Available online at www.sciencedirect.com

Nur s Ou t l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3www.nursingoutlook.org

Attrition of full-time faculty from schools of nursing withbaccalaureate and graduate programs, 2010 to 2011

Di Fang, PhD*, Geraldine D. Bednash, PhD, RN, FAANAmerican Association of Colleges of Nursing, Washington, DC

a r t i c l e i n f o

Article history:Received 14 February 2013Accepted 3 December 2013Available Online 22 February2014

Keywords:Faculty attritionFaculty retentionFaculty retirementFaculty shortageMinority facultyMale faculty

* Corresponding author: Dr. Di Fang, AmeriDC 20036.

E-mail address: dfang@aacn.nche.edu (D.

0029-6554/$ - see front matter � 2014 Elsevihttp://dx.doi.org/10.1016/j.outlook.2013.12.00

a b s t r a c t

The shortage of qualified faculty has been consistently reported as a majorbarrier impeding acceptance of all qualified applicants into nursing programs. Inaddition to faculty recruitment, the attrition of faculty is also a concern forschools of nursing. In this study, we found that nationally 11.8% of full-timefaculty who worked in 2010 left their full-time jobs by 2011. Nearly half oftotal attrition, or 5.7% of full-time faculty members, were related to leaving fornonacademic nursing positions, whereas another 20% of attrition, or 2.4% offull-time faculty, resulted from retirement. Nearly 20% of faculty egressions, or2.2% of full-time faculty, was due to leaving for nursing administrative positionsor full-time faculty positions in an academic setting. Leaving for part-time fac-ulty positions made up slightly more than 10% of faculty attrition or 1.3% of full-time faculty. Our bivariate analysis identifies distinctive academic anddemographic profiles of faculty who left full-time positions for different reasons,and our multivariate analysis further shows that different individual andinstitutional attributes are significantly associated with different types ofattrition.

Cite this article: Fang, D., & Bednash, G. D. (2014, JUNE). Attrition of full-time faculty from schools of

nursing with baccalaureate and graduate programs, 2010 to 2011. Nursing Outlook, 62(3), 164-173. http://

dx.doi.org/10.1016/j.outlook.2013.12.002.

The faculty shortage is consistently reported as aleading barrier to enrollment growth in nursing pro-grams. For example, for more than a decade, theAmerican Association of Colleges of Nursing (AACN)has reported an insufficient number of faculty mem-bers as the major reason for turning away qualifiedapplicants to baccalaureate and graduate nursingprograms. Many previous studies on the facultyshortage issue focused on difficulties in facultyrecruitment, such as an insufficient pool of nursingdoctoral graduates and a lack of resources for nursingschools to compete with the practice sector for quali-fied faculty candidates (American Association ofColleges of Nursing, 2005; Anderson, 2000; Berlin &

can Association of Colleg

Fang).

er Inc. All rights reserved2

Sechrist, 2002; Brendtro & Hegge, 2000; Yordy, 2006).Additionally, studies existed that reviewed differentaspects of faculty retention or attrition, such asretirement, and identified measures that couldenhance retention (Falk, 2007; Horat, 2008; Williamson,Cook, Salmeron, & Burton, 2010; Foxall, Megel, Grigsby,& Billings, 2009; Berent & Anderko, 2011; Cash, Daines,Doyle, & Tettenborn, 2009; Cash, Doyle, Tettenborn, &Faria, 2011).

In comparison, studies quantifying types offaculty attrition and examining individual and insti-tutional characteristics associated with faculty attri-tion are limited. An early study estimated the ratioof faculty retirement to resignation was three to one

es of Nursing, One Dupont Circle, NW, Suite 530, Washington,

.

Nur s Ou t l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3 165

(Ryan & Irvine, 1994). Another study reported themeannumber of full-time faculty and faculty departures in2006 was 20.1 and 1.7 per institution, respectively,which would yield an annual attrition rate of 8.5%. Ofthe faculty who departed, 26% left because of retire-ment, whereas the others left for other reasons,including career change, family obligations, salary is-sues, and workload (Kovner, Fairchild, & Jacobson,2006). However, the findings, based on data from 256nursing schools, may not reflect the experience of allschools with baccalaureate and graduate nursing pro-grams. Based on the AACN’s annual survey, 722institutions offered baccalaureate and graduate degreeprograms in nursing in 2006 (Fang, Wisniewski,& Bednash, 2007).

An accurate assessment of the magnitude andcharacteristics of faculty egressions would help thenursing education community and policy makers bet-ter estimate the size of the faculty pool needed, un-derstand the demographic and academic attributes offaculty who leave their positions, and, consequently,take appropriate measures to reduce faculty attritionthat is costly for nursing schools and limits enrollmentcapacity. Accordingly, we conducted this study toanalyze attrition of full-time faculty from nursingschools offering baccalaureate and/or graduate pro-grams at the national level between 2010 and 2011 toaddress the following research questions:

1. What are the overall attrition rate and attrition ratesby reasons for leaving?

Are different individual and institutional charac-teristics associated with different types of facultyattrition? Specifically for nonretirement attrition:

2. Are faculty without a doctoral degree more likely toleave than doctorally prepared faculty?

3. Are faculty with junior rank or clinical rank morelikely to leave than faculty with senior rank?

4. Are faculty who are not tenured more likely to leavethan faculty with tenure?

5. Are nonwhite minority faculty more likely to leavethan white majority faculty?

Table 1 e Attrition Rates of Full-Time Faculty, 2010 to 2

Number of SchoolsTotal Active Full-time Faculty in 2010

Faculty Left Full-time Positionsbetween 2010 and 2011

1. Left for retirement2. Left for nonacademic nursing (all other reasons)3. Left for administrative positions at schools of nursing4. Left for full-time faculty positions at schools of nursing5. Left for part-time faculty at schools of nursing6. Left because of illness or deathTotal attrition 1

