First-Generation College Student Dissertation Abstracts: Research Strategies, Topical Analysis, and Lessons Learned

First-generation college students are students whose parents or guardians did not obtain a four year college degree (Davis, 2012). As a group these students make up a large part of the college student population and are often reported to encounter difficulties in their campus experience. While the topic of first-generation student has received much attention over the past years, no research effort has been reported that examines dissertations on the topic. This article utilizes a bounded qualitative synthesis study framework to examine the 133 dissertation abstracts found by searching the ProQuest Dissertation and Theses TM digital database for dissertations abstracts from 2009 through 2013 using the search terms “first-generation college students” and “higher education.” The research question for this study was: What can we learn from the examination of doctoral dissertation abstracts that focus on the experience of first-generation college students regarding research strategies, topics addressed, and lessons learned? The study’s findings provide an overview of researcher attributes and the characteristics of the research in terms of methodology and topical focus. “Lessons learned” from the abstracts as well as the omissions in the research are presented. A major finding of the investigation was that very few of the dissertations have entered the academic conversation regarding first generation students – major books on the topic do not reference the dissertations and in a search of academic journals only three of the 133 dissertations were found to have been published.


Introduction
The category of "first-generation college student" was introduced over 25 years ago (Davis, 2012).Students fit within this category by the following definition: "Students can claim first-generation student status if neither one of their parents or guardians possesses a four year degree."(Davis, 2012, p.xiv).Examination of this group of students over the past years has established: (a) there are a large number of these students in higher education and more will be coming to campus; and (b) they often encounter difficulties in the campus experience (Davis, 2012;Housel & Harvey, 2009;Ward, Siegel, & Davenport, 2010).Not only do they often struggle with the academic classroom experiences leading to graduation, they also struggle with the social cultural challenges of campus life.(Davis, 2012;Housel & Harvey, 2009;Ward, Siegel, & Davenport, 2010).Despite the research activity associated with the first-generation college student experience, no studies have reported and synthesized the doctoral research that has occurred on this topic.
The purpose of this study is to fill this gap by describing, analyzing, and interpreting the doctoral dissertation abstracts focusing on the first-generation college students in higher education published in the past five years (2009 thru 2013).The overarching research question for this study is "What can we learn from the examination of doctoral dissertation abstracts that focus on the first-generation college experience regarding research strategies, topics addressed and lessons learned.This central question was addressed by exploring the following sub-questions: What are the attributes of the researcher and the research?What are the guiding frameworks for the research?What are the research strategies and topics associated with the dissertations?What dissertations have published in academic journals?What are the lessons to be learned?A bounded qualitative meta-synthesis using document and template analysis guided the efforts to respond to these key questions.Following a discussion of the methodology, the results are presented using the Wolcott (1994) format for transforming qualitative data: description, analysis, and interpretation.

A Bounded Qualitative Synthesis Framework
New strategies are emerging (Lipsey & Wilson, 2001) to explore and synthesize research questions of interest using available published research.An often used strategy has been to pose a research question and then conduct a comprehensive search of all available relevant research that use a variety of quantitative methods.This approach is referred to as a quantitative meta-analysis (Glass, 1976) and carries out procedures using "effect size data [to] permit meaningful numerical comparisons and analysis across the studies" (Lipsey & Wilson, 2001, p. 8).Studies that seek to provide a synthesis of published research on a topic, however, need not be only quantitative.Qualitative analysis is also available as a strategy to explore and synthesize published research.Noblit and Hare (1988) first introduced procedures on how to proceed with a qualitative meta-analysis and more recently a variety of procedures have been well documented (Hannes & Lockwood, 2012;Major & Savin-Baden, 2010;Paterson, Thorne, Canam, & Jillings, 2001;Saini & Shlonsky, 2012;).This study uses a bounded qualitative synthesis framework (Banning & Folkestad, 2012;Banning & Kuk, 2009) and includes the following characteristics: (a) the study is bounded by a search of a specific research genre.In this study, the genre was the doctoral dissertation abstract; (b) the study is also bounded by a specific time period ; (c) the study is qualitative in its approach and uses qualitative document analysis (Altheide, 1987;Altheide, Coyle, DeVriese, & Schneider, 2008) and template analysis (King, 1998) as the analytic strategy for the study, and (d) the meta-synthesis approach presented by Paterson, Thorne, Canam and Jillins (2001) serves as the structuring device to focus not only on findings, but the methods and theories.The framework for presenting the results follows Wolcott (1994) for transforming qualitative data: description, analysis, and interpretation.The characteristics of the researcher and the research will comprise the description section; a presentation of research strategy by topic will be presented in the analysis section; followed by an interpretation focusing on the implications of the findings and a section regarding lessons learned.

