Rethinking College Student Retention by John M. Braxton et al. (review)

In Rethinking College Student Retention, John Braxton leads a team of researchers to investigate and build on the empirical support for Vincent Tinto’s “interactionalist” theory of student retention. With co-authors William Doyle, Harold Hartley, Amy Hirschy, Willis Jones, and Michael McLendon, Braxton assesses extant evidence for the positive relationship between the construct of “social integration” and near-term reenrollment of students in college.

The research team draws on its published work and that of others to propose modifications to Tinto’s theory, which they describe as “paradigmatic” (2014, p. 3). The authors do a service to scholars by summarizing the findings of a generation of research between the covers of one book.

Recognizing that the external environment and the nature of a campus community differ by institutional type, this text examines a retention model for students enrolled at residential colleges and another for students at commuter colleges and universities. Revisions to Tinto’s residential model of student persistence focus on identifying factors that “function as antecedents of social integration” (p. 84). The commuter student model takes student entry characteristics such as motivation, control, self-efficacy, empathy, and “anticipatory socialization” as the starting point (p. 111). The objective of synthesis in this book is to provide strong guidance for institutional action based on the “explanatory power of empirically tested theory” and a “rock-bed of findings of empirical research” (p. 3).

Rethinking College Student Retention has an inverted structure in which Part 1 sets the policy context and presents recommendations; Part 2 presents the theoretical model and research design; and Part 3 summarizes the findings. The findings conclude with a six-point summary judgment of the cumulative knowledge base produced by empirical studies framed by Tinto’s original model and by these revised models. The details of the statistical analyses, variables, and empirical results that are the basis of the authors’ judgments appear in technical appendixes.

This chapter structure, with an exhaustive set of recommendations for state policymakers in Chapter 2 and recommendations for institutional policy and practice in Chapter 3, highlights the book’s appeal to policy and practitioner audiences. The authors certainly include higher education researchers among their intended readers. The later sections will draw in those who have followed the decades-long development, from Tinto’s first journal article on the subject of undergraduate student retention (1975) to his later book-length treatments (Tinto 1987, 1997, 2012).

Unlike Reworking the Student Departure Puzzle (2000), Braxton’s (2000) earlier edited book that brought together authors with diverse perspectives to directly debate the validity of the interactionalist model, Rethinking College Student Retention seeks synthesizing coherence. The authors use a technique they describe as “analytical cascading” to sift through decades of research—much of it their own—to support decisions about which theoretical propositions and variables to include in their final revised models. Analytical cascading, they explain, “entails empirically identifying influences on those theoretically based factors . . . [that] shape student persistence decisions” (p. 5). It “implies discovering what factors shape the environment for improving student persistence decisions in the manner specified by theory” (p. 5). The purpose is to formulate “levers of action” (p. 3).

Using analytical cascading, the authors assess 13 testable propositions of Tinto’s 1975 statement of the interactionalist theory (p. 75), using the earliest formulation due to its alignment with the bulk of the studies analyzed. Though the 13 propositions are “logically interrelated” and “possess internal consistency,” they question: “Does Tinto’s theory also possess empirical internal consistency”? (p. 76). Taking the position that the theory’s validity rests on empirical internal consistency, they use a “box score” method (p. 77), previously reported by Braxton, Sullivan, and Johnson (1997), to count the “percentage of tests of that proposition [in the empirical literature reviewed] that statistically affirm that proposition” (p. 77). They conclude that a proposition that is statistically significant in “66 percent...