    Footnotes
    1 Dr. Wanchek is an assistant professor, Department of Public Health Science, University of Virginia, P.O. Box 400206, Charlottesville, Va. 22904
    Search for articles by this author
  • Author Footnotes
    2 Dr. Nicholson is a professor, Department of Policy Analysis and Management, Cornell University, N.Y.
    Sean Nicholson
    Footnotes
    2 Dr. Nicholson is a professor, Department of Policy Analysis and Management, Cornell University, N.Y.
    Search for articles by this author
  • Author Footnotes
    3 Dr. Vujicic is the managing vice president, Health Policy Resources Center, American Dental Association, Chicago.
    Marko Vujicic
    Footnotes
    3 Dr. Vujicic is the managing vice president, Health Policy Resources Center, American Dental Association, Chicago.
    Search for articles by this author
  • Author Footnotes
    4 Ms. Menezes is a manager, Dentist Surveys, Survey Center Health Policy Resources Center, American Dental Association, Chicago.
    Adriana Menezes
    Footnotes
    4 Ms. Menezes is a manager, Dentist Surveys, Survey Center Health Policy Resources Center, American Dental Association, Chicago.
    Search for articles by this author
  • Author Footnotes
    5 Dr. Ziebert is the senior vice president, Education and Professional Affairs, American Dental Association, Chicago.
    Anthony Ziebert
    Footnotes
    5 Dr. Ziebert is the senior vice president, Education and Professional Affairs, American Dental Association, Chicago.
    Search for articles by this author
  • Author Footnotes
    1 Dr. Wanchek is an assistant professor, Department of Public Health Science, University of Virginia, P.O. Box 400206, Charlottesville, Va. 22904
    2 Dr. Nicholson is a professor, Department of Policy Analysis and Management, Cornell University, N.Y.
    3 Dr. Vujicic is the managing vice president, Health Policy Resources Center, American Dental Association, Chicago.
    4 Ms. Menezes is a manager, Dentist Surveys, Survey Center Health Policy Resources Center, American Dental Association, Chicago.
    5 Dr. Ziebert is the senior vice president, Education and Professional Affairs, American Dental Association, Chicago.

      ABSTRACT

      Background

      The authors examined the association between educational debt and dental school seniors' intended activity after graduation.

      Methods

      The authors used multinomial logit regression analysis to estimate the relationship between dental educational debt and intended activity after graduation, controlling for potentially confounding variables. They used data from the 2004 through 2011 ADEA (American Dental Education Association) Survey of Dental School Seniors.

      Results

      Fourth-year dental school students with high levels of educational debt were more likely to express an interest in choosing to go into private practice, although the magnitude of this effect was relatively small. For each $10,000 increase in debt, the likelihood of choosing advanced education relative to private practice was 1.5 percent lower (relative risk ratio [RRR], 0.985 [95 percent confidence interval {CI}, 0.978-0.991]). For the same $10,000 increase in debt, the probability of choosing teaching, research and administration was 3.1 percent lower than that for choosing private practice (RRR, 0.969 [95 percent CI, 0.954-0.986]) and was 8.4 percent lower than that for choosing a government service position (RRR, 0.916 [95 percent CI, 0.908-0.924]).

      Conclusions

      Although educational debt was statistically significant for predicting intended activity after graduation, the magnitude of influence of other variables such as sex, race and whether a parent is a dentist was substantially larger.

      Practical Implications

      Concerns regarding rising educational debt and its effect on the dental labor market may be misplaced. The characteristics of the dental school student body may be a more accurate predictor of employment choices that dental school seniors are making than are total educational debt levels.