Data Source: 2010 and 2011 AACN’s Annual Survey of Baccalaur

6. Are male faculty more likely to leave than femalefaculty?

7. Are faculty at small colleges more likely to leavethan faculty at large universities?

For retirement:

8. What is the mean age of full-time faculty atretirement?

Methods

In this study, faculty attrition is defined as faculty wholeft full-time positions between 2010 and 2011 fromschools of nursing where they were employed in 2010.The faculty data are obtained from the 2010 and 2011AACN’s Annual Survey of Baccalaureate and GraduatePrograms in Nursing. The survey collects data on sal-aries as well as demographic and academic character-istics of full-time faculty at the individual level. Thesurvey also asks each participating school to identifyfaculty members who left full-time positions based onfaculty records submitted by the school to the survey inthe previous year and to report reasons for the de-partures and subsequent activities after the departures.

Respondents from 665 schools submitted full-timefaculty data to the AACN annual survey in both 2010and 2011, which represent 83% of the total 801 schoolsoffering baccalaureate and graduate nursing programsin 2010. Respondents reported a total of 15,365 full-time faculty members in 2010, and of them, 1,806were identified in the 2011 survey as having left full-time positions. Accordingly, the attrition rate of full-time faculty from the 665 schools was 11.8% between2010 and 2011 (Table 1).

In this study, we consider faculty attrition to be aloss to the nursing faculty workforce in general.Accordingly, if a faculty member leaves her/his ownschool of nursing and takes a full-time faculty positionat a different school of nursing, we do not consider it aloss to the faculty workforce as awhole but instead as aloss to the school of nursing in which the faculty

011

66515,365

% of Total FacultyAttrition

Attrition Rate (% of TotalFull-time Faculty)

362 20.0 2.4870 48.2 5.782 4.5 0.5

260 14.4 1.7203 11.2 1.329 1.6 0.2

,806 11.8

eate and Graduate Programs in Nursing.

Nur s Out l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3166

member was previously employed. Given this consid-eration, we classified full-time faculty attrition intodifferent categories largely based on whether thedeparted individuals remain in academic nursing.These categories include (a) faculty who left to take anadministrative position at own school or a differentschool of nursing, (b) faculty who left to take a full-timefaculty position at a different school of nursing, (c)faculty who left to take a part-time faculty position atown school or a different school of nursing, (d) facultywho left for retirement, (e) faculty who left for all otherreasons, and (f) faculty who left because of illness ordeath. Categories 1 and 2 include those who remain inacademic nursing fully. Category 3 includes those whoremain in academic nursing partially. Because facultymembers who retired have distinctively differentcharacteristics from faculty who left for nonretirementreasons and because their departures insert a signifi-cant impact on the nursing faculty workforce, wegrouped all retired faculty into category 4. Facultymembers who left for all other reasons were groupedinto category 5 and are considered in this study asleaving academic nursing completely because theywere not reported by their schools as working for ac-ademic nursing after their attrition. Category 6 in-cludes faculty who left because of illness or death,which occurs randomly and cannot be mitigated.

Bivariate analyses using the chi-square test wereperformed to examine individual and institutionalcharacteristics comparing faculty who left with facultywho stayed. Multinomial logistic regression analysesalso were conducted to examine the association be-tween a characteristic and a type of faculty attritionwhile controlling for other variables. All analyses wereperformed using SAS 9.3 (SAS Institute, Cary, NC).

Characteristics of Faculty Who Retired

The profile of full-time faculty in 2010 who stayed orleft for different reasons by 2011 can be found inTable 2. Of the 1,806 faculty who left, 363 (20%) left theirteaching roles for retirement. In general, faculty whoretired are older than faculty who stayed. (The meanage of faculty who retired was 64 years in 2010,whereas the figure for faculty who stayed was52.6 years). Because young faculty members have notreached the eligible age for retirement, it is appropriateto compare faculty of similar ages who either retired orstayed in faculty positions. We found that 98% of fac-ulty who retired were 55 years old or older in 2010;therefore, we compared them with faculty in the sameage group who stayed (Table 3). In this comparison,faculty who retired still showed a much higher con-centration in the age group 65 or older than facultywhostayed (46.7% vs. 13.7%). In addition, faculty memberswho retired were more likely to be full professors(35.4% vs. 19.5%) and hold tenured positions (53.8% vs.34%). On the other hand, they were less likely to have a

nondoctoral degree (32.6% vs. 40.5%) and, interestingly,come from noneAcademic Health Center (AHC) privateinstitutions (17.3% vs. 25.3%).

The mean age of individuals who retired was 64 in2010 (Table 2), and their retirements were reported tothe 2011 AACN annual survey. It is likely that somefaculty retired early, whereas others retired late duringthe 1-year interval between 2010 and 2011. Accord-ingly, we assumed that on average they spent a half ofa year at their work before their retirement; therefore,their mean age at retirement would be 64.5 years.

Characteristics of Faculty Who Left AcademicNursing

Survey respondents reported another 870 faculty wholeft for nonretirement reasons and did not remain inacademic nursing. These individuals were groupedinto one category and accounted for nearly one half ofthe total faculty who left. Of them, nearly 30% werereported by their schools as leaving to take nursingservice positions, whereas another 15% were reportedas leaving for personal or other reasons. In addition,15% left because of position termination or dismissal.For the remaining 40%, they were reported by theirschools as leaving for unknown reasons. Because these870 individuals were reported as either taking nursingservice positions (30%) or not taking academic nursingemployment (70%) after their departures, we assumethe majority of them took nonacademic nursing posi-tions. Their egressions are not only a loss to their ownschools of nursing but also a loss to academic nursingas a whole.

