Agency, Socialization, and Support: A Critical Review of Doctoral Student Attrition Kenneth L Rigler Jr, Ph.D. Fort Hays State University Linda K. Bowlin, Ph.D. Southeastern University Karen Sweat, Ed.D. Northcentral University Stephen Watts, Ph.D. Oracle University Robin Throne, Ph.D. Northcentral University Rigler, K. L., Bowlin, L. K., Sweat, K., Watts, S., & Throne, R. (2017). Agency, socialization, and support: A critical review of doctoral student attrition. Paper presented at the 3rd International Conference on Doctoral Education, University of Central Florida. Agency, Socialization, and Support: A Critical Review of Doctoral Student Attrition Abstract Almost universally, residential doctoral programs have reported attrition rates of up to 50% for face-to-face programs and 50-70% for online doctoral programs. The purpose of this critical review was to explore current literature for doctoral attrition and persistence to explore reasons and attributes for improved persistence to completion. We analyzed a final sample of 79 studies for context with doctoral attrition and persistence and, follow- ing coding, pattern matching, and synthesis, four final constructs emerged: (a) chair agency and chair-candidate relationship; (b) candidate socialization and support systems; (c) candidate preparedness; and (d) financial considerations. These constructs indicated that internal factors from within the doctoral program have affected attrition and we en- courage doctoral educational leadership to implement strategies to improve dissertation chair practices through professional development, increased candidate support and social- ization opportunities, creation of clearer pathways from academic to dissertation research coursework, and improved financial opportunities and support for candidates. We rec- ommend quantitative explanatory studies to further examine the four factors within the doctoral program to examine the effects on doctoral candidate completion. Keywords: doctoral student attrition, doctoral student persistence, all but dissertation Introduction The prominence of a doctorate degree to both an individual candidate and Ameri- can society as a whole has never been greater. The individual candidate’s return on in- vestment of a doctoral education can be significant for career advancement, career 2 change, compensation, leadership development, and life quality intimations resulting from the attainment of the doctoral degree (Brill, Balcanoff, Land, Gogarty, & Turner, 2014). Individuals with doctoral degrees hold the lowest unemployment rate (1.9%) of all degree levels and the highest median weekly income (Bureau of Labor Statistics, 2010). Gains for society have been seen for the 1.7 million research (practitioner and tra- ditional) doctorates awarded and measured since 1957 across all sectors of higher educa- tion and have enhanced the personal and professional lives of executive leadership across academic, governmental, institutional, employer, and employee perspectives (National Science Foundation, 2009). The return to a knowledge-driven society from a prepared doctoral-level scholar-practitioner has value in the use of formal inquiry to solve the real problems of practice and discover innovations that benefit the field (Offerman, 2011; Throne, 2012). Even though benefits to both the individual and larger society have been seen, doctoral program attrition rates have continued to be a problem across doctoral programs worldwide (Council of Graduate Schools [CGS] 2010, 2014, 2015a, 2015b). Almost uni- versally, residential doctoral programs have reported attrition rates of up to 50% and nearly 50-70% for online doctoral programs (Terrell, Snyder, Dringus, & Maddrey, 2012), and this loss of potential doctorates has limited applicants for positions requiring a doctorate and depleted resources from universities and students (Willis & Carmichael, 2011). As attrition rates maintain unacceptable levels, doctoral-granting institutions are faced with diagnosing the contributing factors for attrition and developing corrective measures in order to demonstrate continuous improvement for both internal and external 3 accountability entities (Bogle, 2010; Levitch & Shaw, 2014). The purpose of this study was to critically review the current doctoral attrition literature in order to identity the rele- vant themes relating to the worldwide doctoral attrition phenomenon, most specifically within the U.S., as a study of doctoral attrition with the intent to improve doctoral educa- tion completion may benefit the various sectors of society and provide doctoral-granting programs evidence-based themes for continuous improvement. Source, Survey, and Analysis The total initial articles sourced included 87 that were identified using key words doctoral attrition and doctoral persistence from pertinent general and education aca- demic databases that included ProQuest, EbscoHost, and Google Scholar. Four were ex- cluded in the initial evaluation as studies did not meet the recency requirement of 2010 to present, and the remaining 83 sources were surveyed and coded for pattern matching from which four final constructs were determined for analysis and synthesis: (a) chair agency and chair-candidate relationship; (b) candidate socialization and support systems; (c) candidate preparedness; and (d) financial considerations. Chair, advisor, and mentor were often used synonymously, and these synonyms were retained in the analysis. An additional five sources were excluded following coding as no patterns were identified to align with the four constructs. The final 79 studies were analyzed for context with doc- toral attrition and persistence and key findings, and were collectively synthesized follow- ing analysis in context with the construct patterns that emerged in preliminary and