Objectives Evaluate the impact of a place-based, community and problem-oriented policing intervention in Amsterdam, the Netherlands. Methods This evaluation is guided by the EMMIE framework, and its methods and hypotheses are pre-registered on the OpenScienceFramework: <a href="https://osf.io/7kyr3/">https://osf.io/7kyr3/</a>. The effect is estimated using a quasi-experimental, difference-in-difference design, comparing treatment areas (high-treatment, low-treatment, and catchment area) to synthetic control groups constructed from comparable neighbourhoods in Amsterdam. Outcome measures included the number of fast responses (<3 min), crime-, disorder-, traffic- and accident-related calls-for-service counts, and survey-based assessments of perceived neighbourhood disorder, fear of crime, victimisation, and citizen satisfaction with (local and general) police and the municipality. Results In line with pre-registered hypotheses, we observed a substantial decrease in crime- (RIRR = 0.77) and disorder-related calls-for-service (RIRR = 0.84) in the intervention area and a small but significant increase in fear of crime in the low-treatment area (SMD = <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11292_2025_9678_Article_IEq1.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">$-$</EquationSource> <EquationSource Format="MATHML"><math> <mo>-</mo> </math></EquationSource> </InlineEquation>0.11). Contrary to pre-registered hypotheses, we observed a substantial and statistically significant increase in perceived neighbourhood disorder (SMD = <InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11292_2025_9678_Article_IEq1.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">$-$</EquationSource> <EquationSource Format="MATHML"><math> <mo>-</mo> </math></EquationSource> </InlineEquation>0.31) and decrease in citizen satisfaction with police in the neighbourhood (SMD = <InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11292_2025_9678_Article_IEq1.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">$-$</EquationSource> <EquationSource Format="MATHML"><math> <mo>-</mo> </math></EquationSource> </InlineEquation>0.42) and citizen satisfaction with the municipality (SMD = <InlineEquation ID="IEq4"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11292_2025_9678_Article_IEq1.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">$-$</EquationSource> <EquationSource Format="MATHML"><math> <mo>-</mo> </math></EquationSource> </InlineEquation>0.47) in the high-treatment area. Over all treatment areas (referred to a full-treatment: high, low and catchment combined), the intervention only yielded a small but significant increase in fear of crime (SMD = <InlineEquation ID="IEq5"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11292_2025_9678_Article_IEq1.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">$-$</EquationSource> <EquationSource Format="MATHML"><math> <mo>-</mo> </math></EquationSource> </InlineEquation>0.07) compared to synthetic controls. Conclusions This study yielded mixed results. We observed a substantial decrease in crime- and disorder-related calls-for-service in the high-treatment area. However, observed over the full-treatment area, the reduction disappears, indicating crime displacement dynamics. Moreover, we observed worsened public perceptions in the high-treatment, low-treatment, and the catchment area, highlighting a disconnect between improved local conditions and worsened public perceptions in the high-treatment area.

Evaluating place-based policing in Amsterdam

期刊论文
出版模式：

开放获取
发表日期：

2025年6月23日

查看全文

Tim Verlaan,
Stijn Ruiter

摘要

Objectives

Evaluate the impact of a place-based, community and problem-oriented policing intervention in Amsterdam, the Netherlands.

Methods

This evaluation is guided by the EMMIE framework, and its methods and hypotheses are pre-registered on the OpenScienceFramework: https://osf.io/7kyr3/. The effect is estimated using a quasi-experimental, difference-in-difference design, comparing treatment areas (high-treatment, low-treatment, and catchment area) to synthetic control groups constructed from comparable neighbourhoods in Amsterdam. Outcome measures included the number of fast responses (<3 min), crime-, disorder-, traffic- and accident-related calls-for-service counts, and survey-based assessments of perceived neighbourhood disorder, fear of crime, victimisation, and citizen satisfaction with (local and general) police and the municipality.

Results

In line with pre-registered hypotheses, we observed a substantial decrease in crime- (RIRR = 0.77) and disorder-related calls-for-service (RIRR = 0.84) in the intervention area and a small but significant increase in fear of crime in the low-treatment area (SMD = $-$ $-$ 0.11). Contrary to pre-registered hypotheses, we observed a substantial and statistically significant increase in perceived neighbourhood disorder (SMD = $-$ $-$ 0.31) and decrease in citizen satisfaction with police in the neighbourhood (SMD = $-$ $-$ 0.42) and citizen satisfaction with the municipality (SMD = $-$ $-$ 0.47) in the high-treatment area. Over all treatment areas (referred to a full-treatment: high, low and catchment combined), the intervention only yielded a small but significant increase in fear of crime (SMD = $-$ $-$ 0.07) compared to synthetic controls.

Conclusions

This study yielded mixed results. We observed a substantial decrease in crime- and disorder-related calls-for-service in the high-treatment area. However, observed over the full-treatment area, the reduction disappears, indicating crime displacement dynamics. Moreover, we observed worsened public perceptions in the high-treatment, low-treatment, and the catchment area, highlighting a disconnect between improved local conditions and worsened public perceptions in the high-treatment area.