7
Control Voluntary Turnover by Understanding its Causes
MARION B. EBERLY, BROOKS C. HOLTOM, THOMAS W. LEE, AND TERENCE R. MITCHELL2
It’s no secret, of course, that many people quit their jobs. Whether voluntary employee turnover is a problem, however, “depends.” Sometimes, an individual’s volitional quitting can be a major problem. In certain parts of the food and beverage industry, for example, turnover is routinely well over 100% annually, and many managers must often worry about simply having enough employees to keep their operations running. In segments of the software industry, moreover, losing a single key employee can not only decrease the likelihood of a project’s success, but it can also reduce investors’ confidence in the firm and thereby dramatically lower that firm’s stock price. Sometimes, employee turnover can be far less of a problem, and in some cases it is even desired. If employees can be readily hired, trained, and integrated into the company’s operations and culture, quitting might only be a minor nuisance. Relatedly, turnover may even be desirable because it serves as an immediate constraint on labor costs. Plus, the voluntary exit of a poor performer may be a welcome opportunity to strengthen the organization’s human capital through selection. Thus, employee turnover is not necessarily bad or good. Instead, it is an ongoing organizational issue that needs to be managed. Thus, understanding how employees decide to leave is essential to gaining control.
Empirical research has established that there are four complementary approaches to understanding the turnover process. Two of these approaches focus on who quits and how they quit. In the first approach, leaving is traditionally described through a process initiated by an individual’s feelings and beliefs. More specifically, job dissatisfaction is theorized to initiate a variety of job-search behaviors and corresponding comparative evaluations on the identified employment opportunities that, in turn, set the stage for an employee’s quitting. The second approach is called the “unfolding model of voluntary turnover.” In contrast to traditional ideas, the unfolding model (a) describes multiple quitting processes, (b) includes non-cognitive and external-to-the-person factors, and (c) explains how relative job satisfaction can prompt an employee’s departure. Rather than focusing on why people leave, the third approach examines the reasons why people stay. The broad influences that determine an employee’s choice to remain within a job are represented by the job embeddedness concept. Job embeddedness captures the extent to which the employee is tied to the job, considering both on- and off-the-job factors. Finally, in the fourth approach, understanding focuses on the company’s definition of the circumstances surrounding a person’s leaving. Viewed from the manager’s perspective, the questions are whether an employee’s quitting is functional or dysfunctional as well as whether it is avoidable or unavoidable.
Based on these four approaches, we clarify in this chapter how turnover might best be understood and managed. First, we describe each of these four approaches and their corroborating evidence. Second, we summarize this body of evidence to identify the signals that management should monitor in order to predict future employee quitting. Third, we explain what to do after observing the signals of the enhanced likelihood of turnover. Fourth, we highlight notable exceptions. That is, we suggest conditions under which our recommended signs and actions may be more or less meaningful.
Monitor job attitudes and withdrawal cognitions
The theorized linkages. Based on the evidence to date, managers have good reason to expect links between job (dis)satisfaction and employee turnover. (See Hom and Griffeth (1995) for an excellent, comprehensive academic treatment of the major psychological theories of turnover using this perspective.) First, the lower the level of an employee’s job satisfaction, the lower will be the level of his or her organizational commitment. Second, the lower these two job attitudes are, the stronger will be the initial thoughts, feelings, and expected positive outcomes of quitting. Third, according to this traditional approach, these thoughts, feelings, and expected outcomes lead most people to search for another job. The job opportunities found during this job search are then evaluated. Finally, actual quitting occurs when a “better” job opportunity is found. We hasten to note that “better” has rarely been fully defined. Most often, improved economic outcomes are implied (e.g. compensation and financial opportunities) but seldom explicitly stated.
The evidence. Perhaps equally enlightening as the traditional dissatisfaction-induced leaving processes themselves, the statistical findings provide compelling reasons for a manager to expect the following stable and consistent relationships. The statistics that are provided in parentheses below are: (a) averaged across multiple studies; (b) weighted by large sample sizes; and (c) statistically significant. Furthermore, these values are taken from reports that quantitatively summarize the data from numerous empirical studies, by Griffeth, Hom, and Gaertner (2000), Meyer, Stanley, Herscovitch, and Topolnytsky (2002), and Spector (1997). Our description of a statistic’s size (e.g. large, moderate, or small) derives from the common standards in the social sciences.
First, overall job satisfaction strongly correlates with affective organizational commitment (r = .65), and the intentions to search for another job (r =—.60) and intentions to leave (r =—.46). More specifically, lower scores on job attitude measures are consistently associated with more thoughts of leaving and job-search activities. Second, job satisfaction (r =—.19), organizational commitment (r =—.23), thoughts of quitting (r = . 24), and intentions to search (r = .29) and intentions to leave (r = .38) correlate moderately well with actual turnover behavior. In other words, these variables are consistently associated with employee withdrawal. Third, until people quit, less satisfied employees are absent slightly more often (r’s between—.10 and—.15), somewhat less helpful to co-workers (r’s between .22 and .26), and perform their jobs more poorly (r = .25). Similarly, less affectively committed employees also perform fewer citizenship behaviors (r = .32), perform their jobs more poorly (r = .17), and have higher turnover intentions (r =—.51). Thus, employees with poor job attitudes often exhibit multiple negative behaviors and can be difficult to manage.
What to monitor first: job attitudes. Every traditional theory on employee turnover includes job dissatisfaction as either the primary starting point or as an early stage in the turnover process. Moreover, the evidence indicates consistent and large correlations (cited above) between job satisfaction and the intermediate variables (e.g. organizational commitment, intention to search or to leave) before actual employee turnover occurs. In turn, the evidence reveals consistent, moderate-sized, and predictive over time correlations between the intermediate variables and actual quitting. When considered together, job satisfaction and organizational commitment should be monitored as clear and compelling indicators of future employees’ quitting.
Merely analyzing employee job satisfaction and organizational commitment in a static manner is not enough. It is imperative to examine attitude trends over time. Employees may experience significant decreases in attitudes over time and these changes may lead to an increase in search and quit intentions and actual turnover. For example, employees who appear to be satisfied with their jobs may decide to quit because some aspect of their jobs changed over the past few weeks and made them be less satisfied than before. A static approach to monitoring job attitudes would miss these nuances, and managers might therefore inaccurately assume that these employees will most likely not leave. Eventual leavers will likely experience a decline in job satisfaction and organizational commitment over time, therefore rendering a dynamic monitoring of these attitudes essential to managers (Bentein, Vandenberg, Vandenberghe, and Stinglhamber, 2005; Kammeyer-Mueller, Wanberg, Glomb, and Ahlburg, 2005).