final analysis. The interrelationships among findings that bridged the constructs are presented in the discussion and conclusion. Chair agency and the chair-candidate relationship 4 The high attrition rate in doctoral programs has driven university officials to in- tensify efforts to identify why students are exiting the program prior to completion and closely examine the relationship between the advisor and student (Cockrell & Shelley, 2011). A positive and non-heirarchical relationship between the doctoral committee chair and the doctoral candidate has been identified as one of the most critical factors in the successful completion of a doctorate degree (Bitzer, 2011; Stallone, 2011; Storms et al., 2011); however, only 26 of 457 (6%) candidate respondents identified their supervisor in a positive light (Bégin & Gérard; 2013) and when surveyed, doctoral students repeatedly identified the most consistent and significant barrier to their success in the degree pro- gram was a problematic relationship with the dissertation chair (Willis & Carmichael, 2011). Doctoral program leaders also confirmed the importance of a positive relationship between dissertation advisor and doctoral candidate as they identified insufficient super- visor support as a significant factor in doctoral attrition (Herman, 2011; Kyvik & Olsen, 2014). Doctoral candidates need access to regular communication with the dissertation chairs (Holmes, Trimble, & Morrison-Danner, 2014), and Holley and Caldwell (2011) found that students in doctoral programs were more likely to complete their program and experience greater satisfaction when they were engaged in meaningful interaction with the advisor. Chairs who were overly involved in their own research agenda and thus not available for regular candidate supervision and feedback were a detriment to candidate success (Herman, 2011; Holmes et al., 2014; Van de Schoot, Yerkes, Mouw, & Sonne- veld, 2013); whereas, advisors who were not only available but also initiated regular stu- dent meetings had advisees with higher completion rates (Stock & Siegfried, 2014). 5 As doctoral candidates had the opportunity to interact with the dissertation chair, it was important for a cooperative relationship to be established based on clear expecta- tions between the candidate and chair (Hardre & Hackett, 2015). Moxham, Dwyer, and Reid-Searl (2013) found supervisors and students had similar viewpoints for topic knowledge and work ethic, but found major differences in expectations for research methodology support and the initiation of communication resulting in a break-down in the successful partnership between the candidate and chair. Within this cooperative rela- tionship, Brill et al. (2014) noted a candidate needed a dissertation chair who was willing to go beyond the supervisory role to focus solely on task completion, and a chair willing to provide individualized learning through effective faculty mentoring. The dissertation chair was found to be a pivotal determinant of student success in a program (Hyatt & Williams, 2011); yet, few doctoral programs required, or even pro- vided, professional development opportunities for dissertation chairs to improve their mentoring skills. Dissertation chairs need not be experts in the specific topic of study, but should also be experts in the areas of the research process and effective mentoring (Storms et al., 2011). The establishment of faculty mentoring improvement was neces- sary to foster deeper mentor-mentee relationships, which may enhance the doctoral expe- rience for candidates and improve overall persistence (Fountaine, 2012). In Norway, a country where doctoral programs are government subsidized, federal regulations were implemented involving the inclusion of doctoral faculty mentoring training and as a result there was an increase in doctoral completion each year (Kyvik & Olsen, 2014). Not all faculty make good dissertation chairs but they are one of the greatest as- sets to the doctoral program (Hyatt & Williams, 2011); therefore, continuous professional 6 development expectations are needed for all dissertation supervisors and should include elements of improving interactive communication (Moak & Walker, 2014), rapport build- ing skills (Holley & Caldwell, 2011), service and collegueship (Hyatt & Williams, 2011), and relational collaboration (O’Meara, Knudsen, & Jones, 2013). Additionally, Hyatt and Williams (2011) emphasized the importance of the supervisors’ pedagogical and re- search skills to provide accurate and effective guidance through the dissertation journey. Candidate socialization and support systems Previous research has identified social isolation as one of the main factors associated with doctoral attrition indicating effective doctoral programs needed to incorporate positive social experiences and structures for candidates (Jairam & Kahl, 2012). West, Gokalp, Peña, Fisher, and Gupton (2011) commented on the struggle doctoral students experienced when they were socially and intellectually isolated and the effect of isolation on the time to degree and program completion. Faculty, doctoral students, and universities remain concerned about factors that cause delays and/or attrition and how social integration factors, such as the nature of advising and dissertation topic, are viewed in importance (Wao & Onwuegbuzie, 2011). Espino (2014) examined the problems of minority graduate students, and observed that to be successful in graduate school, students benefitted from socialization into the academic environment as well as resources available from their community, family, or the graduate environment. Stubb, Pyhalto, and Lonka (2011) noted it was not unusual for a Ph.D. program to take several years to complete and involve an experience with stress, doubt, and exhaustion; however, a mismatch