Establishing the Data Set
Dissertation citations and abstracts were found by searching the ProQuest Dissertation and Theses TM digital database from 2009 through 2013 using the search terms "first-generation college students" and "higher education."Foreign language dissertations and master theses were excluded from the study.These procedures established a data set of 133 dissertation citations and abstracts.The abstract was used for the unit of analysis.The abstracts were downloaded from ProQuest TM to Endnotes TM and then using the "Show All Fields" filter the data was imported to a qualitative research software to assist in the analysis (NVivo 10.0 TM ).

Method of Analysis
Qualitative document analysis (QDA) (Altheide et. al., 2008) provided the methodological structure for the study.The purpose of document analysis is to analyze and interpret data from the examination of documents (Schwandt, 2007).In this study, the documents were the doctoral dissertation abstracts related to first-generation college students in higher education.QDA is a qualitative content analysis which examines documents for "patterns" (Altheide, 1987, p. 68) rather than the strict statistical approach of classical content analysis of documents (Krippendorff, 2004).Attention was given to the topic, method, and theory noted in the abstract (Paterson, Thorne, Canam & Jillins, 2001).Within the QDA framework, the specific approach to coding utilized the template analysis method (King, 1998).In this method both a priori codes are utilized along with new codes produced by an inductive approach to the data.The a priori codes were utilized for researcher and research attributes categories and the classification system for the guiding frameworks associated with the research abstracts.The research strategy and topical analyses utilized the inductive coding strategy of the constant comparative method (Corbin & Strauss, 2008).Each dissertation was examined and then coded for the research strategy used and each dissertation abstract was given a topical name (open code) and then the topical names were grouped into larger categories (axial coding).Finally, the coding for the research strategy and topical categories were brought together to form the thematic structure for the data.

Trustworthiness
Two strategies (rich, thick description of methods and triangulation) were utilized to increase the trustworthiness of the study (Creswell, 2007).The thick description of the details of the methods and procedures used in the study helps to establish transparency of methods.When the abstract did not present sufficient information for assigning a code then a second source of information (the full dissertation) was examined.For the question of whether the dissertation has been published a triangulation by source procedure was used.Each article was searched for possible publication in the Academic Search Premier TM database as well as a search of recently published books (Davis, 2012;Housel & Harvey, 2009;Ward, Siegel, & Davenport, 2010) reference sections.

Researcher and Research Attribute Results
Table 1 presents the findings in response to the question of what are the attributes of the researcher and the doctoral research related to first-generation students in higher education.Of the dissertations completed, 75% were authored by females and 25% were authored by males; 51% were completed by students receiving Ph.D. and 47% were completed by students receiving the Ed.D.In addition, there was one each representing the D. B. A. and Psy.D. of the dissertation abstracts analyzed, 35% were quantitative and 60% were qualitative.Five percent were mixed methods dissertations.Table 2 shows the distribution of dissertations by year from 2009 through 2013.The year 2011 was the most productive year with 37 dissertations.
In regard to the degree granting institution, 94 different institutions awarded degrees.Table 3 lists the eight institutions that awarded three or more doctoral degrees during the study's time period.Cappella University awarded the highest number of degrees on this topic with eight.

