Abstract
This article continues a series of reviews of rehabilitation doctoral dissertation research. In 2012, there were 30 doctoral dissertations completed from identified doctoral programs in rehabilitation counseling. Dissertations were indexed by research topic, methodology used, model, and statistical analysis. An annotated bibliography is provided. A consideration of institutional productivity and doctoral graduate employment outcomes were also included in the analysis. Results of the current analysis show that a majority of studies comprised quantitative descriptive research designs. Unlike analyses from previous years, advanced and basic statistics were utilized equitably. The proportion of dissertations focusing on attitudes toward persons with a disability was greater than any of the prior reporting periods. Predictive outcome studies and clinical population research also remain prevalent. Implications for the practice of rehabilitation counseling, education, and future research are discussed.
Dissertations are the principal product culminating from doctoral study. It represents years of training and development while also conveying something of the priorities and competencies of the institutions and programs that serve as gatekeepers. For many doctoral students, the dissertation is first viewed as an intimidating requirement for graduation. However, with training and effort, the dissertation becomes a critical developmental milestone and a path toward content expertise and professional stewardship. Dissertations have long been recognized in rehabilitation counseling as an important source of new information and a compass to the future (Leahy et al., 1989). All claims for professionalization are built, in part, on a solid foundation of esoteric knowledge to guide practice (Cruess et al., 2004; Cullen, 1978; MacDonald, 1995). Each dissertation represents a potential building block in that foundation while also conveying something of the capacity for the academic institutions and programs to provide the education and development necessary for protecting, promulgating, and advancing the discipline of rehabilitation counseling (Golde, 2006).
The significance of the dissertation combined with its relative inaccessibility and limited visibility have led rehabilitation counseling researchers to collect, analyze, and summarize rehabilitation counseling dissertations for nearly four decades (Leahy et al., 1989). Initial analyses were conducted by Lofaro (1981a, 1981b, 1982, 1983a, 1983b) and traced dissertations back to the beginnings of the profession in 1954. Many analyses have been conducted since that time, providing information about trending areas of research, methodological training and sophistication, and future directions for study within a discipline (e.g., Beck et al., 1994, 1996; Leahy et al., 1988, 1989, 1990, 1992). Tansey, Zanskas, Phillips, and colleagues again began publishing these summaries of the rehabilitation counseling research after a 16-year gap from 1996 to 2012 (Tansey, Phillips, & Zanskas, 2012; Tansey, Zanskas, & Phillips, 2012; Zanskas et al., 2014). These most recent reviews are a partial replication of the approach used by Leahy et al. (1988, 1989) and uses a consistent framework for categorization of dissertation characteristics, allowing for a consideration of trends from 2005 until 2011. These trends suggested an increasing use of predictive outcome studies over this period while also conveying a very limited use of experimental and quasi-experimental designs. Tansey, Phillips, and Zanskas (2012) also noted an increasing use of advanced statistics in dissertation research during the years of 2008–2010. That trend, although not linear, seems to have continued to the latest review of dissertations from 2011. The year 2011 also brought an uptick in attitudinal dissertation research even as predictive outcome studies remained prevalent. The use of archival databases for dissertations dipped in 2011 from the previous years (2005–2010) for reasons that were not clear.
The purpose of this current review of doctoral dissertation research was to summarize dissertations from the year 2012 and to consider results in relation to previous summaries as part of a series of annual reviews of doctoral dissertation research in rehabilitation counseling. The present analysis employs the method used in Tansey, Zanskas, & Phillips (2012) and previous dissertation analysis studies. As with previous reviews dating back to Lofaro (1981a), this article provides an annotated bibliography summarizing the content, reference citations, observations, and research trends, while also including institutional productivity and employment trends of doctoral graduates. McCarthy (2014) described the short collective memory of research activity in rehabilitation counseling and the issues it creates for building a strong research foundation and for contributing to the needs of people with disabilities. We believe it critical as rehabilitation counseling researchers to make time for summarizing, synthesizing, building on, and applying the work being done within the discipline.
Method
We based the content review for this study on rehabilitation programs listed in the 2011–2012 National Council on Rehabilitation Education (NCRE) Member Directory. Dissertations were collected using a sequential method paralleling the approach used in previous reviews (Tansey, Phillips, & Zanskas, 2012; Tansey, Zanskas, & Phillips, 2012; Zanskas et al., 2014). The ProQuest Dissertation and Theses database was used to search for dissertators by program faculty in each of the identified parent universities. This process resulted in the collection of 41 dissertations from 15 universities. The researchers then contacted representatives of the doctoral programs to verify the accuracy of any listed dissertations that were in question. We made additions and deletions based on this feedback. The final list for analysis and summary included 30 dissertations.
Once compiled, dissertations were distributed among the authors for initial coding. The authors served as pairs of independent raters for each dissertation. Where discrepancies existed between independent raters, a third researcher, acting as an auditor, made final coding determinations. The content was categorized by methodology, type of data, type of analysis, and research design in addition to a categorization of the general content. We provide a brief description of these categories used for the coding process and refer the reader to Tansey, Zanskas, and Phillips (2012) for the full description. Dissertation methodology was categorized as quantitative, qualitative, or mixed. Quantitative studies were primarily deductive in nature and relied on numerical data to address research questions or hypotheses; dissertations coded as qualitative studies were inductive and used case studies focused on individual meaning of research topics. Dissertations coded as using a mixed method combined quantitative and qualitative research designs (Creswell, 2009). We categorized the research data either as original or archival; dissertations were categorized as using archival data in all cases where the data used for the analysis existed prior to the beginning of the study. The analysis used by dissertators was categorized as either basic or advanced. As with previous reviews, basic statistical analysis included correlation, chi-square, t-test, analysis of variance (ANOVA), and analysis of covariance (ANCOVA), while advanced statistics included factor analysis, structural equation modeling, multivariate analysis of variance (MANOVA), multivariate analysis of covariance (MANCOVA), and path analysis. Type of research design could be categorized as descriptive, quasi-experimental, or true experiment. Dissertations were coded as using a descriptive research design when no intervention was included in the study; descriptive-designed dissertations included quantitative and qualitative methodologies. Dissertations were categorized as quasi-experimental if an intervention was applied but random selection of participants or random grouping of participants into treatment and control groups was not part of the study procedures. Finally, dissertations were categorized as true experimental research designs if an intervention was applied with randomly selected participants who were randomly assigned to treatment and control groups (Creswell, 2009).
Results
Figure 1 shows the number of dissertations published by students graduating from doctoral programs in rehabilitation counseling for each year since data were collected since 2005. The total of 30 dissertations for 2012 is nearly identical to the average (