What to monitor second: global withdrawal cognitions. These traditional approaches most often conceptualize factors such as thoughts of quitting, expected utility of job search, cost of quitting, intention to search, evaluation of alternative jobs, comparisons between current and alternative jobs, and intention to leave as separate entities. Correspondingly, organizational behavior researchers have sought to measure and study these variables separately. An enduring problem, however, is that these intermediate variables (between job satisfaction and turnover) are difficult to separate accurately (e.g. using various statistical methods). As a result, several turnover scholars have sought to simplify the meaning of these “intermediate steps.” More specifically, these numerous factors have been reconceptualized and empirically measured as a global and broader-based variable labeled “withdrawal cognitions.”
When measured directly, global withdrawal cognitions consistently correlate with actual turnover (r = .30). Thus, it is advisable to monitor employees’ global withdrawal cognitions. Similar to job attitudes, managers will benefit from monitoring changes in withdrawal cognitions over time through surveys and questionnaires. Eventual leavers are likely to increase their withdrawal behaviors and cognitions under this traditional approach while stayers likely do not change (Kammeyer-Mueller et al., 2005). In short, the “evidence” informs us to dynamically monitor job attitudes and global withdrawal cognitions as leading indicators of potential turnover.
CASE EXAMPLE
According to hotel survey firm Market Metrix, on average, US hotels will replace two thirds of their workers every year. The annual costs of recruiting, selecting and training these replacements as well as the associated productivity losses range from $100,000 at a small property to more than $1,000,000 at a large hotel. In this competitive industry where labor costs account for nearly one half of total operating costs, expenses incurred because of high staff turnover can mean the difference between making a profit or loss in any given year. Given this context, consider the following example.
In just a few years, Joie de Vivre Hospitality has been able to drive staff turnover at the Hotel Carlton (San Francisco, California) from more than 50% a year to below 10% a year (Dvorak, 2007). While pay and benefits are largely unchanged, simple things that make a difference in morale have changed. The former management company at the hotel did not like to replace aging vacuums despite staff complaints. When Joie de Vivre took over in 2003, the new manager bought each of the 15 housekeepers a new vacuum - and he replaces them annually. A Hotel Carlton employee mentions this gesture as one of many that “seems to show that this company cares about us more.” Other low-cost, high-impact changes include English as a second language and Microsoft Excel courses as well as regular dinners for employees and their bosses to communicate informally.
Monitor shocks and paths as well as job attitudes and global withdrawal cognitions
The overall model. By design, the unfolding model offers a broader perspective than the traditional theories of turnover. (See Lee and Mitchell (1994) for a detailed description of this model.) Based on the evolving evidence, managers should have confidence that the unfolding model broadly and accurately describes the leaving process. More specifically, the unfolding model informs us about four basic patterns of thoughts and actions (or psychological “paths”) for leaving organizations. In one particular pattern, which is labeled “path 4,” leaving is seen as quite similar to the traditional ideas of turnover described previously. Job dissatisfaction prompts thoughts of leaving that lead, in turn, to job search, evaluation of alternatives, and eventual departure from the job. Thus, much of the evidence on the traditional ideas about turnover also pertains to path 4 leaving. From an applied perspective, both traditional and path 4 approaches inform the manager to monitor employees’ negative or declining job attitudes.
In paths 1, 2, and 3, a fundamentally different pattern is asserted. In contrast to dissatisfaction-induced quitting, the leaving process begins with a shock - an external-to-the-person event that causes him to think about leaving. Shocks serve to shake people out of their daily, habitual, ongoing patterns and routines. Although all shocks are mediated by an individual’s perceptual processes, their jarring nature renders them easily identifiable, describable, and understandable by the employee and manager. Thus, shocks are a conceptual tool that keeps monitoring and observations firmly grounded on work behaviors and the employee’s immediate situation. In our research, for example, people report that shocks can be: (a) positive (e.g. acceptance into graduate school) or negative (e.g. a nurse being forced into a life-threatening emergency surgery for which she was unprepared); (b) expected (e.g. a spouse’s retirement) or unexpected (e.g. being denied a six-month leave of absence); and (c) organization related (e.g. the hospital shifts from individual-based to team-based nursing) or personal (e.g. becoming pregnant with one’s third child). As a result, shocks augment job attitudes and global withdrawal cognitions as key antecedent signals to subsequent employee turnover and are easily identifiable by employees.
The unfolding model offers a unique lens through which to examine turnover, because it focuses on the reasons why people leave, and oftentimes these reasons are unrelated to attitudes such as job satisfaction and organizational commitment. Thus, exploring turnover reasons provides researchers and managers with a more complete and accurate picture of why employees leave, which in turn facilitates turnover management and control. Maertz and Griffeth (2004) developed a typology of turnover motives and argued that these motives capture the most proximal causes of turnover cognitions and therefore are the best predictors of actual turnover behavior. Combining these eight motives with the unfolding model, Maertz and Campion (2004) show that the reasons systematically relate to specific decision paths. Exploring turnover reasons therefore can tell us which paths employees will choose.
Monitor for specific paths. Whereas monitoring for job attitudes, global withdrawal cognitions, and shocks is fairly straightforward, determining which specific path employees may take can be more complicated; but such monitoring offers the manager greater understanding of turnover. In path 1, a shock triggers a person to use a pre-existing action plan (or what social psychologists call a script). Minimal mental deliberations occur. A person leaves without considering his or her current attachment to the organization (e.g. organizational commitment) and without considering his or her current job alternatives. Unlike traditional ideas and path 4, job satisfaction is essentially irrelevant. In path 1, a satisfied person can appear to leave abruptly but is in fact following a pre-planned course of action. For example, an interviewee reported the following: “It was very simple; I am going to graduate school . . . It had nothing to do with my job. It had everything to do with my personal situation . . . Very quickly, as soon as I got the letter that I was accepted into graduate school, the decision was made” (Lee, Mitchell, Wise, and Fireman, 1996, p. 17). Note that path 1 involves a shock (e.g. the acceptance letter), script (e.g. return to graduate school), no job search, and no evaluation of alternatives.
In path 2, a shock prompts a person to reconsider his or her basic attachment to the organization because of violations to one’s basic values, personal or professional goals, and/or plans for goal attainment (“images”). After completing these mental deliberations, a person leaves the organization without a search for work alternatives. Unlike traditional ideas and path 4 but like path 1, job satisfaction is also largely irrelevant. In path 2, a satisfied person can leave abruptly because she is reacting to the shock itself. For example, an employee recently left a bank where we were conducting research. In her words, she told us the following:I was working and minding my own business. Then, a new person was hired and she was pretty vocal. One day she was complaining about her pay. While that made me mad, what was worse is that I had been at the bank for over a year and she was making a dollar an hour more than me while doing the same job. I quit the next day.