      Key Words

      ABBREVIATION KEY:

      ADEA (American Dental Education Association), GPA (Grade point average)
      Dental school students' debt levels are rising. The average educational debt of graduating dental school seniors, including both dental school debt and prior educational debt, rose from $122,000 in 2004 to $179,000 in 2011 according to calculations we performed by using data from the ADEA (American Dental Education Association) Survey of Dental School Seniors. Dentists' income has not risen as rapidly over the same period. As a result, average educational debt rose from 70 percent of dentists' median income in 1996 to 103 percent of dentists' median income in 2011.
      • Asch DA
      • Nicholson S
      • Vujicic M
      Are we in a medical education bubble market (published online ahead of print Oct. 30, 2013)?.
      This rise suggests an increase in the burden placed on dental school graduates.
      The rising debt relative to income may be exacerbated by an influx in new dental school graduates. The application to enrollment ratio reached a historical high of 2.9 applicants for each first-year dental school student spot in 2007 before easing slightly to 2.5 in 2008, but it remained well above the 1950-2008 average of 1.7.
      • Wanchek TN
      • Rephann T
      • Shobe W
      • Valachovic RW
      Workforce issues. PowerPoint presentation to: NGA Policy Academy for State Officials Improving Oral Health Care for Children meeting; May 7, 2001; Nashville, Tenn.
      • American Dental Association
      The high demand for dental education is spurring an increase in the number of dental schools. Thirteen new dental schools have been planned or proposed in the past few years.
      • Wanchek TN
      • Rephann T
      • Shobe W
      • Valachovic RW
      Workforce issues. PowerPoint presentation to: NGA Policy Academy for State Officials Improving Oral Health Care for Children meeting; May 7, 2001; Nashville, Tenn.
      • American Dental Association
      If the increased supply of dentists is not met with a comparable increase in demand for dental services, then the higher supply of dentists is expected to further limit wage gains and exacerbate the declining return to dental education.
      Rising educational debt without a comparable increase in expected income has the potential to affect graduates' career choices—including specialty training, willingness to work in underserved areas and willingness to treat Medicaid patients—as well as practice ownership options.
      • Parker MA
      New practice models and trends in the practice of oral health.
      • Merchant VA
      Are we willing to change?.
      In theory, rising educational debt is expected to encourage students who are choosing between different employment options to seek the position with a higher income. The higher income is needed for the graduate to maintain a standard of living comparable with those of previous dental school graduates. The result is that educational debt may increase the likelihood of choosing private practice instead of government service; teaching, research and administration; or public health positions. The effect of educational debt on advanced education is more ambiguous. Higher debt could influence a graduate to seek employment sooner to start repaying loans, or it could encourage the pursuit of advanced education to achieve a higher salary later.
      Empirical evidence for the effect of total educational debt is mixed. Typically, educational debt plays a central role in the decision-making process. One exception is students who already were obtaining advanced education. In one study, surgical residents reported that total educational debt did not play a role in their choosing to go into an academic program instead of private practice, except at the highest debt levels.
      • Lanzon J
      • Edwards SP
      • Inglehart MR
      Choosing academia versus private practice: factors affecting oral maxillofacial surgery residents' career choices.
      Residents with debt greater than $301,000 were more likely to choose private practice. On the other hand, more than one-half of practicing dentists in several graduate cohorts from one dental school reported that educational debt levels influenced their career choices.
      • Altman DS
      • Alexander JL
      • Woldt JL
      • Hunsaker DS
      • Mathieson KM
      Perceived influence of community oral health curriculum on graduates' dental practice choice and volunteerism.
      Investigators in studies have used the results of the ADEA's annual Survey of Dental School Seniors to examine the role of educational debt in employment choice.
