|Home | About | Journals | Submit | Contact Us | Français|
Jay Hertel, PhD, ATC, and Thomas F. West, PhD, ATC, contributed to conception and design; acquisition and analysis and interpretation of the data; and drafting and final approval of the article. W. E. Buckley, PhD, ATC, and Craig R. Denegar, PhD, ATC, PT, contributed to conception and design, analysis and interpretation of the data, and critical revision and final approval of the article.
The study had 3 objectives: (1) to assess the educational history of doctoral-educated certified athletic trainers (ATCs) who work at academic institutions, (2) to determine the current employment characteristics of doctoral-educated ATCs who work at academic institutions, and (3) to identify which competencies doctoral-educated ATCs feel are important for new doctoral graduates to possess upon graduation.
Multiple sources were used to identify doctoral-educated ATCs who work at academic institutions. These individuals were surveyed to assess their educational histories, current employment characteristics, and opinions on desired competencies for new doctoral graduates. Data were analyzed using descriptive and inferential statistics.
Surveys were sent to 130 individuals, and the response rate was 89.2% (n = 116).
Subjects answered questions regarding their educational history and employment characteristics. A 5-point Likert scale was used to assess the importance of 22 competencies for new doctoral graduates to possess upon graduation. Comparisons were made between program directors and non–program directors, respondents employed at doctoral-granting institutions and non–doctoral-granting institutions, and doctoral student advisors and non-advisors.
Subjects reported several different educational backgrounds, job titles, and job responsibilities. Significant differences in job responsibilities and assessment of desired competencies were found between program directors and non–program directors, employees of doctoral-granting institutions and non–doctoral-granting institutions, and doctoral student advisors and non-advisors.
As new doctoral programs are established in athletic training, students should receive training as classroom instructors and program administrators, in addition to learning the skills necessary to perform independent research in athletic training.
The nature and scope of athletic training education in higher education have evolved considerably over the past 3 decades.1 After beginning as a minor component of a physical education degree, athletic training has developed into a distinct academic major at the undergraduate and graduate levels, with degrees often granted by a school or college of health professions. Educational reforms have led to attempts to establish many more accredited undergraduate athletic training curriculum programs as the internship route to certification is eliminated by the year 2004.1 With the establishment of new curriculums, more doctoral-educated certified athletic trainers (ATCs) are needed to fill the tenure-track faculty positions that are being created.
The National Athletic Trainers' Association (NATA) Education Task Force has published several recommendations to reform athletic training education.2 Provision 8 called for the formation of an Education Council. One proposed function of this council was to act as a resource for the development of doctoral programs in athletic training.2 Generally, the rationale for this recommendation was the need for more doctoral-educated ATCs. Specifically, the rationale included 3 reasons for increasing the number of doctoral-educated ATCs. First, if the athletic training profession is to affect higher education policy, there must be a critical mass of doctoral-educated ATCs in senior faculty and administration positions. Second, doctoral programs are needed because of the leadership that these programs provide in guiding research pertinent to the advancement of the athletic training profession. Third, doctoral programs will provide the next generation of athletic training educators.2
Literature regarding the professional preparation and employment characteristics of current doctoral-educated ATCs is lacking. Therefore, the purposes of our study were (1) to assess the educational history of doctoral-educated ATCs, (2) to determine the current employment characteristics of doctoral-educated ATCs, and (3) to identify the importance of competencies that future doctoral-educated ATCs should possess as perceived by current doctoral-educated ATCs. Comparisons were made between athletic training program directors (PDs) and non–program directors (NPDs), as well as between those working at research-oriented, doctoral-granting institutions (DGIs) and non–doctoral-granting institutions (NDGIs). Additionally, because of their pivotal role in the mentoring of future doctoral-educated ATCs, ATCs who were currently advising ATC doctoral students were asked about the importance of competencies.
We mailed a survey to a convenience sample of 130 ATCs with earned doctoral degrees who currently work at institutions of higher education. We gathered the names and addresses of subjects from recent sports medicine journals, NATA committee lists, and the list of accredited programs listed on the NATA World Wide Web site. The survey was approved by an institutional review board, and informed consent was provided by all those who responded. If the survey was not returned within 3 weeks of the initial mailing, we sent a follow-up mailing, which included another copy of the survey. The final response rate was 89.2% (n = 116).
