Factors that Influence the Transition of University Postdocs to Non-Academic Scientific Careers: An Exploratory Study

Full text

1
Factors that influence the transition of university postdocs to non-academic
scientific careers: An exploratory study
Christopher S. Hayter*
Center for Organization Research and Design, School of Public Affairs,
Arizona State University
Phoenix, AZ 85004, USA
chayter@asu.edu
Marla A. Parker
Department of Political Science
California State University – Los Angeles
Los Angeles, CA, USA, 90032
mparke17@calstatela.edu
Abstract
While postdoctoral fellowships are viewed as positions that prepare PhD students for academic
careers, studies show that most postdocs will not find tenure-track employment within universities.
Postdocs consequently pursue non-academic jobs that differ in the degree to which they utilize a
postdoc’s scientific training, yet we know little about how this occurs. To help address this gap,
this study inductively investigates factors that may influence a postdoc’s transition to a non-
academic career. The study uncovers multiple individual, PI, as well as organizational and policy
factors, including the lack of relevant skills, absence of support—and in some cases opposition—
from their principal investigators, and poor availability of non-academic career preparation
opportunities, among others. Viewed collectively, these elements likely hinder a move to non-
academic scientific positions and thus have consequences for postdoc career trajectories and, by
extension, the utilization of new knowledge. The paper opens the door for future research,
theorization, and policy action that might smooth the transition of postdocs into non-academic
careers and potentially improve the impact of publicly-funded research.

2
1. Introduction
University postdoctoral fellow or scholar positions (hereafter called “postdocs”) have long
been acknowledged as critical to the scientific progress of society (e.g. Conti and Liu, 2015).
Defined as individuals holding a PhD engaged in a temporary period of mentored research or
scholarly training (Lin and Chiu, 2015), postdocs are relatively common among science and
engineering PhDs: over 40 percent of PhDs become postdocs within three years of graduating from
their respective doctoral programs (National Science Board, 2018).
The postdoc position is generally viewed as preparation for tenure-track academic careers.
The position is not only thought to improve a PhDs chances to obtain a tenure-track job (Akerlind,
2005, 2009; Conti and Liu, 2015; Helbing et al., 1998; Horta, 2009; Melin, 2004; Recotillet, 2007;
Su, 2013), it may also be a prerequisite for academic employment in some research fields such as
biology (Miller and Feldman, 2015; Nolan et al., 2004). Postdocs have strong preferences for
tenure-track academic employment (Akerlind, 2005) and thus likely possess what Sauermann and
Roach (2011) term a heightened “taste for science.” This compares with individuals who lose
interest in an academic career over the course of their PhD training (Roach and Sauermann, 2017)
or undertake a PhD in preparation for a career in industry (e.g. Mangematin, 2000).
However, most postdocs in the United States will not find employment as tenure-track
academic researchers. This statement is based on available data from the National Science
Foundation (NSF) 2016 Science Indicators. Figure 1 illustrates the rapid growth in academic
postdoc positions from 4,200 in 1973 to approximately 20,200 in 2013. According to Puljak and
Sharif (2009), this growth combined with a slow-growth or flat academic labor market means that
less than 15 percent of postdocs will find employment as tenure-track faculty. Academic
employment also varies by scientific field. Sauermann and Roach (2016) report, for example, that

3
only 10.6 percent of PhDs graduating in the past five years from life and biological science
programs are employed in tenure track faculty positions. In short, PhDs and postdocs have faced—
for at least 20 years—dwindling job prospects for tenure-track faculty employment, a situation that
is unlikely to improve in the near-term (Cyranoski et al., 2011; Fox and Stephan, 2001; Sauermann
and Roach, 2010, 2012; Stephan, 2012; Smaglik, 2014).
<INSERT FIGURE 1 ABOUT HERE>
Postdocs who do not obtain tenure-track academic employment will transition to non-
academic careers that vary in the extent to which they utilize a postdoc’s individual scientific
capabilities and thus differ in quality and compensation. For example, while not specifically
focused on postdocs, Stenard and Sauermann (2016) highlight varying degrees of “educational
mismatch” among scientists and engineers, a situation whereby an individual’s job responsibility
differs from their training and education. On one hand, a low degree of educational mismatch
means scientists and engineers may obtain technical management positions that align with their
training and education, therefore offering high pay and responsibility, new challenges, and
potential for career growth (Sauermann and Cohen, 2010; Stenard and Sauermann, 2016).
On the other hand, a high degree of educational mismatch among scientists and
engineers—that is, a situation where they do not fully utilize their scientific talents—may result in
negative outcomes, such as diminished wages, productivity, and satisfaction (Bowlus, 1995; Judge
et al., 2001; Tsang and Levin, 1985). Scholars highlight similar negative outcomes among PhDs
and postdocs who maintain low-paying non-tenure-track positions, cannot translate their scientific
training into careers in industry, or, even worse, find themselves in jobs unrelated to their scientific
training (Cyranoski et al., 2011; DeGrande et al., 2014; Hancock and Walsh, 2014; Lee et al.,
2010; Mangematin, 2000; NSF, 2016; Salminen-Karlsson and Wallgren, 2008). This situation is

4
not only detrimental to individual postdocs, it represents the underutilization of publicly-funded
human capital therefore limiting knowledge-based economic growth (Acs et al., 2010).
While many postdocs transition to non-academic jobs, we know little about factors that
influence their transition, especially in the United States; several gaps in the extant literature
motivate this investigation. First, relative to research on PhD students and faculty, few studies
focus on postdocs (Stephan, 2012). Miller and Feldman (2015), for example, posit that “there has
been little empirical investigation of the individual postdoc experience…” (p. 698). When postdoc
research exists, it often assumes that postdocs will necessarily take academic jobs (e.g. Su, 2013).
Second, when scholars examine non-academic employment outcomes among scientists-in-
training, they focus on PhD students (e.g. Hancock and Walsh, 2014; Mangematin, 2000). While
PhD students and postdocs have similar characteristics, critical differences exist between the two
positions making it conceptually problematic to draw inferences about postdoc populations based
on existing PhD-focused research. In contrast to postdocs, PhD candidates are advanced graduate
students who must complete courses, take exams, serve as teaching assistants, and demonstrate
their ability to conduct independent research through the dissertation process (Gardner, 2008).
Discussed in the next section, when related postdoc research exists, it focuses on postdoc
perceptions of career prospects (Puljak and Sharif, 2009), or postdoc career outcomes in other
countries, such as Australia (Akerlind 2005) and Germany (Fitzenberger and Schulze, 2014).1
1 Like the United States, Germany and Australia are advanced economies and possess world-class research universities
and limited academic positions available for postdocs. However, several contextual differences exist that motivate
investigation of postdocs and their transition to non-academic careers within the United States. Not only does the
United States spend considerably more on total R&D in both absolute and relative terms, academic R&D expenditures
(OECD, Research and Development Indicators, 2017) in the United States ($59.5 billion) are also far greater than
those in Germany ($17.6 billion) and nearly ten times those in Australia ($6.2 billion). Further, studies show that,
compared to the U.S., government and industry in Australia and Germany exert relatively more policy influence on
postsecondary institutions in those countries resulting in a stronger emphasis on practical skill building in higher
education institutions (Altbach et al., 2017).

5
Absent detailed analyses that consider the individual perspectives of postdocs, scholars are
left to speculate about, for example, the rationale for why PhDs pursue the postdoc position
(Sauermann and Roach, 2013) or the steps required to “improve the transition of postdocs to
regular career positions” (National Research Council, 2000, p. 99). Further, systematic empirical
research is critical to informing specific policy actions taken by departments, universities, and
funding agencies to improve the postdoc experience (National Research Council, 2014).
This paper consequently explores factors that may impact the transition of postdocs in the
United States to non-academic employment relative to their own a priori career goals. Given the
paucity of related micro-level data, the paper employs an inductive, qualitative approach to identify
these factors among a theoretically relevant sample of university postdocs at five Carnegie-
classified Research I universities within the United States. Postdocs in the sample represent a
variety of disciplines, contexts, and locations, supplemented by interviews with university
administrators and industry employers.
An inductive, qualitative approach is appropriate given the paucity of research relating to
postdocs, especially factors that may impact their pursuit of non-academic careers. According to
Lodico et al. (2010), inductive data collection includes: (1) systematic observation of phenomena
under investigation, (2) iterative analyses of patterns and themes emerging from the data, and (3)
the development of generalizations of the study population based on these analyses.2 Inductivity
usually relies on qualitative research methods. Though qualitative research is not intended to
produce generalizations of broader populations, it nonetheless provides the conceptual foundation
2 In contrast to inductive methods, deductive methods are based on general statements or hypotheses that researchers
seek to test by seeking data that support or disconfirm (Lodico et al., 2010).

6
for subsequent quantitative research that can test the external validity of those insights; that is,
qualitative research results can be used to generate testable hypotheses.
In so doing, the paper offers several distinct contributions. First, the paper is among the
few empirical investigations based on the premise that most postdocs will not obtain tenure-track
academic jobs. This premise not only reflects the realities of the academic job market, it is intended
to be of use to policymakers seeking to improve career outcomes among postdocs and, by
extension, the utilization of publicly-funded human capital. Second, the investigation examines
career motivations among university postdocs, mirroring recent research findings among PhD
students (Roach and Sauermann, 2017), that show interest in academic employment diminishes
over time. Third, the paper finds a myriad of individual, principal investigator (PI), as well as
organizational and policy-related factors which, on balance, slow a postdoc’s transition to non-
academic employment. Finally, the paper offers implications for research, theory, and policy in
hopes that future scholarship will subsequently test and build upon the findings.
The remainder of the paper is outlined as follows. Section 2 reviews factors in the literature
that may impact non-academic employment among PhDs and postdocs. Section 3 discusses the
methodological approach of the paper. Section 4 provides the career goals of interviewed postdocs
and how they have changed, while the fifth section reviews factors that influence the transition of
postdocs to non-academic careers as reported by respondents. Section 6 concludes with a
discussion of findings as well as implications for research, theory, and policy.
2. Conceptual Background
This section reviews the literature related to factors that may affect the transition of
postdocs to non-academic careers. In contrast to the general population of scientists and engineers
with at least a bachelors degree who, according to Stenard and Sauermann (2016), find commercial

7
employment with relative ease, postdocs face significant hurdles finding tenure-track academic
research positions, jobs for which they were trained (Cyranoski et al., 2011; Fox and Stephan,
2001; Sauermann and Roach, 2010, 2012; Stephan, 2012; Smaglik, 2014). While studies
acknowledge the dearth of tenure-track academic career opportunities available to postdocs due to
labor market dynamics, little research explores factors associated with the transition of postdocs
to non-academic careers (Stephan, 2013; Su, 2013; Miller and Feldman, 2015).
Related to the job search itself, scholars have examined the importance of search methods
and early career mobility for PhDs and postdocs. For example, Haley et al. (2018) and Wei et al.,
2012) examine academic job search methods employed by postdocs and find that several factors
are critical, including PhD advisor recommendations, inquiries of postdoc advisors about open
positions, attending conferences, and aggressively responding to job advertisements. Wei et al.
(2012) also find that American postdocs tend to favor utilization of personal networks during their
academic job search while foreign postdocs tend to rely upon job advertisement responses3, the
latter resulting in few placements or placements in less desirable positions. In other words, foreign
postdocs generally have fewer social networks in the United States yet focus their job searches in
that country. Related to the present inquiry, scholars have yet to examine how postdocs conduct
non-academic job searches, including how these searches differ from their academic pursuits.
When non-academic searches are mentioned in the literature, they are characterized as an outcome
of an unsuccessful academic job search, highlighting the inability for postdocs to align their desired
careers with their available options (McAlpine, 2018; McAlpine & Amundsen, 2016).
3 These studies draw upon a broader job search literature that focuses on the efficacy of search methodologies
undertaken by job seekers. This literature dichotomizes individual job searches into formal (e.g. applying directly to
job postings) and informal methods (e.g. using social networks during the application process) finding that informal
approaches are generally more effective (Autor, 2001; Granovetter, 1974, 1977, 1983; Kuhn & Skuteurd, 2000).