Note that path 2 involves a shock (e.g. learning about the pay of a fellow employee), no script, violations (e.g. images of personal and professional goal attainment), no job search, and no evaluation of alternatives.
In path 3, a shock provokes a person to consider whether an attachment could form with another organization because of violations to one’s basic values, personal or professional goals, and/or plans for goal attainment. The mental deliberations due to the shock and violations lead an individual to search for another job and to evaluate specific alternatives and one’s current job. Unlike traditional ideas and paths 1 and 2, a reasonably job-satisfied person can leave for a more satisfying job. In path 3, job dissatisfaction may or may not be present. Previously, we reported the following example of path 3 leaving:Sammy Lew (a fictitious name) was a forty-five year old former electrical engineer who had worked for a major military-defense company. [Mr. Lew] . . . had worked for the same firm since graduation twenty-three years earlier. For the last ten years, Mr. Lew supervised a group of twelve engineers. When asked why he left his former company to become a real estate developer, Mr. Lew said that . . . it was the departure from the firm of another work group member. That departure left Mr. Lew as the work group’s oldest and clearly most out-of-date engineer. [He realized that] he had minimal advancement prospects and that “it was time to move on.” As such, he evaluated his interests, aspirations, and available opportunities. Although he had no particular connection to his new profession, real estate development appeared, on balance, his best choice. He quit the defense company upon finding an acceptable job in real estate. (Lee and Maurer, 1997, p. 249)
Note that path 3 involves a shock (e.g. co-worker’s quitting), no script, violations (e.g. images of professional goal attainment), job search (e.g. finding real-estate work), and evaluation of alternatives (e.g. in comparison to his interests, aspirations, and available alternatives).
The evidence
ACCURATE CLASSIFICATIONS. A fundamental test of the unfolding model is whether its four paths can accurately describe employees’ actual leaving. In a set of initial studies, Lee et al. (1996) and Lee, Mitchell, Holtom, McDaniel, and Hill (1999) reported that 33 of 44 nurses (75%) and 212 of 229 former accountants from Big 6 public accounting firms (93%) could be accurately classified into paths 1-4. In independent tests of the model, 86% of public accountants (Donnelly and Quinn, 2006) and 77% of nurses were successfully classified (Morrell, Loan-Clarke, Arnold, and Wilkinson, in press). Holtom, Mitchell, Lee, and Inderrieden (2005) illustrated that precipitating events, or shocks, more often are the immediate cause of turnover than job dissatisfaction. Specifically, across 1200 “leavers” from a variety of industries, 60% reported that a shock precipitated their turnover, of which 59% were unexpected, 40% were personal, and 64% were positive. Similarly, Morrell, Loan-Clarke, and Wilkinson (2004) reported that 44.3% of the nurses in their sample considered a shock their primary reason for leaving while 43.8% of nurses did so in another study (Morrell, 2005).
PATH SPEED. In addition to accurate classifications, a second testable attribute of the unfolding model involves the speed with which the four paths unfold. More specifically, paths vary by their levels of mental deliberation. For example, extensive evaluation of alternatives should take more time than doing no or minimal evaluations. Furthermore, paths differ by whether their basic features are readily available for mental deliberation. If job search is required, for instance, information about alternative employment is less readily available than if no job search is involved. Thus, the path speed should vary systematically. Among nurses, Lee et al. (1996) reported that the durations of paths 1 and 2 were significantly shorter than those of path 4. Among accountants, Lee et al. (1999) reported shorter durations for paths 1 and 2 than for 3 and 4, and a shorter duration for path 3 than for path 4.
SHOCK CHARACTERISTICS. In their sample of 44 nurses, Lee et al. (1996) reported that: (a) path 1 was significantly associated with expected and personal shocks (e.g. your spouse relocates); (b) path 2 was significantly associated with organizational and negative shocks (e.g. you are passed over for promotion); and (c) path 3 was significantly associated with organizational shocks (e.g. you receive a competitive job offer). With the sample of 229 former Big 6 accountants, Lee et al. (1999) replicated two of the earlier findings. In particular, path 1 positively associated with personal shocks and path 3 positively associated with organizational shocks. When taken together, these two significant associations may be robust across industries, occupations, and gender. Key findings of the Morrell et al. (2004) study indicate that: (1) shocks that are expected are more likely to be positive, personal, and lead to unavoidable leaving, (2) shocks that are negative are more likely to be work related, associated with dissatisfaction, affect others, and lead to avoidable leaving, (3) shocks that are more work related are less potent, associated with dissatisfaction and search for alternatives, and lead to avoidable leaving, and (4) shocks tend to cluster into work and non-work domains. In another study, Morrell (2005) reported three distinct clusters of leavers. Cluster 1 (n = 103) leavers had a work-related shock that was unexpected, negative, and affected other workers. Cluster 2 (n = 50) leavers had a personal shock that was expected, positive and private. Cluster 3 (n = 196) leavers had no shock and followed a more traditional dissatisfaction-induced process.
In sum, monitoring for shocks is not enough. Managers must also analyze a shock’s content. Different types of shocks will require different interventions. Rich information about shocks can help a manager decide how quickly to act (e.g. in paths 1 or 2) and what factors might inhibit leaving. In comparison to the traditional approach, the unfolding model identifies additional considerations and adds richness to our understanding of the turnover process.
CASE EXAMPLE
As the following example illustrates, not all turnover is job dissatisfaction induced. Indeed, we have seen many path 1 quits in the news. One involves Nirav Tolia, 27, who left Yahoo! as a senior manager of its e-commerce marketing and $10 million in unvested options. The reason he left: upon hearing a “really interesting start-up idea from a friend” (the shock), he decided in only one day’s time to pursue his dream (a script) of a net start-up e-commerce company. Joining Tolia, Ramanathan Guha, 34, left America Online as one of its senior ranking engineers and $4 million in unvested options also to pursue this shared dream (Bronson, 1999). In short, managers need to be prepared for the possibility that some turnover will be precipitated very quickly. Hence, potential responses need to be considered in advance for a variety of possible scenarios - especially those involving key employees.
