The paper should focus majorly on voluntary employee turnover and not involuntary employee turnover.The paper should include- title- a clear description of the topic- background- why this topic is important- the purpose of the research paper- thesis (argument or expected findings) – value-added componentI want the paper to talk about- Why is it important to mitigate or reduce the turnover of good employees – What causes them to leave voluntarily – What the organization can do to improve- Why is it important to mitigate or reduce the turnover of good employees
I have attached 2 articles and 2 links I would like to use. please feel free to add any other links to make up the paper.For the value-added component please follow option 3
Further instructions for your research paper and proposal
At the 4th year level, it is a learning outcome of the course for students to demonstrate their ability
to not only read, understand, and compile literature from academic sources, but also to
demonstrate upper level skills such as synthesis of diverse literature bases, applying that literature
base to a practical application, or conducting primary research to add to what is written in the
existing literature. I call this the “value-added” component. The “value-added” component is
something that you do that goes beyond what has already been written in the literature on the
topic. So, when you are considering your research paper topic and the proposal, your proposed
work needs to go beyond summarizing what you read in the already published literature on the
subject, and you should detail how you plan to do so in your proposal.
There are many ways in which to provide the “value-added” component. Some suggestions are
as follows:
1.
by:
Conduct primary research – i.e., collect information or data directly. This can be done
a)
doing a small sample of interviews and incorporating the results of those interviews into
your paper in the analysis and recommendations sections;
b)
creating and conducting a short written/online survey and incorporating the results of the
survey responses into your paper in the analysis and recommendations sections; or,
Please Note: For these options, you must comply with the ethics training and approval protocols
of the University of Winnipeg.
2.
Conduct secondary research that brings value and relevance to your topic. This could be
done by:
a)
collecting/finding case scenarios, situations, publically accessible legal cases, etc. and
assessing them against your argument/topic, and then finding similarities/differences, and making
inferences/recommendations based on those cases.
b)
applying your concept/topic/argument to a particular organization or scenario, and
coming up with specific recommendations for that organization/scenario. (Note: This option
may require compliance with the ethics training and approval protocols of the University of
Winnipeg, if you are collecting the information from the organization directly. Please see me
regarding clarification on this based on your proposed research methods.)
3.
Comparing and contrasting two literatures, models/theories, or constructs to come to a
fuller understanding of both. For example, this can be done by demonstrating similarities,
differences, and how and why this is so. Then, coming up with specific suggestions for inclusion,
improvement, or moving the concepts forward.
Employee Voice and Employee Retention
Author(s): Daniel G. Spencer
Source: The Academy of Management Journal , Sep., 1986, Vol. 29, No. 3 (Sep., 1986), pp.
488-502
Published by: Academy of Management
Stable URL: https://www.jstor.org/stable/256220
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to
The Academy of Management Journal
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
CAcademy of Management Journal
1986, Vol. 29, No. 3, 488-502.
EMPLOYEE VOICE AND EMPLOYEE RETENTION
DANIEL G. SPENCER
University of Kansas
This study investigates the relationship between the extent to wh
employees have opportunities to voice dissatisfaction and voluntar
turnover in 111 short-term, general care hospitals. Results show th
whether or not a union is present, high numbers of mechanisms f
employee voice are associated with high retention rates. Implicatio
for theory and research as well as management practice are discuss
The relationship between the job satisfaction and employee
been one of the most widely studied but least understood r
the organizational behavior literature. The consistently low
between job satisfaction and turnover found in empirical re
topic clearly indicate that the relationship is not direct. Many
intercede either to cause satisfied employees to leave organizatio
fied employees to remain with organizations. Previous rese
posed and tested a number of moderators that might clarify t
between job satisfaction and turnover (March & Simon, 1958; Mo
Hand, & Meglino, 1979; Price, 1977). The two most prominent h
the number of employment alternatives employees who are con
ing an organization have, and (2) nonwork-related influence
spouse’s immobility. Also, research has recently centered ar
Mobley’s (1977) model of the process of employees’ decisions on
attempts to better understand those cognitive and behavioral p
mediate the relationship between affective reactions to jobs
But even though scholars have expended much effort in th
and researching the relationship between job satisfaction and t
sistently weak empirical relationships are still found.
Steers and Mowday (1981) suggested a heretofore neglected
may intervene in the relationship in question: employees’ effor
dissatisfying work situations that would otherwise be major fa
decisions to leave their organizations. Employees who succee
dissatisfying work situations will lower their dissatisfaction; w
University of Kansas General Research Fund Grants 3616-0038 and 3219
support for this research. In addition, it was supported by the University o
Business Research Fund provided by the Fourth National Bank & Trust Comp
ideas and opinions expressed herein are solely those of the author. The comme
Mowday, George F. Dreher, and Morris M. Kleiner on earlier drafts are gr
Portions of this paper were presented at the 42nd Annual Meeting of
Management, New York, 1982.
488
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Spencer
1986
489
leave changed to intent to remain, the likelihood increases that these employees will ultimately remain with their organizations. Up until recently, then,
the literature on turnover has focused on only one of several alternatives
open to individuals faced with dissatisfying work situations, leaving; it has
ignored the alternative strategy of attempting to change such situations.
HIRSCHMAN’S CONCEPTS OF EXIT AND VOICE
The proposed relationship between employees’ efforts to
satisfying work situations and turnover is implicit in much of
Albert Hirschman (1970, 1974), whose perspective served as the
foundation for the present research. He extensively studied
declining firms in terms of exit and voice. The two terms refer
that members or clients of an organization have when responding to an
economically dissatisfying relationship with that organization. The source
of dissatisfaction might be declining quality of the organization or declining
quality of its products. To voice dissatisfaction is to try to change a problematic situation, and to exit is simply to withdraw from and leave the problematic situation. Most of Hirschman’s work is grounded in economics and thus
focuses primarily on the clients of an organization using voice or exit rather
than on its members using these options. Hirschman’s concepts are gaining
increasing attention in the organizational behavior literature (Farrell, 1983).
This attention is not limited to studies of employee turnover; other relevant
work includes research on absenteeism (Hammer, Landau, & Stern, 1981),
political behavior (Farrell & Peterson, 1982), organizational due process (Aram
& Salipante, 1981), organizational dissent (Graham, 1986), and the cathartic
effects of voice (Greenberg & Folger, 1983).
EMPLOYEE TURNOVER AND THE LABOR RELATIONS LITERATURE
Unions can be viewed as vehicles for the collective voicing o
dissatisfactions. Previous research has demonstrated that union
associated with employee stability. Freeman and Medoff (19
their own and others’ analyses of individual behavior and indust
demonstrated that unionization is consistently and significa
with retention of employees if wage rates and other known
employee exit are controlled for. The control of wage rates is cr
to differentiate the effects of unions in terms of voice from
effects due to union control of equivalent jobs.
The research Freeman and Medoff (1984) reviewed is withou
based on analyses of very large data sets like the National L
Survey.1 Such data sets constrain the specific relationships tha
ied and the levels of analysis at which such relationships can
In particular, very little research exists on the effects of u
1 This survey was conducted by the Center for Human Resource Resear
University.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Academy of Management Journal
490
September
quitting rates in organizations. Becker (1978) conducted such an organization level study, examining the effects of unionization on rates of quitting for
individuals in low-wage occupations in short-term, general care hospitals.
Contrary to results at other levels of analysis, Becker found a weak, marginally significant relationship (p < .10) between unionization and rate of turn-
over while controlling for monopoly effects and a wide variety of other
known predictors of employee exit.
The generalizability of research like Becker's (1978) is limited; such
work is usually specific to the occupation, organization, and region studied.2
However, when an investigator controls data collection, a study is more
likely to include predictors of exit that are not captured in studies using
existing data bases. For example, Becker examined the effects of fringe bene-
fits on rates of quitting in addition to the effects of wages. Freeman and
Medoff (1984) noted that a major criticism of research on unions' effects in
terms of voice is that most existing data sets do not include information on a
principal monetary reward such as fringe benefits and thus do not fully
control for monopoly effects. Another significant criticism of research on
voice effects of unions is Ulman and Sorensen's (1984) demonstration that,
for unionized employees, the effects of strikes and the threat of strikes provide other alternatives to quitting besides voice realized through the union
grievance mechanism.
Thus, accurately assessing the effects of employee voice on numbers of
exits requires more than assessing the effect of unionization. It requires
exploring the effects of mechanisms, whether union mandated or not, that
organizations put in place to enhance their employees' opportunities for
voicing dissatisfaction. This approach necessitates collecting data specifically for this purpose instead of using an existing data base. This study
employed a data collection strategy similar to that used by Becker (1978).
HYPOTHESES AND STUDY DESIGN
Study One
Two studies provided data for this research. The first sought to determine if there is a direct relationship between (1) the amount of opportunity
an organization gives its employees to voice dissatisfaction and change dis-
satisfying work situations, and (2) the organization's rate of employee
retention. Thus,
Hypothesis 1: There will be a significant and negative
relationship between the total number of voice mechanisms for employees that an organization has and the
voluntary turnover rate among the organization's employees.
2 Becker (1978) surveyed hospitals in three contiguous states in the north central United
States.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
1986
Spencer
491
The focus here was on the amount of opportunity available for voice and not
on the quality of that opportunity; the second may, in fact, be a more viable
determinant of retention. Employee voice mechanisms examined in this
research included grievance procedures, suggestion systems, employeemanagement meetings, counseling services, ombudsman services, nonmanagement task forces, question and answer programs, and survey feed-
back.
The following known predictors of rate of employee turnover were con-
trolled for in this study: wage rate, the ratio of fringe benefits to total
compensation, unemployment rate, number of grievances filed (a surrogate
for job dissatisfaction), percentage of minority employment, organizational
size, and occupation-specific employment opportunities in the region (see
Becker, 1978). The effect of unionization of the workforce on the rate of
employee retention was examined separately, but for the purposes of this
study, it is included as a control variable. This study examined only the
retention rates of an occupation dominated by women that has a documented shortage of workers: registered nursing. Short-term, general care
hospitals that universally employ continuing education for registered nurses
were studied. This population was chosen to control for unwanted variance
due to differences in total employment accounted for by women, shortage of
workers, availability of continuing education, occupational type, and organizational type.