      • Davidson PL
      • Nakazono TT
      • Carreon DC
      • Gutierrez JJ
      • Shahedi S
      • Andersen RM
      Reforming dental workforce education and practice in the USA.
      • Okwuje I
      • Anderson E
      • Valachovic RW
      Annual ADEA Survey of Dental School Seniors: 2009 graduating class.
      The results of a multivariate analysis showed that debt reduced the likelihood of planning to enter public service, and entry into a loan repayment program increased the likelihood of public service relative to other plans.
      • Davidson PL
      • Nakazono TT
      • Carreon DC
      • Gutierrez JJ
      • Shahedi S
      • Andersen RM
      Reforming dental workforce education and practice in the USA.
      The results of a study of 2009 dental school graduates showed a correlation between total educational debt and plans to enter private practice.
      • Okwuje I
      • Anderson E
      • Valachovic RW
      Annual ADEA Survey of Dental School Seniors: 2009 graduating class.
      The authors of the study did not, however, control for other student characteristics. In the same study, no clear majority response emerged when questions on the survey asked students to what extent educational debt influenced their choice of activity after graduation.
      • Okwuje I
      • Anderson E
      • Valachovic RW
      Annual ADEA Survey of Dental School Seniors: 2009 graduating class.
      Twenty-eight percent of seniors with debt indicated that their debt did not influence their choice, 39 percent indicated that it influenced their choice somewhat or moderately, and 33 percent indicated that it influenced their choice much or very much.
      Empirical evidence regarding whether educational debt influences the decision to pursue advanced education is mixed. The authors of two studies who conducted their own surveys of dental school students found opposing results.
      • Dhima M
      • Petropoulos VC
      • Han RK
      • Kinnunen T
      • Wright RF
      Dental students' perceptions of dental specialties and factors influencing specialty and career choices.
      • Scarbecz M
      • Ross JA
      The relationship between gender and postgraduate aspirations among first- and fourth-year students at public dental schools: a longitudinal analysis.
      The results of a multivariate analysis of data from surveys distributed to both fourth-year students and advanced-standing students at one dental school showed that respondents' debt levels reduced their intentions to pursue specialty training, even after controlling for sex, age and class year.
      • Dhima M
      • Petropoulos VC
      • Han RK
      • Kinnunen T
      • Wright RF
      Dental students' perceptions of dental specialties and factors influencing specialty and career choices.
      The result is consistent with those of an annual ADEA survey, which showed that senior dental school students with higher debt loads were less likely to have intentions of pursuing specialty training.
      • Okwuje I
      • Anderson E
      • Valachovic RW
      Annual ADEA Survey of Dental School Seniors: 2009 graduating class.
      Alternatively, investigators in a survey that tracked 138 students from six publicly funded dental schools found that students' plans to pursue advanced education were not influenced by higher debt levels, higher than expected debt, sex, race or whether they had children.
      • Scarbecz M
      • Ross JA
      The relationship between gender and postgraduate aspirations among first- and fourth-year students at public dental schools: a longitudinal analysis.
      Rather, the decision to pursue advanced education was influenced by whether the student had a dental school mentor, a high grade point average (GPA) or encouragement from significant others.
      • Scarbecz M
      • Ross JA
      The relationship between gender and postgraduate aspirations among first- and fourth-year students at public dental schools: a longitudinal analysis.
      However, for the latter result, investigators controlled only for whether debt was less than or greater than $100,000, which may not have provided sufficient variation to identify an effect on graduation plans.
      We conducted a study in which we used the ADEA's dental school senior surveys to extend the previous research in several ways. We examined several years of data to increase the sample size and look at the correlation between the total educational debt and employment choice over time. We used multinomial logit regression analysis to control for various (observable) confounding factors, and we examined the impact of educational debt on several distinct career choices.