Guiding Frameworks Results
Nearly all research is built upon or guided by previous work.In this study, the theory component of the meta-synthesis model (Paterson, Thorne, Canam, & Jillings, 2001) led to the following question of the data: what were the guiding frameworks important enough that the researcher included their mention in the abstract of the study?To answer this question each abstract was examined for the presence of a guiding framework and then subsequently was coded by a previously established guiding framework classification system (Banning & Folkestad, 2012).The guiding framework classification system used in this study is comprised of three components: theoretical guiding framework; instrument based guiding framework, and a literature based guiding framework.The theoretical guiding framework is defined as any use of a theory or named model as guiding the research.The instrument guiding framework is defined as a framework built on an existing research instrument, for example, a specific published survey or inventory (Banning & Kuk, 2009).Finally, the literature guiding framework is defined as research that invokes mention of key background literature in the abstract.Table 4 presents the results of applying this classification system to the 133 abstracts of the study.Seventy-three or 55% of the abstracts did not mention a guiding framework for the study.Thirty percent of the abstracts mentioned a theory or model for the guiding framework.Seventeen percent of the abstracts mentioned important literature and five percent use an instrument as the guiding framework.Within the theory/model category, 25 different frameworks were mentioned.The most often mentioned theoretical work was Tinto's (1975Tinto's ( , 1982Tinto's ( , 1987Tinto's ( , & 1993)).It was noted by 13 dissertation abstracts and Astin's work (1984, 1993, & 1999) was named five times and six models relating to social/cultural capital theory were also used as the guiding framework.

Publishing of the Dissertations
To answer the question of how many of the study's dissertations were eventually published in academic journals a search was made using the Academic Search TM Premier database.The Academic Search TM Premier database includes over 8,000 journals.Of the 133 dissertations in the study, only three were found to be published in academic journals (Garriott, Flores, & Martens, 2013;Olive, 2010;& Stevens, Townsend, Markus, & Phillips, 2012).To be counted as a published article the topic of the article needed to be similar to that of the dissertation and the dissertation author needed to be first author of the article as noted in the American Psychological Associations (APA)'s Ethics Code Standard 8.126, Publication Credit, Dissertations (APA Publications Manual, 2010).In the data set including the years 2012 and The dissertations completed in 2012 and 2013 would not be expected to be published at the time of the present study (January 2014).In order to offset this issue of delay, a second search of dissertation abstracts using the same search terms "first generation college students" and "higher education" but including the years of 2003 through 2008 was conducted.Seventy additional abstracts were located and of the second data set only one dissertation had been found to be published in an academic journal.To further triangulate the question of publication of the dissertations in both data sets (203 dissertations), the references sections of the three most recent books with comprehensive coverage of the first-generation student in higher education topic were examined (Davis, 2010;Housel & Harvey, 2009;Ward et.al., 2012 ).One of the 201 abstracts was referenced and it was also published (Moschetti and Hudley, 2008).After triangulation of data sources, it is concluded that dissertations on first-generation college students in higher education are only very minimally being published (four) and referenced (one).