Respondents were asked to identify the majors and minors of all higher education degrees (bachelor's, master's, and doctoral degrees) earned. Subjects were asked if they had graduated from NATA-approved or Committee on Allied Health Education and Accreditation (CAHEA)–accredited or Commission on Accreditation of Allied Health Education Programs (CAAHEP)–accredited bachelor's or master's degree athletic training programs (or both) and whether they pursued their doctoral degrees as full-time or part-time students. They were also asked to identify the funding sources for their doctoral studies. Descriptive statistics were calculated for the educational history items. All statistics were calculated using the SPSS 8.0 for Windows software package (SPSS Inc, Chicago, IL).
We asked subjects to identify their current job title (eg, program director, assistant athletic trainer), academic rank (eg, assistant professor, associate professor), and the department and college in which they were currently employed. We classified respondents as working at DGIs or NDGIs using the Carnegie Foundation for the Advancement of Teaching classification system.3 We also asked subjects if they were tenure-track employees and if they were currently advising any doctoral students who were ATCs. Descriptive statistics were calculated for the employment characteristic questions.
We asked subjects to indicate the number of hours per week they spent on classroom instruction, classroom preparation, clinical instruction, clinical instruction preparation, advising undergraduates, advising master's students, advising doctoral students, clinical duties, research, grant writing, professional service, administrative responsibilities, and other job-related responsibilities. We also asked subjects whether their current position allowed them the time necessary to maintain their clinical athletic training skills. Each time variable was analyzed using 2 separate 1-factor analyses of variance to compare the time spent in each of the above areas for (1) PDs vs NPDs and (2) employees of DGIs vs employees of NDGIs. The level of significance was preset at .05.
We asked subjects to rate the importance of 22 educational competencies (Table (Table1)1) for ATCs earning new doctoral degrees upon completion of their degree program. A 5-point Likert scale was used to rate the competencies, with 1 representing “not important” and 5 representing “extremely important.” The competencies were abstracted from a previous survey of allied health deans4 that examined the competencies of new graduates of allied health doctoral programs. The competencies were altered to be specific to the athletic training profession.
The mean for each competency was compared with the overall mean of the 22 competencies using a 1-sample t test to determine which competencies were deemed more and less important by the entire respondent pool. Three separate 1-factor analyses of variance were performed on the responses to each competency to examine differences in the perceived importance of each competency between (1) PDs and NPDs, (2) employees of DGIs and employees of NDGIs, and (3) doctoral student advisors and non-advisors. The level of significance was set at .05 for all analyses.
Eighty-six (74.1%) of the respondents were men, and 30 (25.9 %) were women. The mean ± SD age of the respondent pool was 42.0 ± 7.2 years. Of the 116 respondents, 90 (77.6%) had earned a bachelor of science degree at the completion of their undergraduate education, 11 (9.5%) had earned a bachelor of arts degree, and the remaining 15 (12.9%) had earned some other form of bachelor's degree. Seventy-three (62.9%) respondents had earned a master of science degree, 23 (19.8%) had earned a master of arts, and 20 (17.2%) had earned a master of education. Four individuals (3.4%) had earned a second master's degree. Forty-five (38.7%) respondents had completed an accredited (CAHEA or CAAHEP) or approved (NATA) undergraduate athletic training program, and 38 (32.7%) had completed an NATA-approved graduate athletic training program. The types and percentages of doctoral degrees received are shown in Figure Figure1.1. The areas of study in which these degrees were earned are presented in Figure Figure22.
Forty-nine percent (n = 57) of the respondents were full-time students while pursuing their doctoral degree, 26% (n = 30) pursued their doctorate part time, and the remaining 25% (n = 29) attended for a combination of part time and full time. Sixty-seven percent (n = 78) had a graduate assistantship, and 35% (n = 41) received funding from an employer for at least a portion of their time as a doctoral student. The duties performed by those with assistantships included teaching (49%, n = 57), clinical athletic training services (33%, n = 38), research (24%, n = 28), or some combination of these duties.
