Keywords: PIE, persistence, survey, variables, questions
Development of the Persistence in Engineering (PIE) Survey Instrument
Method and Background
The Persistence In Engineering (PIE) survey was developed to identify and characterize the fundamental factors
that influence students’ intentions to pursue an engineering degree over the course of their undergraduate career,
and upon graduation, to practice engineering as a profession. The PIE survey is one of five data-gathering methods
used in the APS. This research brief discusses the initial development and refinement of the PIE survey instrument
and its role as part of the Academic Pathways Study (APS) of the Center for the Advancement of Engineering Education
The PIE survey aims to identify salient correlates of persistence in engineering – both academic and professional—and
thereby extend the research base on engineering student learning and retention. Variations of the survey instrument
will be given to participants in three cohorts of students. The initial version of the PIE instrument was given to
160 students at four of the CAEE partner universities (referred to as Cohort 1).
Development of the PIE survey has been an iterative process, and its scope and content are evolving. As each
administration of the survey reveals new insights, the survey variables and items are evaluated and refined.
Development of the PIE instrument focuses on formulating and categorizing key research questions clustered around
the concept of persistence in engineering. Most of these operational research questions were based on findings
identified in an initial literature review of factors influencing retention in engineering. The operational research
questions were analyzed further, initially by the survey authors, and then by the greater APS team. Based on these
discussions, several key variables were identified.
|Development of the PIE survey has resulted in the formulation of a knowledge base which
forms the basis of a mixed-methods approach in which key issues can be identified and explored as they arise.
In addition to the literature review and APS discussions, existing national surveys on undergraduate education were
also reviewed. Several variables directly related to the goals of the APS survey were identified and incorporated.
These processes ultimately yielded the 26 PIE survey variables (see full text article at
for the complete list of variables and the rationale as to why they relate to persistence in engineering). In order
to measure each variable, a set of survey items were formulated and/or borrowed from the literature.
Conceptually, each PIE survey variable is associated with one of the four fundamental APS research question categories
(skills, identity, education, and workplace). The PIE survey is administered in fall and spring of each academic year
and the variables are being refined after each administration. Responses to each survey are being analyzed both
qualitatively and quantitatively. Consideration of the responses has resulted in the refinement or deletion of
existing variables and the addition of new ones.
What We Found
Reliability analyses were performed on the Spring and Fall 2004 surveys. The Cronbach’s alpha scores for the
variables from the Fall 2004 survey yielded stronger evidence for internal consistency on 13 of the 21 PIE
variables compared to the Spring 2004 survey variables, suggesting that the majority of survey variables were
Implications of Findings
In summary, the PIE survey has contributed the following important outcomes to engineering education research:
- The identification of a set of key variables characterizing core issues related to retention in engineering
education today in the US, and the synthesis of those variables in a survey instrument. The resulting PIE survey
is based on the literature, comprehensive, and responsive to new insights and variables as they emerge from ongoing
- A dynamic survey validation process that draws upon itself as well as other data sources within the APS for
continued evolution (as indicated by the improvement in the internal consistency results over time).
- The formulation of a knowledge base which can inform and enrich the structured interview and ethnographic data
sets within the study. This forms the basis of a mixed-methods approach, in which key issues can be identified and
explored as they arise.
Authors: Ozgur Eris, Helen Chen, Tori Bailey, Kimarie Engerman, Heidi Loshbaugh, Ashley Griffin,
Gary Lichtenstein, and Angela Cole
Source: Proceedings of the 2005 American Society for Engineering Education Conference
The full paper, including references, is located at the ASEE Web site.
For a printable pdf of this research brief, click here.
Brief created April 2007
Back to Research Briefs