Analysis of Results
Often qualitative data analysis follows the pattern of looking at case attributes in relation to induced themes in the data set (Edhlund & McDougall, 2012).In this study the case attributes (deductive codes) have been outlined in the description section of the results highlighting various descriptive categories of the first generation data set.
All the dissertations sought to understand some aspect of the first generation experience, but in reading the abstracts several different strategies were used to pursue this understanding.Helpful to this process of inducing codes for understanding research strategies is the foundational work of Willems (1969) regarding characteristics of research activities.He suggests that how a researcher goes about their work can fit into a "research space" (p.46).The concept of research space focuses on how the researcher goes about their methodology rather than on the topic or phenomenon of the study.The space is two dimensional and is based on the following two dimensions: "the degree of the investigator's influence upon, or manipulation of, the antecedent conditions of the behavior studied" (p.46) and "the degree to which units are imposed by the investigator upon the behavior studied" (p.46).These two dimensions outline the space for research activities.Some research methods place the research low on both of dimensions and this research would be typical of basic exploratory qualitative methods.
A research method focusing on the high side of both dimensions would more likely be found in structured quantitative research focusing on a set of operationally defined variables.The two dimensions form a space; therefore, some research approaches can have varying degrees of both dimensions, for example, low-high, med-med, & high-low.Using Willems (1969) as a starting point, four research descriptive categories of research activity were induced from examination of the dissertation abstracts: The first category is referred to as voice strategy.This strategy was used in understanding the first generation student's experience from their own words without manipulating the preconditions other than setting up a sample frame and topic of interest.The use of pre-defined variables is labeled variable strategy.This strategy is best reflected by many of the quantitative studies where both independent and dependent variables were well defined by the researcher and often included the use of a survey instrument.A third research strategy was found and labeled program/intervention strategy.In this type of research activity the focus is on what can be understood about the first generation students by looking at the impact of interventions and/or programs.Program evaluation studies would fit within this category.
The final category of the induced classification system is the comparative strategy.These studies focused on comparing different aspects of the first generation student's experience with other groups and/or experiences.These four induced classifications help to understand the characteristics of first generation research, but the classifications are not crisp categories, in fact, they can be quite "fuzzy" and a particular study might have elements that reflect more than one category, but each dissertation study was given only one classification code for its research activity.
To move the analysis beyond the discovery of research strategies each abstract was examined for topical focus.
Using the method of constant comparative analysis (Corbin & Strauss, 2008) each dissertation topic was given a descriptive code and then compared and contrasted with all other assigned codes.From this list of codes a thematic categorical structure was induced that organized the dissertation topics into three major categories: academic journey, social/cultural experience, and supportive others.The academic journey category included the largest number of dissertations and focused on the variety of academic progress topics including: college choice, access, admissions issues, retention, persistence, attrition, graduation, and transition from campus.The category of social/cultural experience included the topics of campus involvement, identity issues, cultural influences, and social supports.The final category of supportive others included dissertations focusing on family influence, involvement, and support and also includes studies focusing on the support of faculty and staff.While some dissertations related to several topic areas each dissertation was assigned only to the most representative topic.Table 5 presents the results of examining the dissertation abstracts using the classification of research strategies by topical structure.See appendix for a listing of the dissertation by the foregoing categories of research strategies and topical structure.Comparative Strategy 8 5 1

Discussion: Interpretation
Before presenting a discussion of the interpretations of the results, the third aspect of Wolcott's (1994) transformation strategy, a cautionary note needs to be stated.An abstract of a dissertation is not the dissertation.
Only dissertation abstracts were used, not full dissertations.This limits the full discussion on the content of the dissertation, for example, the traditional look at the specific details of "what works" in meta-analytic studies is missing.The focus is more on the researcher, the research attributes and strategies, broad topical categories, and lessons learned from the studies.However, the analysis of abstracts can give guidance to those seeking a more specific understanding and application of particular issues.The interpretation of the results suggests several items to consider: quality of abstracts, framework issues, intersectionality, and absence from the academic conversation of first-generation college students in higher education.

Quality of Abstracts
The quality of the abstracts is a weakness of the data set.In many cases, they were not as useful as they could have been in terms of providing important information about the study.This issue of abstracts lacking important information is not specific to those in this study (Banning & Folkestad, 2012) and has become an agenda on the national research scene.Kelly & Yin (2007) call for the strengthening of structured abstracts for education research.They call for improving the writing of abstracts to enhance their usability to develop and support arguments.They did not note the importance of well written abstracts to support qualitative bounded synthesis studies, but the concern of lack of clear information is paramount for both uses.They call for abstracts to address the following areas: Background and content, purpose/objective/research questions of the study, setting, population/participants, intervention/program, research design, data collection and analysis, findings/results, and conclusions/recommendations.This need for improvement is underscored and is especially important to synthesis studies such as this current study where the abstract is the major source of data.