Job title data are presented in Table Table2.2. Note that respondents were allowed to report more than 1 job title. Fifty-seven percent (n = 66) held jobs within a department of exercise science or kinesiology, 35% (n = 41) within a department of allied health or medicine, and 8% (n = 9) in some other department. These departments were housed in a variety of different colleges, including colleges of health, physical education, and recreation (28%, n = 32); education (28%, n = 32); and allied health or medicine (22%, n = 26). Eighty-four percent (n = 97) of the positions held by ATCs with doctoral degrees were tenure track.
Nineteen (16%) respondents were currently advising ATCs who were pursuing doctoral degrees. These respondents were advising a mean of 2.65 ± 1.81 doctoral students, 2.25 ± 1.45 of whom were pursuing their degrees full time when the survey was completed. Eighty-seven percent of the current doctoral students were funded by the institution they attended. Most often, these students held teaching assistant or research assistant positions.
In the performance of their jobs, ATCs with a doctoral degree spent time in a variety of different areas (Table (Table3).3). Thirty-nine percent of respondents felt that they were unable to maintain their clinical skills in their current positions. Relative to the NPDs, the PDs spent significantly (P < .05) more time advising undergraduate students and performing administrative duties. Relative to respondents working at NDGIs, respondents working at DGIs spent significantly (P < .05) more time performing research and advising doctoral students and less time teaching in the classroom and preparing for classroom instruction.
The overall mean rating for all competencies for all respondents was 4.10 ± 0.36. Competencies deemed significantly more or less important by the entire subject pool, differences in competency importance between PDs and NPDs, differences in competency importance between employees of DGIs and NDGIs, and differences in competency importance between doctoral student advisors and non-advisors are shown in Tables Tables4,4, ,5,5, ,6,6, and and7,7, respectively.
As athletic training continues to emerge as a distinct entity in health care, doctoral programs designed specifically for ATCs must continue to evolve. The importance of different allied health professions developing discipline-specific doctoral programs has been previously emphasized.5 Additionally, the NATA Education Council has identified the need for development of more doctoral programs in athletic training.2 Our results demonstrate that more than half of our sample obtained doctoral degrees in majors other than the exercise sciences. Nearly one quarter of respondents earned degrees relating to education and administration. The need to expand the knowledge base of athletic training will require the training of doctoral-educated ATCs who have the skills necessary to perform independent research specifically in the area of athletic training. The creation of more doctoral programs specifically in the area of athletic training is one potential way to address this need.
Nearly one half of our respondents pursued their doctoral degrees exclusively on a full-time basis. This finding is of potential interest because Sakalys et al6 demonstrated different motivations and goals in registered nurses pursuing doctoral degrees on full-time and part-time bases. Full-time doctoral students were more likely to be open to a variety of different employment settings after completing their degrees, including teaching and research positions, administration and management positions, and clinical positions. Part-time students were more likely to have the goal of retaining their current positions and were more likely to want to teach full time upon completion of their degrees.6 This disparity between full-time and part-time doctoral students may be of interest to doctoral student advisors.
We found that doctoral-educated ATCs tend to be in the lower academic ranks in their institutions. Forty-four percent (n = 51) of respondents were at the instructor, adjunct, or assistant professor rank. This disproportionate number of athletic training faculty at the lower academic ranks could be due to the relatively young mean age of athletic training faculty. Administratively, only 17 (15%) held department chair positions, and none held higher administrative positions (eg, dean, provost). The NATA Education Task Force2 recommended that ATCs need to become involved in determining policy in their institutions by advancing to upper academic ranks and higher administrative positions. At this point, it appears that this goal is not being met on a national level.
The NATA Education Task Force2 also recommended that athletic training education programs move to align themselves with academic departments revolving around allied health professions. Only 35% of our respondents were employed within departments of allied health, and only 22% were housed in colleges of allied health. Athletic training faculty were most likely to be housed in departments of kinesiology or exercise science and colleges of health, physical education, and recreation, or education. Tenure and promotion decisions in the allied health and medical professions are typically based on the medical model, which includes a review of teaching, service, and scholarly activity that consists of research and clinical care expertise.7 In contrast, promotion and tenure decisions for non–health care faculty are typically based on the academic model, which only assesses a faculty member's teaching, service, and scholarly activity record; consideration of the clinical care expertise is quite variable among individual institutions.
