October 9, 2008
Satisfaction Among Current Doctoral Students in Special Education
By Wasburn-Moses, Leah
Despite the growing demand for professionals with doctoral degrees in special education, doctoral programs are not producing enough graduates to fulfill this need. Although large attrition rates exist in doctoral study across discipline, very little is known about the attrition or satisfaction of doctoral students in special education. This article reports the results of a nationwide satisfaction survey of 619 students from 78 doctoral programs. Findings indicate that students appear generally satisfied with their programs. However, areas of concern include program structure, overall workload, and quality of preparation in research. Implications for doctoral programs are presented. Keywords: doctoral education; survey research; program effectiveness
The federal government responded to these research findings by increasing funding for doctoral training in special education to enlarge the pool of potential faculty (R. Gilmore, personal communication, 2005). This funding was offered in the form of Special Education Leadership Personnel Preparation Grants. Institutions of higher education applying for these grants were required to ensure that (a) recipients would be prepared to work with children with disabilities from diverse backgrounds, (b) the program would attempt to limit time to degree, and (c) the program would limit the amount of work (e.g., graduate assistantships) required to complete the training program (U.S. Department of Education, 2005b). Currently, 81 federal leadership grants are in effect around the country at 41 different institutions of higher education (U.S. Department of Education, 2005a).
However, simply increasing existing numbers of doctoral students may not be sufficient to ameliorate shortages. It is well known that attrition is a serious problem in doctoral study across discipline and across institution. Nationwide, doctoral student attrition has been estimated to be 40% to 70% (Gardner, 2004). Attrition and feelings of dissatisfaction among doctoral students have been linked to a variety of issues, including a heavy workload, unclear program expectations, financial burdens, and a general lack of mentorship and support (Gardner, 2004). The report of the Vanderbilt study concludes with seven recommendations to resolve the shortage of special education faculty, including the improvement of doctoral student recruitment and funding and the increasing of doctoral program capacity. One of these strategies is to enhance mentoring of doctoral students: "any successful efforts to recruit more students . . . will be diluted if these individuals change their goals as a result of their doctoral training experience, or even worse become so disillusioned that they fail to complete the degree" (Deutsch Smith et al., 2001, p. 47).
The purpose of the current study was to expand upon the Vanderbilt study to include both a quantitative and qualitative analysis of doctoral student satisfaction. Also of interest was the relationship of satisfaction to specific individual and program variables targeted for change by the Vanderbilt study. These variables include (a) whether the student received a Leadership Personnel Preparation Grant, (b) student status (part-time versus full-time), and (c) whether the student relocated to attend doctoral study. The current study addressed two research questions:
1. What are strengths and weaknesses of doctoral study in special education, according to current students?
2. How does satisfaction vary by the individual and program elements targeted for change by the Vanderbilt study?
This study sampled current doctoral students in special education. The sample consisted of 619 individuals representing 78 of the 94 identified institutions offering doctoral degrees in special education (83%). The number of respondents represented approximately 38.2% of the current doctoral student population at those 78 institutions. The population estimate used in this study was based partially on department report and partially on an estimate of the average number of students per program. Departments at 66 of the 78 institutions provided the number of doctoral students in their programs (n = 1,372). The average number of doctoral students from these institutions (20.6 per program) was used to estimate the total number of doctoral students among the 12 remaining programs (n = 247). This calculation resulted in a total estimated population of 1,619 in the 78 responding institutions. Sixty-seven of the 78 participating institutions were research extensive institutions; the remainder were research intensive institutions (see Note 1). Thirty-seven of the 78 responding institutions were institutions that had received at least one Special Education Leadership Personnel Preparation Grant (U.S. Department of Education, 2005a).
The majority of respondents were female (83%) and married or living in a similar relationship (65%). A little under half (46%) had dependents. The majority were Caucasian (70%); the largest minority groups represented were African American (8%) and Asian (8%). Four percent were Hispanic and 3% Other. Respondents' average age was 34.7 years, with a range of 21 to 59. Eighty-one percent of respondents were attending research extensive institutions, and 19% were attending research intensive institutions. Approximately 40% of the participants of the study had received a Leadership Grant.