Monitor the reasons for staying in addition to the reasons for leaving
The concept of job embeddedness. To avoid voluntary turnover, it is at first intuitive for managers to identify the reasons why employees are leaving and to attempt to eliminate or minimize the impact of those reasons. The first two approaches to turnover have taken this perspective. In contrast, a more recent approach focuses on why employees are staying and what might prevent them from leaving. The concept of job embeddedness is focused on the broad array of factors that influence a person’s staying in a job. Mitchell, Holtom, Lee, Sablynski, and Erez (2001) identified three ways an employee might be embedded within an organization: links, fit, and sacrifice. (Sacrifice refers here to loss of a value or values, not necessarily to sacrifice to others as in the case of altruism.) Links refer to employees’ formal and informal connections to other individuals or institutions (e.g. friendship ties to co-workers). Fit refers to the extent to which employees’ jobs and communities are compatible with their personal values, goals, and plans (e.g. congruence with organizational culture, satisfaction with local amenities). Research on employee fit has established that it significantly lowers turnover (e.g. Hoffman and Woehr, 2006). Finally, sacrifice captures the perceived losses that individuals may suffer when leaving their job (e.g. loss of retirement benefits, loss of friendship support). Together, links, fit, and sacrifice reflect an employees’ embeddedness within a job. The more embedded they are, the less likely they will quit. What is particularly novel about this approach to turnover is that job embeddedness considers not only links, fit, and sacrifice on-the-job, but also off-the-job - in employees’ personal lives within the local community. For example, an employee with children in middle school and a wife who is engaged in local charities would have many close links to the community and therefore be less likely to leave - especially if getting a new job would entail relocation.
In addition to considering an individual employee’s level of job embeddedness, recent research suggests that turnover might be “contagious” and that co-workers’ job embeddedness may influence an individual’s decision to leave (Felps, Mitchell, Hekman, Lee, Holtom, and Harman, in press). The less embedded one’s immediate co-workers are, the more likely it is that they engage in job search behaviors. In an environment characterized by low job embeddedness, it may be more acceptable to discuss leaving or talk about other job opportunities. Thus, the salience and viability of leaving may increase when observing a number of co-workers looking for other jobs.
The evidence. A fundamental test of the concept of job embeddedness is whether it can predict actual turnover. A variety of studies have shown that this is indeed the case (cf. Allen, 2006; Crossley, Bennett, Jex, and Burnfield, 2007). Specifically, Mitchell et al. (2001) reported that overall job embeddedness (on- and off-the-job) was negatively correlated with intention to leave and predicted subsequent voluntary turnover in two different samples. In a study with bank employees, Lee, Mitchell, Sablynski, Burton, and Holtom (2004) found that off-the-job embeddedness predicted volitional absenteeism and voluntary turnover. Mallol, Holtom, and Lee (2007) compared the relative strength of job embeddedness and its relationship to turnover among Hispanics and Caucasians and concluded that while job embeddedness may vary in strength across different demographic groups, it is nonetheless a robust predictor of employee retention. Tanova and Holtom (in press) further found that notwithstanding differences in labor laws, cultural factors, and unemployment rates between Europe and the USA, job embeddedness predicts turnover effectively on both continents. In summary, the diverse nature of the samples used in job embeddedness research so far suggests that the negative relationship between job embeddedness and turnover may generalize to many occupations, organizations and contexts.
In addition to establishing that job embeddedness predicts employee retention, it is important to show that it provides additional insights over and above the information the other two approaches offer. Only then would it be justified and meaningful to separately monitor embeddedness signals in addition to work attitudes, withdrawal cognitions, shocks, and paths. Mitchell et al. (2001) showed that job embeddedness predicted turnover even after taking into account the influence of gender, job satisfaction, organizational commitment, job search, and perceived alternatives. In an independent study with a diverse sample, Holtom and Inderrieden (2006) found that job embeddedness significantly improved the prediction of turnover above gender and job satisfaction. In addition, they showed that those employees who stayed on the job had the highest levels of job embeddedness followed by employees who left due to a shock followed by employees who left with no precipitating shock. Not only does job embeddedness supplement the traditional approach to turnover research, but it also complements the unfolding model by showing how job embeddedness may buffer the effects of shocks on turnover.
CASE EXAMPLES
Organizations do not have to be large to develop customized approaches for creating job embeddedness. Moreover, the benefits of such efforts not only include improved employee retention but also positive reputations as great employers that generate word-of-mouth advertising and favorable public relations. Following are a number of examples from the Wall Street Journal’s Top Small Workplaces report (Spors, 2007).
Jere Cowden, chief executive of Cowden Associates, an actuarial and human resources consulting firm, keeps a running list of consultants he deems to be the best in the business. Then, when hiring, he draws from this list and evaluates how well each prospective hire would fit into the culture. Most also complete personality assessments to further assess fit with the organization.
FRCH Design Worldwide, an architectural and interior design company, works very hard to promote strong corporate culture in the face of rapid growth. A big part of the company’s effort is devoted to giving employees many chances to have fun together and participate both at work and in their free time. It hosts an annual summer picnic for employees and their families, a trip to a Kentucky horse-racing track, a yearly anniversary party, and a Christmas party for families, along with some sporting events. Every other Friday evening, the company hosts hour-long “wing-dings,” social gatherings at the office with food and drinks. While not all organizations will see the need to socialize so frequently, many would benefit from consciously considering how to increase the links among people in the organization as FRCH has.
At Exactech, an orthopedic device manufacturer, an employee’s training regimen depends on his or her employee-development plan - a document updated at least twice a year by the employee and his or her supervisor that lays out the employee’s key development area and “action plan.” Sometimes the action plan includes reading a book on a particular topic. Other times it includes taking classes. For Anacielo Vale-Grogan, it meant going to weekly French lessons and a six-week language course in France so she could better communicate with representatives from the company’s large French distributor. Such a tailored approach to employee development is relatively unique and represents a sacrifice an employee would have to make if he or she were to leave Exactech.
While not everyone would want to work in rural Vermont, employees of NRG Systems, a wind-measurement equipment maker, clearly do. Only one of 81 employees working there quit last year. Part of the appeal is the unique setting. The company is housed in a newly built solar-powered building replete with skylights, a commons area reminiscent of a ski lodge and an indoor swimming pool. Further, employees are eligible for as much as $2300 annually for purchases of items such as Toyota Priuses, backyard wind turbines, solar-powered water heaters and energy-efficient light bulbs. No doubt these benefits attract unique employees who fit with the local culture and want to be linked in such a community.
Finally, Corporate Ink Public Relations seeks to help its people reside in the local area. Employees buying a home within 20 miles of the office can get loans of as much as $10,000 at 2% interest, to help make the down payment - and the loans are forgiven after five more years of employment. This is clearly an example of creating sacrifices related to the community where employees live and work.
Monitoring for signs of voluntary quitting may not be enough
Is turnover functional and avoidable? A prevailing belief among many organizational behavior scholars and practicing managers is that employee turnover should be minimized. From the manager’s perspective, turnover can substantially increase the costs of staffing, training, and general administration. More subtly, turnover can also often disrupt a business unit’s operations and increase the workloads of remaining employees.
