The fundamental goal of PhD programs is to cultivate independent and leading researchers who are able to adapt to diverse work in academic, industrial, and other professional environments.1 PhD student training is important for health care education and academics.2 Traditionally, the PhD held a remarkable position among academic credentials as a symbol of advanced study and academic training, and as a key element in new knowledge creation. Further studies of doctoral students have focused on several topics, including supply and demand,3 incentives for students and supervisors,4 diversity of doctorates,5,6 and other factors associated with the completion of doctoral programs such as gender,7 supervisory mentoring,8 and financial support.9 However, to our knowledge, there is still a dearth of research that specifically assessed factors that influence individuals to apply to health care doctoral programs. Despite the rapid growth of interest in master's programs in Taiwan, little is known about whether and why these students intend to pursue PhD degrees after completion of their master's degrees. Therefore, the aim of this study was to identify the factors associated with the intention of master's degree graduates in health care to obtain a PhD, through data analyses of the nationwide Graduate Destination Survey (GDS) in Taiwan. This annual follow-up survey provides a good source of information to explore factors related to the intention of master's degree graduates to enroll in health care PhD programs. For example, Drennan10 has conducted a survey to investigate the professional and academic destinations of master's degree graduates in nursing. After a review of the literature, it is postulated that the intention to pursue a PhD degree in health care is associated with graduates' attributes, academic performance, socioeconomic status, and the extent of financial support. Through the analysis of GDS 2007, factors related to the goal of becoming a PhD in health care could be identified, and valuable information could be generated to provide insights into PhD training in the health care professions.
2.1. Data collection
The data analyzed in this study were obtained from GDS 2007, part of the Taiwan Integrated Postsecondary Education Database (https://www.cher.ntnu.edu.tw), which is constructed and maintained by the Center for Educational Research and Evaluation, National Taiwan Normal University, and sponsored by the Department of Higher Education of the Ministry of Education since 2005. The aim of the GDS project is to collect longitudinal data of higher education graduates and construct a valid and reliable database to illustrate their present circumstances, and the trends of their destinations before and after graduation, to be used as a reference for related decisions and reforms in education, human resource, and the economy and industry in Taiwan. Information is gathered annually through distributed questionnaires. In this study, factors associated with the intention of master's degree students to enroll in health care PhD programs were investigated by using data from the 2007 GDS. The fields of health care in this study include medicine, public health, pharmacology, rehabilitation, nutrition, nursing, biotechnology and laboratory science, and dentistry.
The 2007 GDS was a census survey which included 51,427 graduates who earned their master's degrees that year. Note that according to the statistics from the Ministry of Education, the number of master's degree graduates in 2007 reached 54,387 (http://www.edu.tw/statistics/index.aspx). The questionnaire survey was conducted soon after graduation. There were 2587 (5%) health care graduate students recruited into this survey, and 2036 of them responded to the survey questionnaire, with a response rate of 78.7%. Of the respondents, 368 male graduates (18.1%) were preparing to serve their conscription-based military duty after graduation and were excluded from further analyses. The remaining 1668 respondents were further classified into two groups. One group was composed of graduates who would choose to pursue a PhD degree (14.4%), and the other group contained the remaining 1428 graduates. Factors included in the survey were school type, gender, graduate school admission, years in graduate school, attendance of cram school for entrance examination, work during graduate study (i.e., receiving in-service graduate education), multidisciplinary thesis, participation in academic community, conference paper submission, journal article submission, award-recipient designation, granted patent, source of tuition and living expenses, and the educational level of parents. These baseline characteristics are presented in Table 1.