Compared with faculty who stayed, the 870 in-dividuals were more likely to be instructors (30.7% vs.19.8%), hold clinical/other rank (23.3% vs. 14%), havenontenure track positions (54.7% vs. 43.1%), and havepositions without a tenure system (19.9% vs. 13.7%).They were also more likely to have a nondoctoral de-gree (70.8% vs. 52.5%), be nonwhite minorities (18.2%vs. 12.2%), and bemale (9.2% vs. 5.8%). In addition, theywere more likely to come from small schools (11.4% vs.8.8%) or non-AHC private schools (30% vs. 25.7%). Witha mean age of 49.8 years, individuals in this categorywere the second youngest among all attrition groups.

Characteristics of Departed Faculty WhoRemained in Academic Nursing

Of the 1,806 faculty who left, 82 (4.5%) took adminis-trative positions at schools of nursing. Of the 82individuals, 26 were promoted to administrative posi-tions within their own schools of nursing. Althoughindividuals in this group left faculty positions, theyremained in academic nursing full time. As expected,before their departures, these faculty members were

Table 2 e Characteristics of Full-Time Faculty in 2010

Characteristics FacultyStayed

Faculty Left Full-time Positions between 2010 and 2011

Left forRetirement

Left for NonacademicNursing (all Other

Reasons)

Left for AdministrativePositions in Schools

of Nursing

Left for Full-timeFaculty Positions inSchools of Nursing

Left for Part-TimeFaculty Positions inSchools of Nursing

N 13,559 362 870 Chi-squareValue

pValue

82 Chi-squareValue

pValue

260 Chi-SquareValue

pValue

203 Chi-SquareValue

pValue

Professor (%) 11 2.2 70.4 <.0001 18.3 4.0 .0447 5.8 7.7 .0054 4.9 8.1 .0045Associate professor (%) 19.2 10.1 44.5 <.0001 39.0 20.5 <.0001 15.0 13.3 4.5 .0337Assistant professor (%) 35.8 33.7 24.4 4.6 .0318 46.9 13.7 .0002 30.5Instructor (%) 19.8 30.7 59.6 <.0001 8.5 6.5 .0106 22.3 32.0 18.7 <.0001Clinical & other rank (%) 14.0 23.3 57.8 <.0001 9.8 10.0 19.2 4.6 .0326Tenured (%) 21.9 3.9 160.9 <.0001 24.4 9.2 24.2 <.0001 7.9 23.2 <.0001Tenure track (%) 21.2 21.5 28.1 32.7 19.9 <.0001 9.9 15.5 <.0001Nontenure track (%) 43.1 54.7 44.4 <.0001 26.8 8.9 .0029 44.6 61.6 27.6 <.0001No tenure system (%) 13.7 19.9 25.8 <.0001 20.7 13.5 20.7 8.2 .0042Nondoctoral degree (%) 52.5 70.8 110.4 <.0001 30.5 15.8 <.0001 53.5 61.6 6.6 .0100Nonwhite (%) 12.2 18.2 26.4 <.0001 13.4 21.2 18.8 <.0001 14.3Male (%) 5.8 9.2 16.9 <.0001 9.8 10.8 11.5 .0007 11.3 11.2 .0008Mean age (year) 52.6 64.0 49.8 53.8 49.1 52.2Small school (%) 8.8 11.4 6.5 .0105 12.2 14.2 9.2 .0024 6.4Non-AHC public school(%)

32.5 28.5 6.1 .0136 29.3 27.3 27.1

Non-AHC private schoolAHC (%)

25.7 30.0 8.1 .0045 31.7 25.8 20.7

AHC school (%) 33.0 30.1 26.8 32.7 45.8 14.8 .0001

Notes: Chi-squire values and p values are reported at p < .05 level.Data Source: 2010 and 2011 AACN’s Annual Survey of Baccalaureate and Graduate Programs in Nursing.

Nurs

Outlook

62

(2014)164e173

167

Table 3 e Characteristics of Full-time Faculty Ages 55 or Older in 2010

Characteristics Faculty Stayed Faculty Left for Retirement

N 6,577 353 Chi- square Value p Value

Professor (%) 19.5 35.4 52.9 <.0001Associate professor (%) 26.8 25.8Assistant professor (%) 29.3 21.0 11.3 .0008Instructor (%) 12.1 6.8 9.0 .0027Clinical & other rank (%) 12.4 11.1Tenured (%) 34.0 53.8 57.6 <.0001Tenure track (%) 17.2 7.9 20.5 <.0001Nontenure track (%) 36.1 32.3No tenure system (%) 12.7 6.0 14.1 .0002Nondoctoral degree (%) 40.5 32.6 8.7 .0032Nonwhite (%) 9.1 11.0Male (%) 4.7 3.4Age 65/older (%) 13.7 46.7 282.0 <.0001Mean age (year) 60.1 64.3Small school (%) 7.6 7.7Non-AHC public school (%) 32.7 37.4Non-AHC private school AHC (%) 25.3 17.3 11.5 .0007AHC school (%) 34.4 37.7

Notes: Chi-squire values and p values are reported at p < .05 level.Data Source: 2010 and 2011 AACN’s Annual Survey of Baccalaureate and Graduate Programs in Nursing.

Nur s Out l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3168

more likely to have senior ranks as professors orassociate professors than full-time faculty who stayed.For example, 39% of the faculty members in this groupwere associate professors, whereas the figure was19.2% for faculty who stayed. Additionally, they wereless likely to be on nontenure track (26.8% vs. 43.1%)and have a nondoctoral degree (30.5% vs. 52.5%). Onaverage, individuals in this group also were older (witha mean age of 53.8 years) than any other attritiongroups, except for those who retired.