      METHODS

       Data sources

      We obtained the data for our study from the annual ADEA survey of senior dental school students. The surveys are provided to dental schools in March and administered to students during the spring or summer shortly after graduation. The Institutional Review Board for Health Sciences Research at the University of Virginia (Charlottesville) reviewed our study and determined it to be exempt from institutional review board oversight. Response rates ranged from 62 percent in 2009 to 86 percent in 2005, with an average of 75 percent. We used t tests to compare the survey respondent characteristics with the American Dental Association's survey of dental school graduates from 2004 through 2010.
      • American Dental Association
      We found no statistical difference in sex or race, with the exception of 2010 when a higher proportion of white students took the ADEA survey than did non-white students.
      We pooled data from 2004 through 2011 to increase the sample size. We excluded students who reported they were younger than 24 years or older than 50 years (329 respondents) and students who reported that they had more than $600,000 of debt (74 respondents) to remove observations that were more likely to be inaccurate. The average age of senior dental school students who completed the survey was 28 years ((Table 1 and (Table 2). Average total educational debt was $168,300 (adjusted to 2011 dollars by using the consumer price index, which is a common method for adjusting for inflation).
      TABLE 1Summary statistics of dental seniors, 2004–2011.
      VARIABLEOBSERVATIONS, NO.MEAN (STANDARD DEVIATION)
      Age27,12228.28 (3.48) years
      Total Educational Debt
      Total educational debt includes dental school debt and prior educational debt.
      27,716$15.67 ($96.52)
      Dollar amounts were inflation adjusted and divided by $10,000.
      Total Educational Debt27,716$16.83 ($10.18)
      Dollar amounts were inflation adjusted and divided by $10,000.
      Female27,57244 (0.50) percent
      Parent a Dentist23,75914 (0.35) percent
      * Total educational debt includes dental school debt and prior educational debt.
      Dollar amounts were inflation adjusted and divided by $10,000.
      TABLE 2Summary statistics of dental school seniors, according to race or ethnicity, 2004–2011.
      RACE OR ETHNICITYFREQUENCY, NO. OF OBSERVATIONSPERCENTAGE
      Native American1660.62
      Asian6,19023.15
      African American1,2874.81
      Hispanic1,6146.03
      White17,48765.39
      In terms of validity, the annual survey has been conducted by ADEA for many years, and the results have been used in studies for many years. Therefore, we assumed that experts had sufficient time to examine the content of the questions and that they were valid measures. In regard to reliability, we did not conduct repeat trials to determine the accuracy of the answers each participant gave. Because students were estimating current debt, rather than debt from several years in the past, we assumed that the estimates were relatively accurate. Investigators who conducted a study in which they used the same ADEA survey as we did noted that validity and reliability were indicated by the high response rates, high number of responses for each item, consistency in responses across items and lack of concerns in the literature over the decades in which the ADEA survey has been used.
      • Davidson PL
      • Nakazono TT
      • Carreon DC
      • Gutierrez JJ
      • Shahedi S
      • Andersen RM
      Reforming dental workforce education and practice in the USA.

       Research design

      We hypothesized that as educational debt increased relative to income, students on the margin of choosing between different employment options would be more likely to choose the higher-paying option. Higher income is necessary to maintain the same standard of living that the graduates had expected and that previous dental school graduates enjoyed. Therefore, other things being equal, we expected a larger number of graduates to seek more lucrative employment in private practice or potentially though advanced education as educational debt increased relative to income. The effect of educational debt on advanced education is ambiguous because it delays loan repayment but leads to greater income in the future. At the same time, many other factors influence intended employment, such as sex, race or ethnicity, age and socioeconomic status. Some of these variables were available in the data, and we controlled for them in our analysis. Other variables that may matter—such as a person's natural ability to do well enough to pursue advanced education, marital status, importance placed on job satisfaction, availability of loan repayment plans or opportunities to participate in internships—were not available in the data, and we did not include them in our analysis. The hypothesis predicted that total educational debt influenced employment. We used the hypothesis to test the relative importance of educational debt compared with other explanatory variables for different employment choices.
      We tested the hypothesis by using a multinomial logistic regression analysis. The dependent variable was a student's intended primary activity after graduation. The data for this variable came from responses to survey question that asked students “[i]mmediately upon graduation from dental school do you intend your primary activity to be: …” followed by a multiple-choice list of possibility activities. From 2004 through 2008, there were seven possible activities, and from 2009 through 2011, there were 11 possible activities. We consolidated activities into five intended activities: private practice; advanced education; government service; public health; and teaching, research and administration.
      The independent variables for the main hypothesis were whether a student's parent was a dentist, the student's race, age, sex, father's and mother's educational level, graduation year and whether the student attended a public or private dental school. We clustered standard errors according to dental school to allow for a correlation in unobserved school characteristics that may have affected employment intentions among students who attended the same school.
      In addition to the main hypothesis, we investigated a number of other hypotheses by using different combinations of control variables. One possibility was that schools had an important influence on the intended employment choices of students, independent of the schools' influence on educational debt. For some hypotheses, we included an indicator variable for each dental school rather than only including an indicator variable for attending a public school. Another possible hypothesis was that the control variables were correlated with debt and therefore may have captured some of the influence of debt. We considered a hypothesis that included educational debt as the only explanatory variable. For one hypothesis, we omitted the indicator variables for each graduation year to see if those variables were capturing some of the influence of the rise in debt over time. Finally, we added a variable for educational debt squared to explore whether debt had a nonlinear effect on intended primary activity.