Twenty percent of respondents were in their 1st year of doctoral study. Twenty-six percent were in their 2nd year, 27% in their 3rd year, 15% in their 4th year, and 13% in their 5th year or higher. The majority of respondents (67%) were attending doctoral study full- time. Thirty-five percent had relocated to begin their doctoral study. The most common reasons why students selected the university they were attending were the opportunity to work with specific faculty (59%), the amount of financial support offered (52%), and the fact that they would not have to relocate (50%) (see Table 1).
Participants reported that their financial support came mainly from a fellowship, scholarship, or graduate assistantship (51%); their spouse's earnings (40%); other personal earnings (40%); and/ or a research assistantship (37%) (see Table 2). The instrument allowed participants to select multiple reasons and multiple sources of support. The majority of respondents (62%) reported that they would be seeking faculty positions upon completion of their programs. Fifteen percent stated that they would be seeking administrative positions, and a smaller percentage sought positions in teaching, research, or another area (see Table 3).
Participants were identified through their doctoral programs. The list of doctoral programs was compiled from a variety of sources. Programs identified by the National Center for Special Education Personnel and Related Service Providers (www.personnelcenter.org) as offering a doctoral degree (EdD or PhD) in special education were crosschecked with programs listed by the Higher Education Consortium for Special Education (HECSE) and by the Office for Special Education Programs (OSEP) in their list of Special Education Leadership Personnel Grant recipients (HECSE, 2004; Personnel Center, 2005; U.S. Department of Education, 2005a). This search uncovered 97 potential doctoral programs.
Data were collected online during the fall of 2005. Electronic (e- mail) invitations were sent to the programs chairs of each of the 94 identified doctoral programs in special education. As per the methodology utilized in the Vanderbilt study, direct contacts to departments and colleges (via phone and e-mail) were made to determine whether identified doctoral programs were current and active. Of the 97 institutions, 94 indicated that they (a) had a formal major in special education (i.e., either a PhD or an EdD in special education) and (b) currently had doctoral students (Deutsch Smith, Pion, Chowdhury Tyler, & Gilmore, 2003). Program chairs at the 94 institutions were sent an e-mail invitation and asked to forward the message to all of their doctoral students in special education. Follow-up phone calls were made 3 weeks following the initial invitation to nonresponding program chairs. The invitation contained a brief description of the instrument and a link to the online survey. One follow-up was conducted, also via e-mail, at a 3- week interval after the first participant response from each institution. The data collection period was extended to 3 months because late responders in a web survey tend to bring characteristics of respondents closer to characteristics of the population being sampled (Vehovar, Batagelj, Manfreda, & Zaletel, 2002).
The Satisfaction Survey of Special Education Doctoral Students is a new research instrument that was constructed using (a) Deutsch Smith et al's Survey of Doctoral Students in Special Education; and (b) the wider nationwide Survey on Doctoral Education and Career Preparation, which encompassed the humanities, social sciences, biological sciences, and physical sciences (Deutsch Smith et al., 2001; Golde, 1998). The survey is divided into four sections.
The first section of the survey instrument addresses various aspects of the students' program and is based loosely on the two survey instruments listed above. The questions are items intended to measure overall satisfaction and satisfaction in a variety of areas, including advisor, coursework, research experiences, and teaching experiences. The section also includes items about preparedness for responsibilities of the position respondents have planned to enter upon completion of their doctoral programs. One open-ended question asks participants to provide comments on their satisfaction or dissatisfaction with their programs.
The second section examines issues that the student would highlight as possibilities for improvement. The items in this section are adapted from the nationwide Survey on Doctoral Education and Career Preparation. It asks students which decisions they would reconsider if they could go back in time and start their doctoral programs again. Participants select a response of "no,""maybe," or "yes." These seven questions include selecting a different field, advisor, university, and dissertation topic.
The third and fourth sections of the instrument cover program information and demographic information. All of the questions from these sections are identical to those found in the 1999 Survey of Doctoral Students in Special Education (Deutsch Smith et al., 2001). This part of the study was replicated for purposes of sample comparison. Questions in the program information section include student status (full-time, part-time), area of concentration, funding sources, and postdoctoral position sought. Demographic information was collected on gender, marital status, number of dependents, race, and age.
