oblJournal of Applied Psychology 1977, Vol. 62, No. 2, 237-240 Intermediate Linkages in the Relationship Between Job Satisfaction and Employee Turnover William H. Mobley University of South Carolina The relationship between job satisfaction and turnover is significant and consistent, but not particularly strong. A more complete understanding of the psychology of the withdrawal decision process requires investigation beyond the replication of the satisfaction-turnover relationship.
Toward this end, a heuristic model of the employee withdrawal decision process, which identifies possible intermediate linkages in the satisfaction-turnover relationship, is presented. Previous studies relevant to the hypothesized linkages are cited, and possible avenues of research are suggested. A schematic representation of the withdrawal decision process is presented in Figure 1. Block A represents the process of evaluating one’s existing job, while Block B represents the resultant emotional state of some degree of satisfaction-dissatisfaction.
A number of models have been proposed for the process inherent in Blocks A and B—for example, the value-percept discrepancy model (Locke, 1969, 1976), an instrumentalityvalence model (Vroom, 1964), a met-expectations model (Porter & Steers, 1973), and a contribution/inducement ratio (March & Simon, 1958). Comparative studies -that test the relative effiMuch more emphasis should be placed in the cacy of these and other alternative models of future on the psychology of the withdrawal satisfaction continue to be needed. process. . . Our understanding of the manner Most studies of turnover examine the direct in which the actual decision is made is far relationship between job satisfaction and turnfrom complete, (p. 173) over. The model presented in Figure 1 suggests The present paper suggests several of the pos- a number of possible mediating steps between sible intermediate steps in the withdrawal decision dissatisfaction and actual quitting. Block C sugprocess (specifically, the decision to quit a job). gests that one of the consequences of dissatisPorter and Steers (1973) suggested that expressed faction is to stimulate thoughts of quitting. intention to leave” may represent the next log- Although not of primary interest here, it is recogical step after experienced dissatisfaction in the nized that other forms of withdrawal less extreme withdrawal process. The withdrawal decision than quitting (e. g. , absenteeism, passive job beprocess presented here suggests that thinking of havior) are possible consequences of dissatisfaction (see e. g. , Brayfield & Crockett, 195S; Kraut, quitting is the next logical step after experienced 197S). issatisfaction and that “intention to leave,” folBlock D suggests that the next step in the lowing several other steps, may be the last step withdrawal decision process is an evaluation of prior to actual quitting. the expected utility of search and of the cost of quitting. The evaluation of the expected utility of search would include an estimate of the Preparation of this paper was supported by a chances of finding an alternative to working in grant from the South Carolina Business Partnership the present job, some evaluation of the desirFoundation.
Requests for reprints should be sent to William ability of possible alternatives, and the costs of H. Mobley, College of Business Administration, search (e. g. , travel, lost work time, etc. ). The University of South Carolina, Columbia, South evaluation of the cost of quitting would include Carolina 29208. suc’h considerations as loss of seniority, loss of 237 Reviews of the literature on the relationship between employee turnover and job satisfaction have reported a consistent negative relationship (Brayfield & Crockett, 19SS; Locke, 197S; Porter & Steers, 1973; Vroom, 1964).
Locke (1976) noted that while the reported correlations have been consistent and significant, they have not been especially high (usually less than . 40). It is probable that other variables mediate the relationship between job satisfaction and the act of quitting. Based on their extensive review, Porter and Steers (1973) concluded the following: 238 SHORT NOTES clft iJ ,Pi 1^ : i 1 1 A. i * i* Evaluation of Existing Job f,-; i« 1! J 1! L B . l L Experienced Job Satisfaction4… 1 * Dissatisfaction a) Alternative forms of withdrawal, e. g. (a )Aite absenteeism, passive job behavior absents 1 Thinking of Quitting J Evaluation of Expected Utility of Search and Cost of Quitting L E. IL LJL-: 1 Intention to Search for Alternatives < (b )Nor (b) Non-job related factors e. g. , I 1 transfe of spouse, may stimulate transfer intent! (c) Unsolicited or highly visible alternatives may stimulate evaluation F. I-L-. Search for Alternatives G. 1 1 1 J Comparison of Alternatives vs. Present Job d) One alternative may be withdrawal from labor market »1_ 1. 1 Intention to Quit/Stay 1 i Figure 1. The employee turnover decision process. vested benefits, and the like. This block incorporates March and Simon’s (1958) perceived ease of movement concept. If -the costs of quitting are high and/or the expected utility of search is low, the individual may reevaluate the existing job (resulting in a change in job satisfaction), reduce thinking of quitting, and/or engage in other forms of withdrawal behavior.
Research is still needed on the determinants of alternative forms of withdrawal behavior and on how the expression of withdrawal behavior changes as a function of time and of changes in or revaluation of the environment. If there is some perceived chance of finding an alternative and if the costs are not prohibitive, the next step, Block E, would be behavioral intention to search for an alternative (s). As noted by Arrow (b) in Figure 1, non-job-related factors may also elicit an intention to search (e. g. , transfer of spouse, health problem, etc. ). The intention to search is followed by an actual search (Block F).
