Academic journal article Journal of Comparative Family Studies

State-to-State Differences in Social Inequality and Social Bonds in Relation to Assaults on Wives in the United States

Academic journal article Journal of Comparative Family Studies

State-to-State Differences in Social Inequality and Social Bonds in Relation to Assaults on Wives in the United States

Article excerpt




Most research on the causes of wife-beating has used data on the characteristics of the persons involved (such as their education or personality) or the characteristics of the relationship (such as the degree of male-dominance in the marriage). When the research is done by sociologists, they generally make inferences from these individual or "micro level" characteristics to the "macro" or societal level characteristics from which the characteristics presumably spring. For example, macho men and male dominance in marriage are seen as a reflection of a violent and male dominant society. However, just as studies using data on social units are a questionable basis for making inferences about individuals -- what Robinson (1950) called the "ecological fallacy" -- there is also danger of what can be called the "individualistic fallacy" because one cannot be sure that relationships found by studying individuals apply to social units. Few studies have directly measured the characteristics of the society that presumably produces these macho men and male-dominant marriages.

The research reported in this paper was undertaken in response to the need to test theories about the social causes of marital violence with data on the characteristics of society itself (Menzel, 1950). It uses data on rates of wife-assault in each of the 50 states of the U.S. as the dependent variable and seeks to explain why those rates differ from state to state. The theoretical focus of the paper is on three of the most frequently discussed explanations of differences between societies in the rate of wife-beating. Two of these theories concern social stratification, and specifically the violence-producing effect of inequality, and the third theory focuses on social integration versus disorganization. Investigation of these three theories was partly a matter of what data were available and should not be taken to imply that these are necessarily the most important theories. Indeed, Gelles and Straus's review (1979) covers many other theories and there is at least some evidence supporting most of them.


Both conflict theory and its more specific version in the form of the feminist "patriarchical society" theory hold that the lower the status of women in a society, the greater the frequency of wife beating. Feminist scholars have used historical and case study data to show the link between gender inequality at the social system level and wife-beating (Dobash and Dobash, 1979; Martin, 1976). There is also evidence from Levinson's cross-cultural anthropological study (1989:58) showing that wifebeating tends to be less common when women have independent social and economic resources. These propositions need to be tested using comparative data on contemporary industrial societies.

Previous research by my colleagues and I found a higher rate of assault on wives in male-dominant marriages than in equalitarian marriages (Allen and Straus, 1980; Coleman and Straus, 1986; Straus, 1973; Straus, Gelles and Steinmetz, 1980). However, these studies do not directly test the theory that a male-dominant society tends to use violence to keep women in their place because it uses data on individual persons and families. That void led Kersti Yllo and me to carry out a study using the states of the United States as the societal units (Yllo, 1983; Yllo and Straus, 1990).

The Yllo and Straus research was based on an index designed to measure state-to-state differences in the "status of women." This index was composed of indicators of the extent to which women approached equality with men in their economic, political, educational, and legal statues. Extremely large differences between the states were found using this measure. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.