Study Two
The first study did not assess the quality of the hospitals' voice
mechanisms. A second analysis was conducted to examine the relationship
between the number of mechanisms offering employees the option of voice
in an organization and employees' perceptions of the effectiveness of these
mechanisms. From four hospitals participating in the first study, data were
collected on registered nurses' expectancies for resolving work related problems and their perceptions of the general effectiveness of their organizations'
voice mechanisms. It was proposed that,
Hypothesis 2: A high number of employee voice mechanisms will be positively related to high expectancies of
problem resolution among employees and high levels of
effectiveness for the organization's problem resolution
procedures.
STUDY ONE
Methods
Sample. Organizations were the units of analysis in this study. Participa
ing organizations were short-term, general care hospitals ranging from 50
over 1,000 beds in size. A systematic sample of presidents and executi
directors of hospitals in the north central United States was drawn from
American Hospital Association membership directory. These administrator
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Academy of Management Journal
492
Septernber
received questionnaire packages with a cover letter requesting them to forward the questionnaire and the return envelope provided to their hospitals'
personnel directors. In most cases, personnel directors or members of person-
nel staffs completed the questionnaires. Out of a sample of 278 hospitals,
individuals in 129 of them completed and returned questionnaires for a
response rate of 46.4 percent. Previous studies using similar strategies for
data collection have yielded response rates below 30 percent (Becker, 1978).
Of the hospitals participating in the study, 13.9 percent that did not supply
data on employee turnover were removed from the final sample. Data from
111 short-term, general care hospitals, 15 of which had unionized registered
nurses, were analyzed.
Employee voice mechanisms. The questionnaire asked whether or not a
hospital's registered nurses were subject to the following employee-relations
practices: (1) formal grievance procedure, (2) suggestion system, (3) employee-
management meetings, (4) counseling service, (5) ombudsman, (6) nonmanagement task forces, (7) question and answer program, and (8) survey
feedback. A composite index of employee voice mechanisms was created by
simply summing the number of these mechanisms that an organization
employed (Kuder-Richardson Formula 20 = .55).
Turnover among registered nurses. Each hospital administrator was asked
to supply the rate of voluntary turnover for its registered nurses for the
previous 12-month period.
Control variables. The questionnaire sought the following numerical information on a hospital's registered nurses: (1) wage rate per hour, (2) ratio of
fringe benefits to total compensation, (3) number of grievances they filed,
and (4) percentage of minority employment. It also asked the hospital's size
(number of beds) and the number of hospital beds in the county to appraise
alternative employment opportunities in the region. In addition, it asked if
registered nurses were unionized. Finally, unemployment rates for each
hospital's county and standard metropolitan statistical area (SMSA) for the
12-month period under study were obtained from the research and statistics
departments of state employment security divisions.
Results
Table 1 gives means and standard deviations of the study's variables
The mean rate of turnover among registered nurses was 21.1 percent (s.d. =
11.5). Table I also provides Pearson correlations between variables. All correl
tions between variables and turnover were in the predicted direction with
the exception of the positive correlation3 between wage rate and turnover.
3 There is a potential explanation for this finding that Becker (1978), who found simil
results, did not consider. For hospitals with production functions that require high levels
staffing at all hours, average hourly wages are higher because they pay premium differentials
staffing undesirable shifts. In such circumstances, particularly with a shortage of nurses,
would not be surprising that turnover would be high as employees would leave for mo
desirable shifts with other organizations.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
TABLE 1
Means, Standard Deviations, and Intercorrelations for Variables
Variables
1.
Means
Turnover
s.d.
1
2
3
4
5
6
7
8
9
10
among
registered nurses 21.1% 11.5 Employee voice mechanismsb
2. Grievance procedurec 1.94 .23 -.28* 3. Survey feedback 1.50 .50 -.24* .07
4. Ombudsman 1.11 .31 -.13 -.03 .04
5. Question-and-answer
program 1.53 .50 -.17* .05 .31* .04 6. Employee-management
meetings 1.81 .39 -.01 .15 .13 .10 .24* 7. Counseling service 1.78 .41 -.15 .22* .16* .12 .17* .27* 8. Suggestion system 1.60 .49 -.02 .01 .15* .19* .22* .17* .03
9. Nonmanagement task
forces 1.38 .49 -.02 -.10 .24* .01 .23* .24* .21* .10
Control variables
10. Hourly wage rate 8.3 0.8 .27* .15 -.07 -.07 .01 .17* .11 .01 -.09 11. Fringe benefits 25.2 6.6 -.01 .11 .18* .04 .09 .11 .20* -.09 .15 .
12. County unemployment
rate 8.1 3.0 -.12 -.06 .12 .02 -.02 -.09 --.07 -.04 -.09 -.02
13. Grievances filed 9.6 14.0 .13 -.43* .06 -.01 .15 -.03 -.08 .00 .16*-
14. Minority employment 8.4 16.7 .17 .06 .10 .09 -.04 .12 .13 .04 -.02 .2
15. Beds in hospital 375.2 258.1 .13 -.06 -.01 -.04 .09 .10 .10 -.07 .11 .3
16. Hospital beds in
3592.4 4297.7 .18* .13 -.06 .02 -.06 .17* .05 -.07 -.02 .
county
17. Unionizationb
1.1 0.3 -.05 .09 -.10 .04 -.04 .05 -.04 -.10 .17 .10
a Descriptive data are based on Ns ranging from 1
b Coded 2 if present, 1 if not present.
c Mean turnover rates: with grievance procedures,
rates in hospitals with grievance procedures: union
*p < .05.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Academy of Management Journal
494
September
It was hypothesized that the more opportunities employees have to voice
dissatisfaction and have input into changing dissatisfying work situations,
the greater the number of employees that will remain with an organization.
The composite index of employee voice mechanisms correlated negatively
with turnover among registered nurses (r = -.24, p < .05); the greater the
number of employee voice mechanisms that a hospital employed, the lower
the turnover. This significant relationship between the composite index and
rate of turnover holds when a number of known predictors of employee
turnover are statistically partialled out (see Table 2).
Ordinary-least-squares regression was considered appropriate for multivariate analysis of the data (cf. Kochan & Helfman, 1981). As indicated in
Table 2, the multiple correlation of the linear combinations of all control
variables, except for unionization, with turnover was R = .36. Adding unionization to the predictive equation did not increase the multiple correlation
with turnover. Finally, adding the composite index of employee voice
mechanisms, including the unionization variable, to the predictive equation
significantly increased the multiple correlation with turnover to R = .45 and
the explained variance in the dependent variable to AR2 = .07. This significant (t = 2.6, p < .05) increase is reflected by the twofold increase in R2
corrected for shrinkage. Therefore, number of employee voice mechanisms
is significantly associated with rate of turnover among registered nurses
TABLE 2
Results of Hierarchical Regression Analyses of Unweighte
Composite Index of Employee Voice Mechanisms, Unionizati
and Control Variables on Registered Nurse Turnovera
Step 2: Step 3:
Step 1: Assessment of Assessment of
Control Variables Unionization Effect Voice Effect
Independent Variables b 3 tb b 3 tb b e tb
Hourly wage rate 3.31 .23 1.8 3.37 .23 1.8 3.55 .24 1.9
Fringe benefits -.10 -.06 -0.5 -.10 -.06 -0.5 -.03 -.02 -0.1
County unemployment rate -.53 -.14 -1.3 -.52 -.14 -1.2 -.73 -.19 -1.7
Grievances filed .04 .15 1.4 .04 .15 1.3 .04 .16 1.4
Minority employment .09 .13 1.1 .09 .13 1.1 .11 .16 1.3
Beds in hospital .00 .03 0.2 .00 .02 0.2 .00 .05 0.4
Hospital beds in county .00 .03 0.2 .00 .03 0.2 .00 .00 0.0
Unionization
-.72
-.02
-0.2
-.66
-.02
-0.2
Employee voice mechanisms -1.92 -.28 -2.6*
Constant
R
-2.67
.36
R2
.13
Adjusted
a
-3.02
.36
.13
R2
Deletion
.05
of
3.68
.45*
.20
.04
.10
missing
were conducted to 81.
b Partial t.
*p < .05.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
data
r
Spencer
1986
495
when a large number of known predictors of employee turnover are controlled.
In the composite index employed, each employee voice mechanism
received equal weight. When a weighted linear combination of the components of this index was created through regression analysis, the multiple
correlation with rate of turnover among registered nurses was R = .41. Little
confidence can be placed in this weighted combination of variables, however,
because when the sample was split into two subsamples composed of even
and odd cases, the regression coefficients were not stable and the two subsamples failed to double-cross-validate at an acceptable level (r = .16, n.s.;
r = .31, p < .05).
STUDY TWO
Methods
Sample. Four hospitals that participated in study one volunteered t
allow a survey of nonsupervisory registered nurses' perceptions of their
organizations' voice mechanisms. In two hospitals, one with 53 beds and the
other with 250, questionnaires were administered on site during working
hours; response rates were 44.4 and 23.7 percent. In the second two hospitals (107 and 112 beds), personnel departments distributed questionnaires
with return envelopes addressed to the researcher to be filled out at the
employees' convenience; response rates were 42.8 and 27.3 percent. All
nonsupervisory registered nurses currently on staff at the hospitals received
questionnaires; participation was voluntary.
Nurses' perceptions. Nurses estimated how often they felt they would be
effective in resolving eight work-related problems. The response format ranged
from 10 to 100 percent of the time on 10-percent intervals. Problems given
were related to supervision, tasks, co-workers, pay, or promotions. Two
items represented each of the first three types of problems and one item
represented each of the last two types. The reliability for the eight-item
composite measure of expectancy of problem resolution was satisfactory
(( = .86).