Framework Issues
Two interpretations of the framework data are presented.One focuses of the lack of mention of a guiding framework by over half of the studies and the second raises a caution regarding the theoretical frameworks used in the studies.The lack of a guiding framework (Ravitch & Riggan, 2012) to be mentioned in the abstract suggests its lack of importance.Without a guiding framework, particularly the connection of the study to theory and past work can leave a dissertation abstract disconnected and isolated.
The most often mentioned theory/model for a framework was Tinto's work and the work of Astin.These are two of the most prolific and cited researchers regarding retention issues in higher education.It would be a shortcoming not to mention the work of these authors in any study that focuses on the academic journey in higher education.In fact, the lack of their mention in over 87 percent of the abstracts is puzzling.The note of caution regarding the use of these theories for first-generation college students is founded on Tierney's (2008) concern regarding the limitation of the notion of social integration, an important concept to Tinto's (1993) work.Tierney calls for looking at the students that typically fall within the category of "diversity" not from social integration with the campus frame, but more of a multicultural frame "where difference are highlighted and celebrated."(Tierney, 2008, p. 66).This caution is especially important since it is likely that a large majority of first-generation students are students of color (Saenz, Hurtado, Barrea, Wolf, & Yenug, 2007).Intersectionality of identity of first-generation college students should not be overlooked.

Intersectionality
Intersectionality (Crenshaw, 1991) underscores the multi-dimensionality of identity and the importance of interdependence of the identity categories of gender, race, class, ability, sexual orientation is important to this study of first-generation college student.While the methodological issues in addressing intersectionality are complex (McCall, 2005) only a few of the authors noted the issue in their abstract.Of the 133 dissertation abstracts, twenty-two dissertations looked at the intersection of first-generation and Latino, seven explored firstgeneration and African American, and three looked at first-generation and White, and three explored low-income and first-generation status.Several of these intersectional approaches also included gender, and one study looked at the intersection of first-generation, deafness, and Latino.Even given these dissertations that addressed some form of intersectionality, only three abstracts directly used the concept of intersectionality or intersection in addressing the purpose of the study.Tran (2011) addressed the intersectionality of first-generation and science identity in relation to STEM education, Lin (2011) examined the intersection of race, SES, and first-generation status, and Williams (2009) explored the intersection of race, SES, gender, and first-generation.
An issue related to intersectionality focuses on the question who is doing the intersectional research.As found in another study using similar procedures (Banning & Folkestad, 2012), the identity of the researcher often is similar to the identity of the participants.For example, in this study, the nine dissertation that explored the intersection of Latina and first-generation, all nine were women and six of the nine included Spanish in their name structure.This finding raises the question of "marginalization."Is the intersectionality of the Latino, first-generation, and female only a Latina issue?The data set did not allow full exploration of this question for other intersectionalities, but to have only important issues regarding the intersectionality of the first-generation students be addressed by researchers that intersect the study's participant's identities appears rather limiting and marginalizing.

Lack of Publication: Absent from the Academic Conversation
In two popular dissertation guide books (Clark, 2006;Foss & Waters, 2007) the conversation metaphor is used to help illustrate outcomes of the dissertation process.Clark (2006), for example, notes the successful dissertation allows entering the conversation.Foss & Waters (2007, p. 3) state: "Academia is a conversation about ideas, and you are able to contribute ideas to a conversation …" Given the lack of the first-generations in this data set that have been published or referenced by major manuscripts, absent from the conversation is one way to describe this phenomena.This study did not seek an answer to why so few dissertations were published or referenced.The lack of motivation to publish and the lack of acceptance by a journal are two obvious possibilities, but this question needs to be explored in depth in future studies on the role of the dissertation in higher education.The findings of this study, however, do suggest that there is considerable and valuable research being conducted in our graduate schools under the supervision of designated graduate faculty, but this research remains relatively hidden and absent from the important academic conversation regarding the first-generation student in higher education.