Another 260 individuals (14.4% of total attrition) lefttheir schools to take full-time faculty positions atdifferent schools of nursing. This group also includes26 who left to pursue further education. Although theirdepartures were a loss to their own schools of nursing,these individuals remained in academic nursing full-time. Individuals in this group were more likely thanfaculty who stayed to be assistant professors (46.9% vs.35.8%), be on tenure track (32.7% vs. 21.2%), benonwhite minorities (21.2% vs. 12.2%), be male (10.8%vs. 5.8%), and come from small schools defined asbaccalaureate or small master’s institutions based onthe 2010 Carnegie Classification of Institutions (14.2%vs. 8.8%). Individuals in this group were the youngeston average (with a mean age of 49.1 years) of all attri-tion groups.

Of the 1,862 faculty who left full-time positions, 203individuals (11.2%) changed to part-time faculty posi-tions at the same school or a different school ofnursing. These individuals also remained in academicnursing, at least partially. Compared with faculty whostayed, they were more likely to be instructors (32% vs.19.8%), be on clinical/other rank (19.2% vs. 14%), havenontenure track positions (61.6% vs. 43.1%), and havepositions without a tenure system (20.7% vs. 13.7%).

They were also more likely to have a nondoctoraldegree (61.6% vs. 52.5%), be male (11.3% vs. 5.8%), and,interestingly, come from schools at AHCs (45.8%vs. 33%).

Survey respondents also reported 29 (1.6% of totalattrition) individuals left full-time faculty positionsbecause of illness or death. Because severe illness anddeath are likely to occur at random and the number offaculty in this group is small, these individuals wereexcluded from further analyses in this study.

Because a large number of chi-square tests for eachattrition group were performed in the bivariate ana-lyses described previously, the type I error, rejectingthe null hypothesis incorrectly, could occur. Althoughmost of the tests are highly significant with very smallp values, readers need to take the number of tests intoconsideration.

Multivariate Analyses on Faculty Attrition forDifferent Reasons

Our bivariate analyses show a number of individualand institutional attributes, such as a junior rank asinstructor, nontenure status, nondoctoral degree, mi-nority status, and working at small schools, are asso-ciated with different types of faculty attrition. Toexamine the relationship between an attribute and atype of faculty attrition while controlling for the im-pacts of all other covariates, we applied multinomiallogistic regression because our outcome variable, typesof faculty attrition, is polytomous. We treated stayingin a full-time faculty position at one’s own school ofnursing as the reference category and leaving a full-

Table 4 e Multinomial Logistic Regression on Attrition of Full-time Faculty, 2010 to 2011

Characteristics Global Test Left forNonacademicNursing (All

Other Reasons)

Left forAdministrativePositions inSchools ofNursing

Left for Full-timeFaculty Positionsin Schools of

Nursing

Left for Part-timeFaculty Positionsin Schools of

Nursing

D.F. pValue

OddsRatio

pValue

OddsRatio

p Value OddsRatio

pValue

OddsRatio

pValue

Assistant professor 4 <.0001 1.259 0.237 <.0001 1.185 0.97Instructor/clinical/

other rank4 <.0001 2.144 <.0001 0.203 <.0001 0.986 1.19

(associate professor/professor)

Tenure track 4 <.0001 4.271 <.0001 2.507 .0055 3.034 <.0001 1.324Nontenure track/system 4 <.0001 4.228 <.0001 2.097 .0244 2.224 .0026 3.603 <.0001(tenured)Nonwhite 4 <.0001 1.528 <.0001 1.198 1.813 .0001 1.176(white)Male 4 <.0001 1.609 .0001 1.759 1.895 .0018 1.886 .005(female)Small school 4 .0006 1.495 .0013 1.720 1.563 .0272 0.609Non-AHC public school 4 0.976 1.263 0.858 0.697 .0401Non-AHC private school 4 .0006 1.322 .0023 1.508 1.000 0.592 .0056(AHC school)Intercept (coefficient) �4.6672 <.0001 �5.1717 <.0001 �4.9375 <.0001 �5.0213 <.0001N: 14,974Likelihood ratio

chi-square(36):507.1, p < .0001.

Notes: The reference category for the dependent variable is staying in full-time faculty positions. The categories in paren-theses are reference categories for the related variables. p values ofWald chi-square are reported for p< .05. Small schools aredefined as baccalaureate and small master’s institutions based on the 2010 Carnegie Classification of Institutions. AcademicHealth Center (AHC) schools are defined based on the AHC classification of the Association of Academic Health Centers.Data Source: 2010 and 2011 AACN’s Annual Survey of Baccalaureate and Graduate Programs in Nursing.

Nur s Ou t l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3 169

time faculty position from own school of nursing asfour attrition categories: (a) leaving for an administra-tive position at own school or a different school ofnursing, (b) leaving for a full-time faculty position at adifferent school of nursing, (c) leaving for a part-timefaculty position at own school or a different school ofnursing, and (d) leaving for a nonacademic nursingposition. We conducted a separate logistic regressionanalysis on faculty retirement, including faculty aged55 or older only, for the reasons given in our bivariateanalysis.

For multinomial logit models, the adequacy ofnumber of explanatory variables is affected by the sizeof sample cells (SAS/STAT(R) 9.3 User’s Guide). Toensure the percentage of sample cells with a small size(N < 5) is acceptable for our model, we combined sixexplanatory variables into three variables: associateprofessor with professor, instructor with clinical/otherrank, and nontenure track with nontenure system. Inaddition, we excluded nondoctoral degree as anexplanatory variable in our model because of a highassociation (or multicollinearity) between this variableand the variable instructor/clinical/other rank.