      RESULTS

      The main hypothesis predicted that as total educational debt levels rise, more students would choose to enter private practice. One way to analyze the hypothesis is to look at change in debt and intended employment choice over time.
      Our observation of the data over time did not show a clear correlation between educational debt and choosing private practice. Adjusted for inflation, debt among students at public dental schools rose from an average of $118,915 in 2004 to $160,803 in 2011, and among students at private dental schools, debt increased from $179,533 in 2004 to $213,237 in 2011 ((Table 3). Over the same period, we did not see a trend in students intending to enter private practice. The percentage of students intending to work in private practice ranged from 47.45 percent in 2008 to 53.56 in 2009 ((Table 4).
      TABLE 3Total educational debt
      Total educational debt includes dental school debt and prior educational debt.
      over time for public and private schools (unadjusted and adjusted for inflation), according to year of graduation.
      YEARDEBT UNADJUSTED FOR INFLATION, $DEBT ADJUSTED FOR INFLATION, $
      Public SchoolsPrivate SchoolsPublic SchoolsPrivate Schools
      200499,929150,868118,915179,533
      2005104,738162,979120,448187,426
      2006123,293176,466138,088197,642
      2007136,378187,539147,288202,542
      2008142,582205,407148,285213,623
      2009148,584199,651156,013209,633
      2010156,441204,380161,134210,511
      2011160,803213,237160,803213,237
      * Total educational debt includes dental school debt and prior educational debt.
      TABLE 4Intended primary activity after graduation, according to year of graduation.
      YEARPRIMARY ACTIVITY, %
      Private PracticeAdvanced EducationTeaching, Research and AdministrationGovernment ServicePublic Health
      200449.6539.750.487.782.35
      200550.0039.940.776.352.93
      200650.8539.110.576.313.16
      200749.5639.440.416.494.10
      200847.4540.900.846.114.71
      200953.5634.000.399.852.21
      201051.1138.370.607.662.26
      201151.7438.100.437.532.20
      Alternatively, when we looked cumulatively at all students and a range of debt levels, we observed a correlation between people with high educational debt and intent to enter private practice (Figure 1). This result is similar to those of other studies.
      • Okwuje I
      • Anderson E
      • Valachovic RW
      Annual ADEA Survey of Dental School Seniors: 2009 graduating class.
      • Okwuje I
      • Anderson E
      • Valachovic RW
      Annual ADEA Survey of Dental School Seniors: 2008 graduating class.
      • Woolfolk MW
      • Price SS
      Dental education: evolving student trends.
      However, that total educational debt changed substantially, moving from left to right on the x-axis in Figure 1, whereas the percentage of students choosing a career path did not change nearly as much on the y-axis. Thus, the association between debt and intended employment choice was not as large as it might appear visually.
      Figure thumbnail gr1
      Figure 1Employment plans, according to debt level (not adjusted for inflation).
      To reconcile these two observations, we used a multinomial logit regression model to control for other factors that influence intended employment choice ((Table 5). We used private practice as the baseline category for intended employment choice. The baseline for whether a parent was a dentist was no, for race it was white and for sex it was male. Relative risk ratios (RRRs) greater than 1.0 indicated a greater likelihood that the student would choose the respective intended employment rather than private practice as the explanatory variable increased. An explanatory variable that had an RRR of less than 1.0 indicated that there was a negative relationship between the explanatory variable and the particular intended employment option.
      TABLE 5Factors influencing intended primary employment choice in private practice compared with alternative activities.
      The total number of observations was 25,176.
      EXPLANATORY VARIABLES
      A category for the employment choice of other/unknown (n = 532) was included in the regression analysis but is not presented in the table.
      ADVANCED EDUCATION, RELATIVE RISK RATIO (ROBUST SE
      SE: Standard error.
      Relative risk ratios were estimated by using robust SEs.
      )
      TEACHING, RESEARCH AND ADMINISTRATION, RELATIVE RISK RATIO (ROBUST SE)GOVERNMENT SERVICE, RELATIVE RISK RATIO (ROBUST SE)PUBLIC HEALTH, RELATIVE RISK RATIO (ROBUST SE)
      Total Educational Debt
      Graduation year indicator variables were included in the regression analysis but were omitted from the table. Standard errors were clustered according to school. Total educational debt was adjusted for inflation and divided by $10,000.
      0.985
      Significant at the 1 percent level.
      (0.003)
      0.969
      Significant at the 1 percent level.
      (0.008)
      0.916
      Significant at the 1 percent level.
      (0.004)
      0.998 (0.005)
      Parent Is a Dentist (No = 0; Yes = 1)0.850
      Significant at the 1 percent level.
      (−0.037)
      1.095 (−0.233)0.531
      Significant at the 1 percent level.
      (−0.054)
      0.724
      Significant at the 5 percent level.
      (−0.113)
      Race or Ethnicity (Baseline = White)
      Native American0.77 (−0.155)0.000
      Significant at the 1 percent level.
      (0.000)
      2.546
      Significant at the 5 percent level.
      (−0.93)
      2.343
      Significant at the 5 percent level.
      (−0.849)
      Asian, South Pacific Islander0.952 (−0.069)1.034 (−0.289)0.440
      Significant at the 1 percent level.
      (−0.05)
      0.807 (−0.109)
      African American1.989
      Significant at the 1 percent level.
      (−0.396)
      1.74 (−0.592)2.074
      Significant at the 1 percent level.
      (−0.414)
      3.342
      Significant at the 1 percent level.
      (−0.589)
      Hispanic1.144 (−0.179)0.893 (−0.348)0.90 (−0.097)1.533
      Significant at the 5 percent level.
      (−0.276)
      Age0.910
      Significant at the 1 percent level.
      (−0.008)
      1.02 (−0.028)0.980
      Significant at the 5 percent level.
      (−0.009)
      0.979
      Significant at the 10 percent level.
      (−0.012)
      Sex (Male = 0; Female = 1)0.963 (−0.035)1.580