8
Given the scarcity of relevant research, this section also reviews publications that focus on
non-academic career outcomes among PhDs. Though most publications reviewed in this section
focus on PhD students, it is important to understand this literature given that all postdocs possess
a PhD. Additionally, PhD advisors may also serve as postdoc PIs and PhDs as well as postdocs
work in similar academic research environments. The results of this review are parsed into
individual, faculty, and institutional and policy-related sections.
2.1. Individual factors
Individuals who pursue a PhD or postdoc have a heightened “taste for science”, including
a preference for the freedom to choose research projects, ability to publish, and desire to conduct
basic research (Astebro and Thompson, 2011; Stern, 2004). PhD education and the postdoc
position help develop this taste through the building of expertise and transformation of identity to
that of a scholar, strongly influenced by the need to belong and contribute to an academic discipline
(Barnacle and Mewburn, 2010; Gardiner et al., 2014; Golde, 1998; Henkel, 2005).
PhD training “glorifies the academic career at the expense of other scientific career paths”
(Gaughan and Robin, 2004, p. 574), thus individuals with a heightened taste for science generally
prefer academic careers over other alternatives (Roach and Sauermann, 2010). Akerlind (2009),
for example, found that 73 percent of postdocs (in Australia) want an academic career, compared
to 10 percent who want a career in industry. PhD career aspirations are reinforced by
misinformation regarding the existence of large numbers of academic jobs that may never
materialize (Golde, 2005).
When PhDs realize the difficulties associated with obtaining an academic job, they often
drop out of their respective programs (Golde, 2005), while others fail to understand job market
dynamics until they become postdocs (Akerlind, 2009; Felisberti and Sear, 2014). Other studies

9
show that more PhD students would consider pursuing non-academic careers in lieu of a postdoc
position if they knew more about these opportunities (Garrison et al., 2016; Gibbs and Griffin,
2013; Puljak and Sharif, 2009; Scaffidi and Berman, 2011). Regardless, postdocs remain in their
positions—often extending them several times—because they cannot otherwise find a tenure-track
academic position (Gaughan and Robin, 2004; Helbing et al., 1998; Puljak and Sharif, 2009).
While modest in number, some individuals enter PhD programs to obtain a job in industry
(Mangematin, 2000) or choose to do so later upon completion of their PhD (Gibbs et al., 2014;
Harman, 2002; Lee et al., 2010; Recotillet, 2007; Sauermann and Roach, 2012; Stephan et al.,
2004). Taking a job in industry is also associated with waning interest in increasingly unattractive
faculty careers (Fitzenberg and Schultze, 2014; Sauermann and Roach, 2012).
If PhDs decide to explore non-academic career options, they generally do not possess the
requisite networks or experience (Mangematin, 2000), nor do they understand how to translate
their scientific knowledge into commercial opportunities (Hancock and Walsh, 2014). Thus, a
mismatch exists between a PhD student’s skills as an academic scientist and those required to
obtain a job in industry (DeGrande et al., 2014; Hancock and Walsh, 2014; Salminen-Karlsson
and Wallgren, 2008).
In sum, scholars have focused on individual factors that may impact the transition of PhD
students and, to a lesser degree postdocs, to non-academic jobs. These studies focus on the
heightened taste for science among PhDs and postdocs, employment trends in the academic job
market, and the degree to which advisors and PIs have the knowledge and networks needed to
provide relevant career mentoring.
2.2. PI-related factors

10
Individual tenured faculty (PIs) not only recruit and supervise postdocs, they also provide
mentoring critical to postdoc career development and job placement (Chen et al., 2015; van Balen
et al., 2012; Miller, 2012). Faculty develop attitudes over their career concerning the purpose of
PhD education (Gardiner et al., 2014) and placing one’s PhD or postdoc into a tenure-track faculty
position is viewed as a successful end-goal. Thus, most faculty encourage PhD students to pursue
academic research careers and eschew non-academic career alternatives (Akerlind, 2005;
Sauermann and Roach, 2012; Stephan, 2012).
While many faculty view postdocs as a training position, others possess an instrumental
view of postdocs “…not as apprentices but as skilled, bargain-rate assistants, who become
increasingly valuable with time” (Singer, 2004, p. 232). Similarly, Puljak and Sharif (2009) find
that PIs often lack incentive to adequately mentor postdocs and, instead, use them primarily to
produce their grant applications and papers. Faculty often favor foreign postdocs for their strong
work ethic and, in some cases, because their visa status prohibits them from pursuing other
employment opportunities (Cantwell and Lee, 2010; Cantwell and Taylor, 2013).
Of course, some faculty wish to help PhD students (and postdocs) find good jobs but
remain uninformed about academic job market realities (Golde, 2005). Further, relating to non-
academic career paths, faculty are unlikely to have the skills or networks required to effectively
mentor PhDs or postdocs for non-academic careers (Akerlind, 2005; Scaffidi and Berman, 2011).
While the studies reviewed in this section do not relate to career outcomes per se, PI-related factors
may nonetheless affect the ability for postdocs to pursue and find non-academic employment.
2.3. Institutional and policy considerations
Most discussions of policy focus on the implications of research findings such as the ones
discussed above, rather than analyzing policy through empirical investigation. Specifically,

11
scholars recommend that research universities provide applicants with academic job market
information before they commit to a PhD program (Sauermann and Roach, 2012); and PhD
students should be encouraged and trained to pursue non-academic scientific career alternatives
(Akerlind, 2005; Harman, 2002; Sauermann and Roach, 2010). Further, several scholars have
recommended that PhDs and postdocs be informed by universities and faculty that their academic
and non-academic job opportunities may diminish with age, thus, timing is critical (Felisberti and
Sear, 2014; Fitzenberger and Schultze, 2014; Helbing et al., 1998).
Mentoring that emphasizes career possibilities beyond academia is critical to improving
career outcomes among postdocs (Davis, 2009; Kaslow and Mascaro, 2007; Scaffidi and Berman,
2011; Singer, 2004). Mentoring should be supplemented with experiential learning, including the
opportunity to work in industry and other non-academic organizations (Fitzenberger and Schulze,
2014; Gardiner et al., 2014; Mangematin, 2000), affiliation with industry cooperative research
centers (Harman, 2002), and interdisciplinary research training (Boden et al., 2011; Holley, 2009).
The National Research Council (2000, 2014) recommends that institutions play a greater role in
facilitating structured career-related mentoring; Leshner (2012) recommends the establishment of
a national body to collect macro data, document best practices, and set standards that would
improve postdoc career outcomes. Given that most discussions do not investigate the impact or
efficacy of policies per se, we have little understanding of what policies and programs meant to
assist a postdoc’s transition to a non-academic career exist and—if they do—their design and
efficacy.
The sections above demonstrate that scholars have investigated factors associated
transition of postdocs and PhDs to non-academic employment. These studies focus on PhD
students (e.g. Hancock and Walsh, 2014; Mangematin, 2000), postdoc perceptions of their career

12
prospects (Puljak and Sharif, 2009), or postdoc career outcomes in other countries, such as
Australia (Akerlind, 2005) and Germany (Fitzenberger and Schulze, 2014). Given the paucity of
relevant research and the aforementioned contextual differences within the United States as
compared to Germany and Australia (footnote 1), it is possible that factors important to the
transition of postdocs to non-academic employment have been overlooked. This paper thus seeks
to build upon and extend the research reviewed in this section.
3. Research design and methods
3.1. Research setting
This paper employs an inductive, qualitative approach to identify factors that influence the
transition of postdocs to non-academic careers (Corbin and Strauss, 2008; Eisenhardt, 1989).
According to Patton (2002), qualitative research provides researchers with tools to inductively
explore a phenomenon about which little systematic empirical research may exist.4
Qualitative researchers necessarily trade external validity for finer-grained insights—and
gain a stronger conceptual basis for, for example, subsequent quantitative survey work (Creswell,
2003). Scholars may otherwise overlook explanatory factors important for understanding a
phenomenon. Once qualitative data have been collected, they can subsequently be compared to
related research and theory (Creswell, 2003).
To collect data, the research team selected five research sites in the United States: four of
the five research universities rank among the top 100 (two in the top 5 percent) R&D performing
institutions and the fifth is a specialized medical university. Further, the combined postdoc
4 For example, Perkman and Schlidt (2015) employ an inductive, qualitative approach to examine the role of boundary
organizations in mediating open data partnerships between universities and industry. They motivate their investigation
by discussing a relatively robust literature on university-industry partnerships then identify a specific element—the
exchange of open data—that has otherwise been overlooked by scholars. The authors articulate aspects of open data
partnerships, such as nuances within intellectual regimes and loose governance structures that promote data exchange,
that can be considered in future studies, quantitative or qualitative.

13
population of the five universities account for a large proportion (20-25 percent) of the total
number of postdocs in the United States. Table 1 provides basic characteristics of the five
universities included in the study.5 Research sites were chosen based on disciplinary variety, the
willingness of senior administrators to participate in the study, and the ability of each site to
accommodate requirements set forth by institutional review boards (IRBs).
<INSERT TABLE 1 ABOUT HERE>
As is the purpose with inductive, qualitative research, the selection of research sites and
participants sought to capture a relevant cross-section of postdocs in order to understand factors
associated with their transition to non-academic employment. Interviews included current
postdocs, faculty and administrators, and representatives from startups and corporations who have
employed postdocs from participating universities.
3.2. Data collection
The study team identified one individual at each university as the primary point of contact
to facilitate data collection and ensure that the study met that university’s IRB requirements. The
points of contact hold positions in offices responsible for providing program and career-related
assistance to postdocs (e.g. Office of Postdoctoral Affairs). Except for the specialized medical
university whose postdocs work within the life-sciences, each point of contact provided a list of
postdocs categorized by discipline. The study team, together with university representatives, used
these lists to generate a stratified random sample of 180 individuals. According to Trochim et al.
(2016), stratified random sampling allows researchers to ensure representation from key subgroups
of a population—in our case, disciplines which account for a relatively low proportion of the
national postdoc population (e.g. on average, the social sciences and humanities account for
5 Table 1 is de-identified to meet the IRB requirements of three universities in the study.

14
approximately two percent of postdocs nationally). Further, university contacts created a
purposeful sample of approximately 42 individuals, including PIs, industry employers,
entrepreneurs, and administrators with whom the research team could speak.
The total number of number of invited postdoc participants (180) reflected three
interrelated considerations: (1) benefits of capturing insights from a variety of disciplinary and
institutional contexts, (2) need for a population size that would allow the research team to achieve
theoretical saturation during data coding6, and (3) time and resource limits requested by university
coordinators.
Out of the 180 individuals invited to participate, 97 responded fully to our request with 35
other individuals interviewed including PIs (9), university administrators (10) and industry
employers (16) (see Figure 2). Figure 3 illustrates the disciplinary distribution of postdocs; 52.6
percent are in life sciences; 29.9 percent are in engineering; 9.3 percent are in physical and earth
sciences; 5.2 percent are in social sciences and humanities; and 3.1 percent are in STEM
education.7 Figure 4 presents the region of origin associated with postdocs participating in the
study: a plurality (36 percent) of postdocs are from the United States, 25.7 percent from Europe
and Asia, respectively, with the remaining 12 percent split evenly between the Middle East and
Mexico and South America.8 Relative to gender, 40.2 percent of postdocs are female.
6 Theoretical saturation occurs when researchers continue sampling and analyzing data until emergent concepts are
well-developed and no new insights appear (Strauss and Corbin, 1998). Scholars differ as to the sample size
recommended to reach saturation when employing inductive approaches. For example, Creswell (1998) recommends
20-30, Morse (1994) 30-50, while some researchers may choose several hundred (Mason, 2010). The research team,
based on the prior experience of university contacts, hoped to achieve a 30-35 percent response rate required to provide
about 60 responses. By receiving 97 responses, the study team instead achieved a 53.8 percent rate exceeding its
desired sample size.
7 The national averages of relevant postdoc categories include: 65 percent in the life sciences, 14 in physical and
earth sciences, 13 percent in engineering, 2 in social sciences and humanities, and 5 percent among other disciplines.
In comparison, the study population has higher relative representation from engineering and social sciences with less
from the life sciences. See https://ncsesdata.nsf.gov/gradpostdoc/2016/html/GSS2016_DST_32.html (accessed April
2, 2018).
8 Study data do not include the nationality of all participating postdocs. However, in 2016, citizens of the United
States and permanent residents accounted for 46 percent of the 64,712 university postdocs, while temporary visa