In contrast to this prevailing belief, a small body of empirical research indicates that turnover should be more carefully managed rather than minimized. (See Maertz and Campion (1998) for a thorough discussion of these issues.) Importantly, employee turnover can be beneficial to a company if marginal performers voluntarily leave. Whereas “truly bad” performers might be fired, having too many marginal performers (e.g. those persons who are not sufficiently bad to fire) can minimize a firm’s productivity. Similarly, employees who may be labeled “bad apples” because they are chronically negative, do not do their fair share of the work, or bully their co-workers, may have a detrimental effect on how groups work together, and “spoil the barrel” (Felps, Mitchell, and Byington, 2006). The value accrued by encouraging these people to resign can often offset that individual’s overall replacement costs. In a related vein, the new employees can oftentimes be a source of new ideas and mechanisms to shake remaining employees out of their inertia (i.e. “getting new blood”). In those situations where labor costs are substantial and salaries are expected to increase consistently over time, turnover can result in substantial savings and thereby enhance organizational profitability (e.g. replace higher-wage customer service agents with lower-paid new workers). In short, employee turnover should not simply be minimized nor dismissed as an unmanageable process.
The evidence. Two empirical studies directly address the functionality and avoidability of leaving. Based on termination records examined for a seven-month period, Dalton, Krackhardt, and Porter (1981) identified the volitional quitting of 1389 former tellers from 190 bank branches. Next, each former teller’s immediate supervisor reported judgments about (a) their preference on rehiring that person, (b) the former employee’s job performance, and (c) the ease of replacing that leaver. These voluntary leavers represented a 32% overall turnover rate. When voluntary leavers were classified into dysfunctional (e.g. good employees quit) versus functional (e.g. marginal employees quit) based on the supervisor’s judgments on preference for rehire and job performance, the turnover rate of dysfunctional leaving dropped to 18%. When classified by ease of replacement, dysfunctional turnover (e.g. hard to replace) dropped to 9% of the overall turnover rate. These data indicate that the quitting of some people is actually functional or good for the company.
Using a sample of nurses, Abelson (1987) compared the job attitudes among 136 stayers, 30 avoidable leavers, and 16 unavoidable leavers over a one-year period. Levels of job satisfaction and organizational commitment were significantly lower for avoidable than unavoidable leavers and stayers. That is, people who left for unavoidable reasons (e.g. a spouse relocates) had satisfaction and commitment levels comparable to those who stayed. People who left for avoidable reasons (e.g. bad work schedules) had worse attitudes than those who stayed or those who left for unavoidable reasons. In addition, levels of thinking of quitting and intentions to search and to leave were significantly higher for avoidable leavers than unavoidable leavers and stayers.
SHOULD TURNOVER BE ENCOURAGED OR DISCOURAGED?
When considered together, these data indicate the desirability of further classifying voluntary turnover by functionality and avoidability. Nevertheless, it remains the manager’s decision as to: (a) whether dysfunctional turnover should be discouraged; (b) whether functional turnover should be encouraged; and (c) whether such turnover is also avoidable or unavoidable. These decisions focus the manager on differentiating between people he wants to keep or lose.
CASE EXAMPLE
As Lars Dalgaard, CEO of Successfactors, explains, organizations have an interest in not keeping nasty and demeaning people (Sutton, 2007). Companies that tolerate such malicious employees have trouble recruiting the best talent and have problems keeping clients. This may lead to damaged reputations and reduced investor confidence. Other aspects of a business may also suffer. For example, creativity and innovation may be impaired. And cooperation with others within and outside the organization may suffer. In short, the loss of a bad apple may be one of the best examples of functional turnover.
ACTION IMPLICATIONS
When all evidence is considered together, the results suggest value in monitoring for job attitudes, global withdrawal cognitions, shocks, paths, and job embeddedness as well as examining the functionality and avoidability of voluntary turnover. Although each manager’s situation will be different, general action implications can be inferred from the four approaches reviewed. Initially, a company needs to determine the “value to the firm” of each employee. That is, managers need to decide - based on a sound definition of job performance and importance - which individual’s leaving should be seen as a functional quit and which should be seen as a dysfunctional quit.
Job performance and the issue of functionality
To judge whether a particular employee’s quitting would be functional, managers must have a solid definition of job performance. Job performance is commonly determined by a job analysis and often operationalized by the content of the firm’s employee evaluation form. Historically, a host of traditional variables have been applied: observable work behaviors (e.g. cooperation), measurable employee actions (e.g. call time availability for tele-marketers), supervisory judgments (e.g. quality, quantity, and overall performance), tangible work outputs (e.g. number of widgets produced each day), or revenues generated (e.g. total sales dollars).
An emergent issue among organizational behavior scholars is whether job performance has been too narrowly defined. Underlying these various and traditional performance variables is the standard job. More specifically, performance is often tied very closely to the job’s formally required tasks. Scholars of organizational behavior have recently asked whether measuring performance based on a job’s required tasks and behaviors is too restrictive.
Substantial research evidence shows that: what gets measured directs employee attention (e.g. dimensions of job performance); setting standards or goals directs employee efforts and behaviors (e.g. good versus bad job performance); and providing feedback allows for corrective actions toward the standard (e.g. the semiannual job performance evaluation meeting). We know that by focusing attention toward one’s narrowly prescribed job tasks, work effort, behavior, and performance follow. It is not surprising then that focusing on one’s own job can also distract such work effort, behavior, and performance away from other beneficial organizational actions. For a lower-level employee, for example, a focus on one’s job may decrease the likelihood of helping another employee do his job. At the immediate cost of lost opportunity of doing one’s own prescribed job, that lower-level employee may impose larger costs on organizational effectiveness by not helping another employee perform or learn to perform a job. For a higher-level manager, a focus on one’s narrowly defined job may decrease the likelihood of meeting the larger organizational role of representing the company at industry and community events. At no cost to doing one’s own job, for instance, this manager may voluntarily represent the company’s commitment to community well-being or individual citizenship. Thus, employees may overly focus on the immediate job and thereby actually hurt larger organizational functioning.
Organizational behavior scholars have labeled our traditional focus on well-defined and prescribed job behaviors as task performance, in-role performance, or job performance. In contrast, the broader focus that includes non-job-specific effort, behavior, and performance but still benefits the larger firm is labeled extra-role performance, contextual performance, or organizational citizenship behavior. In making decisions about an employee’s functional versus dysfunctional and avoidable versus unavoidable quitting, managers should also decide whether individual and organizational effectiveness can be enhanced by a traditional focus on the narrowly defined job functions or by a broader focus on task and contextual performance. In our judgment, the broader focus is typically better than the narrower view toward individual and organizational effectiveness.