Table 4 shows the results of multinomial logisticregression analyses. Controlling for other variables, wefound that the odds of leaving for a nonacademic

nursing position for an instructor or faculty holdingclinical/other rank are significantly larger than theodds for an associate professor/professor. The likeli-hood of such leaving is also higher for faculty on tenuretrack or on nontenure track/system compared withfaculty who were tenured. For example, the odds ofleaving for a nonacademic position for faculty ontenure track are 4.271 times the odds for tenured fac-ulty. Being male, nonwhite, or working at a smallschool or non-AHC private school is also positivelyassociated with leaving for a nonacademic nursingposition compared with being female, white, or work-ing at an AHC school.

The odds of leaving for an administrative position atschools of nursing for an assistant professor are lessthan one third (0.237) the odds for an associate pro-fessor/professor. Having a position as instructor/clin-ical/other rank also decreases the likelihood of leavingfor an administrative position at schools of nursingcompared with having a position as an associate pro-fessor/professor. On the other hand, being on tenuretrack or on nontenure track/system increases the oddsof leaving compared with being tenured. For example,the odds of leaving, rather than staying, for faculty whowere on tenure track are 2.507 times the odds for fac-ulty who were tenured.

Table 5e Logistic Regression on Retirement of Full-time Faculty, 2010 to 2011, Ages 55 or Older Only

Characteristics Left for Retirement

Odds Ratio p Value

Assistant professor 0.904Instructor/clinical/other rank 0.835(associate professor/professor)Tenure track 0.405 <.0001Nontenure track/system 0.727(tenured)Small school 1.066Non-AHC public school 1.127Non-AHC private school 0.684 .0192(AHC school)Age 65/older 4.984 <.0001(Age 64/younger)Intercept (coefficient) �3.0293 <.0001N: 6,930Likelihood ratio chi-square(8):250.7, p < .0001.

Notes: The reference category for the dependent variable isstaying in full-time faculty positions. The categories inparentheses are reference categories for the related vari-ables. p values of Wald chi-square are reported for p < .05.Small schools are defined as baccalaureate and smallmaster’s institutions based on the 2010 Carnegie Classifi-cation. Academic Health Center (AHC) schools are definedbased on the AHC classification of the Association of Aca-demic Health Centers.Data Source: 2010 and 2011 AACN’s Annual Survey ofBaccalaureate and Graduate Programs in Nursing.

Nur s Out l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3170

Not having tenure increases the odds of leaving for afull-time faculty position at a different school ofnursing compared with achieving tenure. For example,the odds of such departures for faculty who were ontenure track are 3.034 times the odds for faculty whowere tenured. Beingmale or nonwhite is also positivelyassociated with this type of departures compared withbeing female or white. Working at a small school,defined as a baccalaureate or small master’s institu-tion, also increases the odds of leaving compared withworking at a school affiliated with an AHC.

Having a position without a tenure track/system orbeing male is positively associated with leaving for apart-time faculty position compared with havingtenure or being female. On the other hand, working at anon-AHC public or private school is negatively associ-ated with such attrition compared with working at anAHC school. For example, the odds of changing to apart-time faculty position rather than staying for fac-ulty who worked at non-AHC private schools areslightly more than half (0.592) as large as for facultywho worked at AHC schools after controlling for othervariables.

As we discussed in the bivariate analysis, youngfaculty are not at the “risk” of experiencing retirement,and it is appropriate to compare faculty of similar ageswho retired or stayed. Accordingly, we conducted aseparate multivariate analysis on faculty retirementincluding faculty whowere age 55 or older in 2010 only.Because there is no need to analyze other types offaculty attrition again, we used logistic regression forthis analysis because it is reasonable to estimate themultinomial logit model by running a set of binary logitmodels, and the logistic regression coefficients arealmost as consistent as the multinomial regressioncoefficients (Begg & Gray, 1984; Hosmer & Lemeshow,1989; Allison, 1999). We found that including the vari-ables nonwhite minority and male does not improvethe model (measured by difference in likelihood ratiochi-square values between a model including the twovariables and amodel excluding the two variables), andthe two variables were therefore excluded.

Table 5 shows the results of the logistic regression.We found the odds of retiring rather than staying arealmost five times for faculty whowere 65 years or olderthan the odds for faculty who were 64 years old oryounger after controlling for other variables. On theother hand, being on tenure track or working at a non-AHC private school decreases the likelihood of retire-ment compared with being tenured or working at anAHC school.

Discussion

A 2002 study of medical school faculty conducted bythe Association of American Medical Colleges foundthe annual attrition rate for any reason was 8.4% forclinical MD faculty and 7.4% for clinical PhD faculty at

medical schools in the period of 1995 to 1999(Yamagata, 2002). Another study reported that the5-year retention rate for clinical MD faculty across the1999 to 2003 cohorts ranged from 77.4% in emergencymedicine departments to 63.2% in obstetric/gyneco-logic departments, which would yield an annual attri-tion rate of 4.5% for emergency medicine faculty and7.4% for obstetric/gynecologic faculty (Corrice, Fox, &Bunton, 2011). Compared with medical faculty, theannual attrition rate of 11.8% for nursing faculty in2010 to 2011 found in this study is substantially higher.Although the attrition rate of medical faculty may notbe an ideal benchmark to assess the attrition rate ofnursing faculty, given themagnitude of nursing facultydepartures and the needed resources for recruiting andtraining new faculty to replace those departingacademia, it is meaningful to consider the nursingfaculty attrition rate in the context of other healthprofessions.

