estatistical words

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

ECE/CES/4
iii
preface
This manual is the first attempt to develop methodological guidelines at the international level for the
design of victimization surveys.

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

The Manual does not attempt to be prescriptive about which methods to use, but rather illustrates
examples of the types of methodologies available, including the pros and cons of the various methods.

When it is difficult to gather a complete list of elements in the target population or when the
population is geographically dispersed, grouping elements into subpopulations, called clusters, and
sampling within the clusters is an efficient way to draw a sample.

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

Examine user requirements
• Identify internal and external stakeholder and user needs
(what are the policy/research questions that need to be
answered?)

This sort of information can be useful in stratifying the sample (see
section below), identifying sub-groups, data processing, imputation, estimation, record linkage,
quality assessment and analysis.

Therefore one
option may be to outsource some or all of the components of a survey.
95.

Flowchart on questionnaire design 105
Figure 8.

For example, of the 58 victimization surveys inventoried in 2005 by UNODC-UNECE, 11
(19%) utilized simple probability samples, while 42 (72%) used some kind of multistage stratified
sample.
145.

• Prioritise information needs
• Determine broad survey parameters, sample and
technology
• Determine broad output requirements
• Identify data constraints and data quality requirements
• Identify issues and risks
• Obtain first stage approval
5.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

It is usually characterized in terms of error in statistical
estimates and is traditionally decomposed into bias (systematic error) and variance (random error)
components.

The UNODC
and UNECE also appreciate the support of the Statistical Office of the European Union (Eurostat)
which facilitated the editing of the present Manual.

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Factors commonly used in screening for minority respondents, and their scope 44
Table 3.

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

It addresses both
the protection of respondents by means of informed consent and protection of privacy and confidentiality,
as well as data dissemination standards.

Target Population – Frame – Sample – Estimates 164

ix
summary of chapters
Chapter I. Introduction to victimization surveys
Chapter I of this Manual provides an introduction to the manual and its purpose, a brief history of
crime victimization surveys, key messages for policy makers and some discussion on the limitations of
crime victim surveys.

Chapter X. Closing the circle – evaluating completed surveys
Chapter X of this Manual demonstrates the importance of the evaluation process in the life cycle of
crime victimization surveys.

Process used to translate user requirements into survey methodology 13
Figure 2.

Still others fear publicity or news coverage that could result from police
contact.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Chapter III. Methodological issues
Chapter III of this Manual provides an overview of the survey process, the main differences in the type
of victimization survey that are available to the survey manager, the key goals and objectives which
could be identified for any victimization survey along with some of the methodological issues in relation
to achieving such objectives using a social survey.

The Manual does not attempt to be prescriptive about which methods to use, but rather illustrates
examples of the types of methodologies available, including the pros and cons of the various methods.

A sub-population may
be defined by characteristics such as age, gender, citizenship status or ethnicity.

The Manual provides a comprehensive source of information for developing
national victimization surveys and will be particularly useful for those countries who may be
embarking on a survey of this type for the first time.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

These need to be identified
in order to ensure equal probability of selection.

Joint publication can contribute towards providing the public with a
BOX 2: Typical differences between police stati stics and victim surveys
Police data Victim survey
Data source
Administrative police records Personal reports from individual citizens
Data collection method
Comprehensive coverage of police agencies Sample survey
Data recorded in an operational logbook or computer system Data captured by a survey interviewer
National, according to the scope of the police service Natio...

How should interviewers be selected and trained?

The
principal information to be used in determining these objectives will be the goals of the survey, the
information required to meet the data needs of stakeholders, and how this data needs to be utilized.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

The probability distribution that governs the sample selection, along with the
stages and units of sampling, the stratification, and so on, is collectively called the sampling design or
sample design.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Criminal offences that are not as prevalent
in the community will require large enough samples to be representative of the population, however,
data for low prevalence offence types can still be subject to higher error rates than offences that are
more common, and sometimes will not be reliably measured by survey methods.

It is usually characterized in terms of error in statistical
estimates and is traditionally decomposed into bias (systematic error) and variance (random error)
components.

These include
32 victimization surveys (conducted in 23 countries), 24
multipurpose surveys with the inclusion of a module on
victimization (conducted in 16 countries), and two surveys
of other types.