The questionnaire also asked the extent to which nurses agreed or d
agreed with a series of ten statements (see Appendix) designed to reflect t
extent to which the management of their organizations effectively respond
to and used employee voice (a = .89). Finally, three one-item measures asked
employees the extent to which they agreed or disagreed with three statements (see Appendix) designed to assess their perceptions of the effective-
ness of voice mechanisms, the utility of efforts to change dissatisfyi
situations, and their preference for leaving versus attempting to resolve
dissatisfaction.
Results
Table 3 gives means and standard deviations of these variables as well as
Pearson correlations. All correlations are in the expected direction.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
TABLE 3
Means, Standard Deviations, and Intercorrelations f
Variables
Means
s.d.
la
lb
Ic
1. Expectancy of problem resolu
4.2 2.3 a. Supervisory problems
5.3 2.6 .39
b. Task problems
c. Co-worker problems 5.4 2.5 .29 .53
d. Pay problems 3.2 2.3 .24 .30 .37
e. Promotion problems 3.8 2.6 .46 .30 .37 .56
f. Composite expectancy 4.6 1.8 .70 .77 .76 .59
2. Management's responsiveness to voice 3.5 1.2 .31 .40 .34
3. Effectiveness of voice mechanisms 2.8 1.5 .07 -.02 .08
4. Utility of efforts to change dissatisfying situations 4.0 2.0 -.13 -.15 -.19
5. Preference for leaving versus attempting to resolve
dissatisfaction
2.7
1.3
-.06
-.27
-.23
-.16
-.23
a Correlations are based on the combined standardized data from each of the four hospitals
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Spencer
1986
497
TABLE 4
Registered Nurses' Perceptions and Relevant Organizational Char
of Four Nonunion, Short Term, General Care Hospitals with Vary
Numbers of Employee Voice Mechanismsa
High Opportunity Low Opportunity
for Voice
for Voice
I Hospital Hospital
Hospital Hospital
Variables
1
2
3
4
Relevant organizational characteristics
Number of beds
52 250
112
107
Yes
Yes
Yes
Yes
Yes
No
No
Yes
Yes
Yes
No
Yes
No
Yes
No
No
Ombudsman
Yes
No
No
No
Nonmanagement task forces
Question-and-answer program
Survey feedback
Yes
Yes
No
No
No
No
Yes
No
Yes
Yes
Yes
No
6
5
3
3
20.7%
10.8%
21.6%
26.4%
33.3%
40.0%
45.0%
Grievance procedure
Suggestion system
Employee-management meetings
Counseling service
Total number of mechanisms
8.3%
Turnover among registered nurses
Turnover among licensed practical nurses 16.6%
15.0%
Total organizational turnover
Perceived expectancy and effectiveness
Problem resolution expectancies
Supervisory problems
Task problems
Co-worker problems
Pay problems
Promotion problems
Composite expectancy
Management's responsiveness to voice
19.4%
23.1%
28.6%
59.4%
43.5%
34.5%
67.5%b
75.6%b
58.7%
41.3%
55.0%
56.2%
47.7%
49.7%
28.7%
30.9%
36.2%
42.4%
43.3%
58.7%b
48.7%
30.9%
31.5%
38.7%
4.20b
3.88
2.81
3.45
3.48
1.91
2.60
3.68
2.16
2.47
2.87b
3.88b
Utility of efforts to change dissatisfying situations
Effectiveness of voice mechanisms
37.3%
47.5%
Preference for leaving versus attempting to resolve
dissatisfaction
2.62b 2.20 3.34 2.80
a Ns = 8, 46, 33, 15, respectively, for Hospitals
b t-test significant at the p < .01 level (combined
to combined perceptions for Hospitals 3 and 4).
Table 4 reports levels of expectancy
ness of voice mechanisms for hospitals
mechanisms. Hospitals 1 and 2 had six and five mechanisms respectively,
and hospitals 3 and 4 had three mechanisms apiece. When these hospitals
are treated as two groups, t-tests for differences between means indicate
significantly higher levels of expectancy and perceptions of voice effectiveness in the pair of hospitals with the larger numbers of voice mechanisms.
These differences emerged for all variables except expectancies for resolving
problems related to supervision, pay, and promotion. These findings sup-
port the hypothesis that employees of hospitals with large numbers of
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
498
Academy of Management Journal
September
employee voice mechanisms will have high expectancies for problem resolution and will perceive a high level of effectiveness in their organizations'
responses to employee voice.
These results should be interpreted with caution. First, the measures of
employee perception are intercorrelated (see Table 3). Second, Table 4 reports
separate means for each hospital in order to indicate the extent to which
these means differ within each of the two designated groups. As these data
clearly indicate, hospital 1 tended to have the highest levels of expectancy
and perceptions of voice effectiveness and hospital 3 consistently had the
lowest levels. More important, although hospital 2 had more voice mecha-
nisms than hospital 4, the levels of expectancy and perceived effectiveness
of voice mechanisms were only slightly higher for hospital 2 in most cases.
The rate of turnover among registered nurses was also higher for both hospi-
tals 2 and 4 than for the other hospitals. Although extrapolation to other
turnover figures is somewhat problematic, Table 4 does indicate that voluntary turnover for licensed practical nurses and the organization as a whole
was higher for the hospitals with the fewest employee voice mechanisms.
DISCUSSION
The results of study one suggest that the more an org
employees the opportunity to voice dissatisfaction over aspe
in order to change dissatisfying work situations, the greate
that its employees will remain with the organization. Re
turnover rates in short-term, general care hospitals were si
in hospitals with many mechanisms for the voicing of em
faction. In addition, this relationship held with statistical co
variety of known predictors of rate of employee turnover.
The results of study two, although only suggestive, in
numbers of voice mechanisms are associated with high level
expectancies for problem resolution and high perceived effe
organization's procedures for resolving problems. These f
credibility to the construct, number of employee voice m
implies that when employees have many opportunities to
tion they (1) appear to be predisposed to take advantage of th
and (2) appear to consider procedures for problem resolution
Although previous research has consistently found an ef
ization on retention rate, this research did not find such an
ness in the types of organization and occupation examined m
these negative results. Unionization's effects on voice may h
out. Data indicated that unionized hospitals averaged 4.42 mechanisms and
nonunionized hospitals averaged 4.52 mechanisms; in addition, no signifi-
cant differences emerged when each mechanism was considered individually.
Perhaps this finding is a result of a union-threat effect; nonunionized hospi-
tals may provide voice mechanisms that would otherwise be provided by
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Spencer
1986
499
unions in order to reduce dissatisfaction and thus reduce the threat of
unionization.4
The strengths and limitations of this research are worth consider
This research extended previous work on unions' effects on rates of empl
retention by looking at the potential effects of voice mechanisms themselv
Organizations might institute these mechanisms as a result of unionizatio
or in attempts to improve management process. Additional strengths are
this research controlled for a wide variety of potential predictors of rate
turnover and included critical information on employees' perceptions of
organizational characteristics under investigation.
A limitation is that this research does not represent a complete picture
the phenomena involved in the relationship between the options of v
and exit. Two omissions are worthy of note. First, Hirschman's (1970) mo
focused on exiting in silence versus staying and voicing concerns. Ad
tional options open to employees, such as staying in silence and exiting w
voice, need to be incorporated into the model (Barry, 1974; Birch, 1975). I
fact, Birch (1975) argued that in many cases people are more likely to
with voice than to remain and engage in voice, particularly when there is
possibility of retaliation. Second, this research did not address the issu
loyalty. Hirschman (1970) hypothesized that those clients or employees w
have greater loyalty to a product or an organization will be more likely t
others to respond with voice to correct a perceived decline in organizatio
performance. Barry has criticized the loyalty concept as an "ad hoc equat
filler" (1974: 95) that could be inversely as well as positively correlated w
voice (see also Laver, 1976). But it is difficult to deny that loyalty in the f
of behavioral commitment to an organization-entrenchment due to lengt
service-is likely to have an inverse relationship with employees' exit r
whether or not employees have successfully changed dissatisfying states
affairs (Mowday, Porter, & Steers, 1982).
An important limitation, addressed in the introduction, is that the act
quality of the voice mechanisms under study was unknown. The seco
study indirectly addressed this issue by assessing employees' perceptions
the effectiveness of voice mechanisms. This assessment, however, cannot
substitute for an approach that would ideally involve in-depth content analy-
sis of the nature and quality of each employee voice mechanism at each
organizational site under study. This research assessed employees' percep-
tions of the effectiveness of voice mechanisms at only four of the organiza-
tions represented in the larger study; in addition, sample sizes for these
analyses were small and measures whose psychometric properties were relatively unknown were constructed and used.
4 Curtin (1970) found that unsuccessful unionization attempts generally result in improved
communication practices; in over 60 percent of the companies these practices usually included:
"establishment of regular meetings with the employees, institution of a formal grievance
procedure, formation of an office committee, concerted effort to deal promptly with employee
complaints" (1970: 67).
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
500
Academy of Managemenet Journal
September
Another limitation of this research is that although significant relationships emerged from the analysis, they are not very strong. Also, these relationships are purely associational and thus causality cannot be assumed. A
related problem concerns synchronicity of the data; assessment of number of
voice mechanisms was made at the end of the one-year period for which
retention data was collected. Another limitation is that the research was
conducted in one particular type of organization and on one specific o
tional classification. A further limitation is the relatively low respon
however, comparing the responses obtained against population demog
provided by each of the hospitals and bed sizes obtained from the Na
Center for Health Statistics indicated that the responses were represe
of the population of individuals and of hospitals involved. Future rese
should incorporate tests of the hypotheses in different populations a
research methodologies where causality could be inferred as well a
gies of data collection that would improve rates of response.