15
Most postdocs in the study are within the first three years of their position, yet we found
postdocs who had been in their position longer, up to six years. Interviews were conducted in 2016
and 2017 in person or over the phone utilizing an open-ended interview template based on the
literature review. Interviews ranged in length from one-half hour to two hours with most lasting
approximately forty-five minutes in duration.
<INSERT FIGURE 2 ABOUT HERE>
<INSERT FIGURE 3 ABOUT HERE>
<INSERT FIGURE 4 ABOUT HERE>
During the interviews respondents were asked several open-ended questions guided by
Polkinghorne’s (1988) narrative approach, whereby responses are followed by probing questions
meant to clarify meaning and capture in-depth detail related to the question. Postdocs were initially
asked to describe (1) career rationales for undertaking the postdoc position, (2) how those views
had changed over time (if at all), and (3) enablers and barriers (i.e. factors) that might influence
their transition to non-academic careers. The first two questions illustrated the extent to which
postdocs thought it important to consider non-academic careers, while the third—the primary focus
of this paper—focused on specific challenges and enablers to doing so.
The research team took several steps to increase internal validity. In contrast to quantitative
techniques, the validity of qualitative research is defined as “…how accurately the account
represents participants’ realities of the social phenomenon and is credible to them” (Creswell and
Miller, 2000, p. 124-125). Maxwell (1992) similarly frames qualitative research validity in
“descriptive” and “interpretive” terms, among other considerations: descriptive is the extent to
holders accounted for 53 percent. See https://ncsesdata.nsf.gov/gradpostdoc/2016/html/GSS2016_DST_38.html
(accessed April 10, 2018). Please note that the visa status of individual invited to participate in the study were
unknown at the time contact lists were generated.

16
which researchers accurately capture the responses of the study participants while interpretive is
the extent to which researchers capture their meaning. Following procedures outlined in Creswell
and Miller (2000), individual interviews were followed by in-person and over-the-phone meetings
with participants to validate their responses.9 The study team also sent the study findings to
university points of contact for review and comment. Finally, the manuscript was sent to five
scholars with experience in postdoc and scientific career-related research for comment.10
In addition to supplementary interviews with administrators and PIs, university contacts
also provided contact information for individuals who had hired postdocs into non-academic
positions. These contacts were similarly asked to describe which factors might inhibit the ability
of postdocs to obtain and succeed in a non-academic career.
3.4. Data Analysis
All interview data were recorded and a memo was created after each interview
summarizing insights and exploring theoretical implications, including how responses fell into
general categories of factors that influence the transition of postdocs to non-academic careers.
Information was triangulated across multiple sources, including postdocs, faculty, administrators,
postdoc coordinators, and employers.11
Once the data collection phase was complete, interview transcripts and memos were coded
inductively by the study team12, according to procedures recommended by Kuckartz (2014) and
9 Creswell and Miller (2000) term this approach “member checking” whereby researchers are “taking their data and
interpretations back to the participants in the study so they can confirm the credibility of the information and
narrative account” (p. 127).
10 Creswell and Miller (2000) term this stage the “validity audit”.
11 Triangulation is the use of multiple sources of data and/or multiple research methods to provide varying
perspectives on the phenomenon of interest as well as validate study findings during data analysis. For an in-depth,
foundational discussion of triangulation, see Jick (1979).
12 The research team was comprised of three individuals, including the two authors, both tenure-track professors and
a research assistant. The two authors coded the data, asking the graduate student to resolve ties related to divergent
data interpretations. Inter-coder reliability, the extent to which independent coders evaluate reported data and reach

17
Saldana (2012). Specifically, an initial round of first-order (open) coding (Corbin and Strauss,
2008) was conducted by hand, supplemented with an analysis utilizing NVivo software. The
research team then moved to second-order (axial) data coding to understand the relationship among
open codes within and across different types of interviews. Throughout, the research team moved
between the data and the emerging data categories for refinement. Finally, second-order codes
were placed into a conceptual model (see Figure 5 below) that visually illustrates our findings. The
results are reported in the next section.
4. Postdoc career rationales
4.1. Original career motivations
Illustrated in Table 2, a majority (86.6 percent) of postdocs report that they originally
pursued a postdoc to improve their likelihood of obtaining a tenure-track faculty position. Nearly
all respondents in the life sciences (51.5 percent of the study population) indicated that the postdoc
is a prerequisite for tenure-track employment within their discipline. Other postdocs saw the
position as an opportunity to establish a publication record, learn valuable research skills, and work
with well-known scientists to improve their chances on the academic job market. About half of
postdoc respondents unsuccessfully applied to academic positions prior to the postdoc; many
described the postdoc as a “holding pattern” where they could continue their research until they
found an academic job.
<INSERT TABLE 2 ABOUT HERE>
Six individuals also described the postdoc position as a holding pattern, yet did not view
academic employment as their primary career goal. These individuals realized during their PhD
the same conclusion (Neuendorf, 2002), improves data quality. Using (1) percent agreement and (2) Krippendorff’s
alpha, coded variables exceed accepted thresholds of inter-coder reliability, 90 percent and 0.800, respectively.

18
that they did not want to become tenure-track faculty, yet chose to take the postdoc position while
they explored their career options and continued to develop scientific skills and networks.
A total of seven individuals reported that they took the postdoc position specifically to
support non-academic career ambitions. Two postdocs took the position after several years of
experience working in industry and secondary education, respectively. One individual sought to
work on cutting edge science and engineering research to be more marketable to companies in
their field. Three individuals took postdoc positions after their PhD to gain exposure to
entrepreneurship skills and programs in hopes of either establishing their own spinoff company—
or to be hired by one. Finally, one individual sought to obtain a position as an administrator in
higher education.
4.2. Evolving career goals
Postdocs were also asked if their career goals had changed since taking the postdoc position
(also presented in Table 2). At the time of writing, five postdocs in the study had accepted tenure-
track positions at research universities, thus realizing their original career goals. However, in
addition to the thirteen individuals mentioned above, 31 additional postdocs reported that they had
decided to no longer pursue an academic position. Postdocs attribute their decision to the
realization that it was unlikely that they could obtain tenure-track employment or that their interest
had waned, or both. Responses may also be a function of experience: first-year postdocs were
generally more optimistic about their career prospects than third and fourth year postdocs who had
participated in academic searches for several years, including during their PhD.
Among postdocs who had decided not to pursue an academic career, several described
being in a state of uncertainty and indecision about their future employment options. Other
postdocs had clearly decided to pursue research-oriented careers in industry or teaching-oriented

19
positions in primary or secondary education, while others elected to investigate career paths
unrelated to their scientific training. In short, while most postdocs in the study originally viewed
the position as preparation for an academic career, many eventually realized and accepted that they
would be unable to obtain a tenure-track research position.
5. Factors influencing the transition to non-academic careers
This section includes factors that influence the transition of postdocs into non-academic
careers. Reflective of emergent multi-level themes, findings are grouped into individual, principal
investigator (PI), and organization and policy-related factors. Findings are supplemented by quotes
articulated in Tables 3 and 4 that follow sections 5.1 and 5.2, respectively.
5.1. Individual factors
The ease by which a postdoc can transition into a non-academic career is partially mediated
by factors associated with postdocs themselves. Discussed in detail below, these factors include
individual-level motivations, information asymmetries, capabilities and skills required to thrive in
a non-academic career, as well as stress attributed to the realization that one will no longer achieve
their career goals.
5.1.1. Career expectations
As discussed, most postdocs in the study undertake the postdoc to improve their chances
of obtaining a tenure-track research position. Many postdocs expressed surprise that they could
not obtain academic employment given the reputation of their advisors, stature of their PhD and
postdoc programs, and scientific accomplishments. According to respondents, affiliation with
world-renowned universities and PIs built confidence in their scientific abilities, which led to
skepticism of the need to explore non-academic career options. However, these expectations

20
eventually became a liability as postdocs realized, often after several years on the academic job
market, that obtaining tenure track employment was unlikely.
Somewhere in the back of my head, I knew the job statistics and the slog of the academic
life but when you get your PhD at [an Ivy League University] and do your postdoc at [a
well-known research university] you get into the mindset that those numbers don’t apply
to you…that yeah, I can beat the odds. Turns out I couldn’t.
Postdocs making this realization find themselves searching for non-academic jobs that they
otherwise dismissed earlier in their training.
5.1.2. Information asymmetries
Postdocs reported that they did not have accurate information about the academic job
market; or they received contradictory information from their PIs, career services personnel, and
other PhD students. Many also feel that life as a PhD student glorifies tenure-track positions,
glossing over challenges associated with an academic career, including aggressive tenure
requirements, fundraising, supervising students, teaching, work-life balance, and conflict with
colleagues. With time, postdocs not only learn vicariously about life as a tenure-track academic
researcher, they also find that their PhD did not prepare them for being a postdoc.
…[I] wasn’t prepared for all of this…when you take a postdoc, you are kind of thrown into
the deep end…you realize that life as a researcher is pretty solitary. Then there’s all the
other stuff like supervising PhD students…this person was you six months ago…oh, and
grant writing. I now understand why my advisors didn’t sleep at night.
Postdocs also reported either having poor access to data concerning their career prospects
or, discussed below, are “misled to believe that the postdoc will magically overcome the fact that
there are no [academic] jobs.” Postdocs who cannot find academic employment feel that they do
not have information about the value of their scientific training within other contexts and job
options therein. Common questions from postdocs included the following: what do I need to know
to work in government or industry, how do I find out about what jobs exist, and what is the lifestyle

21
and compensation of these individuals? Other postdocs expressed interest about working in or
establishing a start-up company, but knew little about how to do so.
5.1.3. Skills
Related to information asymmetries, postdocs report not being well-prepared to search for
non-academic jobs or associated responsibilities. When searching for non-academic jobs, postdocs
often “[D]on’t know where to start…I feel unprepared for the basic stuff like networking, creating
a resume [as opposed to a curriculum vitae], the job search itself, and how to talk to those people.”
Employers report that postdocs are adept at scientific concepts and research methods, but
lack “the capability to understand some basics of how the real [non-academic] world works.”
According to employers, it is difficult for postdocs to learn how to apply their research skills in
support of product development (in industry) or address a specific applied problem (in government
labs). Postdocs also do not possess leadership and teamwork experience required for industry or
startup teams that integrate multiple functions, such as research, management, manufacturing, and
sales. Associated skills include the capability to work under strict deadlines and budgets, cancel
projects that do not yield results within a specific period, make brief pitches, and communicate
complex concepts to non-scientific audiences.
Industry employers report that postdocs generally require more time, compared to masters
and PhD students, to acclimate to a non-academic career. According to startup personnel, multi-
dimensional skills and swift transitions are particularly important within entrepreneurial ventures,
given the rapid pace of change and need to quickly demonstrate results to investors. Industry and
government respondents lamented that they might consider employing more postdocs if they
possessed non-academic skills and experience.
5.1.4. Personal Crisis