Lessons Learned
While the major focus of this study was the research strategies and topical areas of first-generation college student doctoral dissertations, the dissertation abstracts can also be reviewed from the "lesson learned" framework that is often associated with qualitative synthesis studies.A major and important lesson learned from the dissertation abstracts is that first-generation students still struggle with achieving academic success and finding a sense of belonging within their higher education experience.However, the dissertation abstracts also clearly indicate that support programs can be a valuable intervention for improving academic and social success on campus.Programs at the federal level, for example the TRIO and McNair programs help to promote student success.Institutional programs in the form of summer programs, BRIDGE programs, and especially campus mentoring programs can also present positive experiences helpful to first-generation students.An additional lesson learned from the study is the importance of the family.Positive family relationships that are supportive are important to student success.Also the studies suggested that this support is not always available and that institutions working with the students' families can assist in bringing about not only greater understanding of the first generation student experience, but can assist in a more successful experience.Coupled and associated with the importance of the family are the financial struggles of many of the first-generation families and students.
Another major lesson coming from exploring the dissertations is the complexity of the task of understanding first-generation college students.This complexity is associated with the previously mentioned concept of intersectionality.The studies pointed out the following intersections: gender, race and ethnicity, age, sexual orientation, SES status, geographical location, ability status, and commuter or non-commuter.Even the acceptance of the definition of first generation is cautioned when studies also look at the role of grandparent education and older siblings who have had higher education experience.Coupled with the complexity of individual characteristics and circumstances is the complexity associated with the higher education environment, for example, community colleges, four year colleges, regional universities, and research universities all interface the first-generation student.In addition, within these institutional types are a variety of academic programs.All of these variables interact and impact both the academic journey and the social/cultural experience of the first-generation student.The lesson to be taken away is that research in this arena should not proceed assuming that the first-generation student status acts alone, but intersects a variety of other possible characteristics, circumstances, and situations, both individual and environmental.
Finally, there are lessons for future research.There are both omissions in the current research as well as new concepts that can be explored.A notable omission is the lack of studies directly focused on the experiences of Native American and Asian American students.From a programmatic perspective there was a lack of studies focusing directly on the residential experience.Only one study directly examined the living in the residence hall experience.Given that academic success is related to the campus social experience and support programs are helpful, additional studies are called for to explore the role of student housing and the development of learning communities.Thayer (2000) recognized this potential, but doctoral research has not addressed this issue.A second programmatic research effort that can be enhanced is the study of the transition from high school to college campus.Research attention focusing on the coordination efforts between the two educational environments could assist in insuring a more successful transition.Only a few of the dissertations in this study addressed this issue.The more recent social science concepts of intersectionality, self-authoring, social networks, the imposter syndrome, and place attachment can also point to future research.These concepts plus the omissions suggest possible important directions for future research.

Summary
The overarching question addressed by this study was: What can we learn from the examination of first-generation college students in higher education doctoral dissertations abstracts.One hundred and thirty-three abstracts were examined.Attributes of the researcher and the research were presented and the dissertation abstracts within a matrix defined by research strategy and topical focus.A major concern addressed in the interpretation of the results focused on the lack of publication of the dissertation and therefor, the absence from the academic conversation regarding the first-generation college student in higher education.Finally, a "lessons learned" framework suggested the ongoing need to recognize the complexity associated with the first-generation college students, the importance of programmatic intervention, and the need to address the role of family support.More research is called for to address the role of the college residential experience and the transition from high school to the college campus.

Table 1 .
Distribution of researcher and research attributes

Table 2 .
Distribution of dissertation by year

Table 3 .
Listing of institutions awarding over three degrees or more

Table 5 .
Classification of dissertation abstracts by research strategy and topical focus