Those individuals who left academic roles to retirepotentially represent an early wave of what has beenprojected to be significant retirements from nursingacademic positions. Faculty members who retired aremore likely to be associate professors or professors(61%) and have achieved tenure (54%). After controllingfor age and other variables, we found faculty on tenuretrack are less likely to retire than tenured faculty. It islikely that faculty on tenure track have less years of

Nur s Ou t l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3 171

service than do tenured faculty and therefore are lesslikely to meet institutional requirements on service forfaculty retirement benefits. The departures of manyfaculty members with senior rank and tenured status,especially of those who hold faculty positions atresearch-intensive institutions, not only may reducefunding opportunities for their schools of nursing butmay also limit the availability of mentors for juniorfaculty. Replacement of these individuals will creategreat challenges for academic leaders.

Given the significant contributions of faculty of se-nior rank to academic nursing, the nursing educationcommunity is always interested in the mean age offaculty at retirement (Bellack, 2004). A 2002 study re-ported an average age of 62.5 years at retirement fornursing faculty, and a 2006 study reported the meanage for retirement anticipated by nursing faculty was64.4 years (Berlin & Sechrist, 2002; Kowalski, Dalley, &Weigand, 2006). We found the mean age at facultyretirement was 64.5 years in 2010 to 2011. Our findingmay be further indication that faculty members aredelaying retirement, and, thus, we see an increase inthe average age of retirement for nursing faculty.Similar trends in other professions in the United Statesalso have been documented as results of current eco-nomic difficulties and personal or professional reasons(Levanon, Cheng, & Goldman, 2011; Flaherty, 2013).Nationally, the mean age of nursing professors was61 years, and the mean age of nursing associate pro-fessors was 57.5 years in 2011 (Fang, Li, &, Bednash,2012). In 2011, 25.8% of all full-time faculty memberswere age 60 or older (American Association of Collegesof Nursing, 2012). All these data portend large losses ofthe most seasoned and experienced faculty in bacca-laureate and graduate programs over the next decade.

On the other hand, the loss of this cadre of seasoned,highly positioned faculty also creates an opportunityfor many academic programs to seed newly emerging,doctorally educated nurses in full-time, tenure-trackfaculty positions. However, a significant effort will berequired to mentor and support these replacements toensure they are successful and retained in a full-timefaculty role. Preparing future faculty also will requirevigorous efforts to expand the number of nursesgraduating with doctoral degrees who desire a facultycareer. Additionally, serious attention must be placedon a more rapid progression to the terminal degree bynursing professionals with a focus on elongation of thetime that an individual could fill a faculty role.

The largest departures came from faculty who tooknonacademic nursing positions. These faculty mem-bers are more likely to have a nondoctoral degree andhold the junior rank of instructor or clinical roles. Theyalso are more likely to come from small schools. Theseindividuals left for a variety of reasons. The movementto a practice settingmay be driven by economic factors,such as the salary mismatches between academic andpractice environments. These egressions also may berelated to other factors such as heavy workload anddissatisfaction with the perceived inferior role as

master’s-educated faculty or clinical faculty (Gerolamo& Roemer, 2011; Gazza & Shellenbarger, 2010; Lacey &McNoldy, 2008). Moulton and Wakefield (2007) foundthat a lack of promotion opportunities was a factor forthe attrition of master’s prepared, nontenure trackfaculty. We also found that 63.5% of those facultymembers who left an academic position involuntarilydid not hold a doctoral degree, perhaps giving someevidence of the employing institutions’ preference fordoctorally prepared faculty. This turnover createsinstability for both the faculty member and the aca-demic unit because considerable resources may havebeen invested in the development of these individualsas faculty. The lack of predictability for each of thesepresents long-term career or resource managementchallenges to nursing schools. Further assessment ofthe relevant factors is necessary to design effectiveinterventions.

Among the study’s many findings, it is important tonote that not all faculty departures are a loss to aca-demic nursing. Individuals who left full-time facultypositions to take administrative positions at schools ofnursing should be viewed as having achieved careeradvancement, which presents evidence of commit-ment to an academic, albeit nonfaculty, role in nursing.It is interesting to see that faculty members who wereon tenure track are more likely than faculty memberswho were tenured to take administrative positions.Facultymembers in this group tend to have senior rankas associate professor or professor. These individualsprovide an important asset to nursing programs asacademic leaders and potential mentors for faculty.Identification of these individuals could provide clarityregarding career decision making facilitating theirprogression to administrative roles, their reasons forcommitting to a position in academia, and the rolesthey serve as mentors for new or advancing faculty.

Assuming a full-time faculty position at a differentschool of nursing should not be considered as a loss tothe nursing education community either. However, itmay be a loss to the individual nursing school fromwhich the person left. We found faculty members withsuch departures tend to be on tenure track. Did theyleave because they could not achieve tenure? In addi-tion, faculty in this group are more likely to be male,nonwhite minority, or from small schools. Staying inacademic nursing for male and minority faculty givessome indication of their satisfaction with the facultyrole. Their movement to a different school of nursingcould also be evidence of strong efforts used by otherinstitutions to recruit a more diverse and representa-tive faculty.

Faculty members who switched from a full-timeposition to a part-time position are losses to nursingschools because full-time faculty, as required bynursing schools, often provide more hours of servicesthan do part-time faculty. We found a large majority offaculty in this category of attrition had nontenure trackor nontenure system positions (82%). They also tend tohave junior rank as instructor or on clinical rank (51%)

Nur s Out l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3172

and have a nondoctoral degree (62%). In addition, theyare more likely to come from AHC schools (46%). Thelack of a doctoral degree for many of them may maketheir status at academic institutions, especially AHCschools, less attractive to both the faculty member andthe employing institution that may have created arequirement for the nonedoctorally educated nurse toseek an alternative position.