22
About 20 percent (18 out of 84) of postdocs seeking academic employment during the
study reported that the realization that they will not achieve their long-time career goals resulted
in what one respondent called an “existential crisis.” These postdocs reported periods of chronic
stress, anxiety, anger, and depression, as well as what one respondent described as a “debilitating
period of inaction while I am trying to figure out what I want to do with my life.” Periods of crisis
slow a postdoc’s ability to pursue non-academic career options.
Several factors seem to amplify these crises, including visa status for foreign postdocs,
family, and financial difficulties. Between 1973 and 2013, the number of foreign born postdocs
increased from 17.5% to 47.5% (NSF, 2016). The inability of these postdocs to find an academic
job not only means that they must transition to non-academic employment, it also means they must
do so with an organization willing to sponsor their H2B work visa or otherwise return to their
country of origin. Postdocs may have children and rely on their spouse for financial support. In
some cases, the inability to find tenure-track employment resulted in high levels of conflict with
their spouse or significant other. Further, postdocs are generally paid 40-50 thousand U.S. dollars
a year, many of whom live in high-cost urban areas and possess high levels of student debt.
<INSERT TABLE 3 ABOUT HERE>
5.2. Role of PIs
Faculty PIs play a critical role in the postdoc experience. PIs generally recruit and hire
postdocs, manage their day-to-day research activities, and have significant influence over their
careers. The sections below highlight factors related to PI-postdoc interactions.
5.2.1. Lack of Career Support
While postdocs reported that some PIs view postdocs as “cheap labor”, many PIs also
provide little career mentoring because they believe postdocs should find their own way:

23
I think some of this is just the nature of the beast…I never had someone hold my hand
through my PhD or postdoc…so I had to figure things out on my own, so I now expect the
same of my postdocs…there’s a lot to be said for being a self-starter…let’s see if you can
figure it out and roll with it.
Related to academic job searches, most postdocs reported that PIs are usually willing to support
their candidacy for a tenure track position by writing recommendation letters or making phone
calls. This support stems from pride in placing postdocs in academic positions, which also carries
high value in the academic world.
However, academic-centric definitions of success among PIs can also inhibit non-academic
career pursuits for postdocs. First, advisors generally do not understand contemporary academic
job markets; many began their position under different job market conditions. Second, postdocs
who cannot find tenure-track employment are thought to be at fault for their inability to do so.
Finally, non-academic career pursuits are viewed as a distraction from the research responsibilities
of postdocs. In these cases, PIs either tolerate efforts to explore non-academic job possibilities
(e.g. “do it on your own time”) or they strongly oppose non-academic careers, thus encouraging
postdocs to hide their career preparation activities. In the words of one administrator who
frequently advises postdocs:
All too often PIs try to limit the exposure of postdocs to activities that would help build
valuable skills applicable to any career, even on their own time. While some of these
postdocs do engage in activities clandestinely, many are afraid to so because of the adverse
consequences that come with being found out...which they have seen happen among their
peers.
5.2.2. Skills and Capabilities
In cases where PIs support efforts to investigate non-academic career possibilities, they
may not possess the skills or networks that might otherwise help postdocs. Postdocs reported that
“there seems to be more sympathy among younger faculty…they seem to understand the [job

24
market] challenges we face…but they don’t really know how to help.” PIs interviewed in the study
similarly contemplate how to build skills and networks that would enable them to, for example,
obtain sponsored research funding from industry or commercialize technology. In a few cases,
postdocs attended entrepreneurship support classes to help PIs “figure out how to do a startup.”
Ironically, one PI with a record of placing PhD students and postdocs in non-academic
positions reported that her relationships with former PhDs and postdocs now working in industry
had become an important source of research funding and improved her overall academic
performance. However, this was not the case for most postdocs or faculty within the study.
5.2.3. Socially irresponsible behavior
Socially irresponsible behavior among PIs limits the ability of postdocs to pursue career
opportunities, academic or non-academic.13 Reported behavior includes excluding a postdoc from
co-authoring a paper to which they contributed, repeated use of sexist language, and actions
hampering the renewal of visas for foreign postdocs. Particularly egregious was a PI asking a
postdoc to work full time for 70 percent of their university’s postdoc salary minimum or that they
double their 40-hour in-lab work requirement with the implication that the renewal of their visa
depended on it. J1 and H1B visas are granted for one job (e.g. a postdoc position at a specific
university) and, given uncertainties associated with grant funding, are generally renewed annually.
Thus, foreign nationals chose not to report socially irresponsible behavior for fear of retribution
from their PI.
American and foreign postdocs alike report that some PIs penalize their pursuit of non-
academic career opportunities. Reactions to postdocs telling PIs that they were interested in non-
13 For a review of the relevant social responsibility literature and its application within the context of higher
education, see Hayter and Cahoy (2018); Hayter (2016c) applies a social responsibility framework to analyze
current conceptualizations and practices related to university technology transfer.

25
academic careers or attending related events ranged from “don’t tell me about that” and “do that
on your own time” to not to speaking with a postdoc for several months. In the worst cases,
postdocs are asked to leave their lab. Postdocs attribute these reactions not only to the culture of
science, but as one respondent put it, “the strange notion that PIs own you and all of your time.”
Two postdocs also reported conflicts with their PI relating to intellectual property (IP). One
postdoc, for example, developed a technology during their PhD and continued to work on it as a
postdoc with the knowledge and support of their PI. Following university IP policy, the postdoc
disclosed their technology to the technology transfer office (TTO), which filed for a patent, and
included their PI on the claim. The postdoc established a spinoff company and attracted the interest
of investors. Once the postdoc informed the PI of investor interest, however, their relationship
changed. The PI insisted having control of any related entrepreneurial efforts and threatened
outside legal action if the postdoc did not comply. According to administrators knowledgeable of
the conflict, while university IP policy officially treats postdocs as faculty, TTOs defer to PIs on
postdoc-related IP matters enabling them to, for example, change the distribution of licensing
revenue among inventors not otherwise permissible in peer-to-peer relationships.14 Other postdocs
mentioned this example and another as indication that they should not work on or generate IP
within the university for fear that it would be appropriated by PIs, which consequently limits
entrepreneurship as an non-academic career option.
<INSERT TABLE 4 ABOUT HERE>
5.3. Organization and Policy
14 While the specific percentages differ by university, most IP policies distribute licensing revenue relatively evenly among
inventors, departments or labs, and the university (typically used to support the TTO) itself. For multiple individuals, the proportion
going to inventors is generally divided evenly. However, this was not the case for the postdoc in this example. Further, other
individuals not involved in the technology were added to the patent further diluting the postdoc’s share.

26
This section discusses organizational and policy factors relating to a postdoc’s pursuit of
non-academic careers. Factors include the structure of the postdoc position, university policy
relating to postdocs or lack thereof, and federal policy.
5.3.1. Nature of postdoc position
The postdoc position can itself present a barrier to future non-academic employment
among postdocs. For universities in the study, postdoc fellowships are generally advertised as
temporary two-year positions, extendable for two additional years, with additional years available
with administration approval. Despite university and federal efforts (discussed below), the locus
of decision-making for postdoc career development remains with PIs. Postdoc positions are not
only viewed as salaried, more-than-40-hours-per-week positions that support PI research, PIs
can—following the results above—largely choose not to prioritize the career goals of postdocs.
Thus, as structured, the postdoc position provides little formal occasion to learn about non-
academic opportunities or build corresponding skills and networks.
5.3.2. Institutional policy
For universities in the study, policies or processes for postdocs differ across disparate
departments and disciplines. For example, one lab has an onboarding process, including a two-
week orientation, for incoming postdocs while postdocs in a different department in the same
college spoke of the need for orientation. Differences in policy may be attributed to the
decentralized, faculty-driven culture of academia. University-wide postdoc policies among
universities in the study are generally limited to IP and federal compliance. Further, while funding
agencies provide universities with flexibility for postdoc positions and policies15, administrators
15 For more information, see the NIH Postdoc Handbook:
https://history.nih.gov/research/downloads/stetten_fellow_handbook.pdf (accessed Nov. 29, 2017).

27
report that this guidance is generally disregarded or interpreted conservatively, thus marginalizing
university-wide career-development efforts for postdocs.
Many universities have established postdoc career preparation programs and offices; four
universities in the study have done so, three since 2014. While personnel in these offices
recognized the employment challenges of postdocs, three offices reported being under-resourced
and provided only basic career services such as CV/resume review, job talks, interviewing skills,
and academic job search capabilities. A fourth university also offered communication, leadership,
conflict resolution, time management, basic computer programming, and mentoring services. A
lab within one university in the study offered career services meant to assist postdocs in non-
academic career pursuits.16
Interestingly, an entrepreneurship support program located at one university in the study
had become a de facto career development program for postdocs. Entrepreneurship support
personnel reported that postdocs generally lacked the skill to work within a startup environment,
much less establish a spinoff company, yet possessed valuable scientific knowledge that might be
used to develop and commercialize new technologies. While the entrepreneurship program tailored
its services to meet postdoc needs, it nonetheless faced opposition from faculty “who think the
program distracts postdocs from their primary job.”
5.3.3. Federal policy
Most postdoc positions are funded from research grants made by the National Institutes of
Health (NIH) and NSF, among other sources. Postdoc fellowships funded by NIH generally hold
T32 designation, which are awarded to PIs vis-à-vis their respective universities. However, NIH
16 A chapter of the National Postdoc Association (NPA), which advocates for the interests of postdocs, is also located at each
university in the study. On one campus, the NPA has organized a voluntary network of postdocs and postdoc alumni to support
career development activities, including those supporting non-academic careers, but lacks the resources to scale these efforts.

28
also awards F32 postdoctoral fellowships that go directly to individual postdocs. According to
respondents, F32 awards substantially change the dynamics of the PI-postdoc relationship to PIs,
often providing more autonomy for postdocs though they may receive less attention (e.g.
mentoring) from faculty.
Federal agencies also set salary minimums for postdocs; NIH recently raised salary
minimums for postdocs (Kuo, 2016). In addition to recent programmatic efforts to improve career
outcomes among postdocs17, NIH also requires that postdocs develop individual develop plans
(IDPs) because they help graduate students and postdocs, in cooperation with their advisors and
PIs, identify career goals and resources required to achieve those goals.18 Finally, former President
Barrack Obama signed into law the America Creating Opportunities to Meaningfully Promote
Excellence in Technology, Education, and Science (COMPETES) Reauthorization Act of 2010
(P.L. 111-358), which includes specific language acknowledging and promoting non-academic
career development opportunities among masters and PhD students.19
Following observations relating to the role of entrepreneurship support services in one
study site university, several postdocs discussed the importance of participating in the NSF
Innovation Corps (I-CORPS) program administered by their university. Their participation in the
program allowed them “to understand important basics for working within a private sector
environment…through structured interviews with potential clients [companies].” The I-CORPS
program promotes participation in teams thus postdocs participate based on the interest and
approval of their PI.
17 Through the Broadening Experiences in Scientific Training (BEST) program, NIH provided funding for 17 universities to
establish career-development services, including for non-academic careers. No university in the study is a BEST site.
18 https://nexus.od.nih.gov/all/2013/07/23/individual-development-plans-for-nih-supported-trainees/ (accessed Nov. 29, 2017).
19 Illustrative language in the bill includes: “The Director shall award grants, on a competitive, merit-reviewed basis, to
institutions of higher education to implement or expand research-based reforms in master’s and doctoral level STEM education
that emphasize preparation for diverse careers utilizing STEM degrees, including at diverse types of institutions of higher
education, in industry, and at government agencies and research laboratories.”