2.2. Data analysis
Statistical comparisons of baseline characteristics between the two groups mentioned earlier were performed using chi-square tests. Univariate logistic regression model was used to evaluate the effects of miscellaneous factors included in the survey on master's degree graduates' intentions to pursue a PhD in health care. The odds ratios (OR) and 95% confidence intervals (CI) were calculated for each factor. Individual Wald statistics were also reported. Afterward, a stepwise model selection strategy, based on likelihood ratio test with the entry criterion of 0.05 and removal criterion of 0.1, was used to select statistically significant influential factors of the intention to pursue a PhD degree in health care. In order to evaluate the predictive validity of the selected model, receiver operating characteristics (ROC) curve analysis was applied, and the area under the ROC curve (95% CI) was computed.11 All statistical analyses were conducted using the PASW Statistics 18 (SPSS Inc., Chicago, IL, USA).
3.1. Baseline characteristics of the study population
Table 1 compares baseline characteristics between the PhD-oriented and non-PhD-oriented groups. There was no significant difference in the distribution of school type, graduate school admission, multidisciplinary thesis, and participation in the academic community between the two groups. However, the proportion of male gender and attendance rate of cram schools for entrance examination of graduate schools were significantly higher for the PhD-oriented group. Moreover, the non-PhD-oriented group had significantly higher proportions of respondents spending more than 2 years in graduate school and on the job but had a lower submission rate of conference papers than their counterparts. In general, parents' educational level of the PhD-oriented group is significantly higher than those of the non-PhD-oriented group. It was also noted that a larger proportion of graduates in the PhD-oriented group had their tuition and living expenses paid for by their parents or relatives than those in the non-PhD-oriented group.
3.2. Univariate analysis
Table 2 illustrates the associations of miscellaneous factors with the intention to pursue a PhD degree in the univariate analysis. Among these factors, female gender and additional years attending graduate school and on the job status had a negative effect on the intention to obtain a PhD degree in health care. Attending cram school for entrance examinations, having submitted a conference paper, and an elevated higher education level of parents had positive effects on the intention to pursue a PhD degree in health care. Submission of peer-reviewed journal article was of borderline significance (p=0.05). Regarding the source of tuition and living expenses, those who had to pay for their own tuition and costs also tended to have a weaker intention to pursue a PhD degree in health care.
3.3. Multivariable analysis after the model selection
Table 3 presents the estimated effects of selected factors on intention to pursue a PhD degree in health care after completion of the model selection process. There were seven variables identified as independent factors which were significantly associated with master's degree graduates' intention to obtain a PhD degree in health care. In general, female graduates and those who spent more than 2 years in graduate school had a reduced intention to pursue PhD degrees in health care (adjusted OR=0.18 and 0.46, respectively). Graduates who worked while pursuing their graduate studies also had a lower inclination to pursue a PhD degree in health care (adjusted OR=0.47). However, submission of conference papers and peer-reviewed journal articles had a positive effect on the intention to become a PhD in health care (adjusted OR=1.61 and 1.48, respectively). Regarding the source of tuition, those who paid for their school fees themselves tended to have a weaker intention to pursue a PhD degree in health care (adjusted OR=0.48). Moreover, the higher educational level of the mother was also a promoting factor of the intention to pursue a PhD degree in health care. The effects of other factors in the univariate analysis were not significant after adjustment for these selected factors.
3.4. ROC curve analysis
Fig. 1 depicts the ROC curve analysis of the selected model. The predictive validity of the selected model assessed by area under the ROC curve for intention to pursue a PhD degree in health care is 0.77, with 95% CI 0.73– 0.8.
In this study, seven independent factors were identified to be associated with master's degree graduates' intention to pursue a PhD degree in health care from the data analyses of GDS 2007. A graduate destination survey of health care students is a good source of information to explore the factors related to the intention of health care graduate students to pursue a PhD degree.10 It is well known that dimensions of a person's individuality are essential to understand career choices.12–14 These factors may affect the decision of a master's degree graduate to pursue accomplishment in higher academic, and these findings additionally have implications in health care education. The final model with these selected factors has acceptable predictive validity, which implies that the model may have potential applications to identification of future PhD students and policies associated with health care education.