If no alternatives are found, the individual may continue to search, reevaluate the expected utility of search, reevaluate the existing job, simply accept the current state of affairs, decrease thoughts of quitting, and/or engage in other forms of withdrawal behavior (e. g. , absenteeism, passive job behavior). (e) Impulsive Behavior SHORT NOTES If alternatives are available, including (in some cases) withdrawal from the labor market, an evaluation of alternatives is initiated (Block G). This evaluation process would be hypothesized to be similar to the evaluation process in Block A.
However, specific job factors the individual considers in evaluating the present job and alternatives may differ. (See Hellriegel & White, 1973; and Kraut, 1975, for a discussion of this point. ) Independent of the preceding steps, unsolicited or highly visible alternatives may stimulate this evaluation process. The evaluation of alternatives is followed by a comparison of the present job to alternative(s) (Block H). If the comparison favors the alternative, it will stimulate a behavioral intention to quit (Block I), followed by actual withdrawal (Block J).
If the comparison favors the present job, the individual may continue ‘to search, reevaluate the expected utility of search, reevaluate the existing job, simply accept the current state of affairs, decrease thoughts of quitting, and/or engage in other forms of withdrawal behavior. Finally, Arrow (e) gives recognition to the fact that for some individuals, the decision to quit may be an impulsive act involving few, if any, of the preceding steps in this model. The relative incidence and the individual and situational determinants of an impulsive versus a subjectively rational decision process presents yet another area of needed research.
The model being described is heuristic rather than descripitve. There may well be individual differences in the number and sequence of steps in the withdrawal decision process, in the degree to which the process is conscious, and as noted earlier, in the degree to which the act of quitting is impulsive rather than based on a subjectively rational decision process. One value of such an heuristic model is to guide thinking and empirical research toward a valid descriptive model that can account for such individual differences.
There is a lack of research evaluating all or even most of the possible steps in the withdrawal decision process. There have been a few studies that have tested one or two of the intermediate linkages proposed in the present note. Mobley (Note 1) found high negative correlations between satisfaction and frequency of thinking of quitting (Blocks B and C). Atkinson and Lefferts (1972), who dealt with the association between Blocks C and J, found that the frequency with which people thought about quitting •their job was significantly related to actual termination.
Kraut (1975), looking at the associations among Blocks B, I, and J, found significant cor- 239 relations between expressed intention to stay and subsequent employee participation. These correlations were much stronger than relationships between expressed satisfaction and continued participation. Finally, Armknecht and Early’s (1972) review is relevant to the relationships between Blocks D and/or F and Block J. They concluded that voluntary terminations are closely related to economic conditions. Each of these studies fails to look at a complete withdrawal decision process.
Such research would appear to be sorely needed. Several researchable questions that follow from the withdrawal decision process described in the present note were mentioned earlier, Additional questions include the following. Do individuals evaluate the expected utility of search? If so, what are the determinants and consequences of this evaluation? What are the consequences and determinants of behavior in the face of an unsuccessful search? In such cases, do individuals persist in search, reevaluate their existing jobs, reevaluate the cost of search, or engage in other forms of withdrawal?
Is the process and/or content for evaluating alternative jobs the same as for evaluating the present j o b ? Does satisfaction with the present job change as a function of the availability or evaluation of alternatives? Attention to these sorts of questions rather than a continued replication of the direct relationship between job satisfaction and turnover would appear to be warranted. Particularly useful would be the longitudinal analysis of the variables and linkages suggested by the model.
Such research would be responsive to Porter and Steer’s (1973) conclusion that more emphasis should be placed on the psychology of the withdrawal decision process. Reference Note 1. Mobley, W. H. Job satisfaction and thinking of quitting (Tech. Rep. 7S-3). Columbia: University of South Carolina, College of Business Administration, Management and Organizational Research Center, 1975. References Armknecht, P. A. , & Early, J. F. Quits in manufacturing: A study of their causes. Monthly Labor Review, 1972, 11, 31-37. Atkinson, T. J. , & Lefferts, E. A.
The prediction of turnover using Herzberg’s job satisfaction technique. Personnel Psychology, 1972, 25, 53-64. Brayfleld, A. H. , & Crockett, W. H. Employee attitudes and employee performance. Psychological Bulletin, 1955, 52, 396-424. 240 SHORT NOTES oj industrial and organizational psychology. Chicago: Rand-McNally, 1976. March, J. G. , & Simon, H. A. Organizations. New York: Wiley, 1958. Porter, L. W. , & Steers, R. M. Organizational, work, and personal factors in employee turnover and absenteeism. Psychological Bulletin, 1973, SO, 151176. Vroom, V. H. Work and motivation. New York: Wiley, 1964.
Hcllriegel, D. , & White, G. E. Turnover of professionals in public accounting: A comparative analysis. Personnel Psychology, 1973, 26, 239-249. Kraut, A. I. Predicting turnover of employees from measured job attitudes. Organizational Behavior and Hitman Performance, 1975, 13, 233-243. Locke, E. A. What is job satisfaction? Organizational Behavior and Human Performance, 1969, 4, 309336. Locke, E. A. Personnel attitudes and motivation. Annual Review oj Psychology, 1975, 26, 457-480. Locke, E. A. The nature and consequences of job satisfaction. In M. D. Dunnette (Ed. ), Handbook Received February 5, 1976 •