Conflict Tactics Scale (CTS)

Two general types of incidence surveys exist: the Conflict Tactics Scale (CTS) and the Crime Victim Surveys (CVS). The former requires people to indicate what actions they have taken to resolve family conflicts; the latter requires people to indicate by what crimes they have been victimized. The CVS find the rates of reporting wife assault to the police comparable with the reporting rates for other assaults. However, these surveys have a filtering problem such that people who do not consider their abuse victimization to be a crime do not respond in the affirmative. Hence, incidence rates of reported spousal abuse, which are not defined as criminal by the victim, are low. To circumvent this filtering problem, Straus, Gelles, and Steinmetz devised the CTS, which asks respondents to report modes of conflict resolution in the family. This avoids the problem of whether the respondent defines the action as criminal or not and, therefore, attempts to obtain more accurate estimates of the frequency and incidence of domestic assault in a general population. Straus found violence incidence rates with the CTS were 16 times greater than with the CVS. Presenting the CTS in the context of normal conflict rather than a criminal act reduces filters against reporting.

Surveys of Incidence: Conflict Tactics Surveys

Several surveys using the CTS have been completed. They include (a) a nationally representative U.S. sample of 2,143 interviewed in 1974 by Response Analysis Corporation; (b) a survey of spousal violence against women in the state of Kentucky, which interviewed 1,793 women; (c) a second national survey completed by Straus and Gelles in 1985; and (d) a sample of 1,045 for the Province of Alberta, Canada. These were each obtained by a survey that interviews a representative sample drawn from a general population about experiences of being victimized by violence during family conflicts and the type of actions used to resolve these conflicts. These rates refer to the use of violence at any time in the marriage and may include both unilateral and reciprocal violence. Straus reported approximately equal perpetration rates by gender. This common measure enables some direct comparison between these surveys.

Kennedy and Dutton used a combination of face-to-face meeting and random-digit dialing techniques to survey 1,045 residents in Alberta, Canada, leading to a comparison of American and Canadian rates of wife assault. The “minor” violence rates for the two coun-tries are virtually identical, but the American “severe violence” rates were higher than the Canadian rates. By way of comparison with these North American data, Kim and Cho reported that the Korean intimate partner violence rate was 37.5% for wife assault (any violence) in the preceding year versus 11.6% reported by Straus et al. In 1985, Fumagai and Straus found a lifetime incidence of wife assault in Japan of 58.7% versus 22% in the United States.

Fals-Stewart, Birchler, and Kelley asked 104 men in a spousal violence treatment program and their partners to keep a weekly diary identifying days of physical aggression and a daily CTS checklist for violence. Male-to-female agreement on “violence days” was better after treatment than before, perhaps because couples were aware of tracking. Interestingly, the women were violent on more days than the men, regardless of whose report was read.

Moffitt et al. confirmed Straus’s point in one of the best methodological studies of intimate violence to date, the Dunedin study. When asked about “assault victimization,” they found that respondents reported rates of male violence that were much higher than the rates of female violence, and both rates were quite low. When they asked the same respondents about “relationships with partners,” the rates reported by both genders were much higher and equivalent.

Criticisms of the Conflict Tactics Scale

Some criticisms have been made about the CTS: (a) the CTS ignores the context in which the violence occurred, (b) differences in gender size between men and women make acts scored the same on the CTS quite different in reality, (c) impression management or social desirability factors may preclude people from answering the CTS accurately, and (d) the CTS queries violence occurring in a conflict and may miss “out of the blue” violence.

Straus’s rejoinders to these criticisms are as follows. First, the assessment of context should be done separately because there are so many context variables that they could not all be included on the CTS. The CTS is designed in such a way that any special set of context questions can be easily added. Second, a similar problem, Straus notes, is that repeated slap-ping is highly abusive and dangerous but gets counted as Minor Violence on the CTS. Straus argues that while it is possible to weight actions by differences in size between perpetrator and victim, or to construct an upper limit after which slapping gets counted as Severe Violence, such weightings have rarely led to changes in research results. Third, the social desirability criticism was answered, in part, by a study by Dutton and Hemphill, which correlated scores on two measures of social desirability (the tendency to present a “perfect image” on self-report tests) and scores on the CTS. Social desirability is measured by a test called the Marlowe-Crowne test (MC), which assesses the tendency to present the self in a socially acceptable manner. MC scores did correlate significantly; the higher their social desirability score, the lower their reported rates of verbal abuse. However, it did not correlate with their reports of physical abuse, nor with any reports of abuse (verbal or physical) made against them by their wives. Hence, it seems that reports of physical abuse are largely uncontaminated by socially desirable responding. This means that the incidence survey rates are probably fairly accurate as far as image management is concerned. Finally, the vast majority of violent acts are perceived as emanating from conflict. While the CTS may miss an out-of-the-blue attack, it more than makes up for this with its increased sensitivity over crime victim surveys.


  1. Browning, J., & Dutton, D. G. (1986). Assessment of wife assault with the Conflict Tactics Scale: Using couple data to quantify the differential reporting effect. Journal of Marriage and the Family, 48, 375-379.
  2. Straus, M. A., & Gelles, R. J. (Eds.). (1990). Physical violence in American families. New Brunswick, NJ: Transaction.
  3. Straus, M. A., Hamby, S. L., & Boney-McCoy, S. (1996). The revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316.

See also: