Juvenile curfew effects on criminal behavior and victimization

Additional Info

  • Authors: David Wilson, Charlotte Gill, Ajima Olaghere, Dave McClure
  • Published date: 2016-03-23
  • Coordinating group(s): Crime and Justice
  • Type of document: Title, Protocol, Review, Plain language summary
  • Volume: 12
  • Issue nr: 3
  • Category Image: Category Image
  • PLS Title: Juvenile curfews are not effective in reducing crime and victimization
  • PLS Description: This Campbell systematic review assesses the effects of juvenile curfews on crime and victimization. The review summarizes findings from 12 studies. The search for this review was updated in March 2014, and the review published in March 2016.
  • Title: Juvenile curfew effects on criminal behavior and victimization
Download files:
Background

A juvenile curfew has a common sense appeal: keep youth at home during the late night and early morning hours and you will prevent them from committing a crime or being a victim of a crime. This common sense appeal has led to the popularity of curfews, at least within the United States and Iceland. However, prior reviews have questioned the effectiveness of curfews.

Objectives

The aim of this review was to synthesize the evidence on the effectiveness of juvenile curfews in reducing criminal behavior and victimization among youth.

Search methods

The systematic search was conducted between January 20, 2014 and March 5, 2014. The search strategy yielded 7,349 titles and abstracts. The initial screening identified 100 of these as potentially relevant and in need of a full text review for study eligibility. Fifteen documents representing 12 unique studies were eligible and coded.

Selection criteria

To be eligible, a study must have tested the effect of an official state or local policy intended to restrict or otherwise penalize a juvenile's presence outside the home during certain times of day. This must have been a general preventive measure directed at all youth within a certain age range and not a sanction imposed on a specific youth. All quantitative research designs were eligible. An eligible study must have assessed the effect of a curfew on either juvenile criminal behavior or juvenile victimization. The manuscript, published or unpublished, must have been written in English and reported on data collected after 1959.

Data collection and analysis

The typical evaluation design of an eligible study was a variant on an interrupted time-series. To accommodate these designs, the effect size used in this synthesis was the percent change in the crime or victimization rate during the period of time with a curfew relative to a baseline period, adjusting for any overall linear time trend. The outcomes of interest included crime and victimization, which were categorized by time of day (curfew hours, non-curfew hours, or all hours) and offender or victim age (juvenile or adult). The effects during non-curfew hours and the effects for adults served as control outcomes; that is, outcomes that should be unaffected by a curfew.

Results

The pattern of evidence suggests that juvenile curfews are ineffective at reducing crime and victimization. The mean effect size for juvenile crime during curfew hours was slightly positive (reflecting a slight increase in crime), whereas it was essentially zero for crime during all hours. Both effects were non-significant. Similarly, juvenile victimization also appeared unaffected by the imposition of a curfew ordinance.

Authors’ conclusions

The evidence suggests that juvenile curfews are ineffective at reducing crime or victimization. This is not, however, a conclusive finding. The observational nature of the research designs creates potential sources of bias, and as such the findings need additional replication. However, many of the most plausible biases should have increased the likelihood of finding an effect. In particular, most of the studies reviewed were conducted during a time period when crime was decreasing throughout the United States. Thus, it appears that juvenile curfews either have no effect on crime and victimization or the effect is too small to be reliability detected with available data.

DOI: 10.4073/csr.2016.3

Contact Us

  • P.O. Box 4404, Nydalen,
    N-0403 Oslo, Norway
  • (+47) 23 25 50 00
  • (+47) 23 25 50 10
  • Web: tkristiansen@campbellcollaboration.org
  •  info@campbellcollaboration.org