Interestingly, 45 (39%) respondents stated that their current positions did not allow them the time necessary to maintain their clinical athletic training skills. Given that athletic training is a clinical health care profession, we believe that maintenance of clinical skills should be an important component of the job responsibilities of athletic training educators and researchers. Also, some faculty may have chosen to decrease their clinical workload in favor of a more academic workload as a matter of lifestyle change. Granting that the large majority (86%) of respondents were in tenure-track positions and most were not employed in departments of allied health, it is possible that clinical skills are not given due consideration in the tenure review process of these individuals.8 Further research is needed to examine this professional concern.
The time spent on job-related tasks performed by individual respondents varied widely. Interesting differences were noted in the time spent performing these duties between PDs and NPDs and between those employed in DGIs and NDGIs. Program directors spent significantly more time advising undergraduate students and performing administrative tasks. Doctoral-educated ATCs working at DGIs spent significantly more time advising doctoral students and performing research. These same respondents spent less time preparing for, and instructing in, the classroom than their counterparts at NDGIs. This difference most likely exists because faculty at DGIs have smaller formal teaching loads assigned to them than faculty at NDGIs. Faculty at DGIs are often expected to mentor graduate student research projects as part of their teaching responsibilities. Although those at DGIs may spend fewer hours teaching in the classroom, their total time performing teaching duties may be on par with the time spent by those at NDGIs. The emphases on different job-related duties depend on institution type and on whether or not the person holds the title of PD.
Our analyses used several multiple comparison models. Traditionally, when multiple comparisons are performed, the risk of type I error increases, and a Bonferroni correction should be made to adjust the level of statistical significance and make the analyses more conservative. With our list of 22 competencies, application of the Rom9-modified Bonferroni correction, a more liberal approach than the traditional method, resulted in a suggested P value of .002. We viewed this value as an overly conservative level of significance to be used for survey research, and thus, chose not to apply any method of Bonferroni correction. Given the nature of the data, we felt that the risks of making a type I error were of less concern than the loss of information related to differing perspectives on doctoral education that would have occurred with a more conservative analysis. We advise readers that type I error is likely to have occurred in the analyses of competency comparisons, although the specific comparisons affected are not known. Thus, the results should be interpreted accordingly.
Respondents rated competencies relating to teaching athletic training classes, performing athletic training research, possessing administrative skills, and providing professional leadership as being the most important for new doctoral-trained ATCs to possess. These competencies constitute the common triad of research, teaching, and service often used for tenure and promotion decisions using the academic model.
All of the competencies included in our survey had mean scores greater than 3.33 on a 5-point Likert scale. Therefore, none of the competencies were considered to be “not important” by our respondents. Instead, the competencies with statistically lower means should be considered as being relatively less important in comparison with the higher-rated competencies.
Additionally, a sampling bias may have occurred. We identified ATCs with doctoral degrees through lists of NATA committee members, sports medicine journal contributors and reviewers, and accredited athletic training program directors. As a result, our sample may have neglected to include ATCs with doctoral degrees who were not active in national professional activities. Also, our sample included ATCs with doctoral degrees who were working in academic disciplines other than athletic training (eg, physical therapy, adapted physical education, sports management).
Clinical athletic training skills, teaching classes outside of athletic training, obtaining external funding through grants, and performing research relating to clinical outcomes and athletic training education were rated as being the least important competencies for new doctoral-trained ATCs to possess. These findings may be due to the nature of the model for promotion and tenure evaluation by which most doctoral-trained ATCs are judged. For example, if clinical athletic training skills are not considered a valuable competency by promotion and tenure committees, the tenure-track athletic training faculty will consequently spend less time maintaining their clinical skills and dedicate more time to teaching, research, and service.