The survey was pilot tested with 31 doctoral students at three different doctoral programs in the Midwest. Programs were selected for variety in location (urban, suburban, rural), grant status (one had received Leadership Grants, two had not), and type of institution (two research extensive institutions, one research intensive institution). Reliability on the entire sample was conducted on satisfaction items (a = .84) and on preparedness items (a = .85), the two components of the survey that were not a replication of previous surveys.
First, to validate the sample, chi-square statistics were used to determine differences between the current sample and the Vanderbilt study sample. Analyses of variance (ANOVAs) and t tests were used to determine differences between early and late responders by participant demographics.
Second, several analyses were conducted on satisfaction and preparation items. Differences among specific satisfaction and preparedness questions were analyzed through repeated-measures ANOVA. Correlations also were calculated to determine relationships between satisfaction and preparation items.
Third, qualitative data were coded through inductive analysis (Janesick, 2000). Qualitative data reported in this article are responses to the open-ended question, "Please make any comments you would like regarding satisfaction/dissatisfaction with your doctoral program in the space below." Selective coding was used to generate numerous categories encompassing each response. Categories generated by this process account for most of the individuals' responses. As the coding progressed, categories were compared to subsequent responses and revised to continue to encompass individual responses. Broader categories were then created from the codes to develop a more manageable list (Charmaz, 2000). Quotations reported in this article are illustrative of the responses in each category.
Finally, a factor analysis was conducted to determine factors contributing to overall satisfaction with doctoral program. The factor analysis was conducted on the 21 Likert-type items addressing satisfaction and preparedness. The factors were then analyzed to determine variation in satisfaction and preparedness by the three variables analyzed in this study: grant recipient status (participants who received Leadership Grants versus those who did not), student status (full-time versus part-time), and relocation status (participants who relocated to attend doctoral study versus those who did not).
Demographic and Background Data
Demographics were compared with those measured in the Vanderbilt study in 1999 to establish reliability. Across the two studies, gender, marital status, dependent status (whether participants had dependents), relocation status, and overall satisfaction showed no significant differences. Differences were found in age, race, student status (parttime versus full-time), career aspirations, and financial support. However, these differences could be attributed to the rise in leadership grants, which were intended to provide increased funding for doctoral study, and to target a younger, more diverse student body whose intent was to become faculty (Deutsch Smith et al., 2001; R. Gilmore, personal communication, 2005).
Early responders also were compared with late responders to evaluate representativeness of sample (Vehovar et al., 2002). Early responders were identified as those participants who responded between the first invitation and the follow-up invitation at approximately 3 weeks. Late responders were those who responded after the follow-up. Although late responders were more likely to be at the beginning of their doctoral programs, t(563) = 2.691, p = .007, and anticipated their programs to be shorter than early responders (as might be expended with students at the beginning of their programs), t(563) = 2.015, p = .044, no significant differences were found between early and late responders in terms of sex, race, marital status, student status, relocation status, or satisfaction. Especially with Internet surveys, differences can indicate issues with representativeness in that late responders tend to be more representative of the population as a whole than early responders (Vehovar et al., 2002).
Satisfaction and Preparedness
Quantitative data. When asked about program satisfaction, 25% said they were "completely satisfied" with their doctoral program. Almost half (48%) responded that they were mostly satisfied with their programs, 21% had mixed feelings, and a small percentage were mostly or completely dissatisfied with their programs (see Table 4). As shown in Table 5, of the specific satisfaction ratings, participants felt most satisfied with their advisors (M = 4.25). They felt least satisfied with their ability to juggle work and family (M = 3.47) and their overall workload (M = 3.83). The differences among satisfaction ratings were statistically significant, F(6, 683) = 29.687, p
Participants also were asked to report their overall preparedness for the positions they had planned after doctoral study. As reported in Table 6, about 3% felt "very poorly" or "poorly" prepared for their positions. The majority felt "well prepared" (45%) or "very well prepared" (25%). Participants felt most prepared to teach college courses (M =3.95) and to give professional service (M = 3.94). They felt least prepared to advise college students (M = 3.65) and to lead programs in the schools (M = 3.75). Differences among the specific preparedness questions were also statistically significant, F(4, 131) = 15.833, p
Correlations also were calculated relating specific items with overall satisfaction and overall preparedness for position. Overall satisfaction correlated most strongly with satisfaction with other faculty (r = .479, p = .001), with advisor (r = .477, p = .001), and with courses (r = .441, p = .001) were most strongly correlated with overall satisfaction when all satisfaction factors were tested. Preparedness to design and implement a research agenda (r = .537, p = .001) was most strongly correlated with overall preparation for position.