The Manual does not attempt to be prescriptive about which methods to use, but rather illustrates
examples of the types of methodologies available, including the pros and cons of the various methods.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of sur...

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

Quality or ‘fitness for use’ of statistical information may be defined in terms of six constituent
elements or dimensions: relevance, accuracy, timeliness, accessibility, interpretability, and coherence
(Statistics Canada, 2002c).
85.

The Manual provides a comprehensive source of information for developing
national victimization surveys and will be particularly useful for those countries who may be
embarking on a survey of this type for the first time.

These include simple random samples, systematic random sample, cluster samples, stratified
samples and multistage sampling.

For instance, surveys that rely on telephone interviewing can
encounter difficulties since the advent of mobile phones as primary points of contact.

ECE/CES/4
iii
preface
This manual is the first attempt to develop methodological guidelines at the international level for the
design of victimization surveys.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of sur...

The Manual provides a minimum dataset of suggested
key topics for inclusion in national crime victimization surveys that will further facilitate international
comparability of crime survey results.

Perfect comparability between these different statistics is not possible because victim
surveys reflect the experiences of victimization as perceived by the victims, whereas the other sources
are the product of different administrative systems and operational processes which will vary due to
differences in legal codes, operational systems and the differing cultural contexts.
6.

Depending on the budget availability a range of questions can be answered from a survey of this
type such as:
• How much crime is there and what are its characteristics?

If relying on
public contact lists or random-digit-dialling of landlines, many people who have adopted mobile
phones as their primary contact point will be excluded.

The methodologies employed in design,
enumeration, processing and output processes will vary depending upon the survey’s goals and objectives.

Not only do these studies now provide comparative data, but
thanks to the longitudinal aspect they also provide some data on trends in crime for a number of participating
nations.
22.

• These information needs were prioritized, with core requirements stated as: prevalence of exposure, prevalence of response,
prevalence of financial loss, amount of financial loss.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

The ultimate goal of the Manual is to improve
the comparability of victimization survey results.

It considers the concepts of prevalence and incidence and discusses the issues of multiple and
repeat victimizations.

These surveys can also measure more than criminal incidents and
often include questions about fear and attitudes toward crime and the justice system, as well as various
socio-demographic characteristics that help to assess at risk populations.
9
Chapter II. PLANNING A CRIME VICTIMIZAT ION SURVE Y
Given the sensitive 48. nature of criminal victimization, there may be some reluctance on the part of
victims to report victimizations to a general population survey.

In addition, victims in some countries may not want to get involved with the police, either because
they feel the police could do nothing to help, or because they fear the police would be biased or would
not be willing to intervene.

Police-reported statistics normally collate information on all incidents reported to a variety of
police agencies.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

It is at this step that the level of precision of the data needs to be determined.

It also illustrates the experiences encountered of countries with extensive experience in designing
national crime victimization surveys.

The main assumption in cluster sampling is that there should be similarity across clusters
and heterogeneity within each cluster.

For international comparability, it is important that the topic for comparison is being defined in
the same way (so a country measuring only burglaries without attempts would not be able to compare
its burglary rate including them).

Process used to translate user requirements into survey methodology 13
Figure 2.

The Manual does not attempt to be prescriptive about which methods to use, but rather illustrates
examples of the types of methodologies available, including the pros and cons of the various methods.

The Manual on crime victimization surveys has been developed through a joint effort of the UNODCUNECE
Task Force on crime victim surveys which consisted of the following members:
Soula MacFarlane (Australian Bureau of Statistics, Chair of the Task Force)
Jodi-Anne Brzozowski (Statistics Canada)
Karen Mihorean (Statistics Canada)
Maria Giuseppina Muratore (Istat, Italy)
Frits Huls (Statistics Netherlands)
Beata Gruszczynska (Institute of Justice of Poland)
Alison Walker (Home Office, ...

It considers the concepts of prevalence and incidence and discusses the issues of multiple and
repeat victimizations.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

These questions are particularly relevant for countries that are in the process of developing victim
survey programs for the first time and have limited experience in this field.

Some point out that national averages lack some of the impact that a focus on specific
local areas or problems can bring, or the weaknesses in a sample survey approach in addressing the rarer
but more serious crimes.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of sur...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of survey-...

Victims of robbery: type of weapon used in most recent incident 188
LIST of FIGURES
Figure 1.