Number of dissertations per year from 2005 to 2012.
Regarding the methods used, most dissertations selected a quantitative research design (n = 25, 83.3%). Of the remaining five dissertations, two (6.7%) used qualitative and three (10%) used mixed method designs. Taken as a percentage of the total, Figure 2 illustrates a fairly consistent portion of quantitative, qualitative, and mixed method studies over the years. All dissertations from 2011 used a quantitative design, but the overall total number of dissertations was nearly 11 below the average that year. The 2012 numbers suggest this may have been a random anomaly rather than the beginnings of a trend.

Percentage of dissertations implementing quantitative design from 2005 to 2012.
The analysis of research design involved categorizing quantitative dissertations as descriptive, quasi-experimental, or experimental studies. Of the 28 quantitative dissertations (including both quantitative and mixed method designs), 25 (89.3%) used a descriptive, two (7.1%) quasi-experimental, and one (3.6%) an experimental design. This again represents a fairly consistent pattern since 2005, as shown in Figure 3. Descriptive studies have made up 80% or more of research designs in each year, with an average of 91.2% being descriptive studies.

Percentage of dissertations using experimental, quasi-experimental, or descriptive designs from 2005 to 2012.
Among quantitative and mixed method dissertations, the analysis showed that seven (25%) used secondary data sources for addressing research questions and hypotheses, while the remaining 21 dissertations reported using data original to their study. No two dissertators used the same secondary data set in 2012. Figure 4 shows the use of secondary data sources from 2005 to 2012. There has been a fairly constant decline in use of secondary databases since 2005 when 40% of all quantitative dissertations were based on secondary data sources.

Percentage of dissertations using archival data from 2005 to 2012.
Continued evaluation of quantitative and mixed methods dissertations showed 20 (71.4%) using advanced statistical methods. As seen in Figure 5, use of advanced statistics has fluctuated over the 7 years between 2005 and 2012. The proportion of dissertators using advanced statistics in 2005 (70.0%) was nearly identical to that used in 2012.