29
6. Discussion
Individuals participating in this study generally view the postdoc position as preparation
for tenure-track academic employment, yet our findings echo macro-level data that show most
postdocs are unlikely to obtain tenure-track employment (e.g. Puljak and Sharif, 2009).
Employment outcomes among postdocs in the study reinforce this assumption: only five of the 97
postdocs found academic positions during the two-year duration of the study. While additional
postdocs may find academic employment, many have nonetheless decided that an academic career
is no longer an option.
These individuals are pursuing or plan to pursue non-academic career opportunities and
may therefore face factors uncovered during the investigation. Figure 5 presents a visual
conceptualization of these findings. While the study is qualitative and thus makes no claim to
external validity, these factors nonetheless provide a conceptual basis for future hypothesis testing.
For example, we anticipate that future scholarship might examine the relationship between
identified factors and inter-related career outcomes among postdocs such as skill utilization,
productivity, compensation, and satisfaction (Stenard and Sauermann, 2016). The benevolent
career support intentions of younger PIs and entrepreneurship support programs are examples of
enabling elements, yet most factors uncovered in the study may be classified as barriers, likely
slowing the transition of postdocs to non-academic employment.
The next sections discuss study findings in greater detail. The first section discusses study
results that closely follow the extant research which focuses primarily on PhD students. The second
section highlights unique empirical contributions of the study while the third section, following
Creswell (2003), offers implications for future research.
<INSERT FIGURE 5 ABOUT HERE>

30
6.1.1. Findings following the extant literature
This study documents several factors that comport with previous studies of career
transitions among PhDs and postdocs. Mirroring extant research focusing on PhDs, postdocs in
the study possess a high taste for science and originally pursued the postdoc position to increase
their chances for obtaining tenure-track, academic employment (Sauermann and Roach, 2010,
2011). The opportunity cost of doing so means postdocs forego other training opportunities and
are thus generally unprepared to pursue non-academic careers (e.g. Hancock and Walsh, 2014).
Specifically, postdocs do not possess the requisite skills or networks needed to obtain and succeed
in a non-academic position (DeGrande et al., 2014; Hancock and Walsh, 2014; Salminen-Karlsson
and Wallgren, 2008). Following discussions of information asymmetries among PhDs in the
literature (e.g. Scaffidi and Berman, 2011), postdocs also report that as PhDs and early postdocs
they were not aware of their poor academic career prospects or non-academic career options.
Prior studies (e.g. Singer, 2004) and policy reports (National Research Council, 2014)
discuss the long-standing challenge of PIs viewing their postdocs PI as “cheap labor”, a perspective
also captured within this study. While postdocs view PIs as critical to career preparation and
training (Chen et al., 2015; van Balen et al., 2012; Miller, 2012), most PIs do not necessarily
support non-academic career pursuits among their postdocs, a finding in keeping with research on
PhD advisors (Sauermann and Roach, 2012) or postdoc PIs in Australia (Akerlind, 2005). When
PIs do support the non-academic career pursuits of postdocs, they generally do not understand
non-academic job market dynamics, nor do they possess the skills or social networks that would
otherwise enable the success of their postdocs, a finding in keeping with prior research among
PhD students (Golde, 2005; Scaffidi and Berman, 2011).
6.1.2. Unique Empirical Contributions

31
While this paper validates findings from the extant literature, it also offers several unique
empirical contributions. As discussed, many postdocs expressed concern that they did not
previously have access to information regarding academic job markets and their non-academic
career options. However, as PhDs and early postdocs, other individuals were indeed aware of the
(low) likelihood of obtaining an academic position, yet nonetheless chose to disregard this
information. Other postdocs report that they were led to believe by their PhD advisor and PI that
they would be able to “beat the odds” given their Ivy League PhD education, postdoc at a university
of similar stature, and renown PI. The same postdocs later realized that they could not beat the
odds (and/or that high profile academic research positions are undesirable) and thus needed to
pursue non-academic employment.
While existing studies document the strong emphasis among advisors and PIs on academic
careers for their postdocs, they do not discuss actions PIs might take to slow or inhibit the transition
of postdocs to non-academic careers. This study shows that some PIs in the study view the pursuit
of non-academic jobs by postdocs as a distraction or an indication of failure. Worse, some PIs
penalize their postdocs for exploring non-academic careers by refusing to speak with them or
asking them to leave their lab.
Perhaps the most startling factor, and another unique contribution of the paper, is the
identification and description of socially irresponsible behavior among PIs. Postdocs, faculty, and
administrators reported abuses, the most serious of which involved taking advantage of foreign
postdocs during their yearly visa renewal process. Postdocs similarly reported that PIs
misrepresented IP ownership and misallocated resulting licensing revenue.20 According to one
20 Though not focused on career transitions or postdocs per se, Hayter et al. (2017) finds evidence of IP-related
conflict between graduate students and advisors when they are jointly involved in the establishment and
development of a university spinoff company.

32
administrator, such actions are not only social irresponsible, they are also illegal, may invalidate
associated patents, and could attract the interest of inspectors general at research funding agencies.
Further, situations whereby universities don’t do more to prevent irresponsible behavior—
especially actions that are unethical and/or illegal—may represent institutional failures with
unforeseen legal and political consequences.
Several considerations are worth noting when interpreting these findings. First, the findings
are not meant to suggest that socially irresponsible behavior is commonplace among PIs; most
postdocs in the study work for PIs who care about their wellbeing and wish to see them succeed
professionally. Second, PIs are under enormous pressure to raise money, manage large research
enterprises, publish in esteemed journals, and increasingly demonstrate commercialization-related
outcomes. Third, behavior among PIs is a matter of degree: refusing to allow postdocs to pursue
non-academic training opportunities stands in contrast to violations of patent and immigration
laws. In short, PI-related findings should be interpreted carefully within the complex context of
academic science.
Another unique finding is that some postdocs in the study view entrepreneurship as a
potential alternative to an academic career. Specifically, postdocs at two universities took
advantage of entrepreneurship support services to expand their non-academic career options,
including the opportunity to work in a startup environment.
As discussed, mention of postdoc policies and programs in the literature is generally
limited to the discussion section of empirical papers (e.g. Sauermann and Roach, 2012) or policy-
focused consensus reports (e.g. National Research Council, 2014). Thus, another original
contribution of this study is the documentation of university-level policies and programs as well
as their potential impact on non-academic careers among postdocs. While administrators were

33
concerned about postdoc job prospects, universities in the study take a relatively hands-off
approach to postdoc career development. In the words of one administrator: “[Y]ou can’t think of
the postdoc as a university program…the postdoc is something that happens on campus with little
involvement from the administration.”
While some universities may offer relevant career services, only one college in one
university offers substantive programing to prepare postdocs for non-academic careers. The
efficacy of cross-cutting policies and programs nonetheless relies upon the discretion of individual
PIs who may choose not to allow postdocs to participate in these activities. Interestingly, little
coordination exists in the study’s participating universities among efforts to improve postdoc
employment outcomes (academic and non-academic) and the aforementioned entrepreneurship
support program.
The remaining contribution of the paper relates to the physical and emotional well-being
of postdocs. Many postdocs feel disconnected from social networks that might otherwise support
their personal and professional development. Because of this isolation, combined with the
realization that they will no longer realize their long-standing career goals, some postdocs
experience high levels of stress, anxiety, and depression, culminating in debilitating crises.
6.1.3. Implications for future research
In sum, this paper provides in-depth description of factors that may impact the transition
of postdocs to non-academic careers. As discussed, most postdoc-related research assumes that
these individuals will obtain tenure-track positions while our study assumes that most postdocs
will not. Though beyond the scope of our investigation, this project would have benefitted from a
deeper understanding of institution and mobility-related aspects associated with the postdoc job
search. As discussed, while scholars have investigated academic job search methods employed by

34
postdocs, they have yet to examine how postdocs go about conducting non-academic job searches
(Wei et al., 2012). Future research might investigate to what types of universities postdocs are
applying as well as their location. Given that approximately half of postdocs in the United States
are comprised of foreign nationals, to what extent do postdocs apply to academic jobs in the
country of their postdoc in comparison to other countries? Do United States citizens apply for
tenure-track academic research positions in other countries? How do the experiences of citizens of
the United States differ from foreign nationals during their academic job search? Scholarship
addressing such questions will provide a deeper understanding of the role of geography in national
and international academic labor markets, in the present case among postdocs.
Scholars can also, following Creswell (2003), build on the unique findings outlined in this
study by examining their incidence and impact through quantitative survey work, including the
collection of time-series data. Following Roach and Sauermann (2017), scholars could investigate
how and to what extent career-related goals among postdocs change over time. Opportunities for
theorization likely exist among different disciplines, university types, and personal characteristics
given the relatively heightened taste for science and strong expectations for an academic career
among postdocs.
While this study uncovered evidence of socially irresponsible behavior among PIs, future
research can examine other types of abuse and, more importantly, the prevalence of this behavior.
Challenges among foreign postdocs are mentioned in the literature (e.g. Cantwell, 2011), yet
additional insights on the relationship among visa type and status, types of behavior, and specific
career outcomes, nonacademic and otherwise, is welcome. Researchers might also examine the
extent to which university expectations of PIs related to fundraising, eminent publication, and
management are correlated to irresponsible behavior. Further, future studies could examine the

35
extent to which university administrators understand the nature and incidence of socially
responsible behavior among PIs, institutional failures that enable these behaviors, and, conversely,
organizational processes and cultures that help prevent these abuses.
Related, an opportunity also exists for researchers to empirically investigate the incidence
and composition of policies and programs meant to enable non-academic job placement among
postdocs, as well as their relative efficacy. Of interest is the intersection, real or potential, between
postdoc career development programs and entrepreneurship and technology development support
programs. Scholars might frame this investigation in terms of a postdoc’s capability to contribute
to university economic development goals, along with the interventions and social networks
required to realize their potential. More generally, future research might investigate how non-
academic career mentoring and career guidance differ among faculty and, following the academic
entrepreneurship literature, examine the structure and contributions of their social networks
(Hayter, 2016a; 2016b; O’Gorman et al., 2008).
While this study assumed that impact should be understood in terms of postdoc career
outcomes and related factors (e.g. compensation, satisfaction, opportunity for promotion), other
individual impacts deserve attention, in our case the incidence of chronic stress reported among
postdocs. Following research highlighting mental illness among PhDs (Leveque et al., 2017),
future interdisciplinary scholarship might investigate the nature and incidence of stress and related
symptoms as well as their specific impact on postdoc wellbeing, including their career search.
Exploratory research might also examine enablers and barriers to non-academic scientific
employment from the perspective of individuals already working in these careers. Specifically,
what factors enabled former postdocs to search for, obtain, and succeed in non-academic careers?

36
To what extent did these factors relate to PI behavior and networks, university programs and
policy, and other factors outside the scope of the present study?
Scholars can also investigate interactions among factors. For example, to what extent does
career related stress differ by visa status among postdocs? While the lack of differentiation among
scientific disciplines is discussed below as a limitation of the study, scholars might investigate the
degree to which a postdoc’s discipline influences their ability to build skills and networks
pertaining to non-academic jobs. To what extent does PI-driven academic culture consider or
disregard university and federal policies?
Finally, NSF and other organizations might also collect more and better data, beyond the
periodic survey of earned doctorates, to understand better the transition of postdocs into non-
academic careers. Such data collection might not only provide detailed insights into various factors
that slow or accelerate the career development of postdocs, it could also be used to understand the
efficacy of various university and federal efforts to improve postdoc career outcomes.
6.2. Implications for Theory
This paper finds that multiple factors may influence the transition of postdocs to non-
academic employment. Variation in non-academic career outcomes among postdocs range from
well-paid, in-field technical management positions to jobs that fail to utilize a postdoc’s scientific
talents. These factors should thus be considered for future research efforts that seek to build theory
within at least three contemporary perspectives. First, factors identified here may be
conceptualized as ex ante factors that subsequently impact educational mismatch, a situation
whereby an individual works in a job not closely related to their training and education, within the
context of scientists and engineers (Stenard and Sauermann, 2016). As discussed, the degree to

37
which individuals utilize acquired human capital is related to their capacity for personal and
professional development.
The two primary components of educational mismatch include over-education, whereby
an individual’s education is not utilized during their employment (Freeman, 1976; Halaby, 1994;
Hartog, 2000; Hartog and Oosterbeek, 1988), and deficiency of skills required for said
employment opportunity (Allen and Van der Velden, 2001; Wolbers, 2003). The perspective of
one employer that postdocs “…have all the academic science skills you don’t need, and none of
the organizational skills that you do”, presages the possibility that study findings may be viewed
as ex ante factors predictive of educational mismatch. Factors identified during this study might
not only be used to predict specific career trajectories, they might also be compared to conditional
differences, such as compensation and job satisfaction.
The study also has theoretical implications for the science and technology human capital
(STHC) model (Bozeman et al., 2001; Corley et al., in press), and related knowledge spillover
entrepreneurship theories (Ács et al., 2009; Braunerhjelm et al., 2010). Both perspectives view
individual scientists as the vehicle through which new knowledge is embodied, disseminated, and
utilized. For STHC, scientific career trajectories and productivity are a function of human capital
accrued through formal education and experience, social capital gained through external social
networks, and cultural dimensions associated with social-economic status, gender, nationality, and
academic discipline (Corley et al., in press).
The knowledge spillover perspective similarly views individual scientists as “knowledge
agents” whose ability to disseminate and commercialize new knowledge is mediated by several
factors, including individual, cultural, and network-related factors, as well as uncertainties
associated with the nature of knowledge itself (Audretsch et al., 2015). Similar to the education