Crimes and
victimization Property crime Contact crime / violent
crime
No crime
specified
Household
burglary
Theft of
vehicles Other theft Robbery Physical
Assault
Sexual
offences
Measure of victimization
in the past 12
months
x x x x x x
Measure of repeat
victimization in the
past 12 months
x x x x x x
Reporting to the
police x x x x x x
Crimes involving
weapons x x x
Victims who
suffered physical
injury
x x x
Victim-offender
relationship x x x
Public con...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

At each step, there are decisions that need to be taken, balancing
requirements with resources.
64.

In some cases, it may be
possible to select the respondents based on just one question (e.g. persons who have immigrated to the
country in a given period), but often more questions may be needed.

Comparability challenges of administrative police statistics and crime victim surveys 82
Table 10.

It is important to have in mind the kinds of uses
to which the output will be put when designing the survey vehicle, in order to ensure that outputs will
be fit for the specified purpose when delivered.

The United Nations Office on Drugs and Crime (UNODC) - United Nations Economic Commission
for Europe (UNECE) Manual on Victimization Surveys provides a comprehensive source of
information for developing a national victimization survey and the key issues which need to be considered,
including the approach to take, available methods, some key analytical issues and the presentation
of results.

The proportion of the population experiencing one or more criminal 168
victimizations in the past 12 months
Table 15.

The Manual does not attempt to be prescriptive about which methods to use, but rather illustrates
examples of the types of methodologies available, including the pros and cons of the various methods.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of sur...

Although police statistics and
victim surveys normally cover comparable geographic areas, if appropriately nationally representative,
victim surveys may exclude some categories of victims, such as very young children or persons residing
in institutions such as a prison, hospital, care centre or military barracks.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Crimes and
victimization Property crime Contact crime / violent
crime
No crime
specified
Household
burglary
Theft of
vehicles Other theft Robbery Physical
Assault
Sexual
offences
Measure of victimization
in the past 12
months
x x x x x x
Measure of repeat
victimization in the
past 12 months
x x x x x x
Reporting to the
police x x x x x x
Crimes involving
weapons x x x
Victims who
suffered physical
injury
x x x
Victim-offender
relationship x x x
Public confidence...

The Manual provides a minimum dataset of suggested
key topics for inclusion in national crime victimization surveys that will further facilitate international
comparability of crime survey results.

The ICVS plays an important role in
providing more harmonized victimization data allowing comparisons between countries while national
and some sub-national victim surveys will often have a larger scope and sample size to allow a closer
focus on topics of particular relevance to a specific country.
7.

Increases in evidence-based policy-making and holistic approaches to social problems have driven
demand for social indicators that support the measurement of progress, well being and social inclusion.

A stratified random sample derives
from a sampling plan in which the population is first divided into mutually exclusive and exhaustive
groups, called strata, and then a simple random sample is taken from each stratum.
165.

The Chapter continues with an
examination of survey modes, covering face-to-face interviews, telephone interviews, self-administered
interviews, internet based questionnaires and mixed-mode interviewing.

For example, 70 in Italy);
• Other groups such as ethnic minorities, immigrants, the disabled, the lesbian and gay community
or people living in special circumstances.
123.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of survey-rel...

Some countries have population registers which provide useful sampling frames for different
purposes as they tend to have good coverage of the population, are regularly updated, and contain
useful additional information or can be linked to other databases to acquire additional data.

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

Chapter II. Planning a crime victimization survey
Chapter II of this Manual describes the relationship between information obtained from conducting
crime victimization surveys and that obtained from police-recorded crime statistics.

Transform inputs into statistics
• Produce aggregate estimates
• Impute data
• Identify and correct deviation anomalies
• Identify and treat significant aggregate level anomalies
• Identify and resolve outliers
• Weight data
• Validate weighted data
• Confidentialise data
2.

Multiple or repeat victimizations can be captured, although
there will normally be a limit to the number of such incidents
recorded in order to prevent outliers overly affecting the
estimates
Often includes attempts within the main category Includes attempts, but usually able to separately identify these
Sources of error:
Reporting by the public Sampling error
Processing error, edit failure, non-responding police
department
Non-sampling error related to: coverage, respondent ...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim survey...