Factors related to academic performance in graduate school tend to promote the intention to pursue PhD study as well, including submission of conference papers and peer-reviewed journal articles. Academic writing is an essential task and skill for graduate students,15 primarily because most graduate institutes in Taiwan take academic publication as a compulsory criterion for graduation. Graduate students who are successful in getting their work published in an academic publication show greater academic potential, and a talent for academic publication is also a favorable characteristic when applying for a PhD program. It is not amazing that these factors can encourage master's degree students to pursue higher academic accomplishments since these academic performances reveal their research potential, which is essential for the pursuit of a PhD degree. In contrast, it is not surprising that spending more years in graduate school tended to lower the intention of students to obtain a PhD degree in health care.
Another finding is that female graduates have a lower inclination to pursue a PhD degree. The gender difference in pursuit of academic degree, particularly PhD degrees, is a common phenomenon worldwide.16–21 In this study, a more than 5-fold discrepancy was noted between males and females regarding the odds of pursuing a PhD degree. Gender role expectation in society,22 a chilly climate for women in an academic setting, and disadvantages due to unconscious bias are all responsible for this phenomenon.11 In order to improve diversity in the health care pipeline, more strategies should be considered to promote the enrollment rate of female graduates for PhD programs in health care.11
Because a longer duration of full-time study often postpones the timetable of when students enter the job market, as well as when they become self-supporting, a PhD student may need social and financial support from others, particularly their parents. Socioeconomic background is an important component of the transition of students,23 and parents' educational encouragement has a favorable influence on self-confidence, which certainly leads to a higher probability of educational pursuit and performance.24 Hence, it is rational to anticipate that parental socioeconomic status, which may be reflected by the parents' educational level, will be a significant predictor of an individual's intention to apply to doctoral programs in health care since financial and emotional support from the family makes a substantial contribution to any master's degree graduate's intention to pursue a PhD degree.
Economic considerations have long been regarded as major determinants of whether or not a student pursues higher academic achievement. Previous studies have demonstrated that students who support themselves have a higher attrition rate25 and take more time to complete their degrees.26 It has also been demonstrated that the financial burden is an obstacle to receiving further PhD training.24 It takes a longer time to pursue a PhD degree than earn a master's degree, and doctoral students may have additional obligations (children, elder care, mortgage, auto loan, etc.) during their graduate studies. These financial burdens also have a great impact on their decision to pursue a doctorate. In this study, it is noted that those who had to pay for their tuition tended not to pursue a PhD degree in health care either. For financially disadvantaged students, scholarships and research assistantships can enhance their intention to attain further academic achievement. Note that working during graduate study also lowers one's inclination to obtain a PhD degree. Although work experience may be beneficial for job searching,23,27 it is not necessarily helpful as a mechanism to ensure further academic pursuit. In fact, many students on the job complete a master's program with specific career goals in mind. They are more concerned with the direct application of their knowledge, and the benefits of a diploma to their work instead of academic research. Further academic achievement is not their primary interest.
There are several limitations in this study. First, most of the factors included in the analysis are baseline characteristics. The GDS 2007 did not include data regarding psychological aspects or social support,24 both of which may play critical roles in student decisions as to whether to pursue PhD degrees. However, despite a lack of these data, the selected model still possessed acceptable predictive validity. Future studies in similar topics should pay more attention to these untouched aspects. Second, the GDS did not contain data about the classification of health care professionals. Therefore, it is impossible to evaluate whether factors affecting the intention to pursue a PhD degree were different among the different health care professions, such as medicine, public health, or nursing. Third, this study used the database of GDS 2007, which was a cross-sectional survey. The relationships explored in this study were conditional on a specific period, and it is not possible to evaluate the variation and trend of these relationships with time. A longitudinal follow-up survey will provide more valuable information about the dynamic aspects of these issues. Another validation survey may also be beneficial to test the consistency of our results. Fourth, although the response rate of this survey was high, it was not precluded from the potential of selection bias.