The majority of participants indicated that they would not change most of the aspects of their doctoral programs if they could go back in time and reconsider their decisions (see Table 7). Few specified they would change their decision to go to graduate school or the timing of their program (i.e., taking time off before or during their programs). About one third indicated that they might or would have selected a different dissertation topic (33%) or a different university (31%). Twenty-eight percent might or would have selected a different advisor. The only aspect that the majority of the participants indicated they might or would change was in the area of taking more courses outside their department. To this question, 20% responded "yes," they would take more courses outside their department, and 36% responded "maybe." Qualitative data. Participants were also asked to comment on satisfaction with their doctoral programs. As reported in Table 8, of the 371 comments received, more than half (51%) of comments expressed satisfaction, 40% expressed dissatisfaction, and 9% were neutral. The largest category was in the area of mentoring/support, which included 38% of responses. Participants mentioned the importance of mentoring and support from advisors and other faculty. Positive comments included, "My advisor has been very helpful and appears to have my best interest at heart when advising me"; "I couldn't ask for a 'better' faculty in terms of meaningful advice, encouragement, and availability"; and "The faculty at this university is very supportive, yet they encourage independence with the premise of creating leadership in the field." Responses appeared to define a balance between challenging, independent work and individual guidance and encouragement. Those who expressed dissatisfaction in this area cited faculty with self-centered motives and/or inadequate support: "I have initiated all contact and have had to repeatedly follow up on requests or commitments made to me ... it is dubious if I will be able to finish the program due to lack of support and feedback"; "Typical of ivory tower politics, my advisors were more interested in their own success than mine"; and "I have felt that I have been left on my own to jump many unnecessary hurdles that have been placed there only to see if we can handle the pressure."
Program requirements or program structure were mentioned in 29% of responses. These respondents commented on research and/or teaching opportunities, class or program requirements, and overall program organization. Those who mentioned research and/or teaching opportunities appeared to be concerned with an availability of a range of opportunities: "I feel I will exit this program a well- rounded researcher and clinician"; "I have been very satisfied with the amount and breadth of research opportunities"; and "I feel opportunities might be available, but they are limited and divided among the Doc students, so the students don't get a sampling of all areas." Comments surrounding class or program requirements also showed a need for balance: "The quality of the coursework has been outstanding and has more than made up for the lack of quantity in terms of choices for academic coursework"; "Courses were either extremely valuable or useless"; and "The rules and requirements constantly change however, you are not informed of these changes until after you have not met the requirement." Program organization was another theme in responses in this category: "Satisfaction is a systematically structured doctoral program"; "There has been little structure for really feeling like I'm a part of the whole"; and "lack of program manual . . . different professors giving different directions regarding the program . . . lack of leadership in the department."
Fewer responses discussed balancing work and family (8%), financial issues (6%), or expressed general satisfaction/ dissatisfaction about their programs. Of those who mentioned issues with balancing work and family, some respondents indicated an understanding by advisors, programs, and faculty of those needs: "My advisor has been supportive and flexible in any changes I felt I had to make in my program to juggle . . . school, family, work, etc."; "very flexible, and faculty acknowledge that we all have lives and families and work"; and "My professors have also been very understanding and accommodating regarding family and work. This has made a huge difference." Those who were dissatisfied indicated being overwhelmed by their workload and a lack of accommodation for family needs: "It is difficult to manage a family, full-time teaching, and a doctoral program and it seems that the requirements at (university) are sometimes unreasonable"; "very satisfied (with the program), and my comments regarding my dissatisfaction with time management is due to my responsibilities of working full time, having two small children, and trying to go to school as close to full time as possible"; and "It's a struggle to manage with kids if both work/study at the same time."