between the individual and the environment can lead to negative experiences such as stress and burn-out, and the scholarly community played an 7 important role in the socio-psychological well-being of the doctoral student. A positive relationship in the scholarly community can also be a beneficial buffer in times of stress, and the maintenance of socio-psychological well-being during the doctoral program may be carried forward long after the doctoral degree is finished (Stubb et al., 2011). Doctoral online programs that allow working adults to continue professional ca- reers while completing a doctorate have increased in popularity resulting in the need for universities to take collective responsibility for the success or failure of these candidates by structuring student-to-student support networks for distance learners (Terrell et al., 2012). Cockrell and Shelley (2011) found significant relationships among indicators for the support systems and student satisfaction, and recommended that doctoral programs increase support systems to improve student attrition. Stallone (2011) also emphasized the importance of student–cohort relationships and human factors that remain under the control of the university, which has affected attrition levels, and Holmes et al. (2014) suggested students needed opportunities to learn from each other, such as teambuilding, to reduce social isolation. Gardner (2010) agreed that doctoral student support needed improvement and an increased understanding of doctoral student socialization may assist faculty and administrators to better support doctoral students to higher rates of comple- tion. Levitch and Shaw (2014) reported elements that impacted retention in an online learning environment that included (a) program design, (b) orientation and use of the online environment, (c) faculty and leadership system support, (d) learning style, and (e) recognition of student needs. Doctoral student retention rates may be impacted that could benefit the institution and assist the faculty by providing a satisfactory experience for the 8 student, and the authors noted that when doctoral students were satisfied with the learning environment, they were more likely to complete the educational program pursued, and students not satisfied with the educational experiences were more likely to drop (Levitch & Shaw, 2014). Rockinson-Szapkiw, Heuvelman-Hutchinson, and Spaulding (2014) noted online doctoral students who used web-based technologies for communication had a higher sense of connectedness than doctoral students who used the phone or email, and results highlighted how web-based communication tools can increase connectedness and decrease doctoral attrition. Likewise, Kiley (2011) found student support systems to be an area in need of improvement as student satisfaction impacted attrition, which paralleled findings by Lott, Gardner, and Powers (2010), who posited doctoral students may leave the doctoral program due to a lack of socialization, especially those who are relatively isolated demographically. Luna (2010) focused on cohort programs and the benefits of cohort support at any level of education, not just doctoral, and found that candidates with realistic expectations of the doctoral program tend to persist and complete the degree. Martinez, Ordu, Della Sala, and McFarlane (2013) noted doctoral students typically did not leave the program for one reason but chose to leave due to multiple factors that included support and isolation. Byers et al. (2014) also discovered that students felt that outside support systems were vital in meeting the challenges of doctoral programs, and benefited from coping strategies that included the cohort and support by family, friends, and other doctoral students. Finally, Carter, Blumenstein, and Cook (2013) focused on gender differences as women were significantly overrepresented compared to men in seeking counseling sessions for relationships and family issues, possibly indicating that home and 9 family issues impact women more than school issues. Candidate preparedness Often doctoral candidates did not understand that success in doctoral studies re- quired a different skill set than what they previously possessed (Brill et al., 2014). Candi- dates with certain skills and preparation were found to be more likely to persist than those without these skills and preparation factors that influenced degree completion may be cat- egorized as motivation, psychological, prior academic preparation, program preparation, and demographics (Bitzer, 2011; Brill et al., 2014; Mason, 2012; Stock & Siegfried, 2014). Motivation (and corresponding goal setting) was identified in the past research as a key attribute to persistence (Bain, Fedynich, & Knight, 2007; Bitzer, 2011; Herman, 2011; Spaulding & Rockinson-Szapkiw, 2012) and to the amount of time needed to com- plete a doctoral degree (Wao & Onwuegbuzie, 2011). Offerman (2011) found the rea- sons a candidate had for pursuing the degree was a significant factor in persistence, and other researchers determined that the most effective reasons were desire for development (Bégin & Gérard, 2013), self-improvement (Bain et al., 2007; Bégin & Gérard, 2013), learning (Bégin & Gérard, 2013), better employment opportunities (Holley & Gardner, 2012), improved quality of life (Holly & Gardner, 2012), and achievement (Bégin & Gér- ard, 2013; Holly & Gardner, 2012). Doctoral candidate satisfaction was strongly correlated with motivation (Mason, 2012), as was the amount of time a student spent on task (Gardner, 2010; Moxham et al., 2013), the ability to work independently (Bitzer, 2011; Mason, 2012; Litalien & Guay, 2015; O’Meara et al., 2013; Stock & Siegfried, 2014), and the priority for the doctorate (Santicola, 2013). Hardre and Hackett (2015) identified graduate student desire for 10