Men's Rights Agency - Domestic Violence
Aggression in British Heterosexual Relationships: A Descriptive Analysis
Michelle Carrado, M.J. George, Elizabeth Loxam, L. Jones and Dale Templar
2. MATERIALS AND METHODS
The survey of heterosexual adults was conducted on a sample of 1,978 UK adults aged 15 and over within a regular commercial bimonthly survey ("Omnibus Survey", Market Opinion and Research International [MORI]) seeking to determine consumer and social attitudes. The total survey was conducted as face-to-face interviews in the respondent's home by trained interviewers (75% female) who conduct these surveys regularly. The section containing questions relating to conflict tactics (verbal reasoning, verbal or symbolic victimization, and physical victimization) in personal relationships was administered as a self-completion instrument. Sampling quotas were used to ensure that the sample was representative of sex, age, socioeconomic group, relationship status, and geographical region of the adult population of the United Kingdom. To achieve this, 150 sampling points were used and data obtained were weighted to reflect the known profile of the adult population of the United Kingdom as determined from the most recent national census data.
The questions administered in this section of the survey were derived from the CTS as devised by Straus [1979]. The exact items of the original CTS were not used in every case, although items were derived from it and based on the concept of progressive series of escalating levels of possible conflict as outlined by the CTS [Straus, 1979]. The items used were as follows and are given the letters indicating their position in the 12 items asked:
Items of Physical Victimization
E. Your partner has, with some force, pushed, grabbed, bitten, scratched,
or shoved you (you have, with some force, pushed etc.).
C. Your partner has slapped you (you have slapped your partner).
D. Your partner has punched or kicked you (you have punched or kicked your
partner).
H. Your partner has thrown a heavy object at you, smashed something over
you or hit you with a heavy object (you have thrown ...etc).
J. Your partner has struck you with a sharp or pointed object (you have struck
... etc.).
A total of three questions each using this scale were used to gain data on
conflict tactics. Two questions were asked in the context of any past or
present heterosexual relationship, where one question related to victimization
sustained and one to victimization inflicted. The third question asked for
victimization sustained solely within current relationships. Thus, respondents
supplied data concerning victimization both sustained and inflicted in any
of their heterosexual relationships and victimization sustained in their
current relationship. In each case, the scale items were prefaced with an
introduction that stated:
Thinking about any personal relationships you have ever had (with a member of the opposite sex) which, if any, of the actions below have you ever had done to you by a partner, boyfriend/girlfriend, husband/wife when one or both of you has disagreed or quarrelled?
As appropriate, this wording was changed to "thinking about your current relationship ... " to ascertain victimization sustained in current relationships and " ... have you ever done to a partner ...." to ascertain victimization inflicted against a partner in any relationship.
Eliciting data in terms of relationships, rather than a specified time period, was a similar approach to that adopted in a previous small-scale conflict tactics study conducted in the United Kingdom [Archer and Ray, 1989; Russell and Hulson, 1992]. This particular methodology was preferred as it obviates criticisms leveled when conflict tactics are only asked in context to a period of, for instance, the last year. It was not possible to ask the extent to which respondents had experienced or committed each form of assault on more that one occasion. Some evidence of multiple victimization (sustained or inflicted) has been derived by identifying the numbers of individuals reporting more than one category of assault.
In a further two questions, respondents were asked to ascribe possible context and meaning to the conflict tactics either sustained or inflicted in all past and present relationships. Items used in these two questions were formulated de novo and sought to offer respondents a series of alternatives which included such explanations as self-defense, instrumental or expressive reasons, or the involvement of alcohol. Respondents were allowed to identify any number of items, rather than given a forced choice, as where more than one assault was identified different contexts might be possible. The items used in these two questions are shown in detail below:
Items for Reason and Context
A. He/she thought it was the only way to get through to me / I thought it
was the only way to get through to him/her.
B. He/she was getting back at me for something nasty I said or threatened
to do to him/her I was getting back at him/her for something nasty he/she
said or threatened to do to me.
C. He/she was getting back at me for some physical action I had used against
him/her/I was getting back at him/her for some physical action he/she had
used against me.
D. To stop me doing something/ to stop him/her doing something.
E. To make me do what he/she wanted / to make him/her do what I wanted
F. He/she thought I was about to use a physical action against him/her /
I thought he/she was about to use a physical action against me.
G. He/she was "under the influence" of, for instance alcohol at the time
/ I was "under the influence" of, for instance, alcohol at the time.
H. It is or was in his/her character, that's the way he/she is or was / It
is my character, that's the way I am.
In addition, respondents could cite "other," "no particular reason," or "don't know" to these questions. Although the options offered are by no means extensive, they allow preliminary analysis of the context and reason respondents considered for the conflict tactics reported. All the data for survey forms were collated and computerized and tables of results were produced by MORI in the form of simple descriptive statistics. The levels of sample difference needed for statistical significance at the 5% level for this survey data were 3-4% when comparing samples of between 500 and 1,000 and 5-8% when comparing samples of 200-500, where between 10 and 30% ( or between 90 and 70%) of a sample respond positively.
Go Back to UK DV Survey Introduction/Index page
Go forward to: 3. Results - Overall Incidence of Victimization - Item Victimization