Only 6% of comments related to financial issues. Several of those who expressed satisfaction specifically mentioned grant support. Some commented on the importance of funding to their programs: "The efforts of the faculty to maintain financial support for students through the leadership grant is excellent"; "The lack of funding makes it difficult to have a life outside of the university even when there is time off'; "and "low in funding makes the pursuit of the degree rather dragging and distressing . . . funding opportunities are not well known and advertised."
Relationship to Individual and Program Factors
A factor analysis was performed on the 21 Likert-type satisfaction questions contained in the survey. An exploratory principal components factor analysis yielded a two-factor solution accounting for 42.1% of the variance in participants' overall satisfaction with their doctoral programs. The two factors were preparedness and satisfaction. Components of the first factor included preparedness to (a) publish in refereed journals, (b) teach college courses, (c) advise college students, (d) give professional service, and (e) lead programs in schools. The components of the second factor were satisfaction with (a) advisor, (b) other faculty, (c) financial support, (d) doctoral student colleagues, (e) courses, (f) research experiences, and (g) overall workload.
The two factors were then analyzed to determine any differences by grant recipient status, student status (fulltime versus part- time), and relocation status (students who relocated to begin doctoral study versus those who did not relocate). Several significant differences were found. First, grant recipient status was related to satisfaction, t (502) = 2.621, p = .009. Grant recipients felt more satisfied with their preparation (M = 4.06 on a 5-point scale) than those who did not receive grants (M = 3.90). Second, student status was related to perceptions of preparedness, r(566) = -2.170, p = .03. Part-time students felt better prepared for their future positions (M = 3.91) than full-time students (M= 3.77). Third, relocation status was also related to perceptions of preparedness, t(564) = -2.585, p = .01. Those who did not relocate felt better prepared (M = 3.88) than those who did relocate (M =3.72).
In general, doctoral students appeared to be most satisfied with mentoring and support. However, this relative strength may be masking problems with program structure, workload, and a lack of research-related activities. Balancing independent work with supported activities and providing a range of teaching and research opportunities also may present challenges. Doctoral program planners should consider for what they intend to prepare students and map program requirements and experiences to those outcomes.
Participants in this study appeared to be comparatively well satisfied with their advisors, despite the fact that 31% indicated they would or might select another advisor if they could go back and change their decisions. Many wrote positive comments about the mentoring or other personal support they had received. Clearly, mentoring and successful advisor selection are crucial for doctoral student success (Lovitts, 2001; Nettles & Millett, 2006). Not surprisingly, this study found satisfaction with advisor and other faculty to be highly related to overall satisfaction. These findings support previous research in doctoral study in education as well as across field (Nettles & Millett, 2006).
Because of these positive findings and because of the strong relationship between the presence of mentoring and doctoral student success, one might conclude that satisfaction with other program elements would be similarly high. However, the results of this study indicated that there may be some areas of concern. These issues included a need for a flexible yet clear program structure, greater emphasis on research, and a focus on doctoral student outcomes.
If about one third of doctoral students in special education are part-time, doctoral programs should consider making use of these students' experiences and knowledge and increase program flexibility, rather than forcing part-time students to fit into a doctoral program intended for full-time students. Participants' comments reinforced this need for flexibility, particularly for individuals who held outside jobs and/or who were parents.
Surprisingly, those who did not relocate and those who were part- time students reported feeling better prepared for their future careers. These individuals may be less likely to change positions after graduation, so they may be reporting the confidence they feel with their current positions. They may also be less integrated into the academic community and, as such, may not have a complete picture of faculty roles and responsibilities. Nevertheless, instead of the traditional assumption that doctoral students will seek out and locate the program with the closest fit to their interests, faculty should consider how students with a wide variety of interests could fit into their programs. On the other hand, expanding distance learning options may be a possibility for bringing students to programs that might match their needs. These issues may be a factor in the finding that 31% of respondents indicated that if they could revisit their decisions, they would or might reconsider their choice of university.