      Significant at the 5 percent level.
      (−0.333)
      0.620
      Significant at the 1 percent level.
      (−0.036)
      1.353
      Significant at the 1 percent level.
      (−0.12)
      * The total number of observations was 25,176.
      A category for the employment choice of other/unknown (n = 532) was included in the regression analysis but is not presented in the table.
      SE: Standard error.
      § Relative risk ratios were estimated by using robust SEs.
      Graduation year indicator variables were included in the regression analysis but were omitted from the table. Standard errors were clustered according to school. Total educational debt was adjusted for inflation and divided by $10,000.
      # Significant at the 1 percent level.
      ** Significant at the 5 percent level.
      †† Significant at the 10 percent level.
      The results show that total educational debt was associated negatively with an intention of seeking advanced education; obtaining a teaching, research or administration position; or entering government service compared with entering private practice. This finding was consistent with our hypothesis that higher debt levels could encourage more students to choose to enter private practice. However, the magnitude of the effect was relatively small compared with that of other explanatory variables.
      One concern with using large data sets is that the results may be statistically significant but of such a small magnitude that they are not practically meaningful. To give context to the magnitude of our results, we translated the findings into what an increase in debt would mean for the labor market. Holding all other variables constant, each $10,000 increase in a student's debt decreased his or her likelihood of choosing advanced education by 1.5 percent (RRR, 0.985; 95 percent confidence interval [CI], 0.978-0.991); choosing teaching, research and administration by 3.1 percent (RRR, 0.969; 95 percent CI, 0.954-0.986); or choosing a government service position by 8.4 percent (RRR, 0.916; 95 percent CI, 0.908-0.924) relative to choosing private practice. The intention to enter public health was not significantly correlated with educational debt.
      African American dental school students were approximately twice as likely as were white dental school students to intend to enter advanced education or government service compared with private practice and more than three times more likely to choose public health. The students' sex did not have a significant effect on seeking advanced education compared with entering private practice. However, female students were 58 percent more likely to choose teaching, research and administration, 38 percent less likely to choose a government service position and 35 percent more likely to choose a public health position than were male students. A student was 15 percent less likely to seek advanced education, 47 percent less likely to seek a government service position and 28 percent less likely to seek a public health position if his or her parent was a dentist.
      The association between total educational debt and intended primary activity was similar across the various alternative specifications (results available on request). Total educational debt squared was not significantly associated with choosing advanced education or teaching, research and administration, and it was only significant at the 10 percent level for choosing government service and public health positions. The results also were not statistically different when compared with inclusion of all reported debt levels and all ages rather than exclusion of students who reported being younger than 24 years or older than 50 years and having total educational debt levels greater than $600,000. The fact that the results did not change across different specifications supports the initial finding that total educational debt was associated with a positive but relatively small increase in the likelihood of choosing private practice.
      We used the values shown in (Table 5 to predict how students' choices of employment would change if the mean educational debt per dental school student increased, holding other things constant. The hypothesis predicted that a $10,000 increase in mean debt would result in 1.5 percent (RRR, 0.985) of students changing their intended activity from advanced education to private practice. Because there were 10,503 students intending to enter advanced education, the increased debt would have resulted in 158 fewer people intending to pursue advanced education cumulatively from 2004 through 2011. The increased debt also would have resulted in five (3.1 percent) fewer students pursuing teaching, research and administration and 164 (8.4 percent) fewer students seeking a government service position. Instead of those employment options, the students would have transferred their intended activity to private practice, resulting in 326 (2.5 percent) more people intending to enter private practice over the eight-year period.
      Figure 2 shows the relative importance of different variables. A bar above zero percent indicated that there was a higher probability of choosing private practice. A bar below zero percent meant there was a higher probability of choosing an alternative intended employment option. For a $10,000 increase in debt, there was a small increase in probability that a student would pursue private practice versus seeking advanced education; teaching, research and administration; or a government service position. Female students were more likely than were male students to pursue teaching, research and administration and public health positions over private practice, but they were less likely to choose government service positions. Black students were more likely to choose advanced education, government service and public health positions over private practice than were non-black students. Finally, if a student's parent was a dentist, he or she was more likely to pursue a private practice position than an advanced education, government service position or a public health position.
      Figure thumbnail gr2
      Figure 2Comparison of magnitude of preference for private practice. A value greater than zero percent means private practice is more likely, and a value less than zero percent means private practice is less likely.

      DISCUSSION

      Although the findings of our analysis were consistent with the main hypothesis' prediction and with the findings of prior studies, they address the relative importance of educational debt to a senior dental school student's intent to enter private practice. Once we controlled for other characteristics, the magnitude of the effect of educational debt on intended employment choice was relatively small. These results run counter to the conventional wisdom that rising educational debt is a primary driver pushing students toward private practice. This result is consistent with our observation that students are not choosing private practice more frequently over time.
      One weakness of our study was the lack of data on students' actual employment after graduation, ultimate career path and future earnings. To our knowledge, no studies have been conducted regarding the relationship between intended and actual primary activity after graduation. Because the ADEA's survey was administered close to graduation, we could only assume that intended and actual activities were correlated. The accuracy of this assumption depends on vacancy rates, the difficulty in entering different employment sectors and the availability of information given to senior dental school students regarding the job market. At a minimum, the intended primary activity indicated the demand for different activities after graduation, which can be a useful gauge of the incentives graduating seniors faced. We also do not have information on the extent to which people switch employment sectors midcareer, which would alter expected lifetime earnings and could change willingness to work in employment sectors, such as public health, in the short term.
      There were a number of other variables we omitted that could have influenced employment choice, such as GPA, student loan repayment programs, internships and other opportunities. Some variables, such as GPA, would be difficult to include directly because causality could go in either direction. A student with a high GPA may have better access to advanced education. On the other hand, a student planning on entering a graduate program may choose to study more to obtain sufficiently high grades. However, a proxy for academic ability would be a valuable addition to the hypothesis. Other variables, such as the availability of student loan repayment programs, graduate training, internships and opportunities in the military and public health, would add to the analysis. Including participation in one of these programs in the hypothesis would not clarify what the student would have done otherwise. However, inclusion of the availability or access to these programs in a given year would provide a useful contribution to the analysis. If an omitted variable is correlated with an included variable, then the effect may be attributed incorrectly to the included variable. If the omitted variable is not correlated with the included variables, then the regression analysis results will be accurate but the overall hypothesis will not explain a student's intended employment choice fully.
      The results of our study show that student characteristics have a larger influence on intended employment choice than does total educational debt. Key factors in the decision to pursue advanced education or a government service position were race or ethnicity. Furthermore, female and black students were much more likely to pursue public health positions. One factor that encouraged students to pursue private practice was having a parent who was a dentist. The expectation of taking over a parent's dental practice may have explained the influence of having a parent as a dentist on the intent to enter private practice after graduation.