The findings of this study reinforce Steers and Mowday's (1981) a
tion that investigations of the process of employee turnover need to
employees' efforts to change unsatisfactory work situations. On the o
tional level of analysis, future research should consider not only
voice mechanisms and their quality, but also informal organizatio
tures that create and sustain those mechanisms. It is not inconceivable that
the rough assessment of total numbers of voice mechanisms that this s
used is tangible evidence of a potentially salient organizational comp
such as managerial philosophy. The individual level of analysis need
model of the process of turnover decisions that incorporates the two m
alternatives available to employees: (1) searching for alternative em
ment to exit from a dissatisfying work situation, and (2) attempting to
a dissatisfying work situation in an effort to remain. Mobley's (1977) m
of the process of turnover decisions focused only on the first cou
action. But what causes employees to take one course of action before the
other or to take both courses of action simultaneously? Perhaps employees'
commitment and expectancies concerning the effectiveness of voice mechanisms and procedures for problem resolution would play a significant role in
this process.
If future research supports the hypotheses examined in this study, one
practical implication is that organizations may be able to effectively reduce
employee turnover by increasing the sophistication of their processes for
resolving complaints. Employees who search for alternative employment
without attempting to change dissatisfying work situations give organiza-
tions no indication that anything is wrong until after they find other work
and hand in their resignations. By listening to, encouraging, and providing
mechanisms for employees who want to change dissatisfying work situations,
perhaps organizations could prevent employees' forming an intent to leave.
In addition, organizations that engage in such activities should be able to
obtain information from employees that could improve the effectiveness of
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Spencer
1986
501
these organizations (Freeman, 1976; Hirschman, 1976), information that is
not available when employees silently leave.
REFERENCES
Aram, J. O., & Salipante, P. F., Jr. 1981. An evaluation of organizational due
resolution of employee/employer conflict. Academy of Management Rev
Barry, B. 1974. Review article: Exit, voice, and loyalty. British Journal of Po
79-107.
Becker, B. 1978. Hospital unionism and employment stability. IndustrialRel
Birch, A. H. 1975. Economic models in political science: The case of "Exit, v
British Journal of Political Science, 5: 69-82.
Curtin, E. R. 1970. White collar unionism. Personnel policy study No. 220,
Conference Board, New York.
Farrell, D. 1983. Exit, voice, loyalty, and neglect as responses to job dissatisfaction: A mul
mensional scaling study. Academy of Management Journal, 26: 596-607.
Farrell, D., & Peterson, J. C. 1982. Patterns of political behavior in organizations. Acade
Management Review, 7: 403-412.
Freeman, R. B. 1976. Individual mobility and union voice in the labor market. American Economic Review, 66: 361-368.
Freeman, R. B., & Medoff, J. L. 1984. What do unions do? New York: Basic Books.
Graham, J. W. 1986. Principled organizational dissent: A theoretical essay. In B. M. Staw
Cummings (Eds.), Research in Organizational Behavior, vol. 8: In press. Greenwich,
Conn.: JAI Press.
Greenberg, J., & Folger, R. 1983. Procedural justice, participation, and the fair process effect in
groups and organizations. In P. B. Paulus (Ed.), Basic group processes: 235-256. New
York: Springer-Verlag.
Hammer, T. H., Landau, J. C., & Stern, R. N. 1981. Absenteeism when workers have a voice: The
case of employee ownership. Journal of Applied Psychology, 66: 561-573.
Hirschman, A. 0. 1970. Exit, voice, and loyalty. Cambridge, Mass.: Harvard University Press.
Hirschman, A. 0. 1974. Exit, voice, and loyalty: Further reflections and a survey of recent
contributions. Social Science Information, 13(1): 7-26.
Hirschman, A. 0. 1976. Some uses of the exit-voice approach-Discussion. American Economic
Review, 66: 386-391.
Kochan, T. A., & Helfman, D. E. 1981. The effects of collective bargaining on economic
behavioral job outcomes. In R. G. Ehrenberg (Ed.), Research in labor economics: 321
Greenwich, Conn.: JAI Press.
Laver, M. 1976. "Exit, voice, and loyalty" revisited. British Journal of Political Science,
463-482.
March, J. G., & Simon, H. A. 1958. Organizations. New York: John W
Mobley, W. H. 1977. Intermediate linkages in the relationship betwee
employee turnover. Journal of Applied Psychology, 62: 237-240.
Mobley, W. H., Griffeth, R. W., Hand, H. H., & Meglino, B. M. 1979.
analysis of the employee turnover process. Psychological Bulletin
Mowday, R. T., Porter, L. W., & Steers, R. M. 1982. Employee-organi
psychology of commitment, absenteeism, and turnover. New York: Academic Press.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
Academy of Management Journal
502
September
Price, J. L. 1977. The study of turnover. Ames: Iowa State University Press.
Steers, R. M., & Mowday, R. T. 1981. Employee turnover and post decision accommodation
processes. In L. L. Cummings & B. M. Staw (Eds.), Research in Organizational Behavior:
235-281. Greenwich, Conn.: JAI Press.
Ulman, L., & Sorensen, E. 1984. Exit, voice, and muscle: A note. Industrial Relations, 23:
424-428.
APPENDIX
All items used 7-point response formats; "R" indicates reverse scoring.
Management's responsiveness to voice mechanisms:
The management of my organization is not interested in resolving indivi
problems. (R)
The management of my organization encourages employees to voice their problems.
The management of my organization has not provided enough mechanisms (for example,
suggestion systems, grievance procedures, etc.) to allow employees to effectively voice
their dissatisfaction. (R)
I get the feeling that my superior does not want to hear about my complaints. (R)
The personnel manager is open to receiving complaints.
My boss comes around regularly to keep in touch with any complaints that I may have.
The personnel manager makes an effort to keep in touch with any complaints that workers
have.
I feel intimidated by my superiors when pursuing a grievance. (R)
Management views grievances as a challenge to their authority. (R)
The organization encourages suggestions to improve situations that are dissatisfying to
employees.
Effectiveness of voice mechanisms:
The mechanisms to resolve employee problems of my organization are very effectiv
Utility of efforts to change dissatisfying situations:
Attempting to change something at work that dissatisfies me would be a waste of tim
Preference for leaving versus attempting to resolve dissatisfaction:
I would rather find a job elsewhere than attempt to change something at work that dis
fies me.
Daniel G. Spencer is an associate professor of business at the University of Kansas,
Lawrence, Kansas. He received his Ph.D. degree in organization and management fro
the University of Oregon. His current research interests include turnover, absenteeism
and processes of resolution of employee-organization conflicts.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:02:28 +00:00
All use subject to https://about.jstor.org/terms
How to Keep Your Best Employees: Developing an Effective Retention Policy [and
Executive Commentary]
Author(s): Terence R. Mitchell, Brooks C. Holtom, Thomas W. Lee and Ted Graske
Source: The Academy of Management Executive (1993-2005) , Nov., 2001, Vol. 15, No. 4,
Themes: Business Strategies and Employee Development (Nov., 2001), pp. 96-109
Published by: Academy of Management
Stable URL: https://www.jstor.org/stable/4165789
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/4165789?seq=1&cid=pdfreference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to
The Academy of Management Executive (1993-2005)
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
Academy of Management Executive, 2001, Vol. 15, No. 4
How to keep your best
employees: Developing an
effective retention policy
Terence R. Mitchell, Brooks C. Holtom, and Thomas W. Lee
Executive Overview
The competition to retain key employees is intense. Top-level executives and HR
departments spend large amounts of time, effort, and money trying to figure out
how to keep their people from leaving. This article describes some new research
and its implications for managing turnover and retention. These ideas challenge the
conventional wisdom that dissatisfied people leave and money makes them stay. People
often leave for reasons unrelated to their jobs. In many cases, unexpected events or
shocks are the cause. Employees also often stay because of attachments and their sense
of fit, both on the job and in their community. We discuss these ideas and make
recommendations for integrating them into a comprehensive retention plan.
.
............................................................................................................................................................
tarily, he or she believes, at that moment, that
Voluntary turnover is a huge problem for many
organizations today.' A Wall Street Journal article
leaving the job is the right thing to do. However,
states: "Job hopping prevails amid a cornucopia of
transitions to another job or situation (e.g., stay-at-
vacancies."2 The Society of Human Resource Man-
home parent, additional education) take a per-
agement, a professional organization dedicated to
sonal toll. Going to a new job is stressful. There is
the problems facing managers, says that the reten-
uncertainty and ambiguity. The employee, along
with family members, must make numerous ad-
tion issue is its "hottest topic."3
justments, especially if relocation is involved. New
Some industries or groups are particularly at
living accommodations, schooling for children,
risk. The Big 5 accounting firms are described as
and spousal reemployment are just some of the
having "raging turnover," particularly among fe-
possible hurdles. In addition, friends may be left
male employees. While up to 50 percent of the new
behind. Some people estimate that it can take up to
hires may be women, only about 5 percent will
a year for adjustments to be made and a career to
make it to partner. High-technology companies
get back on track.5
also face acute losses and shortages. One report
At the organizational level, turnover inflicts nu-
suggests the average tenure for people in high-
merous costs. Departing employees often take with
technology positions is one year. In addition, looking at the whole U.S. workforce, approximately
them valuable knowledge and expertise gained
half of the workers expect to leave their jobs in the
through experience. Often they have established
close relationships with clients. In addition to
next five years.4
There are many reasons why people voluntarily
these indirect or less quantifiable costs, organiza-
leave organizations. Some are personal: changes
tions face many costs directly related to turnover,
in family situation, a desire to learn a new skill or
including exit interview time and administrative
trade, or an unsolicited job offer. Other reasons are
requirements, payout of unused vacation time, and
influenced by the employing organization: observ-
the cost of temporary workers or overtime for co-
ing the unfair treatment of a coworker, being
workers asked to fill in. Replacement costs include
passed over for promotion, or being asked to do
advertising, processing of candidates, interview-
something against one's beliefs. Turnover is a
ing, and selection. Finally, training costs-both
problem because it imposes extensive costs on
formal and informal-add to the overall burden.
both individuals and organizations.