Doctoral students felt least satisfied with their ability to juggle work and family and with their overall workload. This is where the need for individualization in a program can be balanced with the need for structure. Many participants expressed frustration with being parttime students, with having to jump through hoops they felt were unnecessary, and with being part of a program with little structure and guidance. Doctoral study is notorious for being a fast- paced blend of courses, teaching, research, exams, outside service, and the dissertation. Just as preservice programs map content and products to professional standards, perhaps some basic standards for advanced graduate study in special education might be in order. At a minimum, programs should ensure that they have a comprehensive, up- to-date manual with sample yearly plans and justification for each required activity, a list of grant-related requirements for students who are grantsupported, and an orientation for new students. The need for clear structure should not conflict with the need for program flexibility. Program manuals could outline several possible paths for degree completion, including different program lengths for part-time students, different research and teaching opportunities, alternative for course delivery (if available), options for summer courses/ employment, and options for those who wish to pursue administrative versus faculty positions.
Another aspect of doctoral program quality raised in the results of this study was the importance of research experiences to satisfaction and to program quality. Nearly 70% of participants in this study reported the intent to pursue faculty or research careers. This finding makes attention to research expectations even more crucial to the improvement of doctoral programs. In the current study, responses to questions surrounding satisfaction with research experiences and preparation for conducting independent research were correlated with both overall satisfaction and preparedness. This finding reinforces previous speculation that "the traditional doctoral program experienced by many in special education may be insufficient preparation for an active and productive research career" (Heward, 1995, p. 145). Furthermore, a full one third responded "yes" or "maybe" that they would have changed their decisions about their dissertation topics if possible. Dissertations can initiate individuals' research agendas; therefore, they and other research experiences must be productive.
After a comprehensive, nationwide study of more than 9,000 doctoral students across fields, Nettles and Millett (2006) concluded that "research productivity proved to be an important predictor of doctoral degree completion in all (areas)" (p. xxi). However, education ranked lowest across discipline in engaging doctoral students in publication-15% of doctoral students in education had published a journal article as compared with 30% overall. Furthermore, only 30% had presented at a national conference as compared with a 37% overall presentation rate (Nettles & Millett, 2006). Clearly, this is an area in which the field of education can improve. Programs are encouraged to be explicit about introducing this new culture and demonstrating the responsibilities and expectations for college and university faculty, as well as immersing students in the actual work of research.
Just as recent attention has been drawn to K-12 student outcomes and outcomes for preservice teachers, we need to focus our attention on doctoral student outcomes. For what are we preparing doctoral students in special education? Participants in the current study reported feeling least prepared to advise college students and to lead programs in the schools. Many wished they had taken more diverse coursework. Based on the results of this study, some doctoral programs may not expose students to the full range of roles and responsibilities of special education leadership personnel. More than half of participants expressed at least some desire to have taken more courses outside their department, perhaps indicating a need for broader preparation.
As such, programs should consider attempting to tie their courses and other experience to futures roles. With respect to leading programs in the schools, some still make the distinction between the PhD and the EdD, with the EdD being more practical and generally not leading to a position requiring intensive research (Shulman, Golde, Bueschel, & Garabedian, 2006). Because many special education doctoral programs are quite small (Deutsch Smith et al., 2001), two separate doctoral programs probably would be unfeasible for most institutions. Regardless, attention should be paid to each student's intended career path to ensure that they are well prepared for their future positions. Even if their courses remain the same, field experiences, research opportunities, comprehensive exam questions, and teaching experiences could differ. For example, future practitioners could engage in action research, provide professional development workshops for teachers, and/or coteach master's-level classes. Although the argument could be made that incoming students may not have an accurate prediction of their future goals, these experiences could benefit all students. Again, providing a balance between supported and independent activities and offering a range of experiences within a clear structure appear to be important to many doctoral students. Flexibility is imperative to adult students with multiple responsibilities, but providing a clear structure that allows students to understand options and select the most pertinent experiences to their career goals is also significant.