It describes the process of organizing a survey
and discusses target populations, sampling options and frequency of enumeration.

ECE/CES/4
iii
preface
This manual is the first attempt to develop methodological guidelines at the international level for the
design of victimization surveys.

The Manual will cover the end-to-end business process for developing and conducting crime victimization
surveys as well as disseminating information from these surveys.

Chapter III. Methodological issues
Chapter III of this Manual provides an overview of the survey process, the main differences in the type
of victimization survey that are available to the survey manager, the key goals and objectives which
could be identified for any victimization survey along with some of the methodological issues in relation
to achieving such objectives using a social survey.

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

Chapter VI. Interviewing (if not a self-response survey)
Chapter VI of this Manual addresses issues associated with the conduct of interviews, including nonresponse,
the use of proxy respondents, training of interviewers and quality control throughout the
interviewing process.

Depending on the budget availability a range of questions can be answered from a survey of this
type such as:
• How much crime is there and what are its characteristics?

The Manual provides a comprehensive source of information for developing
national victimization surveys and will be particularly useful for those countries who may be
embarking on a survey of this type for the first time.

Finally, it proposes a draft list of key topics which could be
considered as elements for international comparability of victim surveys.

Crimes and
victimization Property crime Contact crime / violent
crime
No crime
specified
Household
burglary
Theft of
vehicles Other theft Robbery Physical
Assault
Sexual
offences
Measure of victimization
in the past 12
months
x x x x x x
Measure of repeat
victimization in the
past 12 months
x x x x x x
Reporting to the
police x x x x x x
Crimes involving
weapons x x x
Victims who
suffered physical
injury
x x x
Victim-offender
relationship x x x
Public con...

Chapter III. Methodological issues
Chapter III of this Manual provides an overview of the survey process, the main differences in the type
of victimization survey that are available to the survey manager, the key goals and objectives which
could be identified for any victimization survey along with some of the methodological issues in relation
to achieving such objectives using a social survey.

While the size of the sampling error can be estimated using standard statistical methods, non-sampling
errors are more difficult to quantify.

For example, 70 in Italy);
• Other groups such as ethnic minorities, immigrants, the disabled, the lesbian and gay community
or people living in special circumstances.
123.

Henry Mayhew’s ‘London Labour and the London Poor’ (1851) and
Charles Booth’s ‘Labour and Life of the People’ (1889-1903) involved interviewing their subjects about
their social conditions and attitudes.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

No one source will provide a definitive measure of victimization in society.

The Manual focuses on national household victimization surveys but references are included where
relevant to international surveys such as the International Crime Victim Survey (ICVS — see paragraphs
20-21) and other surveys of sub-national populations.

Factors commonly used in screening for minority respondents, and their scope 44
Table 3.

For example, personal fraud is aimed at deceiving people and a victim may never discover a
fraud that was perpetrated against them, or they may discover the fraud long after it took place.

However, it is not normally a suitable method for creating a representative
sample for the final survey which requires a probability-based sample design.
7 The Kish grid (random digits table) consists of combination of digits, which is unique or almost unique for every household.

When a criminal victimization
occurs, there are a number of ways in which this can be measured and a number of stages
at which a measurement can be taken.

It considers the concepts of prevalence and incidence and discusses the issues of multiple and
repeat victimizations.

In order to avoid non-relevant units from being included in the sample, the first part of an interview
normally includes some questions used to check the eligibility of the person or unit being contacted.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Household surveys that included victimization modules in the 79
UNODC-UNECE inventory
Table 9.

Chapter VI. Interviewing (if not a self-response survey)
Chapter VI of this Manual addresses issues associated with the conduct of interviews, including nonresponse,
the use of proxy respondents, training of interviewers and quality control throughout the
interviewing process.

As the surveys matured and started to produce long-run time series, they became
increasingly useful in the measurement of trends and influential both inside and outside government
as a measure of success or failure in reducing crime.
1 United Nations Crime and Justice Statistics http://www.uncjin.org/Special/history.html
2 UNODC website: http://www.unodc.org/unodc/en/data-and-analysis/United-Nations-Surveys-on-Crime-Trends-and-the-Operations-of-
Criminal-Justice-Systems.html
3 K...

Some countries publish police statistics and victim survey estimates in separate reports, while other
countries have attempted to incorporate both sets of statistics in a more unified presentation of the
pattern of crime in a country.