One well-known New York law firm estimates the
replacement costs for an associate at as much as
At the individual level, if a person leaves volun96
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
2001
Mitchell,
Holtom,
$200,000.6 A BusinessWeek study estimated that re-
and
Lee
97
factors, including pay, benefits, and other rewards,
placement costs alone are over $10,000 for about
influence job satisfaction, as do structural and pro-
half of all jobs, and for 20 percent of all jobs are
cedural factors reflecting autonomy or fairness. In
over $30,000.7 In a separate study, the Hay Group
terms of what initiates the turnover process, job dis-
found replacement costs are about 50 to 60 percent
satisfaction has been described as the most impor-
of a person's annual salary.8 Other costs may not
tant and frequent cause. Thus it is good, solid advice
be so easily quantified. For example, understaffing
to design jobs and manage work environments to
because of excessive turnover among corrections of-
maintain a high level of job satisfaction.
ficers has been blamed for high-profile prison es-
Once dissatisfaction sets in, an employee pre-
capes.9 Reduced effectiveness of cockpit crews that
sumably looks around for other work alternatives.
are reorganized frequently because of turnover is
The employee may conduct a job search and un-
another example of hidden organizational costs.'0
cover some interesting options. Both perceived and
actual alternatives can influence this process. At
A BusinessWeek study estimated that
replacement costs alone are over $10,000
this point, it appears that the underlying thought
is, "I intend to leave." If alternatives are judged to
be favorable in comparison to the present job, the
for about half of all jobs, and for 20
person is predicted to leave. If not, the person
percent of all jobs are over $30,000.
stays. Thus attitudes about one's current job and
Organizations of all sizes and types are recognizing that they are engaged in a struggle to retain
talent, and are actively trying to do something
about it. One recent report declares that one in 10
big businesses has a full-time person assigned to
retention." The big question is-what should this
person do? How can an organization attract and
retain its employees-especially its most valuable
and irreplaceable ones?
the availability of alternatives are seen as the
antecedents for voluntary turnover.'3 Satisfied employees will be less attracted by alternative jobs.
Research over the last 15 years has expanded
this focus somewhat. Other attitudes have been
studied, such as organizational commitment, job
involvement, and perceived organizational sup-
port. How and when people search for alternatives
have also been explored as has how modern conceptions of a career influence propensity to remain
in a firm.'4 It appears that many employees expect
to have numerous jobs during their careers. Some
authors have examined personality predictors of
turnover, such as conscientiousness, while others
The Prevailing Wisdom
Management scholars have learned a lot about
have examined the influence of psychological con-
voluntary turnover over the years; thousands of
tract violations on turnover intentions and actual
studies have been conducted on the topic.'2 Both
of this work, some clear and enduring principles
turnover.'5 Violating initial job expectations can
decrease trust, cause anger, and precipitate the
turnover process. Still, while these studies have
added to our knowledge, the prevailing wisdom
have evolved.
has remained relatively unchanged for 50 years.
the academic and practitioner literatures have
made contributions to our knowledge. And from all
One of the best recent reviews summarizes this
state of affairs: "Together with turnover intentions
Academic perspective
Two major factors have captured most of the aca-
demic attention: job satisfaction and job alternatives. People who are satisfied with their jobs (e.g.,
evaluate positively their pay, supervision, chances
for promotion, work environment, and tasks they
do) will stay, while those who aren't will leave.
Given the same level of dissatisfaction, people
with more alternatives will be more likely to leave
than those with fewer alternatives.
Considerable research has explored these relationships in detail. There are many causes of job
satisfaction, such as job enrichment, good super-
and cognitions, affect and alternatives have been
the prominent antecedents to turnover."'6
Two other comments on this academic literature
are necessary. First, in most cases, staying is seen
as the simple obverse of leaving. That is, people
who are satisfied with their jobs and/or have few
alternatives will remain on the job. The same au-
thors quoted above concluded that "relatively less
turnover research has focused specifically on how
an employee decides to remain with an organization and what determines this attachment."'7 This
point is critical for our work because we believe
vision, clear roles, and met expectations. Dissatisfaction is associated with job stress, repetitive
that staying and leaving involve different psychological and emotional processes.
The second point is that researchers' collective
work, role ambiguity, and role overload. Economic
effort to predict turnover has not been very suc-
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
98
Academy
of
Management
Executive
November
cessful. Even the more complex theories, with mul-
with their jobs? We also knew that many people
tiple attitudes and assessments of perceived alter-
searched for alternatives and compared them with
natives, leave about 75 percent of the variance in
their present job. But were there people who left
turnover unexplained.18 It was this rather gloomy
without searching or making such comparisons? We
assessment that prompted us to pursue new ways
to understand this topic.
spent a substantial amount of time talking to people
about their personal experiences of leaving jobs. We
read the practitioner and scientific literature, interviewed hundreds of people who had left jobs in a
Practitioner perspective
wide variety of occupations, and surveyed thou-
Much of the practitioner literature has also re-
sands of others-many of whom subsequently left
flected the prevailing wisdom. Assessing job sat-
their jobs.
isfaction and making changes based on this infor-
Some clear themes emerged from these investi-
mation has been an organizational strategy for
gations. First, many people said they initially
retaining employees and reducing turnover for de-
thought about leaving in response to some particular event. We call such an event a shock to the
system. Frequently cited shocks were mergers, unsolicited job offers, friends' leaving, having a baby,
spouse relocation, a poor performance appraisal,
and administrative changes. What was of interest
to us were the ways these shocks differed. Some
were seen as positive (e.g., job offer, pregnancy);
some were negative (e.g., friends' leaving, poor
appraisal); and some were neutral (e.g., spouse
relocation, administrative changes). Some were
expected (e.g., being accepted to graduate school)
and some unexpected (e.g., changes in compensation plan). Whether expected or unexpected, when
a shock occurred, serious thoughts about leaving
followed. Some violated mutual expectations (e.g.,
the psychological contract), while many did not.
Some were external to the job, while others happened at work. There was a great variety in what
precipitated thoughts of leaving. Some of these
events, eventually or immediately, caused job dissatisfaction, but some did not.
cades. Sears, for example, has been doing this for
over 50 years.'9 Since perceived or actual alternatives are often beyond an organization's control,
they have received less practitioner attention than
job satisfaction.
We should also point out that much of the practitioner literature on staying and leaving reflects
an economic perspective. In terms of attracting
employees in today's job market, signing bonuses
have become commonplace.20 For keeping employees, "retention bonuses have become the tool of
choice."2' Added to this financial arsenal are housing allowances, profit sharing, and spousal financial and job assistance.22 High-tech firms-which
are extremely dependent on human capital-have
often led this charge. They attract employees with
stock-option riches; when market downturns
threaten this carrot, they reprice or reissue the
options.23 In short, one way to make a job attractive
is to throw money at people.
Before concluding this review, we want to make
it clear that the prevailing wisdom has substantial
truth behind it. Satisfaction is important for staying, and all the good management techniques that
have been espoused over the years probably contribute to it. June Delano, at Eastman Kodak, in addressing the turnover problem, recommends things
like frequent and honest communication, fair and
equitable compensation, and clear performance expectations.24 In addition, pay and other financial incentives work to increase motivation, commitment,
and satisfaction. That having been said, we believe
that the prevailing perspectives on leaving and staying are too narrow. We've spent the last 11 years
developing and testing new ideas about employee
retention. Here is what we've found.
Whether expected or unexpected, when a
shock occurred, serious thoughts about
leaving followed.
A second theme that emerged was that quite a
few people left their jobs without searching for
another one and making comparisons with their
present job. This process seemed to occur in two
distinct ways. Some people seemed to have plans
or scripts that were triggered by some event. They
knew if they got pregnant or got accepted to graduate school, they would leave. Or they already had
a plan in place based on past experience (e.g., to
Shocks and the Unfolding Model
Our initial focus was on why and how people leave
their jobs. We knew that accumulated job dissatisfaction was one reason. But were there other reasons? Did some people leave who were satisfied
leave if a small start-up were bought out by a big
company). Other people did not have plans or
scripts, but left because of an event that was so
shocking or negative or unexpected, that they just
packed up and exited without making comparisons. These people may have believed that they
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
2001
Mitchell,
Holtom,
could get another job elsewhere, but they left fairly
and
Lee
99
place. The event happens and the person thinks
quickly and violated the oft-given advice to never
about leaving, calls up the plan, and leaves. Dis-
leave a job without another in hand.
satisfaction is not the initiating cause, and there is
So it seemed to us that the process of leaving
no search for alternatives.
was far more complex than the accepted model,
where job dissatisfaction and the comparison of
alternatives result in departure. Simply put, there
were many different ways or paths that people
used to leave their jobs. In an attempt to construct
some orderliness out of this variety and complexity, we developed the unfolding model.
The unfolding model describes four paths people
take when they leave a job. These paths unfold over
time at different rates, and they involve different
processes and behaviors.25 We have many examples
of these paths, based on our empirical research in
Path 2: Leaving without a plan
Path 2 also is started by a shock, but there is no
plan or script in place. Again, the person leaves
without searching for alternatives. Bill Russell, a
famous basketball player and coach, left a coaching job soon after a game where he found himself
yelling at his players to kill the opposition.30 The
head of the computer services for a large organization quit the day after a person he didn't like was
hospitals, banks, accounting firms, and retail gro-
named his boss. A nurse was denied her request
cery stores. Where possible, we present high-profile
for six months' leave; an employee whose father be-
examples that have appeared in the news.
came ill needed to move closer to his home; a trans-
Three of the paths are initiated by shocks-
fer request was denied; a friend was laid off; an
events that caused people to question their staying
individual was passed over for a promised promo-
with a firm-not job dissatisfaction. Examples
tion. In all these cases, the people left without con-
abound in the popular press. A man is pressured to
sidering alternative jobs. In Path 2, the shocking
fly to an out-of-town meeting when his pregnant wife
event is often negative and involves such a violation
is three days overdue and he misses the delivery of
of expectations that negative emotions like distrust
their baby. It was the "deal breaker, the event so
or anger accompany the leaving process.
egregious that it drove Mr. Walton to quit."26 A
woman likes her job at a plastics start-up, but starts
Bill Russell, a famous basketball player
looking for another job when she is asked to clean
the bathrooms.27 A man's boss trims $1.25 off a travelexpense reimbursement request for laundering a
shirt while on a trip.28 These events, or shocks,
prompt people to think about leaving.
at his players to kill the opposition.