This study has three major limitations. The first limitation is the low response rate. The low response rate combined with a lack of knowledge about the population ordinarily results in a sample that is difficult to validate. However, (a) the similarity of many of the demographic variables to those measured previously, (b) the knowledge of the substantial subsequent changes in doctoral student funding in the form of Leadership Personnel Preparation Grants, and (c) the lack of differences between early and late responders all help validate the claims of this study. Nonetheless, it is important to note the limitations of the response rate and the restricted ability to contact nonresponders when interpreting the findings of the current study. Furthermore, there are little to no available data regarding current Leadership Grant awardees, so it is not possible to know whether the percentage of participants who obtained these grants is comparable to the population, either within or across institution.
The second limitation is the differences in survey content and methodology between the two studies. Although many of the items on the survey instrument were identical to the ones used by Deutsch Smith et al. (2001) in their 1999 "Study of Special Education Leadership Personnel with Particular Attention to the Professoriate," some of the items on the survey differed from those on the previous instrument. Specifically, the Vanderbilt study survey was longer, but it did not include items from the Survey on Doctoral Education and Career Preparation (Golde, 1998).
Furthermore, the previous survey was mailed, whereas the current survey was Web based. It is well documented that respondents to Internet-based surveys tend to be younger and better educated, are more likely to be single and male, and are less likely to come from an African American or Hispanic background than respondents to a mail survey (Vehovar et al., 2002). Although the current study did draw a much younger sample than the previous study, respondents showed similar marital status and a similar gender distribution and were actually more likely to be from a minority background, particularly African American, than their counterparts in the Vanderbilt study. When all of these factors are controlled, the most important factor predicting participation in a Web survey is computer and Internet usage, which probably varies at least somewhat among the doctoral student population (Vehovar et al., 2002).
Finally, the fact that the study surveyed active doctoral students also limits interpretation of results. Participants' perspectives on their doctoral study may change once they have graduated and entered new positions. Stability of the perspectives measured by this study is unknown due to the lack of data. All of these differences should also be taken into consideration when evaluating and interpreting results.
Implications for Research
Future research should attempt to establish a direct link between specific experiences in doctoral program and doctoral student attrition, as well as between those experiences and career attainment. Although doctoral programs presumably have some common elements, such as advisors, assistantships, and exam and dissertation requirements, specific program elements that might vary across university have still not been documented. It is safe to assume that attrition among doctoral students in special education is high, but a comprehensive study of exactly who leaves doctoral programs, when they leave, and why they leave is crucial if the goal is to close the supply-anddemand gap in special education leadership personnel.
Longitudinal studies following these students from the time of entry into their programs through the job search and into their initial years in their new positions may give insight into several unknowns, such as reasons for attrition and career choices. Universities must vary widely in their program and course offerings, funding, procedures for advisement, and research and teaching opportunities. Longitudinal studies would also offer opportunities to tie some of these program elements to doctoral student outcomes.
Another important angle for future research is to study the doctoral programs themselves. The knowledge of how programs are constructed and structured; the different paths for part-time students, grant recipients, and so on; their recruitment and retention strategies; the way in which faculty use time and interact with doctoral students; and the systems used for assessing doctoral students are all areas of interest when attempting to expand doctoral program capacity, decrease attrition, and increase program quality. In a time of personnel shortages, efficiency is crucial. If elements of "successful" programs (i.e., those with high retention rates and satisfied and successful graduates) can be identified, specific guidelines can be developed to assist programs with restructuring. In conclusion, due to the large attrition rate among doctoral students, efforts need to be made to consider what is known about satisfaction when planning and implementing programming for doctoral students. Based on the results of this study, doctoral programs should consider several elements when planning for reform, including managing competing needs for flexibility and structure and providing a wide range of research and teaching experiences. Students also need explicit orientation to the academic culture. A focus on research productivity while in doctoral programs should strengthen preparation for research-based positions as well as potentially increasing doctoral student retention.
Finally, programs should consider mapping doctoral student experiences and products with desired outcomes. Some basic guiding standards might assist program planners in their efforts. Longitudinal studies of doctoral students may give insight into attrition as well as into desirable program elements that can be implemented to improve program quality. If effective program elements are identified and implemented through research, inroads may be made in improving satisfaction and program quality and perhaps even in decreasing the shortages in special education leadership personnel.