The details of the unfolding model and avoidability
The unfolding model becomes particularly helpful in assessing avoidability because (a) path 4 captures much of the traditional approach, and shocks capture much of the avoidability issue. Because the speed of path 1 quitting can be unpredictable and the speed of path 2 quitting can be quite quick, managers may have minimal opportunity to respond to shocks that may initiate such leaving. Instead, they may need to have already in place a mechanism that allows for proactive and quick actions aimed at encouraging or discouraging an employee’s leaving (e.g. quick counteroffers, rapid and informal grievance procedures, open-door policies by the firm’s top executives). One technique that we’ve developed is to gather examples of (or stories about) actual shocks (e.g. specific events that initiate actual quitting) from leavers and other examples of events that prompted stayers to think about quitting. Managers can then simply list these examples and ask current employees (a) whether these events would prompt thoughts of leaving and (b) what scripts they might have in place and follow if the event were to occur. This information can help managers anticipate the events that likely prompt thoughts of quitting and scripted actions. As a result, the informed manager can elect to be proactive (or inactive) in discouraging (or encouraging) leaving.
In contrast, quitting in paths 3 and 4 is more predictable and slower. Because people search for alternatives and evaluate the located options, managers may have more opportunity to respond. Thus, they may have more opportunity to craft individualized actions that can encourage or discourage quitting (e.g. matching pay increases from external job offers, dealing with accumulated job dissatisfaction, restructuring one’s job responsibilities, reassigning employees to other units within the company, loaning employees to external service organizations like the United Way).
Besides merely reacting to employees’ withdrawal behaviors and cognitions, managers can also take a more proactive stance and reduce turnover by attempting to embed employees within their jobs. Managers may increase their employees’ job embeddedness by hiring employees who already have established ties to the organization (e.g. peer referrals) or the community. Then, for all the newcomers, it is important to effectively integrate them into the social fabric of the organization. For example, training all newcomers together as a group, revealing the exact timing of the progression through the various socialization stages, and providing positive social support through organizational role models are all ways through which new employees may be embedded on the job during their first weeks in the organization (Allen, 2006). Once socialized, managers may continue to influence their employees’ ties, fit, and sacrifices by, for example, designing work around teams, establishing mentoring programs, organizing social events, and identifying clear promotional paths through training and developmental activities.
How to monitor job attitudes, global withdrawal cognitions, shocks, paths, and job embeddedness
Surveys. Job satisfaction, organizational commitment, global withdrawal cognitions, and job embeddedness can be efficiently and validly measured via organizational surveys. In particular, they can be assessed with standard, professionally developed, and well-researched measures (e.g. validated scales). For several of the better-known scales, moreover, they are copyrighted and commercially available. Some of these validated scales measure global job satisfaction, organizational commitment, withdrawal cognitions, and job embeddedness whereas others measure facets of satisfaction, commitment, withdrawal cognitions, and embeddedness. Although valid, standard scales are often quite long. As a result, questionnaire length can discourage survey completion, and less information is actually obtained. Although less reliable and less valid than the longer standard scales, researchers and managers often must compose and use a smaller number of original questions (e.g. one to five items) to measure overall job satisfaction, organizational commitment, global withdrawal cognitions, and job embeddedness. Albeit less desirable, long-standing research practice and enduring empirical evidence indicate an acceptable tradeoff between shorter questionnaire length and lower (though adequate) strength in predicting voluntary turnover (e.g. Arnold and Feldman, 1982).
Management by wandering around (MBWA). Although shocks and paths can be measured with surveys, these concepts are relatively new. As a result, there are no standard and validated scales for their measurement. In addition to the technique of listing actual and potential shocks and scripts (described above), there are other useful techniques for identifying paths. In their delightful book, In Search of Excellence, Peters and Waterman (1982) initially recommended MBWA. The idea is simple and appealing. Managers should prioritize and allocate a certain portion of their day to watching, talking, and generally interacting with their employees. If done consistently and sincerely, understanding, empathy, and trust should develop. As a direct result, managers should readily learn: (a) what individuals interpret as shocks; (b) whether scripts exist; (c) the specific content of images and if image violations occur; (d) how embedded the employee is; and (d) whether job search cognitions and behaviors are engaged. Although performance appraisals may also generate some of this information (assuming managers have established an open and honest relationship with their employees), they are typically only conducted once or twice a year and would therefore miss out on crucial time-dependent information such as the occurrence of a shock. Consistently “wandering around,” however, allows managers to collect data in real time and to intervene immediately as needed.
The research that we’ve reviewed offers some clear advice on managing (or controlling) employee turnover such as monitoring for signs or predictors of employees’ quitting and assessing the functionality and avoidability of employee turnover. With that said, however, there are a number of considerations that should strengthen or weaken our recommended actions.
Importance of workforce stability
Selective retention is less important in situations where workforce stability is less critical; conversely stated, retention is more important in situations where stability is more critical. Four questions immediately arise. First, are replacement employees readily available? If replacement employees are readily available, selective retention should be less important to managers. An immediate indicator of availability is the unemployment rate in the “relevant labor market.” Numerous studies indicate a consistently negative, substantial relationship between aggregate turnover and unemployment rates (r’s commonly .80 and higher). In general, the higher the unemployment rate, the higher the availability of replacement employees.
Second, can the requisite job-specific knowledge, skills, and abilities (as well as broader needs for contextual, citizenship-based performance) be readily taught? If success in training is relatively quick and inexpensive, selective retention should also be less of an issue. Two immediate and related indicators of training success are the job’s organizational level and whether advanced education is required. In general, lower-level (e.g. non-exempt) jobs that do not require advanced educational experiences (e.g. bachelor’s or master’s degrees) should lend themselves to quicker and less expensive training than higher-level (e.g. exempt) positions that require higher education. When considered together, lower-level jobs that are filled primarily with non-college-educated persons and that occur in labor markets with high unemployment rates strongly suggest far less attention might be devoted to selective retention.
Third, it is becoming increasingly clear that many employees are valued for their human and intellectual capital. More specifically, some employees may have specialized knowledge that is critical to organizational effectiveness. With the (incredible) pace of organizational change and the competitive nature of many industries, human and intellectual capital should become increasingly important in a manager’s judgment about what is and what is not functional turnover.
Fourth, as an organization’s success becomes more dependent on relationships, network connections, and interactions with others within and outside of the organization, an employee’s social capital, independent of his or her skills, knowledge, and abilities, becomes a more critical asset. If employees with high social capital quit, they might leave behind a communication gap that cannot easily be filled. Unless other employees have worked closely with the quitter and developed a similar network of connections in both quantity and quality, such social capital losses may lead to immediate decrements in performance that, due to the inherently interdependent nature of today’s work, may ripple throughout the organization (Dess and Shaw, 2001; Shaw, Duffy, Johnson, and Lockhart, 2005). In sum, managers need to consider the availability of replacement workers, the ease of training, and the leaver’s human capital, intellectual property, and social capital in judging the functionality of an individual’s turnover.