Interesting differences were noted in the ratings of competencies by PDs and NPDs. Program directors rated the administrative skills to direct a CAAHEP-accredited athletic training education program and clinical athletic training skills as significantly more important than NPDs did. Non–program directors rated the ability to perform collaborative research with exercise science faculty, to define research priorities in athletic training, and to advise doctoral students who are ATCs as being more important competencies than did PDs. These findings represent seemingly different priorities between PDs and NPDs, with PDs emphasizing competencies related to undergraduate education and NPDs emphasizing research competencies. Again, none of the competencies were rated as “not important” by either group, but rather the relative importance of these competencies differed between groups.
Several differences were also found between respondents currently advising doctoral students who are ATCs and those who were not advising these students. The doctoral student advisors identified 5 research-related competencies as significantly more important than the non-advisors did. The advisors rated the possession of administrative skills necessary to direct a CAAHEP-accredited athletic training program as significantly less important than did non-advisors. Because the doctoral student advisors are mentoring the next generation of athletic training educators, their views are likely to greatly influence new doctoral-educated ATCs. Conflicts appear to exist between the relative importance of competencies as viewed by PDs and doctoral student advisors, with advisors emphasizing research skills and PDs emphasizing administrative and clinical skills. It should be noted that 8 of the 19 doctoral advisors also served as program directors.
Similar findings have been noted among other health care professionals. The goals of nursing doctoral students changed as they advanced through doctoral programs, with more advanced students wanting to attain more research-related positions and fewer advanced students wanting positions at NDGIs upon completion of their degrees.10 The influence of doctoral advisors on their students may affect such career goal changes.
The results of the comparison of the importance of competency between employees of DGIs and NDGIs were very similar to those comparing the ratings of competencies by doctoral student advisors and non-advisors. Those at DGIs rated research competencies as most important and administrative skills as less important. The ability to generate external funding through grant writing was also rated as more important by those working at the more research-oriented DGIs. This finding was expected, as the focus of DGIs and NDGIs is typically different with regard to generating external research funding.
The creation of doctoral programs in athletic training is driven by the need to develop a theoretical and research knowledge base to validate the clinical practice of athletic training. The establishment of this knowledge base will allow athletic training to be viewed as a distinct field of study within academia. Boore5 asserted that the development of doctoral programs in an allied health profession (in this case, athletic training) is driven by the need to develop a research base that validates the common clinical practice techniques in the respective profession. Training more clinicians with research skills who can validate advances in the clinical practice of health care has been recommended.11
As doctoral programs in athletic training are developed at different institutions, graduates must be prepared to master certain competencies. Boore5 suggested that doctoral programs in the allied health professions should have 3 primary foci: theory of clinical practice, advanced care specialty training, and research. Elder and Nick4 reported 21 competencies that were rated as “very important” by deans of allied health colleges for new doctoral graduates to possess. These competencies were divided into 3 categories: general competencies such as logical thinking and leadership skills, teaching competencies, and research competencies.
Jansma and Surburg12 reported a total of 79 competencies for doctoral graduates in adapted physical education developed by members of that profession. They suggested that competencies for doctoral programs should be viewed as guidelines for preparing doctoral students in research, teaching, and service in academia rather than as a means of standardization of doctoral programs at different institutions.
We are not suggesting that a list of competencies like those now used to accredit entry-level athletic training education programs be created to accredit doctoral programs in athletic training. However, as new doctoral programs in athletic training develop, the goals needed to train the next generation of athletic training educators and researchers must be addressed in earnest by existing faculty. New doctoral programs must be built around the strengths of the faculty at each institution on an individual basis. Many doctoral programs will focus more on specific competencies than others. Individual faculty-student mentoring in research activities is one of the keystones of graduate education in the United States, and we believe that this mentoring should remain an integral part of doctoral education in athletic training. However, the mastery of a knowledge base incorporating the breadth of clinical athletic training practice is also extremely important in the development of future athletic training faculty. The potential of “overspecialization” of doctoral students has been suggested by Knapp,13 who cautioned that doctoral students can become so focused on specific areas within their fields of study that they are unable to teach broad-ranging undergraduate courses upon graduation.
