A detection model of college withdrawal
Timothy J. Pleskac
, Jessica Keeney
, Stephanie M. Merritt
, Neal Schmitt
, Frederick L. Oswald
Department of Psychology, Michigan State University, United States
Department of Psychology, University of Missouri-St. Louis, United States
Department of Psychology, Rice University, United States
Received 17 August 2009
Accepted 14 December 2010
Available online 13 January 2011
Accepted by Paul Levy
Signal detection theory
Many students during their college careers consider withdrawing from their respective college or univer-
sity. Understanding why some students decide to withdraw yet others persist has implications for both
the well being of students as well as for institutes of higher education. The present study develops a
model of the decision to withdraw drawing on theories of voluntary employee turnover from organiza-
tional psychology and signal detection theory from the cognitive sciences. The model posits that precip-
itating events or shocks (e.g., changes in tuition) lead students to consider withdrawing from the
university. If the evidence surpasses a criterion then the student decides to withdraw. The model was
used to identify shocks students were sensitive to and to test hypotheses about the underlying decision
process. The theoretical implications of this model in terms of understanding and predicting student
withdrawal decisions and voluntary employee turnover decisions are discussed.
Ó 2010 Elsevier Inc. All rights reserved.
A detection model of college withdrawal
An unavoidable fact in higher education is that some students
persist in obtaining a degree, while others withdraw. The National
Center for Education Statistics reported that only 57% of bachelor’s
or equivalent degree-seekers that began college in 2001 had within
6 years graduated from that same college. This overall completion
rate is qualiﬁed by a number of dimensions. Females have a greater
completion rate than males (60% vs. 54%). Completion rates also dif-
fer by race and ethnicity, with Asian/Paciﬁc Islanders having the
highest rate and American Indian/Alaskan natives the lowest (66%
and 40%, respectively; Knapp, Kelly-Reid, & Ginder, 2009). Under-
standing why some students persist at their chosen institution
and others decide to withdraw has important implications for a
range of institutional processes including student admissions, inter-
vention efforts for at-risk students, directions for federal funding,
and maintenance of a rigorous athletic program (Hagedorn, 2005).
Most descriptive level explanations of student retention are
structural in nature. They focus on how academic, social–psycho-
logical, and environmental factors, predict intermediate attitudes
such as different levels of satisfaction and perceptions of poor ﬁt
with the university setting, which in turn predict college turnover
(Aitken, 1982; Bean, 1985; Braxton & Lee, 2005; Tinto, 1975).
Similar approaches focus on the role and availability of different
support systems and their impact on student persistence (Nora,
2004; Nora & Cabrera, 1996). An alternative approach, one we take,
is to focus on the process students use to decide to withdraw from
college. To do so, we developed a formal cognitive model of the
decision process to withdraw from college.
Formal cognitive models
Before going further we should clarify what a formal cognitive
model is and why it is important in theory development. A formal
cognitive model uses mathematical or computer language to spec-
ify how basic cognitive processes give rise to a phenomenon of
interest (Busemeyer & Diederich, 2010). In the case of this paper,
we are interested in modeling how students decide to withdraw
from university. To be certain, computational models have been
used to address similar questions relevant to industrial/organiza-
tional behavior, but the focus of these models have tended to be
at the level of understanding the interaction of people in the com-
plex systems of organizations (Ilgen & Hulin, 2000). Formal cogni-
tive models provide a different level of analysis then these
Formal cognitive models also differ in important ways from
other models often used in psychology. By formally specifying a
theory in mathematical language the model can synthesize the
process and/or system in an observable and testable form. In other
words, one can use the model to see how the process works (or
does not work) to produce the behavior of interest (e.g., a decision).
One can also then add or subtract features to the model (e.g.,
variability) or change parameter values within the model and then
0749-5978/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
Corresponding author. Address: Department of Psychology, 282A Psychology
Building, Michigan State University, East Lansing, MI 48824-1116, United States.
Fax: +1 517 4322476.
E-mail address: firstname.lastname@example.org (T.J. Pleskac).
Organizational Behavior and Human Decision Processes 115 (2011) 85–98
Contents lists available at ScienceDirect
Organizational Behavior and Human Decision Processes
journal homepage: www.elsevier.com/locate/obhdp