38
mismatch literature that links negative personal and professional outcomes to the underutilization
of human capital (Bolus, 1995; Oyer, 2008), factors unearthed during the study might be used to
predict the degree to which knowledge-producing institutions sufficiently enable knowledge
dissemination and commercialization.
In sum, each of these three conceptual perspectives emphasize the importance of individual
scientists, their ability to generate new knowledge and accrue human capital, and factors that
enable or impede the utilization of that knowledge, in our case through postdoc employment in
non-academic jobs that utilize their significant scientific capabilities. Following from the study
findings, future research can address undertheorized elements in the literature, such as the
relationship between scientific culture and how knowledge is utilized (e.g. Corley et al., in press),
educational mismatch and entrepreneurial behavior (e.g. Stenard and Sauermann, 2016), and
among scientific discipline, employment sector (i.e. industry vs. government), and economic
impact (Audretsch et al., 2015).
6.3. Study limitations
In addition to its contributions, it is important to note this study’s limitations. The most
important is related to the purpose and limitations of qualitative research. The study relies upon
point-in-time data gathered from a small sample of postdocs, among other respondents, associated
with five Carnegie-classified Research I universities located within the United States. While the
study sought to maximize theoretical heterogeneity, as is the purpose of inductive, qualitative
research (Corbin and Strauss, 2008; Eisenhardt, 1989), it limits our ability to generalize about other
populations of postdocs.
More broadly, our approach likely neglects perspectives from other types of universities
not otherwise included in our study. Examples include such as universities that receive less R&D

39
funding, focus on specific populations (e.g. Historically Black Colleges and Universities), and
those within other countries. We also did not consider differences that may exist based on the
demographic or cultural identity of the postdoc (e.g. gender or race), which can impact experiences
and access to important resources in academic environments (e.g. Settles et al., 2006; Griffith,
2010).
Similarly, differences among specific disciplines were not examined, yet are likely to be
relevant. For example, one PI in the social sciences reported that while postdocs are relatively
uncommon in his field due to the relative scarcity of research funding, he had over the past 20
years placed all his postdocs into academic positions. Other PIs in the life and physical sciences
spoke of the importance of having teams of postdocs to run multi-million-dollar research projects,
though most of these postdocs will not find tenure-track employment.
The findings do not necessarily establish strict evidence of a causal relationship between
the factors documented within the study and some type of career outcome. They do suggest,
however, as is the purpose with inductive research, that these factors are likely to be important and
should be investigated in greater depth during subsequent research studies that can test and expand
upon these findings.
6.4. Conclusions and Policy Implications
While postdocs bear at least some responsibility for their career outcomes, it would be a
mistake, based on our research, to say that the burden lies completely on their shoulders. Postdocs
work in a system that not only takes advantage of them, often blurring ethical and legal boundaries,
they do so in some of the most highly regarded universities in the United States. University
administrators and faculty alike might ask themselves to what extent would greater understanding
of socially irresponsible behavior among PIs attract the attention—or ire—of policymakers and

40
the public. For example, policymakers are increasingly interested in improving the nation’s return
on its R&D investments (Mazzucato, 2015). Yet if the results of this study apply to the broader
population of postdocs, then universities vis-à-vis their PIs are not only expending enormous
public resources to “train” postdocs (and PhD students) for jobs that do not exist, postdocs are—
however infrequently—treated irresponsibly in the process, as well as the aforementioned
institutional failures that accompany such behavior.
To be sure, educational mismatch within science and engineering is macro-level problem
in the United States and abroad (e.g. Cyranoski et al., 2011). As discussed, universities in the study
possess little in the way of cross-institutional policies and programs that can support the non-
academic career outcomes of postdocs. Aside from the BEST program (see footnote 17), it is also
unclear to what extent the federal government prioritizes non-academic career outcomes among
university postdocs.21 However, in the near term, programs such as BEST and university-based
entrepreneurship programs may offer valuable lessons in the design and operation of programs
meant to help address these challenges. For example, while a systematic examination of the BEST
program was beyond the scope of this paper, initial inquiry uncovered programmatic examples
that provide postdocs with industry-focused management and communication training as well as
externship opportunities.
Universities and federal funding agencies might also work together to strengthen mentoring
requirements associated with academic research funding. While PIs are currently required to
submit postdoc mentoring plans, several PIs and administrators reported that the impact of these
21 The 2010 America COMPETES Act “[a]uthorizes the Director [of the NSF] to use funds appropriated to carry out
grants to IHEs for the provision of financial support for post-graduate research in fields of study with potential
commercial applications to match any private sector grant of financial assistance to any post-doctoral program in such
a field.” (Sec. 522). Future investigations might explore the extent to which the NSF designed and implemented
programs (or could) that take advantage of this authorizing language and, if programs do exist, their impact.

41
plans is unclear, if not ineffectual. Thus, clearly linking the efficacy of mentoring activities to
academic and non-academic career outcomes among postdocs may provide greater understanding
of the impact of these efforts and perhaps provide incentive to improve mentoring over time.
Further, a cursory examination of BEST programs found that some universities offer coaching and
advice for PIs that is designed to improve career-related postdoc mentoring. Resulting programs
would likely be useful not only to the general population of postdocs, they could also be structured
to address challenges among specific groups, such as older postdocs and foreign nationals.
Postdocs take advantage of entrepreneurship support programs to expand their knowledge
and skills of non-academic possibilities. Studies of academic entrepreneurship—the establishment
and growth of new companies based on academic research—show that universities often lack
effective networks and intermediaries necessary to commercialize their technologies (Hayter,
2016b; Hayter et al., 2018). Thus, postdoc scientific capabilities combined with their interest in
entrepreneurship support programs, along with the opportunity to provide postdocs with non-
academic career options, may offer universities and PIs an opportunity to re-conceptualize some
postdoc positions as organizational intermediaries focused on the commercialization of university
technologies.
Long term, university and government policymakers, in cooperation with non-academic
employers, might rethink both the postdoc position as well as PhD education. Specifically, efforts
to re-conceptualize the postdoc (and PhD) should assume that advanced scientific skills are not
only of value in academia, but that they can also serve other sectors (Sinche et al., 2017). While
this idea is not new (e.g. Abu-Yousif, Hett, Skoczenski, and Hasan, 2010), it is our hope that this
study, along with future research might highlight its urgency. The careers and wellbeing of

42
postdocs, who are the backbone of science, as well as the impact of current and future national
research investments in the United States depend on it.
Acknowledgements: This work was supported by the National Science Foundation under Grant
No. 1723769 and the Arizona State University (ASU) Center for Organization Research and
Design (CORD); the authors are grateful for their support. The authors would also like to thank
Dr. Barry Bozeman, Dr. Elizabeth Corley, Dr. Albert Link, Dr. Spiro Maroulis, Dr. Henry
Sauermann, Dr. Don Siegel, and Dr. Eric Welsh for their helpful comments, significantly
improving the quality of the paper. We also thank Dr. Mary Feeney and PhD students taking her
Spring 2018 PAF 620 seminar for their helpful comments on an early version of the paper. While,
due to IRB requirements, we cannot list the names of research coordinators from the study
universities, we are nonetheless grateful for their substantial efforts throughout the investigation,
including their comments and suggestions for improving the quality of this paper. The authors
would also like to thank Lauren Lynch, a former undergraduate research assistant at ASU, for her
tremendous contributions to this project. Finally, we thank the editor, Dr. Maryann Feldman, and
three anonymous reviewers for their constructive comments and support.

43
References
Abu-Yousif, A. O., Hett, E. C., Skoczenski, A. M., & Hasan, T. 2010. The ABC's of industry: a
postdoc program provides a sneak peek into industry careers. Nature Biotechnology,
28(6), 625-626.
Ács, Z.J., Braunerhjelm, P., Audretsch, D.B., Carlsson, B., 2009. The knowledge spillover
theory of entrepreneurship. Small Business Economics 32 (1), 15–30.
Åkerlind, G.S., 2005. Postdoctoral researchers: roles, functions and career prospects. Higher
Education Research & Development 24, 21-40.
Åkerlind, G.S., 2009. Postdoctoral research positions as preparation for an academic
career. International Journal for Researcher Development 1, 84-96.
Allen, J., Van der Velden, R., 2001. Educational mismatches versus skill mismatches: effects on
wages, job satisfaction, and on‐the‐job search. Oxford economic papers 53, 434-452.
Altbach, Philip G., Liz Reisberg, and Hans de Wit, eds., 2017. Responding to massification:
Differentiation in postsecondary education worldwide. Rotterdam, Netherlands: Sense
Publishers.
Amabile, T.M., 1996. Creativity in context: Update to the social psychology of creativity.
Westview Press, Boulder, Colorado.
Åstebro, T., Thompson, P., 2011. Entrepreneurs, Jacks of all trades or Hobos? Research
Policy 40, 637-649.
Audretsch, D.B., Lehmann, E.E., Hinger, J., 2015. From knowledge to innovation: the role of
knowledge spillover entrepreneurship. In: Antonelli, C., Link, A.N. (Eds.), Routledge
Handbook of the Economics of Knowledge. Routledge, New York.
Autor, D. H., 2001. Why do temporary help firms provide free general skills training? The
Quarterly Journal of Economics 116, 4, 1409-1448.
Barnacle, R., Mewburn, I., 2010. Learning networks and the journey of ‘becoming
doctor’. Studies in Higher Education 35, 433-444.
Bender, K.A., Heywood, J.S., 2009. Educational mismatch among Ph. Ds: Determinants and
consequences, in: Richard, B.F., Daniel, L.G. (Eds), Science and engineering careers in
the United States: An analysis of markets and employment. University of Chicago Press,
Chicago, Illinois, pp. 229-255.
Boden, D., Borrego, M., Newswander, L.K., 2011. Student socialization in interdisciplinary
doctoral education. Higher Education 62, 741-755.
Bowlus, A.J., 1995. Matching workers and jobs: Cyclical fluctuations in match quality. Journal
of Labor Economics 13, 335-350.
Bozeman, B., Dietz, J. S., & Gaughan, M. 2001. Scientific and technical human capital: An
alternative model for research evaluation. International Journal of Technology
Management, 22(7/8), 716–740.
Braunerhjelm, P., Ács, Z.J., Audretsch, D.B., Carlsson, B., 2010. The missing link: knowledge
diffusion and entrepreneurship in endogenous growth. Small Business Economics 34(2),
105–125.
Cantwell, B., 2011. Academic in-sourcing: international postdoctoral employment and new
modes of academic production. Journal of Higher Education Policy and Management
33(2), 101-114.
Cantwell, B., Lee, J., 2010. Unseen workers in the academic factory: Perceptions of neoracism
among international postdocs in the United States and the United Kingdom. Harvard
Educational Review 80, 490-517.