A significant percentage of faculty members wholeft held doctoral degrees. For example, of the facultymembers who left for part-time faculty positions andfor nonacademic positions, 38% and 29%, respectively,had a doctoral degree. Nonetheless, they made thedecision to seek employment, at least partially, outsideof an academic environment. Given the significant in-vestment on these individuals by the doctoral pro-grams where they were educated and the nursingschools where they were employed, their departuresrepresent a great loss to the nursing education com-munity. A number of studies have reported reasons forfaculty attrition such as low salaries, growing work-load, lack of institutional support, and frustrationswith the quality of students who were not prepared fornursing education (Dunham-Taylor, Lynn, Moore,McDaniel, & Walker, 2008; Gerolamo & Roemer, 2011;Cash et al., 2009; Garbee & Killacky, 2008; Gerolamo &Roemer, 2011). Additional studies of reasons for thedepartures of doctorally prepared faculty, economic orotherwise, should be undertaken to determine how todevise appropriate strategies to attract this group offaculty to maintain teaching and research roles.

We also found men are more likely than women toleave, and to a lesser degree, members of minoritygroups also are more likely than white faculty to leave.The differential attrition rates by sex and minoritystatus are not unique to academic nursing. Similarpatterns occurred in academic medicine (Yamagata,2002). One study of medical school faculty found thelack of opportunities for career advancement, lowsalary, and leadership issues were major reasons forthe higher attrition of women andminority faculty at amedical school (Cropsey et al., 2008). Other studiesfound medical school faculty from underrepresentedgroups were less likely to be promoted and more oftenconsidered leaving (Fang, Moy, Bergeisen, & Hurley,2000; Palepu, Carr, Friedman, Ash, & Moskowitz,2000). Further investigation of the motivators for themobility of male and minority nursing faculty couldbring important information to efforts to enhance theretention of faculty from underrepresented groups.

In this study, we quantified the national attritionrate for full-time nurse faculty as well as attrition ratesby reasons for leaving. We also identified certainindividual and institutional characteristics are associ-ated with different types of attrition. Our findings willhelp the nursing education community and the gov-ernment to better estimate the pool of faculty candi-dates needed to replace those departing academia andtake more focused measures to reduce nurse facultyegressions. In addition, we speculated a number of

reasons to explain different types of attrition. Theremay be other factors that could affect nursing facultyattrition such as the downsizing of many public in-stitutions during the current economic difficulties andthe trend toward moving advanced clinical educationfrom master’s level to doctoral level in nursing, whichwould require more doctorally prepared faculty. How-ever, we do not have data to validate these reasons.Therefore, future studies on reasons for faculty attri-tion are needed.

Conclusions

In this study, we found that nationally 11.8% of full-time nursing faculty who worked in 2010 left theirfull-time jobs by 2011. Nearly half of total attrition, or5.7% of full-time faculty, were related to leaving fornonacademic nursing positions, whereas another 20%of attrition, or 2.4% of full-time faculty, resulted fromretirement. We also found the mean age of faculty atretirement in 2010 to 2011was 64.5 years. Nearly 20% offaculty egressions, or 2.2% of full-time faculty, weredue to leaving for nursing administrative positions orfull-time faculty positions. Leaving for part-time fac-ulty positions accounts for slightly more than 10% offaculty attrition or 1.3% of full-time faculty.

The findings have answered our research questionsthat certain individual and institutional characteristicsare associated with a higher likelihood of faculty attri-tion. Faculty members holding a junior rank asinstructor or on a clinical track are more likely thanassociate professors or professors to move to differentschools to assume full-time faculty positions or to leavefor nonacademic nursing positions. Male faculty aremore likely than female faculty to leave for full-timefaculty positions at different schools, switch to part-time faculty positions, or leave for nonacademicnursing positions. Facultymembers at small schools aremore likely than faculty at AHC schools to leave for full-time faculty positions at different schools or nonaca-demic nursing positions. The best predictor of facultyattrition is nontenure track/system status. Faculty onnontenure track/system are more likely than facultywho are tenured to leave for administrative positions,full-time faculty positions, part-time faculty positions,or nonacademic nursing positions. In addition, ourbivariate analyses show having a nondoctoral degreecompared with having a doctoral degree increases thelikelihood of switching to part-time faculty positions orleaving for nonacademic nursing positions.

Acknowledgments

The authors thank Robert Rosseter for editorialassistance.

Nur s Ou t l o o k 6 2 ( 2 0 1 4 ) 1 6 4e 1 7 3 173

r e f e r e n c e s

Allison, P. D. (1999). Logistic regression using SAS system: Theory andapplication. Cary, NC: SAS Institute Inc.

American Association of Colleges of Nursing. (2005). Facultyshortages in baccalaureate and graduate nursing programs:Scope of the problem and strategies for expanding the supply.Retrieved from http://www.aacn.nche.edu/Publications/WhitePapers/FacultyShortages.htm.

American Association of Colleges of Nursing. (2012). 2011 Annualsurvey of baccalaureate and graduate programs in nursing.Washington, D.C.

Anderson, C. A. (2000). Current strengths and limitations ofdoctoral education in nursing: Are we prepared for the future?Journal of Professional Nursing, 1(6), 191e200.

Begg, C. B., & Gray, R. (1984). Calculation of polychotomouslogistic regression parameters using individualizedregressions. Biometrika, 71, 11e18.

Bellack, J. P. (2004). One solution to the faculty shortagedBegin atthe end. Journal of Nursing Education, 43(6), 243e244.

Berent, G. R., & Anderko, L. (2011). Solving the nursing facultyshortage: Exploring retention issues. Nurse Educator, 36(5),203e207.