      CONCLUSION

      In this study, we focused on the relative importance of educational debt on dental school seniors' intended employment after graduation. We controlled for a variety of characteristics and found that although educational debt was significant, the magnitude of its effect was relatively small compared with other characteristics. Despite educational debt receiving a lot attention, demographic enrollment trends appeared to be more important in determining the future dental labor market. Our findings suggest that focusing on the characteristics of the student body may be a better avenue for understanding employment after graduation.
      Key areas for future research include extending the analysis back to when major shifts in student body composition occurred. There were few changes in sex or race during our period of our study, but major changes had occurred in prior decades.
      • Woolfolk MW
      • Price SS
      Dental education: evolving student trends.
      Another area for future research is to examine the relationship between intended and actual employment, long-term career earnings and flexibility in moving between employment sectors in the dental labor market. The findings from our study, nonetheless, can help dentistry gain a better understanding of the factors that influence students' employment decisions.

      References

        • Asch DA
        • Nicholson S
        • Vujicic M
        Are we in a medical education bubble market (published online ahead of print Oct. 30, 2013)?.
        N Engl J Med. 2013; 369: 1973-1975
        • Wanchek TN
        • Rephann T
        • Shobe W
        Oral Health and the Dental Care Workforce in Southwest Virginia. Weldon Cooper Center for Public Service, University of Virginia, Charlottesville, Va2011: 36 (Accessed Feb. 17, 2014.)
        • Valachovic RW
        Workforce issues. PowerPoint presentation to: NGA Policy Academy for State Officials Improving Oral Health Care for Children meeting; May 7, 2001; Nashville, Tenn.
        (Accessed Feb. 17, 2014.)
        • American Dental Association
        American Dental Association dental workforce model: 2008-2030. ADA Health Policy Resources Center, Chicago2010
        • Parker MA
        New practice models and trends in the practice of oral health.
        N C Med J. 2012; 73: 117-119
        • Merchant VA
        Are we willing to change?.
        J Mich Dent Assoc. 2012; 94: 14
        • Lanzon J
        • Edwards SP
        • Inglehart MR
        Choosing academia versus private practice: factors affecting oral maxillofacial surgery residents' career choices.
        J Oral Maxillofac Surg. 2012; 70: 1751-1761
        • Altman DS
        • Alexander JL
        • Woldt JL
        • Hunsaker DS
        • Mathieson KM
        Perceived influence of community oral health curriculum on graduates' dental practice choice and volunteerism.
        J Dent Educ. 2013; 77: 37-42
        • Davidson PL
        • Nakazono TT
        • Carreon DC
        • Gutierrez JJ
        • Shahedi S
        • Andersen RM
        Reforming dental workforce education and practice in the USA.
        Eur J Dent Educ. 2011; 15: 73-79
        • Okwuje I
        • Anderson E
        • Valachovic RW
        Annual ADEA Survey of Dental School Seniors: 2009 graduating class.
        J Dent Educ. 2010; 74: 1024-1045
        • Dhima M
        • Petropoulos VC
        • Han RK
        • Kinnunen T
        • Wright RF
        Dental students' perceptions of dental specialties and factors influencing specialty and career choices.
        J Dent Educ. 2012; 76: 562-573
        • Scarbecz M
        • Ross JA
        The relationship between gender and postgraduate aspirations among first- and fourth-year students at public dental schools: a longitudinal analysis.
        J Dent Educ. 2007; 71: 797-809
        • American Dental Association
        2010-11 survey of dental education academic programs, enrollment and graduates: volume 1. ADA Survey Center, Chicago2012: 39-44
        • Okwuje I
        • Anderson E
        • Valachovic RW
        Annual ADEA Survey of Dental School Seniors: 2008 graduating class.
        J Dent Educ. 2009; 73: 1009-1032
        • Woolfolk MW
        • Price SS
        Dental education: evolving student trends.
        J Dent Educ. 2012; 76: 51-64