44
Cantwell, B., Taylor, B.J., 2013. Internationalization of the postdoctorate in the United States:
analyzing the demand for international postdoc labor. Higher Education 66, 551-567.
Chen, S., McAlpine, L., Amundsen, C., 2015. Postdoctoral positions as preparation for desired
careers: a narrative approach to understanding postdoctoral experience. Higher
Education Research & Development 34, 1083-1096.
Conti, A., Liu, C.C., 2015. Bringing the lab back in: Personnel composition and scientific output
at the MIT Department of Biology. Research Policy 44, 1633-1644.
Corbin, J., Strauss, A., 2008. Basics of qualitative research: Techniques and procedures for
developing grounded theory. SAGE Publications, Thousand Oaks, California.
Corley, E.A., Bozeman, B, Zhang, X., Chin-Chang, T. The expanded scientific and technical
human capital model: the addition of a cultural dimension. Journal of Technology
Transfer, in press, DOI 10.1007/s10961-017-9611-y.
Creswell, J.W. 1998. Qualitative inquiry and research design: Choosing among five traditions.
Thousand Oaks, CA: Sage.
Creswell, J.W. 2003. Qualitative inquiry and research design: Choosing among five traditions.
Thousand Oaks, CA: Sage.
Crewell, J.W., Miller, D.L. 2000. Determining validity in qualitative inquiry, Theory into
Practice, 39(3): 124-130.
Cyranoski, D., Gilbert, N., Ledford, H., Nayar, A., Yahia, M., 2011. Education: the PhD factory.
Nature 472, 276-279.
Davis, G., 2009. Improving the postdoctoral experience: An empirical approach, in: Richard,
B.F., Daniel, L.G. (Eds), Science and engineering careers in the United States: An
analysis of markets and employment. University of Chicago Press, Chicago, Illinois, pp.
99-127.
DeGrande, H., De Boyser, K., Vandevelde, K., Van Rossem, R., 2014. From academia to
industry: are doctorate holders ready? Journal of the Knowledge Economy 5, 538-561.
Eisenhardt, K.M., 1989. Building theories from case study research. Academy of Management
Review14, 532-550.
Felisberti, F.M., Sear, R., 2014. Postdoctoral researchers in the UK: a snapshot at factors
affecting their research output. PloS ONE 9, e93890.
Fitzenberger, B., Schulze, U., 2014. Up or out: Research incentives and career prospects of
postdocs in Germany. German Economic Review 15, 287-328.
Fox, M.F., Stephan, P.E., 2001. Careers of young scientists: Preferences, prospects and realities
by gender and field. Social studies of Science 31, 109-122.
Freeman, R., 1976. The overeducated American. Academic Press, Cambridge, Massachusetts.
Gardner, S. K. 2008. What's too much and what's too little?: The process of becoming an
independent researcher in doctoral education. The Journal of Higher Education, 79(3),
326-350.
Gardner, S.K., Jansujwicz, J.S., Hutchins, K., Cline, B., Levesque, V., 2014. Socialization to
interdisciplinarity: Faculty and student perspectives. Higher Education 67, 255-271.
Garrison, H.H., Justement, L.B., Gerbi, S.A., 2016. Biomedical science postdocs: an end to the
era of expansion. The FASEB Journal 30, 41-44.
Gaughan, M., Robin, S., 2004. National science training policy and early scientific careers in
France and the United States. Research Policy 33, 569-581.

45
Gibbs, K.D., Griffin, K.A., 2013. What do I want to be with my PhD? The roles of personal
values and structural dynamics in shaping the career interests of recent biomedical
science PhD graduates. CBE-Life Sciences Education 12, 711-723.
Gibbs, K.D., McGready, J., Bennett, J.C., Griffin, K., 2014. Biomedical science Ph. D. career
interest patterns by race/ethnicity and gender. PloS ONE 9, e114736.
Golde, C.M., 1998. Beginning graduate school: Explaining first year doctoral attrition. New
directions for higher education 1998, 55-64.
Golde, C.M., 2005. The role of the department and discipline in doctoral student attrition:
Lessons from four departments. The Journal of Higher Education 76, 669-700.
Granovetter, M., 1974. Getting a Job: A study of Contacts and Careers. Chicago, IL: University
of Chicago Press.
Granovetter, M. S., 1977. The strength of weak ties. American Journal of Sociology 78,6, 1360-
1380.
Granovetter, M., 1983. The strength of weak ties: A network theory revisited. Sociological
theory, 201-233.
Griffith, A. L. (2010). Persistence of women and minorities in STEM field majors: Is it the
school that matters? Economics of Education Review, 29(6), 911-922.
Groot, W., Van Den Brink, H.M., 2000. Overeducation in the labor market: a meta-
analysis. Economics of education review 19, 149-158.
Halaby, C.N., 1994. Overeducation and skill mismatch. Sociology of Education 67, 47-59.
Haley, K. J., Hudson, T. D., & Jaeger, A. J., 2018. Career Coherence, Agency, and the
Postdoctoral Scholar. In Jaegger, Al and Dinin, A. (Eds.), In: The Postdoc Landscape,
121-142. London, England: Academic Press.
Hancock, S., Walsh, E., 2016. Beyond knowledge and skills: rethinking the development of
professional identity during the STEM doctorate. Studies in Higher Education 41, 37-50.
Harman, K., 2002. The research training experiences of doctoral students linked to Australian
cooperative research centres. Higher Education 44, 469-492.
Hartog, J., 2000. Over-education and earnings: where are we, where should we go? Economics
of Education Review 19, 131-147.
Hartog, J., Oosterbeek, H., 1988. Education, allocation and earnings in the Netherlands:
Overschooling? Economics of Education Review 7, 185-194.
Hayter, C. S., 2016a. Constraining entrepreneurial development: A knowledge-based view of
social networks among academic entrepreneurs. Research Policy 45(2): 475–490.
Hayter, C. S., 2016b. A trajectory of early-stage spinoff success: The role of knowledge
intermediaries within an entrepreneurial university ecosystem. Small Business
Economics, 47(3): 633–656.
Hayter, C.S., 2016c. A social responsibility view of ‘formal’ university technology transfer:
Motivating knowledge dissemination? Duquesne Law Review, 54(1): 7-52.
Hayter, C. S., Lubynsky, R., Maroulis, S., 2017. Who is the academic entrepreneur? The role of
graduate students in the development of university spinoffs. The Journal of Technology
Transfer, 42: 1237–1254.
Hayter, C.S., Cahoy, D. 2018. Toward a strategic view of higher education social
responsibilities: A dynamic capabilities approach. Strategic Organization 16(1) 12–34.
Hayter, C.S., Nelson, A.J., Zayed, S., O’Connor, A. 2018. Conceptualizing academic
entrepreneurship ecosystems: a review, analysis and extension of the literature. Journal of
Technology Transfer 43: 1039–1082.

46
Helbing, C.C., Verhoef, M.J., Wellington, C.L., 1998. Finding identity and voice: A national
survey of Canadian postdoctoral fellows. Research Evaluation 7, 53-60.
Henkel, M., 2005. Academic identity and autonomy in a changing policy environment. Higher
education 49, 155-176.
Holley, K., 2009. The challenge of an interdisciplinary curriculum: A cultural analysis of a
doctoral-degree program in neuroscience. Higher Education 58, 241-255.
Horta, H., 2009. Holding a post-doctoral position before becoming a faculty member: does it
bring benefits for the scholarly enterprise? Higher Education 58, 689-721.
Jick, T.D., 1979. Mixing quantitative and qualitative methods: triangulation in
action. Administrative Science Quarterly 24, 602–611.
Judge, T.A., Thoresen, C.J., Bono, J.E., Patton, G.K., 2001. The job satisfaction–job
performance relationship: A qualitative and quantitative review, Psychological Bulletin
127, 376-407.
Kaslow, N.J., Mascaro, N.A., 2007. Mentoring interns and postdoctoral residents in academic
health sciences center. Journal of Clinical Psychology in Medical Settings 14, 191-196.
Kuckartz, U., 2014. Qualitative text analysis: A guide to methods, practice and using software.
SAGE Publications, Thousand Oaks, California.
Kuhn, P., & Skuterud, M, 2000. Job search methods: Internet versus traditional. Monthly Lab.
Rev. 123, 3.
Kuo, M., 2016. NIH sets new postdoc stipend levels. Science, August 9,
http://www.sciencemag.org/careers/2016/08/nih-sets-new-postdoc-stipend-levels
(accessed November 29, 2017)
Lee, H.F., Miozzo, M., Laredo, P., 2010. Career patterns and competences of PhDs in science
and engineering in the knowledge economy: The case of graduates from a UK research-
based university. Research Policy 39, 869-881.
Leshner, A.I., 2012. Standards for postdoc training. Science 336, 276.
Lin, E.S., Chiu, S.Y., 2016. Does holding a postdoctoral position bring benefits for advancing to
academia? Research in Higher Education 57, 335-362.
Lodico, M. G., Spaulding, D. T., & Voegtle, K. H. (2010). Methods in educational research:
From theory to practice. San Francisco: John Wiley & Sons.
Mangematin, V., 2000. PhD job market: professional trajectories and incentives during the
PhD. Research Policy 29, 741-756.
Mason, M. 2010. Sample size and saturation in PhD studies using qualitative interviews, Forum:
Qualitative Social Research, 11(3), 1-18.
Maxwell, J.A. 1992. Understanding and validity in qualitative research, Harvard Educational
Review, 62(3), 279-300.
Mazzucato, M. 2015. The entrepreneurial state: Debunking public vs. private sector myths (Vol.
1). Anthem Press.
McAlpine, L., 2018. Postdoc Trajectories: Making Visible the Invisible. In: Jaegger, Al and
Dinin, A. (Eds.), The Postdoc Landscape, 175-202. London, England: Academic Press.
McAlpine, L., & Amundsen, C., 2016. Post-PhD career trajectories: Intentions, decision-making
and life aspirations. London, England: Palmgrave Macmillan.
Melin, G., 2004. Postdoc abroad: inherited scientific contacts or establishment of new
networks? Research Evaluation 13, 95-102.
Miller, J.M., 2012. Postdoctoral appointments: motivations, markets, and experiences (Doctoral
dissertation). University of North Carolina, Chapel Hill, North Carolina.

47
Miller, J.M., Feldman, M.P., 2015. Isolated in the Lab: examining dissatisfaction with
postdoctoral appointments. The Journal of Higher Education 86, 697-724.
Morse, J.M. 1994. Designing funded qualitative research. In Densin, N.K. and Lincoln, Y.S.
(Eds.), Handbook of qualitative research (2nd ed., pp.220-35), Thousand Oaks, CA:
Sage.
National Research Council, 2000. Enhancing the postdoctoral experience for scientists and
engineers: A guide for postdoctoral scholars, advisers, institutions, funding
organizations, and disciplinary societies. National Academies Press, Washington, D.C.
National Research Council, 2014. The postdoctoral experience revisited. National Academy
Press, Washington, D.C.
National Science Board, 2018. Science and Engineering Indicators.
https://www.nsf.gov/statistics/2018/nsb20181/report/sections/science-and-engineering-
labor-force/s-e-labor-market-conditions (Accessed September 23, 2018).
Nerad, M., & Cerny, J. (1999). Postdoctoral patterns, career advancement, and problems.
Science, 285(5433), 1533-1535.
Nolan, S.A., Buckner, J.P., Kuck, V.J., Marzabadi, C.H., 2004. Analysis by gender of the
doctoral and postdoctoral institutions of faculty members at the top-fifty ranked
chemistry departments. Journal of Chemical Education 81, 356.
O’Gorman, C., Byrne, O., Pandya, D., 2008. How scientists commercialise new knowledge via
entrepreneurship. The Journal of Technology Transfer, 33(1): 23–43.
Ohyama, A., 2015. Entrepreneurship and job relatedness of human capital. Economica 82, 740-
768.
Organisation for Economic Co-operation and Development (OECD), 2018. Research and
Development Statistics.
http://www.oecd.org/innovation/inno/researchanddevelopmentstatisticsrds.htm
Oyer, P., 2008. The making of an investment banker: Stock market shocks, career choice, and
lifetime income. The Journal of Finance 63(6), 2601-2628.
Patton, M.Q. Qualitative research and evaluation methods, Sage, Thousand Oaks, CA.
Polkinghorne, D.E., 1988. Narrative knowing and the human sciences. SUNY Press, Albany,
NY.
Preston, A.E., 1994. Why have all the women gone? A study of exit of women from the science
and engineering professions. The American Economic Review 84, 1446-1462.
Puljak, L., Sharif, W.D., 2009. Postdocs' perceptions of work environment and career prospects
at a US academic institution. Research Evaluation 18, 411-415.
Recotillet, I., 2007. PhD Graduates with postdoctoral qualification in the private sector: Does it
pay off? Labour 21, 473-502.
Roach, M., Sauermann, H., 2010. A taste for science? PhD scientists’ academic orientation and
self-selection into research careers in industry. Research Policy 39, 422-434.
Roach, M., Sauermann, H., 2017. The declining interest in an academic career. PloS ONE 12(9),
e0184130.
Saldaña, J., 2015. The coding manual for qualitative researchers. SAGE Publications, Thousand
Oaks, California.
Salminen-Karlsson, M., Wallgren, L., 2008. The interaction of academic and industrial
supervisors in graduate education. Higher Education 56, 77-93.
Sauermann, H., Cohen, W.M., 2010. What makes them tick? Employee motives and firm
innovation. Management Science 56, 2134-2153.

