This study was designed to permit a
"meta-evaluation" of the impact of alternative policing programs on
social disorder. Examples of social disorder include bands of
teenagers deserting school and congregating on street corners,
solicitation by prostitutes and panhandlers, public drinking,
vandalism, verbal harassment of women on the street, street violence,
and open gambling and drug use. The policing methods researched
included community-oriented policing and traditional intensive
The data used in this study were taken from
surveys conducted between 1983 and 1990 in seven cities. For this
collection, a common set of questions was identified and recoded into
a consistent format across studies. The studies were conducted using
similar sampling and interviewing procedures, and in almost every case
used a quasi-experimental research design. For each target area
studied, a different, matched area was designated as a comparison area
where no new policing programs were begun. Surveys of residents were
conducted in the target and comparison areas before the programs began
(Wave I) and again after they had been in operation for a period
ranging from ten months to two-and-a-half years (Wave II).
The original studies used random sampling.
Residents aged 19 years and older in the cities of
Houston, TX, Newark, NJ, Baltimore, MD, Madison, WI, Birmingham, AL,
Oakland, CA, and Denver, CO.
personal interviews and telephone interviews
Survey respondents were asked questions regarding
police visibility and contact, encounters with police, victimization,
fear and worry about crime, household protection and personal
precautions, and neighborhood conditions and problems. Demographic
information was collected including race, marital status, employment
status, education, sex, age, and income. In some cities, not all the
variables were available. Variables in the dataset that begin with a
"Z" represent responses to Wave II questions, while the other
variables are from Wave I.
The percentages of individuals reinterviewed in
each of the seven original studies are as follows. Houston: 57
percent, Newark: 48 percent, Baltimore: 70 percent, Madison: 62
percent, Birmingham: 76 percent, Oakland: 64 percent, and Denver: 80
Responses to the survey questions were primarily
recorded using dichotomous and Likert-type scales.