1. A research extensive institution is defined as an institution with a wide range of baccalaureate degrees that awards more than 50 doctorates annually in 15 or more disciplines. Research intensive institutions are similar but award fewer doctorates in fewer disciplines (McCormick, 2000).
Charmaz, K. (2000). Grounded theory: Objectivist and constructivist methods. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (2nd ed., pp. 509-536). Thousand Oaks, CA: Sage.
Deutsch Smith, D., Pion, G. M., Chowdhuri Tyler, N., & Gilmore, R. (2003). Doctoral programs in special education: The nation's supplier. Teacher Education and Special Education, 26(3), 172-181.
Deutsch Smith, D., Pion, G. M., Chowdhuri Tyler, N., Sindelar, P. T., & Rosenberg, M. S. (2001). The study of special education leadership personnel with particular attention to the professoriate (No. H920T970006-00A). Washington, DC: U.S. Department of Education.
Gardner, S. K. (2004). Doctoral education in the United States: A synthesis of the literature. Unpublished manuscript, Pullman, WA.
Golde, C. M. (1998). Beginning graduate school: Explaining first- year doctoral attrition. New Directions for Higher Education, 101, 55-64.
Heward, W. L. (1995). Introduction to the topical issue: Training special education faculty for the 21st century. Teacher Education and Special Education, 18, 143-146.
Higher Education Consortium for Special Education (HECSE). (2004). HECSE membership list. Available from http://hecse.uky .edu/ hecse/memlist.html
Jablonski, A. M. (2001). Doctoral study as professional development of educators in the United States. European Journal of Teacher Education, 24(2), 215-221.
Janesick, V. J. (2000). The choreography of qualitative research design: Minuets, improvisations, and crystallization. In Y. S. Lincoln (Ed.), Handbook of qualitative research (2nd ed., pp. 379- 400). Thousand Oaks, CA: Sage.
Lovitts, B. E. (2001). Leaving the ivory tower: The causes and consequences of departure from doctoral study. Lanham, MD: Rowman & Littlefield.
McCormick, A. C. (2000). The Carnegie Classification of institutions of higher education, 2000 edition. Menlo Park, CA: Carnegie Foundation for the Advancement of Teaching.
Nettles, M. T., & Millett, C. M. (2006). Three magic letters: Getting to Ph.D. Baltimore: Johns Hopkins University Press.
Personnel Center. (2005). Personnel preparation programs. Retrieved June 8, 2005, from http://www.personnelcenter.org/get.cfm
Shulman, L. S., Golde, C. M., Bueschel, A. C., & Garabedian, K. J. (2006). Reclaiming education's doctorates: A critique and a proposal. Educational Researcher, 35(3), 25-32.
U.S. Department of Education. (2005a). Personnel preparation to improve the education of students with disabilities: Preparation of leadership personnel. Washington, DC: Author.
U.S. Department of Education. (2005b). Preparation of leadership personnel: Grant guidelines. Retrieved December 15, 2005, from http:/ /www.ed.gov.programs/osepprep/2005-325dk.doc
Vehovar, V., Batagelj, Z., Manfreda, K. L., & Zaletel, M. (2002). Nonresponse in Web surveys. In R. J. A. Little (Ed.), Survey non- response. New York: John Wiley.
West, J. (2001). The shortage of special education faculty: Why it is happening, why it matters, and what we can do about it. Nashville, TN: Vanderbilt University Press.
Miami University, Oxford, Ohio
Author's Note: The complete Vanderbilt study data are available in the full report, which can be ordered by e- [email protected] Correspondence concerning this article should be addressed to Leah Wasburn-Moses, Educational Psychology, 201 McGuffey Hall, Miami University, Oxford, OH 45056; e- mail [email protected]
Leah Wasburn-Moses, PhD, is an assistant professor of educational psychology at Miami University. Her research interests include teacher education, special education, and policy.
Copyright PRO-ED Journals Sep/Oct 2008
(c) 2008 Remedial and Special Education; RASE. Provided by ProQuest LLC. All rights Reserved.