In conclusion, this study identified seven independent factors associated with master's degree graduates' decisions to pursue PhD degrees in health care via data analyses of the 2007 GDS. We have demonstrated that the baseline characteristics of master's degree graduates are important determinants of intention to pursue a PhD degree in health care. PhD education in health care prepares individuals for their roles as the next generation of faculty, researchers, and technology leaders. More and more investigations and debates argue that a PhD-level education is of manifest importance and necessity. The results of this study imply that the university has possible strategies to identify potential PhD students and policies to attract other types of PhD students who are needed in this field.
1. Manathunga C, Lant P. How do we ensure good PhD student outcomes? Educ Chem Eng
2. Gill D, Griffin A, Woolf K, Cave J. Twelve tips for studying medical education at doctoral level. Med Teach
3. Basil MD, Basil DZ. The marketing market: a study of PhD supply, demand, hiring institutions, and job candidates. J Bus Res
4. Mangematin V. PhD job market: professional trajectories and incentives during the PhD. Res Policy
5. Neumann R. Diversity, doctoral education and policy. High Educ Res Dev
6. Usher R. A diversity of doctorates: fitness for the knowledge economy? High Educ Res Dev
7. Seagram BC. An investigation of gender and other variables on time to completion of doctoral degrees. Res High Educ
8. Green SG, Bauer TN. Supervisory mentoring by advisers: relationships with doctoral student potential, productivity, and commitment. Pers Psycho
9. Ehrenberg RG, Mavros PG. Do doctoral students' financial support patterns affect their times-to-degree and completion probabilities? J Hum Resour
10. Drennan J. Professional and academic destination of masters in nursing graduates: a national survey. Nurs Educ Today
11. Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem
12. Adya M, Kaiser KM. Early determinants of women in the IT workforce: a model of girls' career choices. Inform Technol People
13. Mau WC. Cultural differences in career decision-making styles and self-efficacy. J Vocat Behav
14. Patton W, Creed P. The relationship between career variables and occupational aspirations and expectations for Australian high school adolescentsc. J Career Dev
15. Swales J, Feak CB., 2004. Academic writing for graduate students: essential tasks and skills, University of Michigan Press, Ann Arbor, MI.
16. Appel ML, Dahlgren LG. Swedish doctoral students' experiences on their journey towards a PhD: obstacles and opportunities inside and outside the academic building. Scand J Educ Res
17. Fox MF. Organizational environments and doctoral degrees awarded to women in science and engineering departments. Women Stud Q
18. Gupta N. Indian women in doctoral education in science and engineering. Sci Technol Hum Val
19. Kurtz-Costes B, Helmke LA, Ülkü-Steiner B. Gender and doctoral studies: the perceptions of Ph.D. students in an American university. Gender Educ
20. McClure J, Meyer LH, Garisch J, Fischer R, Weir KF, Walkey FH. Students' attributions for their best and worst marks: do they relate to achievement? Contemp Educ Psychol
21. Wall S. Of heads and hearts: women in doctoral education at a Canadian university. Women Stud Int Forum
22. Delamont S., 2006. Gender and higher education. In: Christine S, Becky F, Smulyan L, editors., The Sage handbook of gender and education. SAGE Publications, Thousand Oaks, CA, pp. 179-189.
23. Salas-Velasco M. The transition from higher education to employment in Europe: the analysis of the time to obtain the first job. High Educ
24. Stewart M, Williamson I, King J. Who wants to be a business PhD? Exploring minority entry into the faculty "pipeline". Acad Manag Learn Educ
25. Lassibille G, Gómez LN. Why do higher education students drop out? Evidence from Spain. Educ Econ
26. Siegfried JJ, Stock WA. So you want to earn a Ph.D. in economics? How long do you think it will take? J Hum Resour
27. Biggeri L, Bini M, Grilli L. The transition from university to work: a multilevel approach to the analysis of the time to obtain the first job. J R Statist Soc A