This note reexamines the generally accepted belief that persons with discrimination-related grievances are much less likely to complain about their problem than are persons with grievances arising from consumer purchases, torts, or other common kinds of personal problems. We find that previously reported analyses greatly overstate the gap between complaining in discrimination problems and other kinds of problems. Drawing on data from three surveys, each conducted in a different country (the United States, Canada, and Australia), we find that for some types of discrimination problems the level of complaining in fact equals or exceeds complaining in other arenas.

Injury is common in all societies. Americans are perceived as quick to respond to injury by turning to the legal system. This article compares compensation seeking by Americans and Canadians, examining the degree to which cultural factors shape the response of injured parties in the two countries and the extent to which resources and experience influence individual action. Drawing on two large-scale telephone surveys, one conducted in five federal judicial districts around the United States and one conducted in the Canadian province of Ontario, the article looks at the factors that influence claiming and seeking legal assistance. The overall patterns indicate that residents of Ontario are somewhat less likely to claim but more likely to seek legal assistance than are residents of the United States. Moreover, while cultural variations (e.g., religion, type of residence) are good predictors of claiming in Ontario, these factors have little influence on claiming in the United States. As for seeking legal assistance, few predictors are found to influence behavior in the United States while a variety of factors (community size, type of problem, stakes, gender, and education) influence behavior in Ontario.

McEwen and Maiman (1986) have disagreed with my claim that the case characteristic of admitted liability explains more variability in dispute outcome and compliance than whether the case was resolved through a mediation or adjudication forum. Those authors reanalyzed some of my data from an Ontario small claims court and concluded that forum type is the stronger variable. I take issue with them on a number of conceptual and methodological points. In my own reanalysis of the Ontario data I am able to demonstrate statistically that admitted liability is the stronger predictor of outcomes. I also discuss why this should be so and raise some questions about compliance. Whether we can generalize to McEwen and Maiman's data from Maine courts is a matter of speculation, but I am inclined to infer that we can. Our debate raises important issues in the assessment of dispute resolution.

In this paper disputes are seen as varying along a dimension of admitted liability, that is, the extent to which defendants admit some obligation to plaintiffs; they may admit no liability, partial liability, or full liability. This conceptualization was used in an empirical study of a small claims court. The results paint a portrait of the court that is at variance with most of the previous literature. Consumer issues constitute a substantial portion of the court caseload. On average, defendants, including individual consumers, do well when they dispute claims. Among disputed cases, small rather than large businesses predominate. Prior literature has suggested that, in comparison to adjudication, mediation of claims produces compromise outcomes and higher rates of compliance. This research shows that mediation yields a large percentage of all-or-none results, but to the extent that there is compromise and compliance, it can be partly ascribed to admitted liability characteristics. Some data on defaulted cases are also presented.

This paper develops a theoretical framework for investigating the socialpsychological dynamics of punishment reactions to criminal offenders. Two basic types of punishment motives are delineated: behavior control and retribution. The target, or focus, of these motives may be the offender or a broader social audience. Within each of the four cells defined by this classification, we review the literature describing how punishment responses are influenced by characteristics of the rule, the offense, and the offender, as well as by the attitude and personality of the reactor. A large number of empirically testable propositions are generated.