Log in / Register
Home arrow Psychology arrow Personnel psychology and human resource management


What do we know about the correlates, causes, and consequences of absence that we didn't know 1 5 years ago? Because of its less than ideal psychometric properties, absenteeism has particularly profited from the advantages of meta- analytic summary. In the domains of job satisfaction and demographics, we now have a solid base of population estimates of effect sizes, a fact that both stimulates shifts in research emphasis (from job satisfaction to commitment) and challenges researchers to explain these effects (e.g. gender and absence). In the domains of performance and turnover, wide confidence intervals in estimated correlations with absence have inspired good theoretical thinking about likely moderators.

Fifteen years ago, because few had even asked, we knew very little about how people viewed their own absence behavior or that of others. Now, we know that people have a tendency to underestimate their own absence and hold inflated views of the absence of others. Managers and employees hold different standards about how much absence is acceptable, but managers are capable of applying reasonable attributional criteria to judgments about absence. Key variables from the theory of reasoned action account for variance in intentions to be absent. Within-person decision models show that the meanings of absence vary among individuals despite some more universal causal mechanisms.

Fifteen years ago, little was known about the social and cultural dynamics of absenteeism. In my opinion, the enhancement of knowledge on this front represents the greatest advance during the review period. Meaningful crossgroup differences and normative mechanisms are now well established, and it is becoming clear that supervisory groups are often critical in shaping the attendance patterns of their members. The social perspective is important because it forces the researcher to attend to a factor that seems to be critically important in matters of absence and attendance – work context. In this domain, the conflict model provides some especially good examples. As noted at several points earlier, the social perspective also holds excellent promise as a partial explanatory mechanism for phenomena covered under the other models, including how gender affects absence and how absence is or is not legitimated as a response to stress or ill health.

Fifteen years ago, with the exception of some experimental tests of interventions, the standard study of absence passed out a questionnaire in the workplace and obtained absence data from personnel records. Although this is still true, it is important to remember that the advances noted above have stemmed from a remarkable diversity of methods, including the use of observation, interviews, diaries, vignettes, self-reports, supervisory reports, telephone surveys, household visits, within-person designs, content analysis, and meta-analysis. Absence is difficult to study, and one should use what works. This diversity (uncommon in many research areas) is to be commended.

Fifteen years ago, there were very few studies of work stress and absenteeism. Now, there are many, but they do little to enhance our general understanding of absence, either because they fail to probe rigorously the possible adjustive aspects of absence or because they fail to consider when absence might be the preferred reaction to stress over other responses.

Except for the micro-level studies of time valuation, the fruits of the economic model are generally unenlightening, often replicating effects seen in firm-level studies. This is unfortunate, because the economic model seems moderately well equipped to probe some aspects of a very important issue – the impact of nonwork factors on absence. The lack of information on these factors was decried 15 years ago Johns & Nicholson, 1982), and this is still a serious omission from the absence research agenda.

The increasing interest in absenteeism by medical researchers holds promise if absenteeism is viewed as a psychologically mediated response to medical condition and not simply as an economic indicator of the cost of sickness or the efficacy of treatment. This requires that research in the medical domain incorporate more understanding of workplace psychology and behavior.

In my opinion, progress in the past 15 years has been due to creative anarchy in theory and method. In this sense, the lack of a dominant paradigm or model has been helpful. On certain terms and conditions, some organization of this anarchy might be welcome. However, two temptations should be avoided, both of which have to do with putting quantity before quality. At one extreme, I have already alluded to the fallacy of organizing empirically with a regression equation stocked with variables from more and more models. We already have evidence that knowledge accruing from this approach is literally noncumulative (e.g. Johns, 1978) and simply trying to account for more variance is a futile task in any event (Nicholson & Martocchio, 1995). At the other extreme, organizing theoretically with a grand process model has the limitations alluded to at the beginning of the chapter.

What is needed now are probably some small theories that cross-cut two or more of the models described above. For example, one can conceive of a theory that combines differential medical susceptibility, culture, and notions of deviance to explain in part why women are often absent more than men and when they might not be. A series of such small theories would better organize our understanding of absenteeism while still respecting the wide varieties of meanings that absence holds for individuals and groups.


This research was supported by grant 94ER0506 from Quebec's Fonds pour la Formation de Chercheurs et l'Aide a la Recherche and grant 410-920202 from the Social Sciences and Humanities Research Council of Canada.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science