Finally, such analysis is not only important in operating an organization, but is also potentially vital to successful merger and acquisition activity. According to research done by partners at Bain Capital (Harding and Rouse, 2007), people issues are often at the root of failed deals. In analyzing 40 recent merger and acquisition deals, the 15 that were classified as successful, the acquirers had all identified key employees for retention during the due diligence phase or within 30 days of the announcement of the deal. This task was carried out in only one third of the unsuccessful deals.
Organizational types
Selective retention and workforce stability can be less critical in certain organizational types. In particular, turnover is likely unimportant in temporary organizations whose sole purpose is to create a given project or produce a certain service, and then, by design, disband. Common examples include independent motion picture productions (e.g. shooting a movie), political campaigns (e.g. winning an election), and joint ventures intended to spread risk across the multiple participants (e.g. oil exploration). Similarly, bureaucracies, which are relatively buffered from market forces (e.g. state and local governments, public universities), may have limited concern with retention and stability because, for example, of sufficient slack resources, adequate time to forecast accurately human resource requirements, or de facto monopoly position. Finally, those organizations where creativity and innovation are critical to survival might actually consider encouraging departures of less creative or innovative employees (e.g. think-tanks, innovation centers, and advertising firms).
“The more things change, the more things stay the same”
Given that people will and do quit firms, how have companies begun to control the leaving of these kinds of extraordinary and other less extraordinary (aka normal) people? Certainly, there is no single magic bullet that all firms should follow. With increasing frequency, however, many firms have (re)turned to the old fashion and followed the now counterintuitive idea of proactively building loyalty via mutual company and employee commitment (Bernstein, 1998). In other words, they seek to prevent the seeds of leaving (e.g. shock, image violations, and decline in job attitudes) before they get planted.
With these ideas in mind, Booz, Allen, and Hamilton implemented job rotation to help their consultants balance family and work stresses. During periods of unusual family turmoil, for example, consultants can be reassigned to jobs with stable hours and minimal travel; as a result, it is easier for their employees to deal with the work-family stresses. At International Paper (IP) and Citigroup, for instance, career development programs have been implemented. At IP, 13,000 white-collar workers must meet annually (and separately from their performance appraisal meetings) to map their long-term career strategies and their next specific job move. At Citigroup, 10,000 managers are reviewed twice a year to identify their next job placement. In short, labor market imperatives are driving firms to manage the quitting process proactively, and the research in organizational behavior provides strong and compelling managerial tactics.
CONCLUSION
In our view, this chapter mirrors a larger megapoint that underlies this entire volume. More specifically, the research on organizational behavior offers discernible, interpretable, logically consistent, and empirically verifiable foundation principles. With respect to controlling turnover, we confidently advise the following. First, consider whether selective retention and workforce security is sufficiently important in your particular firm’s situation. If it is, then decide whether an employee’s quitting is functional (versus dysfunctional) and avoidable (versus unavoidable). Third, anticipate an employee’s leaving by monitoring job attitudes, global withdrawal cognitions, shocks, paths, and job embeddedness. Fourth, look at the shocks or levels (and reasons) of job dissatisfaction and embeddedness and decide whether there is anything the company can do (e.g. is this turnover avoidable?). Fifth, determine whether that employee is likely to leave more quickly (e.g. paths 1 and 2) or slowly (e.g. paths 3 and 4), which in turn advises a manager to be proactive (for path 1 and path 2 processes) or more reactive (for path 3 and path 4 processes). Finally, gather most or all of this information by routine employee surveys and/or “management by wandering around” (e.g. talk to people).
Simply put, our prescriptions say that turnover is a process that requires more active attention and management than it typically receives. Employee surveys are routinely administered, for example, but this information is infrequently used for specific and proactive interventions aimed at controlling turnover. Similarly, many companies designate their “key” employees whose leaving would be clearly dysfunctional but take little action to understand the events that prompt their thoughts of leaving. Also, companies can better anticipate employees’ reactions to shocks by managing the information immediately before and after the occurrence of a particular jarring event. Job counseling can be made available, for instance, to deflect path 3 and 4 processes. Finally, companies can simply make it harder for employees to leave by “embedding” their key employees in the organization. For example, using teams, having people serve as mentors or having responsibilities for projects maintains attachments (or linkages) to firms. Generous perks and rewards contingent on continued tenure can also render turnover less appealing. In sum, controlling turnover requires substantial understanding of the phenomenon and the willingness to be proactive or reactive in managing the quitting process. Ultimately, appropriate management of turnover can increase organizational effectiveness and the bottom line.
Abelson, M. (1987). Examination of avoidable and unavoidable turnover. Journal of Applied Psychology, 72, 382-386.
Allen, D. G. (2006). Do organizational socialization tactics influence newcomer embeddedness and turnover? Journal of Management, 32, 237-257.
Arnold, H. J., and Feldman, D. C. (1982). A multivariate analysis of the determinants of job turnover. Journal of Applied Psychology, 67, 350-360.
Bentein, K., Vandenberg, R., Vandenberghe, C., and Stinglhamber, F. (2005). The role of change in the relationship between commitment and turnover: A latent growth modeling approach. Journal of Applied Psychology, 90, 468-482.
Bernstein, A. (1998). We want you to stay. Really. Business Week (June 22), 67-72.
Bronson, P. (1999). Instant company. The New York Times Magazine (July 11), 44-47.
Crossley, C. D., Bennett, R. J., Jex, S. M., and Burnfield, J. L. (2007). Development of a global measure of job embeddedness and integration into a traditional model of voluntary turnover. Journal of Applied Psychology, 92, 1031-1042.
Dalton, D. R., Krackhardt, D. M., and Porter, L. W. (1981). Functional turnover: An empirical assessment. Journal of Applied Psychology, 66, 716-721.
Dess, G. G., and Shaw, J. D. (2001). Voluntary turnover, social capital, and organizational performance. Academy of Management Review, 26, 446-456.
Donnelly, D. P., and Quinn, J. J. (2006). An extension of Lee and Mitchell’s unfolding model of voluntary turnover. Journal of Organizational Behavior, 27, 59-77.
Dvorak, P. (2007). Hotelier finds happiness keeps staff checked in. Wall Street Journal (December 17), B3.
Felps, W., Mitchell, T. R., and Byington, E. (2006). How, when, and why bad apples spoil the barrel: Negative group members and dysfunctional groups. Research in Organizational Behavior, 27, 181-230.