Path 1: Following a plan
Path 3: Leaving for something better
Path 1 is characterized by a shock to the system
Path 3 commences with a shock that leads to rela-
and a plan or a script already in place for leaving.
tive, possibly minimal job dissatisfaction. In turn,
Some people know that they will leave if they or
the person considers alternatives and eventually
and coach, left a coaching job soon after
a game where he found himself yelling
their spouse gets pregnant or if they get accepted
leaves, usually with another job in hand (or in
into some sort of educational program. Temporary
some cases, almost a sure thing). Patty Stonesifer
or part-time employees often work for some spe-
was a Microsoft executive. When she turned 40, she
cific amount of money, then quit. Many of today's
decided she wanted to be with her family more.
technical employees plan to leave their jobs after
She lined up some consulting with DreamWorks,
12 months; when the time is up, they quit, confident
and quit her job at Microsoft.3' David Falk, of
that they can find something else whenever they
ProServ, found himself in the hospital with a blood
want.29 One of the best examples we encountered
clot. While he rested and reflected about the diffi-
was the employee in one organization who took a
culties he had with his boss, Donald Dell, he de-
hiring bonus, received the standard three month's
cided to quit and start his own business.32 In our
training, got a bonus for finishing the training,
interviews, we found a considerable number of
then quit, without another job lined up. And that
Path 3 leavers report the shock as being an unso-
was the employee's plan from the start. In sum,
licited job offer. Many of these people are not dis-
Path 1 is precipitated by a shock. The shock can be
satisfied with their jobs. Mike Mitchell, a Phelps
expected (e.g., end of training) or unexpected (e.g.,
Dodge Corp. employee, was "perfectly happy"
spouse has to relocate). It can be positive (e.g., preg-
when he got an offer he couldn't refuse.33 Path 3
nancy), neutral (e.g., merger), or negative (e.g., sex-
dissatisfaction is often relative; the people like
where they are, but the alternative is better. The
ual harassment). The key point is that a plan is in
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
100
Academy
of
Management
Executive
November
two key time periods: the time between some event
or accumulating dissatisfaction and the decision
key is that the shock triggers the relative dissatisfaction, which leads to consideration of alterna-
to be very thoughtful and quite complex.
to leave, and the time between the decision to leave
and actual leaving. The first phase involves checking for a script or plan and seeing how a shock may
Path 4: Leaving an unsatisfying job
happen fairly quickly in Paths 1 and 2, where time is
tives, which results in quitting. This process tends
fit with the rest of one's feelings about a job. These
measured in days or weeks. It can also involve
Path 4 is initiated by accumulated job dissatisfac-
accumulation of discontent, which happens more
tion. There are two different ways in which this
slowly in Paths 4A and 4B, where time is typically
dissatisfaction leads to quitting. In what we call
measured in months and years. The second phase
Path 4A, people become dissatisfied and leave
involves searching for and evaluating altematives.
without searching (similar to Path 2 people, but
This process is often lengthy in Paths 3 and 4B, and,
without the shocking event). Jim Manzi was the
therefore, they take the longest time to unfold.
CEO for Lotus Development Corp. when IBM took it
over. Initially enthusiastic about the merger, he
found over the next half year that a variety of things
happened he didn't like, especially the direction and
The Complexities of Leaving a Job
vision for the new merged group with which he
We gathered data from five separate samples that
worked. He decided to quit-without another job in
address the principles of the unfolding model. In
hand.34 We interviewed people who quit because
the first two studies (one qualitative and the other
over time they got bored, failed to make friends, or
quantitative), we focused on leavers and the paths
changed in such ways that their current job was
they took. In the first study, we interviewed 44
clearly not where they wanted to be.
nurses who left eight hospitals in a large metro-
Path 4B also involves accumulated job dissatis-
politan area.36 In the second study, we surveyed
faction, but here the person does look for and eval-
229 voluntary leavers who left the Big 5 accounting
uate alternatives. He or she leaves upon finding a
firms.37 We then conducted two more studies where
more attractive option. Note that this path reflects
the prevailing wisdom we described above. Dis-
we gave questionnaires to a large sample of employees and then tracked who left and who stayed.
satisfaction leads to search, which leads to leav-
In the third study, we examined 177 employees at
ing. One computer-system consultant reported that
over six months he was shuffled about and felt
eight stores of a retail grocery chain. The fourth
study involved 208 employees of a community-
"like a piece of meat." Eventually his dissatisfac-
based hospital.38 The participants in these last two
tion led him to search for a better job and ulti-
predictive studies were given a list of events that
mately to quit.35 People following this path reported to us examples of cost cutting, increased
workloads, or continuing problems with work
were previously determined to be shocks, based on
schedules or work assignments that over time led
asked to indicate whether they had experienced
to dissatisfaction, search, and leaving.
these events and how much the experience made
interviews and focus groups. These people were
Table 1 presents a summary of the four paths,
them think about leaving their jobs. They were
including the attributes of each and the amount of
also asked whether they had scripts or plans for
time it takes for paths to unfold. There seem to be
leaving in place, in case the event did occur. In our
Table 1
The Unfolding Model Paths
Path
1
2
3
4A
4B
Following a Leaving without Leaving for Leaving without Leaving for
Attribute plan a plan something better a plan something better
Initiating event Shock Shock Shock Job dissatisfaction Job dissatisfaction
Script/Plan
Relative
job
Yes
No
No
No
Yes
No
Yes
No
Yes
Yes
No
Yes
No
Yes
Medium
Long
dissatisfaction
Alternative
No
search
Time
Very
short
Short
Long
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
2001
Mitchell,
Holtom,
and
Lee
101
fifth study, we did both. We had 841 employees at
tion, unsolicited job offer, pregnancy), while 99 left
a large financial center fill out questionnaires,
because of internal shocks (e.g., poor evaluation,
then interviewed 72 of them who subsequently left
merger, disagreement with boss). Thus shocks that
during the next year. We will call these the nurse,
did not involve money and were often out of the
accountant, grocery store, hospital, and bank stud-
organizations' control frequently precipitated the
ies in the following summary comments.
Table 2 presents the paths taken by the employ-
ees who left their jobs in the nurse, accountant,
and bank studies.39 One initial comment is neces-
sary. Using either questionnaires (accountants) or
interviews (bank), we were able to classify over 95
percent of the leavers reliably (i.e., agreement
leaving process.4' Again, the conventional wisdom's reliance on a primarily economic explana-
tion for turnover may also be too simple.
We can look at the reactions to shocks in more
detail by analyzing the data on thoughts about
leaving from the grocery, hospital, and bank studies. The internal organizational shocks that caused
across several independent, impartial judges) into
the most thinking about leaving were: being en-
a particular path. Our data suggest almost everyone leaves via one of these paths.
But how and why people left is what is impor-
couraged to leave (but not fired); having a major
for promotion; or receiving an unexpected negative
tant. Note that Paths 3 and 4B are the most fre-
performance evaluation.
disagreement with one's boss; being passed over
quently used in all three studies. Both of these
Mergers, reorganization into work teams, the
paths involve a search and evaluation of alterna-
completion of a training or education program,
tives, with about 75 percent of people searching
or disagreement with a coworker produced fewer
before leaving. Also note that the process of leav-
strong thoughts about leaving. The external
ing is initiated for more people by a shock (Paths 1,
2 and 3, which account for 63 percent) than by
shocks that caused the most thoughts of leaving
were the unexpected job offer (which probably in-
accumulated dissatisfaction (Paths 4A and 4B,
which account for 37 percent). In combination,
cludes pay as an issue to consider), followed by a
these data help to explain why the traditionally
leaving were significantly related to intention to
studied variables of job dissatisfaction and alter-
leave and actual leaving in these three samples.
spouse's relocation. Importantly, thoughts about
natives aren't strongly predictive of turnover. A lot
We conclude this section with two final points.
of people leave without alternatives, or as a result
First, the job satisfaction of leavers, while signifi-
of some shocking event that may not be associated
cantly lower than stayers, was above 3.0 on a five-
with job dissatisfaction.40 Our findings show that
the leaving process is considerably more complex
than reflected in the conventional wisdom.
The content of these shocks is important to understand what went on. Very few people left specifically because of a shock involving money (e.g.,
lower-than-expected raise, change in benefits,
someone else in a similar job makes more). Interestingly, no nurses, four out of 69 bank employees,
and 14 out of 212 accountants cited a monetary
shock as the reason for leaving. Slightly over half
the people (128) left because of shocks that were
external to the organization (e.g., spouse reloca-
point scale for the four samples where it was measured (accountant, grocery, hospital, and bank).
Thus many people leave who are relatively satisfied with their jobs. Second, the paths unfold at
different rates, much the way we have suggested.
People in Paths 1 and 2 (who experienced a shock)
reported that they left more quickly than those in
Path 3, who experienced a shock, but also searched
for alternatives. It is hard to assess time to leave
for Paths 4A and 4B people because the job dissatisfaction that precipitates leaving (4A) or search
(4B) can take a very long time (in two of our studies
Path 3 occurred more quickly than Path 4B). Fur-
Table 2
Number of Leavers by Paths
Path
1
2
3
4A
4B
Following a Leaving without Leaving for Leaving without Leaving for
Sample plan a plan something better a plan something better Total
Nurses
6
6
14
8
10
44
Accountants
6
7
136
8
55
212
Bank
employees
4
6
19
6
34
69
Total
16
19
169
22
99
325
Percent
5
6
52
7
30
100
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
102
Academy
of
Management
ther, Path 4B, which involves search, takes longer
than Path 4A, which doesn't.