Earning a terminal degree has been recommended for those ATCs aspiring to become PDs.14 Perrin and Lephart8 discussed the paradox for athletic training PDs who try to maintain clinical skills and perform independent research in an effort to achieve tenure and promotion. The importance of undergraduate health care instructors remaining clinically active has been previously discussed in the allied health literature.15 Including clinical practice as part of the service aspect of the tenure review process has been suggested, but often the review of athletic training faculty is not performed by other faculty involved in the clinical practice of health care.8 Additionally, Perrin and Lephart8 expressed the opinion that student perceptions of athletic training faculty who do not maintain clinical practice skills may be less than optimal and may result in lower educational outcomes. We highly recommend the use of the medical model for promotion and tenure reviews of athletic training faculty.
The role of doctoral-educated ATCs as clinical instructors varies extensively among individuals. Formal training in educational methods has been shown to increase the confidence of clinical supervisors during clinical instruction of student athletic trainers.16 These findings, combined with our sample's rating of the ability to teach undergraduate and graduate athletic training courses as important competencies for new faculty to possess, suggest that athletic training doctoral students should be given ample training and practice in performing both classroom and clinical instruction during graduate school. In conjunction with educational facets and advanced clinical knowledge, an athletic training doctoral program should also provide the student with thorough research and administrative training, which will prepare the student for a career in higher education.
The results of this survey give a comprehensive view of the educational history and employment characteristics of doctoral-educated ATCs. Additionally, the results elucidate the importance of specific competencies for new doctoral-educated ATCs upon completion of graduate school as viewed by current doctoral-educated ATCs. These findings may be helpful to institutions developing new doctoral programs in athletic training and to ATCs considering earning doctoral degrees.
“Educational History, Employment Characteristics, and Desired Competencies of Doctoral-Educated Athletic Trainers” is certainly a timely and important paper as the job market for terminally degreed athletic training scholars continues to grow. Dr Hertel and colleagues should be commended for beginning to provide data to describe this population of athletic trainers. They identified 3 purposes for their study: (1) to assess the educational history of doctoral-educated certified athletic trainers (ATCs); (2) to determine the current employment characteristics of doctoral-educated ATCs who work in academic institutions; and (3) to identify which competencies doctoral-educated ATCs felt were important for new doctoral graduates to possess upon graduation. Clearly, all 3 of these purposes are important.
The demographics on educational history and employment characteristics are very valuable for understanding the general trends relating to these issues. However, several sampling issues make it difficult to generalize these findings. The subjects of this study were from a sample of convenience. Samples of convenience are sometimes necessary, but not in this case. The NATA possesses a database that allows identification of ATCs who are members of NATA and possess terminal degrees. There are 2 basic problems in using a sample of convenience for this study. First, participants were identified through authorship in sports medicine journals, NATA committee lists, and a list of program directors. Individuals identified in this way do not necessarily represent the average or typical doctoral-educated ATC but more likely represent leaders in the profession. Although assessing education history, employment characteristics, and competency importance of education leaders is valuable, these data should not be misrepresented as typical. Further, the stated sampling procedure did not ensure that the doctoral-educated ATCs were involved in athletic training education or that they possessed their positions as a result of being doctoral-educated ATCs. Certified athletic trainers with terminal degrees may be primarily physical therapy educators or may work in other settings not directly related to athletic training. Not working in a setting directly related to athletic training is less of a concern than the extent to which other credentials may have influenced the nature of the position (ie, having a physical therapist license was the influential factor in obtaining the employment in that setting). If other credentials or experiences influenced the nature of employment secured, they should be factored out to accurately represent the roles and opinions of doctoral-educated ATCs.
Multiple univariate statistical analyses were used in this study, which presents a problem in interpreting the actual probability of a type I error (the probability that the effects are different due to chance). The overall probability of committing a type I error in an experiment using multiple univariate tests is 1 − (1 − α)c, where α is the accepted level of probability of a type I error (ie, P = .05) and c represents the number of univariate tests performed.1 It appears that 22 t tests and 66 analyses of variance were performed in this study. Therefore, just considering the number of analyses of variance performed, the actual overall probability level was 1 − (1 − α)c = 1 − (1 − .05)66 = 0.97, which means that there was a 97% probability of a type I error. Stated in another way, 97% of the time, any differences detected would be due to chance. Consequently, it is difficult to determine if the differences identified in this study are true differences. Using multiple univariate tests may be appropriate if the effects are independent of one another.2 In this case, the 22 competencies are related.