Berlin, L. E., & Sechrist, K. R. (2002). The shortage of doctorallyprepared nursing faculty: A dire situation. Nursing Outlook, 50,50e56.

Brendtro, M., & Hegge, M. (2000). Nursing faculty: One generationaway from extinction? Journal of Professional Nursing, 16,97e103.

Cash, P. A., Daines, D., Doyle, R. M., & Tettenborn, L. V. (2009).Quality workplace environments for nurse educators:Implications for recruitment and retention. Nursing Economic$,27(5), 315e321.

Cash, P. A., Doyle, R. M., Tettenborn, L. V., & Faria, V. (2011).Working with nurse educators’ collective wisdom:Implications for recruitment and retention. Nursing Economic$,29(5), 257e264.

Corrice, A. M., Fox, S., & Bunton, S. (2011). Retention of full-timeclinical M.D. faculty at U.S. medical schools. Analysis in Brief,11(2).Washington, DC: Association of AmericanMedical Colleges.

Cropsey, K. L., Masho, S. W., Shiang, R., Sikka, V., Kornstein, S. G.,Hamption, C. L., & the Committee on the Status of Women andMinorities. (2008).Whydo faculty leave?Reasons for attrition ofwomen and minority faculty from a medical school: Four-yearresults. Journal of Women’s Health, 17, 111e118.

Dunham-Taylor, J., Lynn, C. W., Moore, P., McDaniel, S., &Walker, J. K. (2008). What goes around comes around:Improving faculty retention through more effectivementoring. Journal of Professional Nursing, 24(6), 337e346.

Falk, N. L. (2007). Strategies to enhance retention and effectiveutilization of aging nurse faculty. Journal of Nursing Education,46, 165e169.

Fang, D., Moy, E., Bergeisen, L., & Hurley, J. (2000). Racial andethnic disparities in faculty promotion in academic medicine.Journal of the American Medical Association, 284, 1085e1092.

Fang, D., Wisniewski, W. S., & Bednash, G. D. (2007). 2006e2007Enrollment and graduations in baccalaureate and graduateprograms in nursing. Washington, DC: American Association ofColleges of Nursing.

Fang, D., Li, Y., & Bednash, G. D. (2012). 2011e2012 Salaries ofinstructional and administrative nursing faculty in baccalaureate

and graduate programs in nursing. Washington, DC: AmericanAssociation of Colleges of Nursing.

Flaherty, C. (2013). Working way past 65. Inside Higher Ed, June 17.Retrieved from http://www.insidehighered.com/news/2013/06/17/data-suggest-baby-boomer-faculty-are-putting-retirement.

Foxall, M., Megel, M. E., Grigsby, K., & Billings, J. S. (2009). Facultyretirement: Stemming the tide. Journal of Nursing Education,48(3), 172e175.

Garbee, D. D., & Killacky, J. (2008). Factors influencing intent tostay in academia for nursing faculty in the southern UnitedStates of America. International Journal of Nursing EducationScholarship, 5(1), Article 9.

Gazza, E. A., & Shellenbarger, T. (2010). The lived experience ofpart-time baccalaureate nursing faculty. Journal of ProfessionalNursing, 26(6), 353e359.

Gerolamo, A. M., & Roemer, G. F. (2011). Workload and the nursefaculty shortage: Implications for policy and research. NursingOutlook, 59, 259e265.

Horat, C. T. (2008). Phenomenological study of new nursing faculty inGeorgia universities: Experiences affecting retention or attritiondecisions. (Dissertation). Phoenix, AZ: University of Phoenix.

Hosmer, D. W., & Lemeshow, S. (1989). Applied logistic regression.New York: John Wiley & Sons.

Kovner, C., Fairchild, S., & Jacobson, L. (2006). A report of the facultycensus survey of RN and graduate programs. New York: NationalLeague for Nursing.

Kowalski, S. D., Dalley, K., & Weigand, T. (2006). When will facultyretire: Factors influencing retirement decisions of nurseeducators. Journal of Nursing Education, 45, 349e355.

Lacey, L. M., & McNoldy, T. P. (2008). Influential factors in the choiceof a nursing education career. Lexington, NC: North CarolinaCenter For Nursing.

Levanon, G., Cheng, B. & Goldman J. (2011). U.S. workersdelaying retirement: What businesses can learn from thetrends of who, where, and why. The Conf BoardExecutiveAction Ser, 350. Retrieved from http://www.financialfinesse.com/wp-content/uploads/2012/01/US_workers_delay_retirement.pdf.

Moulton, P., & Wakefield, M. K. (2007). Nurse faculty recruitment andretention project. Grand Forks, ND: Central Rural HealthUniversity of North Dakota School of Medicine & HealthSciences.

Palepu, A., Carr, P. L., Friedman, R. H., Ash, A. S., & Moskowitz, M.A. (2000). Specialty choices, compensation, and careersatisfaction of underrepresented minority faculty in academicmedicine. Academic Medicine, 75, 157e160.

Ryan, M., & Irvine, P. (1994). Nursing professorate in America:Projected retirements and replacements. Journal of NursingEducation, 33(2), 67e73.

SAS/STAT(R) 9.3 User’s Guide: The CATMOD Procedure. Retrievedfrom http://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_catmod_sect033.htm.

Williamson, M. L., Cook, A. L., Salmeron, L. S., & Burton, D. (2010).Retaining nursing faculty beyond retirement age. NurseEducator, 35, 152e155.

Yamagata, H. (2002). Trends in faculty attrition at U.S. medicalschools, 1980-1999. Analysis in Brief, 2(2). Washington, DC:Association of American Medical Colleges.

Yordy, K. D. (2006). The nursing faculty shortage: A crisis for healthcare. Princeton, NJ: Robert Wood Johnson Foundation.