In summary, we have added to the accepted
principles that dissatisfied people who have options quit. Our research shows that many people
leave as a result of shocks. Many of the shocks are
external and don't involve money. Many people
are relatively satisfied when they leave. All the
paths are taken and occur with different speeds,
suggesting that a rich diversity characterizes voluntary turnover.
Executive
November
scribe as nonaffective or nonattitudinal. Many peo-
ple stay because of attachments they have to people (e.g., coworkers, employee network groups), or
activities like the company softball team or sponsored community-service activities.44 Leaving a job
often requires individuals to sacrifice or give up
perks, routines, or projects to which they have
grown accustomed. People reported to us that they
weren't particularly satisfied or dissatisfied, but
that wasn't the issue that was important. It was the
idea that they couldn't leave right now because too
many things kept them entrenched in their jobs.
Job Embeddedness and Staying
The traditional perspective of why people stay on a
job includes the factors from the turnover research
(e.g., satisfied with job, no alternatives), plus other
attitudes or affective reactions (positive feelings)
about the job, including organizational commitment and perceived organizational support. If we
like our job, are committed to our organization, and
believe they are committed to us, we stay. As we
mentioned earlier, these variables have been
shown to be related to turnover.42 However, our research on the unfolding model, and our reading of
both the scientific and practitioner literature (we will
treat them together in this section), suggested that
other factors were operating to keep people in their
jobs. Two streams of thought seemed important.
First, it was clear that nonwork factors could
keep someone on the job. Family pressures, community commitments, and many other off-the-job
variables can influence an employee's likelihood
of staying with an organization. For example, one
issue that is receiving increased attention is worklife balance. Employees today want time to attend
to their personal and family activities off the job.
Related issues include work-related stress and frequent travel. One study reported that 40 percent of
Americans feel they have excessive workloads
that negatively influence their lives off the job. In
short, accumulating evidence suggests family,
hobbies, and church commitments can keep one on
the job (especially if one would have to relocate
when changing jobs), and lack of attention to these
outside commitments by one's employer might
prompt one to leave.43
Job embeddedness was the term we used to summarize a broad constellation of factors influencing
retention, including those mentioned above. Think
of a person's life as being like a web, with various
nodes or objects on the web representing people,
things, groups, and institutions. One's job is in the
middle of the web. The number of connections may
vary, the distance or strength of the attachments
may vary, and the overall connectedness of the web
(attachments among nodes) may vary. Leaving a job
may greatly disturb that web, and it was our belief
that it was this web-like quality, or embeddedness,
that often keeps people on the job.
Based on our research on the unfolding model,
people's stories about why they stay in a job, and our
ongoing deliberations about these topics, we believe
that job embeddedness consists of three factors:
links, fit, and sacrifice. In addition, each of these
factors can occur on or off the job. A review of these
factors and some examples are presented below.
Links
Links are the connections between a person and
other people, groups, or organizations. Many organizations are coming to see these links as im-
portant on the job. Many CPA firms are using men-
tor systems and designated role models to
increase the attachment of their female employees. They are also encouraging employees to
choose their own clients.45 These are link-building
activities. Engelhard, a diversified manufacturing
company in New Jersey, uses mentors and what
they call a Buddy System.46 We also know that
many organizations have increased their use of
teams, but some organizations, such as the soft-
One study reported that 40 percent of
Americans feel they have excessive
workloads that negatively influence their
lives off the job.
The second big area omitted from the traditional
perspective was all the things that one might de-
ware development firm WRQ, use teams strategically to increase this networking bond.47
Off-the-job links can influence the employee as
well. The number of family, relatives, friends, and
other types of links established through hobbies,
church, or community activities can embed a person.
Organizations can facilitate such links. Some firms
give employees time off to volunteer in their communities.48 SAS Institute, a statistical software firm, of-
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
2001
Mitchell,
Holtom,
and
Lee
103
fers a discount on homes purchased in a nearby
Other employers provide a pleasant organizational
community. Other companies provide new employ-
environment, funds to personalize offices, an atrium,
ees with information on resources and activities in
lovely views, exercise facilities, and massages.55
their communities. Thus a variety of organizational
strategies are available to increase links.
ities with a more long-term or personal focus.
Some of the sacrifices involve programs or activ-
Many companies (e.g., Sears, Citibank, and Inter-
The number of family, relatives, friends,
and other types of links established
through hobbies, church, or community
activities can embed a person.
national Paper) have long-term development plans
they work out with employees.56 They invest in the
employees' training and growth. Other companies
offer personal-development funds to retain top employees, who can use the funds to obtain training
on any topic they feel will help their performance.
Equally important are the things an employee
would give up that relate to their personal, off-the-
Fit
job situation. Child-care support through the use of
Fit is defined as an employee's perceived compat-
on-site centers, vouchers, or assistance is frequently
ibility with job, organization, and community. The
used. The Child Care Action Campaign reports that
research on organizational fit has been extensive
10,000 firms provide some sort of help. Twenty-seven
and clear. The more individuals fit with their jobs,
percent of job applicants to Union Bank in California
said that the day-care center was a big attraction,
coworkers, and larger corporate cultures, the lower
the turnover. Perceptions of fit are especially important during the early stages of adjustment to a
job.49 A recent survey conducted by Caliper of 180
executives who left their jobs reported that "nearly
40 percent of employees essentially said they are
resigning because their jobs just did not f it."50
The fit with one's off-the-job environment is also
important. Does the employee like to ski, fish, sing
in a choir, or attend local theater? Do they have a
child with soccer practice after school, or a parent
who needs to be visited? Both recreational preferences and interpersonal responsibilities can be
and the turnover for those using the center was 2.2
percent, compared with 9.5 percent for non-users.57
Healthcare clinics can also help embed an employee.58 Booz Allen & Hamilton builds loyalty through its
concern for the demands of one's personal life by
providing time off when it is needed.59
In summary, job embeddedness is a relatively
new idea. It reflects those on- and off-the-job factors that keep people in their current positions.
Table 3 summarizes examples of how employers
can help embed employees in jobs using the various dimensions to guide their specific efforts.
better fulfilled if the employee has some choice
over when and where they work. The use of flex-
time is one organizational option that helps this fit
and is widely used. For example, a recent survey of
614 accounting firms worldwide ranked flexible
schedules as the most effective retention tool.5'
Other strategies, like part-time work, telecommuting, and 35-hour work weeks, give people a better
chance to be in sync with their off-the-job activities. One retention champion summarizes these
ideas by saying that companies need to "recruit"
their good performers and that jobs should be customized. "One size fits one" was his motto.52
Embeddedness limits leaving
To capture the impact of these diverse forces that
keep people in their jobs, we developed a measure
of job embeddedness that we have used in three
separate studies.60 It is measured with a survey
designed to assess fit, links, and sacrifice, both on
and off the job. The three studies were mentioned
earlier: a retail grocery story chain (N = 177), a
hospital (N = 208), and a banking center (N = 841).
In each study, we surveyed a large group of employees and measured job embeddedness, job satisfaction, organizational commitment, perceived
job alternatives, job search, and intention to leave.
Sacrifice
Sacrifice reflects the cost of what people have to
We then tracked who left over the next year and
conducted exit interviews. A clear and compelling
give up if they leave a job. Many of these sacrifices
involve foregoing financial incentives tied to lon-
story emerged.
gevity. Retention bonuses, retirement funds, stock
options, and golden handcuffs help to keep people
on the job.53 But many of these benefits can be
matched by competing firms. More subtle approaches are also being used. For example, many
companies have minisabbaticals tied to longevity.54
tention to leave and actual turnover in all three or-
Job embeddedness was negatively related to inganizations. People who reported being more embedded in their jobs were less likely to leave their
organizations. Of greater importance was the fact
that, in all three studies, job embeddedness added to
the prediction of turnover, over and above the pre-
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
104
Academy
of
Management
Executive
November
Table 3
Job-Embeddedness Suggestions by Dimension
Dimension
Organization
Community
Links * Provide mentors * Provide organizational support for community* Recognize team accomplishments; reinforce based service
team identities * Sponsor employee sports teams in local
leagues
Fit * Hire based on fit with the job * Recruit most intensively in local markets
* Hire based on fit with the organizational (minimize relocation)
culture and values * Promote work-life balance programs (e.g.,
flextime, job sharing)
Sacrifice * Provide financial incentives (golden * Promote without requiring transfer
handcuffs)
*
Provide
home-buying
assistance
* Provide nonfinancial incentives (e.g.,
sabbaticals or unique perks)
dictions made by job satisfaction, organizational
commitment, job search, and perceived alternatives
Retention cannot be accomplished purely
combined. Thus job embeddedness captured some-
through money. A host of on-the-job and
thing new and different from the standard measures
off-the-job factors must be considered
used in traditional turnover research.
In addition, we tried to understand how job em-
when developing a retention plan.
beddedness influenced the decision to leave. It
turns out that those who are highly embedded,
when faced with an event we have called a shock
(from the unfolding model), have fewer thoughts
about leaving. They also have fewer scripts or
plans in place for leaving in case a shock occurs.
As we would also suspect, people who are embedded tend to be satisfied, committed, and have high
job involvement. Thus job embeddedness is reflected
not only in traditional job attitudes, but also seems to
buffer the individual from the effects of shocks that
might otherwise prompt one to leave.
Employees go through several stages during the
time they are employed by organizations. There is
an entry phase, where employees learn the ropes,
the norms, and expectations surrounding a job.
There is a period where employees settle in. This
period can be very short or very long, but is char-
acterized by some stability in terms of employees'
lives on the job. Not a lot of new information is
learned about the job or the organization. Finally,
there is a period of withdrawal or detachment that
precedes retirement or voluntary or involuntary turn-
over. The material we have covered so far is very
important for developing a comprehensive retention
Developing a Comprehensive Retention Plan
plan that can influence employees during all the
There is no magic pill for a company's retention
we will organize the key findings to help executives
problems. We know from observing professional
formulate a comprehensive retention plan.
stages of their careers. In the following discussion,
sports or CEO succession, for example, that even
paying a person millions of dollars a year does not
prevent relative job dissatisfaction and lower or-
ganizational commitment, more job search, or an
increased likelihood of quitting. We know from a
century of observing collective bargaining, more-
over, that the positive effects of more pay are often
short-lived. Retention cannot be accomplished
purely through money. A host of on-the-job and
off-the-job factors must be considered when developing a retention plan. Ultimately, a company's
leader must survey these factors and select those
that are most applicable to his or her firm. We do
have some guidelines about how to proceed.
