|Date added||June 16, 2014|
|Category||2014 CMRSC XXIV Vancouver|
|Tags||Research and Evaluation, Session 3B|
|Author/Auteur||Ward Vanlaar, Anna McKiernan, Robyn Robertson|
|Stream/Volet||Research and Evaluation|
Background. In jurisdictions around the world there is an increasing demand for the use of
alcohol ignition interlocks to reduce impaired driving. It is crucial that program administrators
and practitioners understand behavioural patterns of offenders on an interlock to inform
decision-making. This insight can guide administrators with regard to program development,
implementation, and the use of program features. Previous research has focused on the
behaviour of offenders on an interlock, specifically with respect to compliance with device
requirements and program rules. In particular, offenders tend to blow fails or violate the
conditions of the interlock program at a relatively high rate at the beginning of their participation
and this behaviour quickly diminishes after offenders have been on the device for some time
(the “learning effect”). Learning theory and general deterrence theory can be used as theoretical
frameworks to test hypotheses regarding this learning effect.
Aims. The purpose of this current study is to further investigate behavioural patterns of
offenders using interlock data organized by jurisdiction and gender for each violation type (e.g.,
restart violation, running retest violation). As such, the current study aims to extend and bolster
previous findings as well as uncover new patterns using data collected from three jurisdictions in
the United States: Florida, Texas and California.
Methods. Using interlock data provided by LifeSafer (an interlock provider), which were drawn
from the period between 1999 and 2012, events such as breath samples when trying to start the
car, breath samples after having started the car (also known as a running retest) and attempts
to skip the running retest were analyzed in order to uncover relevant behavioural patterns.
Results. The results from this current study corroborate the findings from previous research,
i.e., many offenders on an interlock are not compliant at the beginning of their program
participation, but the majority of them soon learn to become more compliant. It was found that
those learning patterns were most pronounced in two states with stronger and more consistent
monitoring practices (Texas and Florida) whereas these patterns were less pronounced in the
state with less consistent monitoring practices (California). In terms of gender, no substantial
differences between males and females were found. With respect to length of participation, it
became clear that participants who are only in the program for a maximum of one year become
compliant much faster than participants who are in the program for at least one year.
Discussion and conclusions. These findings speak to the need for consistent monitoring of
offenders as well as coupling interlock programs with other interventions like treatment for
higher-risk offenders. The findings further suggest that using not only negative reinforcements
for bad behaviour but also using positive reinforcements for good behaviour may be beneficial.
Ward Vanlaar, Anna McKiernan, Robyn Robertson