48
Sauermann, H., Roach, M., 2012. Science PhD career preferences: levels, changes, and advisor
encouragement. PloS ONE 7, e36307.
Sauermann, H., Roach, M., 2016. Why pursue the postdoc path? Science 352, 663-664.
Scaffidi, A.K., Berman, J.E., 2011. A positive postdoctoral experience is related to quality
supervision and career mentoring, collaborations, networking and a nurturing research
environment. Higher Education 62, 685-698.
Settles, I. H., Cortina, L. M., Malley, J., & Stewart, A. J. (2006). The climate for women in
academic science: The good, the bad, and the changeable. Psychology of Women
Quarterly, 30(1), 47-58.
Sinche, M., Layton, R.L., Brandt, P.D., O’Connell, A.B., Hall, J.D., Freeman, A.M., Harrell,
J.R., Gowen Cook, J., Brennwald, P.J. 2017. An evidence-based evaluation of
transferrable skills and job satisfaction for science PhDs. PLoS ONE 12(9): e0185023.
Singer, M., 2004. The evolution of postdocs. Science 306, 232-232.
Smaglik, P., 2014. Employment: PhD overdrive. Nature 511, 255-256.
Stenard, B.S., Sauermann, H., 2016. Educational mismatch, work outcomes, and entry into
entrepreneurship. Organization Science 27, 801-824.
Stephan, P.E., 2012. How economics shapes science (Vol. 1). Harvard University Press,
Cambridge, Massachusetts.
Stephan, P.E., Sumell, A.J., Black, G.C., Adams, J.D., 2004. Doctoral education and economic
development: The flow of new PhDs to industry. Economic Development Quarterly 18,
151-167.
Stern, S., 2004. Do scientists pay to be scientists? Management Science, 50, 835-853.
Strauss, A., Corbin, J. 1998. Basics of qualitative research: Techniques and procedures for
developing grounded theory, Thousand Oaks, CA: Sage.
Su, X., 2013. The impacts of postdoctoral training on scientists' academic employment. The
Journal of Higher Education 84, 239-265.
Trochim, W., Donnelly, J.P., Arora, K. Research methods: the essential knowledge base (2nd
ed.). Cenage, Boston, MA.
Tsang, M.C., Levin, H.M., 1985. The economics of overeducation. Economics of Education
Review 4, 93-104.
Van Balen, B., Van Arensbergen, P., Van Der Weijden, I., Van Den Besselaar, P., 2012.
Determinants of success in academic careers. Higher Education Policy 25, 313-334.
Wei, T. E., Levin, V., & Sabik, L. M., 2012. A referral is worth a thousand ads: Job search
methods and scientist outcomes in the market for postdoctoral scholars. Science and
Public Policy, 39(1), 60-73.
Wolbers, M.H., 2003. Job mismatches and their labour-market effects among school-leavers in
Europe. European Sociological Review 19, 249-266.

49
Figure 1: Number of University Postdocs Employed by Universities in the United States (1973-2013)
SOURCE: National Science Foundation, National Center for Science and Engineering Statistics, Science and Engineering Indicators 2016, special tabulations (2014) of the Survey
of Doctorate Recipients.

50
Table 1: Basic characteristics of universities participating in the study (2017)
Institution
Ownership
Research Funding
Number of Postdocs
A
Private
$750m - $1b
>1,500
B
Private
<$50m
500-750
C
Public
$750m - $1b
>1,500
D
Public
$500m - $750m
500-750
E
Public
$250m - $500m
<250

51
Figure 2: Distribution of Interview Respondents by Type (n = 132)
9
10
16
97
020 40 60 80 100
FACULTY
UNIVERSITY ADMINISTRATORS
INDUSTRY EMPLOYERS
UNIVERSITY POSTDOCS

52
Figure 3: Distribution of University Postdoc Respondents by Discipline (n = 97)
3
5
9
29
51
010 20 30 40 50
STEM EDUCATION
SOCIAL SCIENCE AND
HUMANITIES
PHYSICAL & EARTH SCIENCES
ENGINEERING
LIFE SCIENCES

53
Figure 4: Distribution of University Postdoc Respondents by Country of Origin (n = 97)
6
6
25
25
35
0 5 10 15 20 25 30 35
MEXICO & SOUTH AMERICA
MIDDLE EAST
ASIA
EUROPE
U.S.A

54
Figure 5: Factors Relating to Postdoc Transition to Non-academic Employment

55
Table 2: Original and Current Career Motivations for Taking the Postdoc Position
RATIONALE
Original
Current
To obtain tenure-track academic research position
86.6%
54.6%
Preparation for specific non-academic career
7.2%
7.2%
Holding pattern; will not pursue academic job
6.2%
38.1%

56
Table 3. Quotes Relating to Individual Factors
**quotes from administrators, PIs, or industry employers
Career Expectations
“Being a scientist is so much part of who I am and who I’ve been. I feel that in some way I have to disconnect with that part of
myself in order to explore [the world] outside of academics…”
“The culture here looks down on jobs in industry…and there is this weird idea that you don’t need to think about other options.
Like, you are at [a prestigious research university], why would you even consider anything else?”
“I’ve always mastered everything I have done, so the notion of not being able to find a job seemed dumb…with the [Ivy-League]
degree, the publications, the world-famous PI, and the postdoc at [their university]. I guess I was the dumb one…”
Information Asymmetry
“I wish I had a better idea of what I was getting myself into. You are told that all you need is a PhD from [a prestigious
university]…when I didn’t get any bites on the job market I was told all I need is a postdoc from [another prestigious university].
Turns out most of us don’t get jobs…isn’t there someone out there who thinks that this would have been good for us to
understand?
“The problem is that we don’t have a network outside of [their university]. We are tied to a physical location 50-60 hours a
week, working weird hours…it’s hard to get off campus. No wonder we don’t know what options are out there.”
“The worst part [is]…we’re not taught anything or given options outside of academics in grad school.”
**“The first thing we should tell people when they apply for a PhD is that you are not going to get a university job…we should
tell them again when they consider a postdoc. But there is no incentive to do that...I’ve heard people say just let them figure it
out when they get here…that just seems negligent to me.”
Skills
“Now that I know I have to leave, I am having a tough time understanding what success looks like…what do I do? Who do I talk
to? How much can I make?”
“Honestly, I am not sure I know how to do anything other than managing this lab and running experiments…if I can find a job
doing that, then I am golden. Otherwise, looks like I will be working at Wendy’s.”

57
** “PhDs and Postdocs really don’t have a grasp of what a real-world job is like…this is not a lab where you roll in at 11:00 am.
You need to be a grown-up here…show up each day, on time, attend meetings, and work in a team…I think it’s a bit of a culture
shock for these guys.”
** “It’s toughest for postdocs to understand that they are no longer trying to build their research portfolio but rather work in a
team toward a common goal. These are smart individuals but you have to get them to play well with others…that’s not how they
are trained.”
Personal Crisis
“I have found coming to term with the idea of leaving academics has been incredibly difficult.”
“When it first hit me that academia was no longer an option, it was gut-wrenching…either I couldn’t sleep or I couldn’t get out
of bed…I would go into [the lab] and just stare my notes or the equipment…I’m better now…well, a little…but I still can’t
focus. I don’t know what to do.”
“Realizing that you did all this work for nothing is existential…I have been in crisis mode…trying to figure things out. I am
getting out of here…I may be working at Starbucks next year…but at this point I just need to get out.”
“Imagine spending your entire life training for something that you then realize is not going to happen. I am 42 years old, have a
wife and two kids… What do I do? I know how to be an academic and that’s it. How did I get here?”
** “These are the things I stay up at night worry about. There is already a lot of pressure on these guys and then add the job
rejection on top. Some postdocs just roll with it [not obtaining an academic job] while others kind of melt down…I am most
worried about the bouts of anxiety and depression…when they don’t know what to do or where to go.”
“[I’m] currently feeling hopeless…I don’t feel that there is a way out of my postdoc.”

58
Table 4. Quotes Relating to PI Factors
**quotes from administrators, PIs, or industry employers
Lack of Career Support
“I mentor doctoral students because my PI doesn’t…I learned that your doomed if you think your PI is going to provide career
guidance, much less anything related to outside [non-academic] jobs.”
“You get contradictory advice from different people here…your professor is your boss and will go to bat for you if it’s an
academic job…but then they disappear if you want to do something else…”
“My PI wouldn’t allow me to attend [a university workshop exploring non-academic careers] and told me to stop
going…because I think he understood they were trying to help me understand my options…look after my interests.”
“A lot of PIs just love to squash any interest you have other than being chained to the bench…and pursuing an academic path
that may never happen.”
“My professor reluctantly let me attend [a university career development event] but told me that the workload in the lab would
not change…I just had to work later at night and on the weekends.”
“With my PI, there is this penalty for thinking about or going into industry…like, you are dead to me.”
“If I need to go to a [non-academic career] workshop, I just lie…it’s just easier that way.”
** “There’s a culture in academia that looks down on non-academic jobs. Faculty fear that you might leave the lab…so they
[postdocs] will suffer if they look at other options.”
Skills and Capabilities
“Faculty do not have proper training to mentor the postdocs [for non-academic careers]. They lack management skills. Even
though PIs have fantastic scientific minds, they lack soft skills.”
“My PI is supportive but he thinks he knows everything. But the reality is that he doesn’t have a clue about how to help me [look
for non-academic employment].
“We’ve come to this understanding where I will learn how to start a company and they will come along for the ride. Of course
they know nothing about it…I can’t find a[n academic] job so maybe this will turn into something.”

59
Socially Irresponsible Behavior
“When I arrived at [the university] my PI explained to me that he approved my visa renewal…he then told me he was going to
pay me 70 percent of the salary he promised before I got here…when I asked him if this is normal, he just asked me if I was
serious about working [at the university].”
“[I] realized that students can really be taken advantage of and this left a bad taste in [my] mouth with academia and decided that
working for a large company and/or starting [my] own is the way to go.”
“Our PI creates this pressure cooker environment in our lab…you see the foreign postdocs sleeping on the floor of the labs,
working 100-plus hours a week…and then you see them crying because they can’t get a job and have to return back to [their
home country]. PIs know what they are doing…they take advantage of these guys.”
“PIs think they own you and everything you do…after we had an invention, we filled out the disclosure form…I found out that
the PI had ‘mistakenly’ left my name off the forms. I confronted my PI and he asked me if I wanted to work in academia. I didn’t
say anything after that.”
“If you come up with anything that can be commercialized, they take it and claim that it’s theirs. I’ve seen this with a couple of
colleagues and learned to hide what I am doing. They don’t want you thinking about a startup or commercialization…just the
science.”
** “I see something bad almost every week and it seems to be getting worse…postdocs come into my office and ask me if this or
that seems wrong to me…the visa issue is a big one because foreign postdocs are afraid to report their PIs…these are small
scientific communities and PIs will blackball their postdocs if you cross them. I just try to help where I can.”
** “Faculty here seem to be getting greedier as they understand the [monetary] value of their inventions…this unfortunately
translates into claiming anything produced in their lab. PhDs do ok because IP ownership is spelled out…it’s the postdocs who
get taken advantage of.”