Felps, W., Mitchell, T., Hekman, D., Lee, T., Holtom, B., and Harman, W. (in press). When staying depends on others: Collective job embeddedness as a predictor of employee turnover. Academy of Management Journal.
Griffeth, R. W., Hom, P. W., and Gaertner, S. (2000). A meta-analysis of antecedents and correlates of employee turnover: update, moderator tests, and research implications for the millennium. Journal of Management, 26, 463-488.
Harding, D., and Rouse, T. (2007). Human due diligence. Wall Street Journal (October 2), A16.
Hoffman, B. J., and Woehr, D. J. (2006). A quantitative review of the relationship between person-organization fit and behavioral outcomes. Journal of Vocational Behavior, 68, 389-399.
Holtom, B. C., and Inderrieden, E. (2006). Integrating the unfolding model and job embeddedness to better understand voluntary turnover. Journal of Managerial Issues, 18, 435-452.
Holtom, B. C., Mitchell, T., Lee, T., and Inderrieden, E. (2005). Shocks as causes of turnover: What they are and how organizations can manage them. Human Resource Management, 44, 337-352.
Hom, P. W., and Griffeth, R. W. (1995). Employee Turnover. Cincinnati, OH: South-Western College Publishing.
Kammeyer-Mueller, J. D., Wanberg, C. R., Glomb, T. M., and Ahlburg, D. (2005). The role of temporal shifts in turnover processes: It’s about time. Journal of Applied Psychology, 90, 644-658.
Lee, T. W., and Maurer, S. D. (1997). The retention of knowledge workers with the unfolding model of voluntary turnover. Human Resource Management Review, 7, 247-275.
Lee, T. W., and Mitchell, T. R. (1994). An alternative approach: the unfolding model of voluntary employee turnover. Academy of Management Review, 19, 51-89.
Lee, T. W., Mitchell, T. R., Holtom, B. C., McDaniel, L. S., and Hill, J. W. (1999). The unfolding model of turnover: a replication and extension. Academy of Management Journal, 42, 450-462.
Lee, T. W., Mitchell, T. R., Sablynski, C. J., Burton, J. P., and Holtom, B. C. (2004). The effects of job embeddedness on organizational citizenship, job performance, volitional absences, and voluntary turnover. Academy of Management Journal, 47, 711-722.
Lee, T. W., Mitchell, T. R., Wise, L., and Fireman, S. (1996). An unfolding model of voluntary employee turnover. Academy of Management Journal, 39, 5-36.
Maertz, C. P. Jr., and Campion, M. A. (1998). 25 years of voluntary turnover research: A review and critique. In C. L. Cooper and I. T. Robertson (eds), International Review of Industrial and Organizational Psychology (Vol. 13, pp. 49-81). New York: John Wiley and Sons .
Maertz, C. P. Jr., and Campion, M. A. (2004). Profiles in quitting: integrating process and content turnover theory. Academy of Management Journal, 47, 566-582.
Maertz, C. P. Jr., and Griffeth, R. W. (2004). Eight motivational forces and voluntary turnover: A theoretical synthesis with implications for research. Journal of Management, 30, 667-683.
Mallol, C., Holtom, B. C., and Lee, T. W. (2007). Job embeddedness in a culturally diverse environment. Journal of Business and Psychology, 22, 35-44.
Meyer, J. P., Stanley, D. J., Herscovitch, L., and Topolnytsky, L. (2002). Affective, continuance, and normative commitment to the organization: A meta-analysis of antecedents, correlates, and consequences. Journal of Vocational Behavior, 61, 20-52.
Mitchell, T. R., Holtom, B. C., Lee, T. W., Sablynski, C., and Erez, M. (2001). Why people stay: Using job embeddedness to predict voluntary turnover. Academy of Management Journal, 44, 1102-1121.
Morrell, K. (2005). Towards a typology of nursing turnover: The role of shocks in nurses’ decision to leave. Journal of Advanced Nursing, 49, 315-322.
Morrell, K. Loan-Clarke, J., and Wilkinson, A. (2004). The role of shocks in employee turnover. British Journal of Management, 15, 335-349.
Morrell, K., Loan-Clarke, J., Arnold, J., and Wilkinson, A. (in press). Mapping the decision to quit: A refinement and test of the unfolding model of voluntary turnover. Applied Psychology: An International Review.
Peters, T. J., and Waterman, R. H. (1982). In Search of Excellence. New York: Harper and Row. Shaw, J. D., Duffy, M. K., Johnson, J. L., and Lockhart, D. E. (2005). Turnover, social capital losses, and performance. Academy of Management Journal, 48, 594-606.
Spector, P. E. (1997). Job Satisfaction. Thousand Oaks, CA: Sage Publications.
Spors, K. (2007). Top Small Workplaces, 2007. Wall Street Journal (October 1), R1.
Sutton, R. I. (2007). The No Asshole Rule: Building a Civilized Workplace and Surviving One that Isn’t. Warner Business Books.
Tanova, C., and Holtom, B. (in press). Using job embeddedness factors to explain voluntary turnover in 4 European countries. International Journal of Human Resource Management.
Shock questionnaire
On a piece of paper, quickly provide answers to the following questions:1. Have you ever left a job in response to a shock (a jarring event that caused you to reconsider your attachment to the organization)?
2. Do you know any people (e.g. friends, family, colleagues, former classmates) who have left a job in response to a shock?
3. What type of shocks did you or they experience?
4. Categorize the shocks according to the following criteria:a. Personal vs organizational
b. Expected vs unexpected
c. Positive vs neutral vs. negative
5. What responses to the shocks were enacted by the organizations?
6. Were these responses effective in stemming turnover? If not, what should have been done?
Job embeddedness
Recently, a new theory has been designed to help organizations reduce voluntary turnover by increasing “job embeddedness.” Empirical evidence suggests that such efforts are typically complementary to the traditional approach (e.g. increasing job satisfaction or organizational commitment) because job embeddedness focuses on different factors. Take a few minutes to think through the following issues with a small group of people from your organization.
How can your organization systematically increase each of the following aspects of job embeddedness among employees?
1. Fit in the organization (perceived congruence with the values, culture, etc.).
2. Links in the organization (number of people, groups and projects employees are meaningfully attached to).
3. Sacrifice related to leaving the organization (specific perks, benefits, or hardships incurred upon leaving).
4. Fit in the community (congruence with the values, amenities, offerings in an area).
5. Links in the community (number of institutions, friends, and family that employees are connected with).
6. Sacrifice related to the community (community-based issues that would arise if an individual leaves an organization; relocation to a new city is not required; for example, new commute patterns).
Part IV