Make strategic decisions
The first steps in developing a retention plan are
largely strategic:
Determine whether turnover is a problem. How
many people are leaving? Who is leaving? Do we
want these people to leave? What does it cost to
replace them?
Determine why people are leaving. Conduct exit
interviews. Consider having outside consultants perform the exit interviews one to three months after
departure to ensure that the reasons provided are not
defensive or protecting those left behind. Having the
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
2001
Mitchell,
Holtom,
HR department or an outsider conduct the interview
also helps the leaver to not fear retribution.
and
Lee
105
Use realistic job previews for new employees.
These previews can reduce the number of events
Investigate reasons why people stay. Conduct
that are unexpected, and clarify the psychological
focus groups with existing employees. Determine
the factors that keep them in their jobs.
Develop top-level support for the plan. The or-
contract and reduce turnover.6'
Attack the unsolicited job offer problem directly.
ganization needs to be willing to devote financial
and human resources to the planning, develop-
outside offers.62 This information lets firms adjust
ment, execution, and maintenance of any plan.
Identify the targets of the plan. Is it for only a
Some companies offer bonuses to people reporting
current compensation and organizational policies
to be competitive. It also provides opportunities for
quick strike counter offers. Employees could be
few? Is everyone involved? If everyone is part of
the program, that means that the initial selection
encouraged to report job offers for spouses (espe-
of employees into the company is critical, because
ment counseling and services for spouses to keep
from that point on the company will attempt to
employees from leaving.
retain almost every employee.
cially if they require relocation). Provide place-
Determine which scripts most frequently appear
for Path 1 leavers. If pregnancy is a major factor,
Draw on conventional wisdom
Once these strategic parameters are set, organizations should continue to pay close attention to basic management practices advocated by the traditional turnover literature.
Routinely assess job satisfaction and organizational commitment. Make the gathering and public
feedback of these data part of the organizational
culture.
Be prepared to make changes based on these
findings. Have resources and strategies at the
ready. Gathering data without action causes cyn-
icism and anger.
Focus on topics like supervision, pay, the work
environment, and company values. These are the
more traditional factors that human resources personnel are trained to deal with.
Apply the unfolding model
In addition, we think the unfolding model highlights some important practices for the development and implementation of a retention plan. For
example, for many people, the leaving process is
initiated by a shock (Paths 1, 2, and 3). Through the
use of surveys of former and current employees as
well as exit interviews, organizations can identify
the types of events that cause people to reconsider
their employment. This information can be helpful in
the following ways:
Learn the distribution of shocks across paths.
This information shows how many people leave
because of dissatisfaction, and how many people
leave without another alternative in hand. It gives
you a feel for what initiates the process.
Analyze the content of the shocks. Are people
leaving because of family responsibilities or educational opportunities? Are they leaving as a result of unsolicited job offers? Are the events expected or unexpected?
consider day care, healthcare support, or extended
maternity leave as policies that may sustain employment. If educational opportunities are causing people to leave, consider the development of educational programs in-house, or provide time off and
tuition support to attend a nearby college or training
facility.
Prepare people for potential shocks. If mergers
or the annual performance appraisal period are
seen as shocks, get employee discussion groups
underway to ascertain anxieties and solicit input
about these events. Make sure the reasons and
procedures for such events are clear.
Be prepared to respond to external personal
shocks. Part-time work, telecommuting, flex time or
minileaves can help an employee deal with an acute
(or chronic) problem such as an ill child or a parent
who comes to live with an employee's family.
Assess the time it is taking people to leave. Some
paths (e.g., Paths 1 or 2) unfold fairly quickly, while
others (e.g., Path 4) take longer. In some cases (e.g.,
low job satisfaction), companies can react to the information and improve conditions over time, while in
other cases they need to anticipate shocks and move
quickly to preempt departure.
Assess job embeddedness
Job embeddedness can be established and maintained through careful attention to the connections
employees make to people, institutions, and activities both inside and outside the organization. Pro-
viding support and/or incentives to get involved in
the local community can have an important impact
on retention that is independent of job satisfaction
or organizational commitment.
Work to ensure a good fit with the job for each
employee. Through the use of personal develop-
ment plans, employers can continue to provide opportunities for employees to fit in their jobs
throughout their caireer. One job fits one.
This content downloaded from
132.174.251.47 on Fri, 01 Mar 2024 04:03:02 +00:00
All use subject to https://about.jstor.org/terms
106
Academy
of
Management
Executive
November
Acknowledgments
One job fits one.
This article was written while the first author was a visiting
professor at New Mexico State University's College of Business
Facilitate fit off the job. Resources about commu-
nity activities can be gathered and made avail-
Administration and Economics. Ren6e Brown was especially
helpful assisting in its completion.
able. Work hours (flextime, part-time) can facilitate a
match with employee preferences. Cafeteria-style
compensation plans allow people to choose compensation options that fit their personal lifestyles (e.g.,
more healthcare, more vacation time).
Endnotes
'Bernstein, A. We want you to stay. Really. BusinessWeek, 22
June 1998, 67; see also Lublin, I. In hot demand, retention czars
face tough job at some top firms. Wall Street Journal, 12 Sep-
Influence links on the job through programs that
create connections between people. Establish
tember 2000, B1.
mentoring or buddy systems. Set up teams and be
in control of careers. Wall Street Journal, 11 September 1997, B 1.
mindful of the composition. Try to put people together who will like one another (similar interests,
hobbies, or life circumstances). Introduce longterm projects and provide project teams with a
sense of identity. Recognize group achievements.
Influence off-the-job links. These can be tougher
to influence than those on the job. However, a company can sponsor (and provide time for) employees
to participate in various activities, such as community cleanup or beautification. Organizations can
sponsor little league teams for children or bowling
leagues for employees. Various off-the-job social activities can increase the opportunity to establish links.
Create sacrifices on the job by introducing financial incentives tied to longevity. Some firms have
sabbaticals available after a set number of years.
Suitable perks, such as access to a beach house or
tickets to the opera or sports events, are good
ideas. Long-term career development and training
are increasingly valued by employees. The best
offices and parking places can be made dependent
on tenure. One investment bank pays the college
tuition for children of employees who have been
with the company for five or more years.
Pay attention to off-the-job sacrifices. Many of
these things would be given up only if one had to
relocate (such as football seats that get better every year). However, other company-related sacrifices can be made salient. Company vehicles (e.g.,
trucks or vans) used off site, weekend day-care
facilities, or support for long-term community ser-
3 Wysocki, B. Retaining employees turns into a hot topic. Wall
Street Journal, 8 September 1997, B1.
vice would disappear if one left.
Winning the Competition for Talent
Employee-retention programs require an overall,
comprehensive, thoughtful process to be effective.
They are expensive and require substantial effort,
and vary across organizations and industries. However, there is growing recognition that these programs compete for talent. Our ideas based on the
unfolding model and job embeddedness have ex-
panded knowledge of the turnover process and suggested important actions to alleviate the problem.
2 Lublin, J., & White, J. Throwing off angst, workers are feeling
4 A number of recent articles make this point. See Gilbertson,
D. Why do people quit their jobs? Because they can. New York
Times, 1 February 1998; and Konrad, R. Tech employees jumping
jobs faster. CNET Newsletter, 14 June 2000.
5 Papers that document these difficulties in adjustment include: Shellenbarger, S. An overlooked toll of job upheavals:
Valuable friendships. Wall Street Journal, 12 January 2000, Bl;
Shellenbarger, S. Many employees falter at helping families adjust to job transfers. Wall Street Journal, 27 September
2000, B1.
6 Schmitt, R. From cash to travel, new lures for burned-out
lawyers. Wall Street Journal, 2 February 1999, Bl.
7 BusinessWeek, 20 April 1998.
8Wysocki, op. cit., addresses this issue of replacement.
9Blase, J. After prison breaks, guards feel the heat: Highprofile escapes nudge states to address an acute shortage of
guards. Christian Science Monitor, 2 February 2001, 2.
'0 Smith, D. 1999. The role of collective efficacy and transactive memory on aircrew performance. Unpublished dissertation.
University of Washington.
1' Lublin, op. cit.
12 Two excellent reviews are: Griffeth, R. W., Hom, P. S., &
Gaertner, S. 2000. A meta-analysis of antecedents and correlates of employee turnover: Update, moderator tests, and research implications for the next millennium. Journal of Management, 26(3): 463-488; Maertz, C. P., & Campion, M. A. 1998. 25
years of voluntary turnover research; A review and critique. In
C. L. Cooper & I. T. Robertson, (Eds.), International review of
industrial and organizational psychology: London: John Wiley &
Sons, Ltd., 49-86.
13 Maertz & Campion, op. cit.
14 Hall, D. T., & Moss, J. E. 1998. The new protean career
contract: Helping organizations and employees adapt. Organizational Dynamics, 26: 22-37.
15 Robinson, S. L. 1996. Trust and breach of the psychological
contract. Administrative Science Quarterly, 41: 574-599; Robinson, S. L., & Rousseau, D. M. 1994. Violating the psychological
contract. Journal of Organizational Behavior, 15: 245-259; Turnley, W. H., & Feldman, D. C. 1999. The impact of psychological
contract violations on exit, voice, loyalty and neglect. Human
Relations, 52(7): 895-922.
16 Maertz & Campion, op. cit., 56.
'7 Ibid., 75.
18 Griffeth, Hom, & Gaertner, op. cit.
19 review of Sears policies was done by Bernstein, D. S.
Satisfaction guaranteed? How Sears keeps its people. Inside
Technology Training, 11 April 1999.
20 Lu...