The authors do not indicate why they decided to exclude sexual violence from their study.
In their analysis, Button and Gealt (2010) use several variables independently. These are: substance abuse, delinquency, aggression, and sibling violence. To measure independent variables, they used several dependant variables for each. For example, to measure substance abuse (which refers to the use of cigarettes, marijuana, and alcohol in the past thirty days), they asked respondents to indicate frequency of abuse, ranging from 0 to 31 times during thirty-day period (0 = none, 1 = less than one, 2 = 1 to 5, etc.). To measure delinquency, they scaled the frequency of six items that ranged from never to almost every day (0 = never, 1 = before but not in the past, up to 7 = almost every day). These items were: stealing something from a store without paying for it; breaking into a car, house, or other building; cheating on a test; sneaking money from an adult's wallet, purpose, or other place; damaging or destroying property that does not belong to them; and skipping or missing classes without permission. And to measure aggression, the authors asked respondents to indicate if they had hit someone with the intention of hurting them in the past month. Independent variables in this equation were 0 = no, and 1 = yes.
The authors measured sibling aggression by asking respondents to indicate if a sibling had experienced any of the following in the past month: verbal abuse; threats (psychological abuse), shoving, pushing, or slapping; fights -- punching, kicking; and fights with the threat of weapon use (physical abuse). They used six scales to measure the frequency...
They used Age as a categorical variable (12 years and younger and 19 years and older). Gender was used as a self-reported dichotomous variable (1 = male, 2 = female). Finally, they asked respondents questions about the occurrences and frequency of child maltreatment and domestic violence they witnessed.
The study used data coming from a sample of Delaware public school students. The survey is based on a census of all eighth and eleventh grade students of all public schools within the state of Delaware. Button and Gealt (2010) used data from the year 2007 school year. They collected data for all students but used sample for students who had a sibling living with them. In total, data for 8,122 students (4,549 eighth grade students and 3,574 eleventh grade students) was used in the study. They also identified the racial category of respondents but the results from that category sample were not analyzed. The data collection was in general comprehensive and the methods used were sophisticated, but there are certain limitations of the study. The authors acknowledge some of them, while others are ignored.