On the whole, Hertel et al present some very thought-provoking information, some positive and some disturbing. Although I am not confident in the inferences from the statistical analyses in this study, some general trends are apparent and should be pursued and discussed. This paper serves as an excellent springboard for discussion on the roles and attitudes of doctoral-educated ATCs. More information is needed to understand this new and growing population of athletic trainers.
We appreciate Dr Ingersoll's commentary on our manuscript and the opportunity to respond. Dr Ingersoll raises 3 central issues in his commentary: sampling, professional preparation in addition to doctoral study, and data analysis. We offer responses to these issues in the order in which they were raised.
Our identification of doctoral-educated ATCs was a labor-intensive effort and more involved than simply requesting a list of ATCs who met our inclusion criteria from the NATA. It would have been possible, as Dr Ingersoll suggested, to access other sources, such as the NATA membership database. This database, however, uses members' self-reported educational preparation. Thus, there is no assurance of the comprehensiveness of this approach. In preparing this response, we requested from the NATA a list of all ATCs with terminal academic degrees. The list returned to us by NATA in August 2000 contained 562 names, including individuals with self-reported credentials of both terminal academic degrees (PhD, EdD, HSD, DA) and terminal professional degrees (MD, DO, DPM, DC, DPT). Interestingly, the list we procured from the NATA also included some individuals who did not have their terminal degrees, as some ATCs with the credentials of MEd were also included. To make matters more confusing, the list did not specify each person's credentials, so we were unable to ascertain which of these individuals held terminal academic degrees, which had earned terminal professional degrees, and which did not, in fact, hold a terminal degree.
Of those appearing on the NATA-provided list, 110 of our sample of 130 who were mailed surveys were listed. This indicates that through our methods of sampling, we were able to identify 20 individuals who met our inclusion criteria but did not appear on the NATA-provided list. We were able to eliminate 126 individuals from the NATA list because of their listed credentials (nonacademic terminal degree) or because their mailing addresses indicated that they were not currently employed in an academic setting. This exclusion left 326 individuals on the NATA-provided list, who may or may not have met our inclusion criteria. Included in these remaining 326 were individuals who had received their terminal academic degrees since our initial mailing in March 1998, who may have had either a terminal academic or a terminal professional degree, or who may not have had a terminal degree at all. Additionally, because many ATCs list their home mailing address with the NATA, it was unclear which of these individuals were employed in an academic setting. We believe that our sample of convenience was as representative a sample of ATCs with doctoral degrees working in the academic setting as would have been available from any other source.
We agree with Dr Ingersoll that factors other than those addressed in this project, such as a subject's concurrent licensure as a physical therapist, may have influenced individual responses to the survey questions. Other factors, such as being a paramedic or having professional preparation before completing doctoral study, are also worthy of assessment. We hope this paper and exchange of commentary will facilitate the inquiry into such issues by future investigators.
The final issue addressed by Dr Ingersoll involves our data analysis and the risk of type 1 error. Although the calculations offered in his commentary are correct, the application of these statistical methods in survey research is not unprecedented.1 Although this statistical approach may be novel in sports medicine research, it is more commonplace in educational research. When faced with the analysis of responses to a large number of questions, investigators must narrow the focus to those responses that emerge as being of greatest interest.2 One could choose an arbitrary criterion, such as the 5 responses with greatest differences or all responses with differences of greater than 1 unit. Neither approach ensures any greater guarantee that items selected represent true population differences than the use of multiple t tests or analyses of variance does. We chose to use parametric statistical tests to identify areas of focus for analysis and discussion. This approach is commonly applied in disciplines such as educational research when the nature of the project is exploratory and the consequences of type I error are negligible, as was the case in our investigation.2 Although other investigators might have selected another approach to identify focus areas, we believe the approach we used was the most appropriate for this project and consistent with contemporary survey research methodology.