Other key factors to consider when embarking on a crime victimization survey include whether
an organization has the infrastructure available and capability to undertake a crime victimization
survey.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

The mode of interviewing may affect the overall
budget, but the proportions of the various components within the budget will remain more or less the
same.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

It
was originally part of the UK Ministry of Information’s Home Intelligence Division and consisted of
a handful of researchers, mostly recruited from the new field of market research.4

ECE/CES/4
iii
preface
This manual is the first attempt to develop methodological guidelines at the international level for the
design of victimization surveys.

Examples of contact and non-contact types of crime that may be included 60
in general household surveys
Table 7.

Changes in the legal system and societal attitudes to certain acts or changes in culture, for example the
reduction of sexual taboos, can have an important impact on the number of police-reported criminal
incidents.

Victim surveys and police
statistics are therefore complementary sources, each of which has specific advantages and weaknesses.
37.

The Manual provides a comprehensive source of information for developing
national victimization surveys and will be particularly useful for those countries who may be
embarking on a survey of this type for the first time.

Countries where census data is the main source of population
data available, as opposed to regularly updated population registers, face particular challenges
when trying to construct an up-to-date sample frame.

In addition, victims in some countries may not want to get involved with the police, either because
they feel the police could do nothing to help, or because they fear the police would be biased or would
not be willing to intervene.

These are some of the methodological issues that clients need to be aware of and which are
described in more detail throughout this Manual.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

As mass marketed frauds and scams generally require the victim to respond or interact to an invitation of some
sort, it was also important to gather a measure of exposure (where people received an invitation that they could respond
to), as well as when they actually did respond, and where they subsequently lost money.

When victimization surveys are conducted in new or emerging areas of crime where data has not
previously been gathered, or the offences are not thought to be well represented in recorded crime
statistics, the objective of the survey may be to establish a baseline crime victimization rate.

It also illustrates the experiences encountered of countries with extensive experience in designing
national crime victimization surveys.

• The clients were presented with a survey proposal, and memorandums of understanding were signed to secure funding for
the survey and outline the required outputs and delivery dates.

These should form part of the goal setting process determined at the
beginning of the survey wherever possible, enabling subsequent methodologies employed to anticipate
constraints and maximise the data that can be generated within those constraints.

Quality or ‘fitness for use’ of statistical information may be defined in terms of six constituent
elements or dimensions: relevance, accuracy, timeliness, accessibility, interpretability, and coherence
(Statistics Canada, 2002c).
85.

For example, topics could be ‘measure of household burglary
victimization in the past twelve months’ or ‘robbery crimes involving weapons’.

In turn, a broad recognition of such needs should be used to develop a detailed specification
of user needs.

At these meetings a factor emerged which has continued
to remain in the foreground through all subsequent efforts, namely, the problem of comparability of
definitions.1
10.

We use the term crime to encapsulate the focus of a survey of crime victims but in some cases
the term can be understood to cover a wider range of events.

The Chapter continues with an
examination of survey modes, covering face-to-face interviews, telephone interviews, self-administered
interviews, internet based questionnaires and mixed-mode interviewing.

A key recommendation
of the Commission was the implementation of a new source of crime statistics – a victimization survey.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim survey...

It is important to be aware of the fundamental differences between
these statistics to understand why these two measures can and do sometimes diverge, and why they
should be seen as complementary rather than mutually exclusive.

This evidence base can also be used to assess the effectiveness of crime prevention programmes
by government.

Simple random samples are generally created without replacement, meaning that no person or
sample unit is selected more than one time for inclusion in the sample.

Types of crime that may be included in general household surveys, 60
by category of victim
Table 6.

These examples will assist people to make informed choices to
suit their particular circumstances.

The random selection is based on a selection matrix of l random numbers defined as follows: l is
the maximum households dimension, each record of the selection matrix represents a household, and
each selection matrix element has l digits.

Ethical considerations
Chapter VIII of this Manual deals with the important area of ethical considerations.

Moreover,
information on consensual or ‘victimless’ crime (i.e., drug use, prostitution, gambling), and corporate
or white-collar crime, is typically not collected through victimization surveys, although it is feasible to
collect some indicators of such activity in this way.
52.