Percentage of dissertations using advanced statistical analysis from 2005 to 2012.
Content Analysis
A content analysis was conducted to better understand research topics covered using the categories established in previous reviews. Inter-rater reliability for the coding of dissertations between raters was found to be Kappa = .867 (p < .001). As shown in Table 1, the distribution of dissertations among the major categories from 2012 was as follows: (a) attitudes and attitudinal change, two (6.7%); (b) predictive outcome studies, 13 (43.3%); (c) clinical population studies, two (6.7%); (d) assessment, four (13.3%); (e) program and policy studies, eight (26.7%); and (f) family studies, one (3.3%). There were no dissertations categorized as professional issues, social and cultural aspects of disability, or technology in the 2012 review. Predictive outcome studies (n = 13) and studies focused on programmatic or policy change (n = 8) were most common. The consistent use of research topics across dissertation analyses allowed for a comparison of dissertation studies beginning in 1990. This comparison across reviews allows for a consideration of potential trends or changes over 20 years. Table 1 provides a comparison between the content analysis provided in this review and reviews conducted on dissertations from 1990–1991, 1992–1993, 2005–2007, 2008–2010, 2011, and 2012.
Historical Comparison of Dissertation Categories 1990–1993 and 2005–2011.
Program of Dissertation Completion
Finally, an analysis of where dissertations were completed show that, for 2012, Southern Illinois University at Carbondale produced the most dissertations (n = 6), followed by Michigan State University and the University of Wisconsin–Madison (n = 5). The University of Northern Colorado produced three. The University of Arkansas–Fayetteville, Virginia Commonwealth University, and Wayne State University, each produced two; and George Washington University, Pennsylvania State University, University of Arizona, University of Iowa, and Utah State, each produced one.
Employment Settings of Doctoral Graduates in 2012
A review of doctoral program websites, an internet search, and email contacts resulted in the researchers being able to identify 29 of the 30 graduates from 2012. When there were questions of whether online information was accurate and up to date, we sought to contact the doctoral student’s advisor for confirmation of their employment setting (i.e., academic, clinical, administration). Of these 29 graduates, 22 obtained employment in an academic setting either in the United States or internationally, three were employed in primarily clinical settings, three in administration or leadership for independent living centers, and one in a private research center.
Retrieval of Dissertations
The retrieval method implemented in this review resulted in 12 dissertations being identified as rehabilitation counseling dissertations that were later removed for a 40.0% initial false-positive rate. This rate is similar to that produced in Zanskas et al. (2014) but much higher than Tansey, Phillips, and Zanskas (2012), which reported a 15.3% false-positive rate. The false-positives were due primarily to master’s thesis incorrectly being included in the initial sample as well as rehabilitation counseling faculty serving as dissertation chairs for doctoral candidates in related programs (i.e., special education, counselor education, counseling psychology).
Conclusion
The analysis of dissertation research provides an increased exposure to the research being completed at the apex of doctoral training while also providing insights into the field of rehabilitation counseling, both its research and preparation of new researchers. On the whole, rehabilitation counseling dissertation research appears to continue on a strong pace, with the number of dissertations and the institutions supporting graduates remaining relatively stable. That said, this analysis also offers some insights for future directions and unmet needs in the field. It is worth noting and monitoring the fact that there were no dissertations covering professional issues as the major content area. This is the first time since these categories of content analysis have been used that professional issues have made up less than 10% of the dissertations. The proportion of quantitative, qualitative, and mixed analyses remained fairly steady, with quantitative dissertations far outpacing use of the other two. There was also a slight uptick in the use of archival data for dissertation analysis after a steady drop over the previous 2 years.
Among the many indicators shared in this study, the trend in research design is among the most telling. Experimental and quasi-experimental research designs have not made up over 10% of dissertations in any year between 2005 and 2012. The financial costs and extended time required to produce quality intervention research for graduate students who typically experience a scarcity of both likely factors into the limited amount of intervention research being conducted. However, in this era of evidence-based practice, it is unclear whether new researchers will develop the skills needed to build on existing evidence with intervention research if mentored research focused primarily on descriptive designs. Perhaps the most likely path to seeing an increase in the proportion of dissertations using experimental and quasi-experimental research designs will be from increasing the amount of experimental research conducted by rehabilitation counseling faculty. A recent systematic review of intervention research in rehabilitation counseling suggested that much of the intervention research related to rehabilitation counseling is being conducted outside of traditional rehabilitation counseling programs (Phillips et al., in press). Faculty who are engaged in intervention research are more likely to feel confident steering doctoral students toward and supervising their intervention research. Also, intervention research is relatively expensive and time-consuming. Faculty engaged in intervention research are more likely to have projects and resources that could be shared with doctoral students in their own dissertation research, thus reducing financial barriers and possibly the time to conduct a quality intervention study. In programs where intervention research is not being conducted, dissertators might be encouraged to seek mentors or even to form a dissertation committee that is co-chaired or shared by intervention specialists from related fields. Given the ongoing push for more intervention research in the field and its potential for informing the applied practice of rehabilitation counseling, this trend should continue to be monitored.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