There is no general model that
brings them together to optimize or to prescribe a level of quality.

This additional information was thought essential
to support the development of crime prevention strategies.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of sur...

Once the expected costs of various components of
survey operation have been calculated, the survey manager is in a good position to determine whether
the goals and objectives of the survey can be met with the likely available resources, and consider
negotiating for or seeking additional funding.

For example, 70 in Italy);
• Other groups such as ethnic minorities, immigrants, the disabled, the lesbian and gay community
or people living in special circumstances.
123.

Illustrative proportion of budget for each survey stage 23
Figure 4.

The ultimate goal of the Manual is to improve
the comparability of victimization survey results.

The sample selection method will depend on the existence or absence of population
lists or registers, such as telephone or address listings.

In either case, it is essential to clearly describe the data, how it has been derived
and its appropriate uses.

ECE/CES/4
iii
preface
This manual is the first attempt to develop methodological guidelines at the international level for the
design of victimization surveys.

There were initial fears
that government asking the public questions about their life and attitudes would be greeted with some
suspicion, but these fears proved to be unfounded.
13.

If the selected person was unavailable or refused the interview, the interviewer
was not allowed to interview any other household member.8

The Manual on crime victimization surveys has been developed through a joint effort of the UNODCUNECE
Task Force on crime victim surveys which consisted of the following members:
Soula MacFarlane (Australian Bureau of Statistics, Chair of the Task Force)
Jodi-Anne Brzozowski (Statistics Canada)
Karen Mihorean (Statistics Canada)
Maria Giuseppina Muratore (Istat, Italy)
Frits Huls (Statistics Netherlands)
Beata Gruszczynska (Institute of Justice of Poland)
Alison Walker (Home Office, ...

Access to confidential unit record data from the survey was also
desirable.

In the international arena, the United Nations Survey of Crime Trends and Operations
of Criminal Justice Systems (also known as the ‘Crime Trends Survey’ – CTS) was established in 1978
and aimed to conduct a more focused inquiry into the incidence of crime worldwide.

Population parameters include values such as the mean,
total, proportions, ratios and percentiles.
187.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of survey-rel...

How should the
quality and limitations of a survey be monitored and reported?
4.

Simple random samples represent the simplest sample design and they provide unbiased estimators
of population statistics.

This far-reaching
worldwide study draws on administrative sources and regularly collects statistics on police, prosecution,
courts and prisons.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

ECE/CES/4
iii
preface
This manual is the first attempt to develop methodological guidelines at the international level for the
design of victimization surveys.

Criminal offences that are not as prevalent
in the community will require large enough samples to be representative of the population, however,
data for low prevalence offence types can still be subject to higher error rates than offences that are
more common, and sometimes will not be reliably measured by survey methods.

It defines the technique and methods of sampling units, the design of
which also affects the estimates and their statistical significance.
130.

It is usually characterized in terms of error in statistical
estimates and is traditionally decomposed into bias (systematic error) and variance (random error)
components.

Chapter IX. Publication and dissemination
Chapter IX of this Manual addresses principles and guidelines for data dissemination and documentation,
including dissemination formats, and standards for metadata and references.

As the surveys matured and started to produce long-run time series, they became
increasingly useful in the measurement of trends and influential both inside and outside government
as a measure of success or failure in reducing crime.
1 United Nations Crime and Justice Statistics http://www.uncjin.org/Special/history.html
2 UNODC website: http://www.unodc.org/unodc/en/data-and-analysis/United-Nations-Surveys-on-Crime-Trends-and-the-Operations-of-
Criminal-Justice-Systems.html
3 K...

Chapter II. Planning a crime victimization survey
Chapter II of this Manual describes the relationship between information obtained from conducting
crime victimization surveys and that obtained from police-recorded crime statistics.

Acquire data
• Prepare survey frame and sample
• Select sample
• Allocate interviewer workloads
• Conduct interviews and field editing
• Manage field operations
• Resolve field queries
• Capture responses and repair data
• Extract management information about the collection
process
8.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

A sub-population may
be defined by characteristics such as age, gender, citizenship status or ethnicity.

The Manual on crime victimization surveys has been developed through a joint effort of the UNODCUNECE
Task Force on crime victim surveys which consisted of the following members:
Soula MacFarlane (Australian Bureau of Statistics, Chair of the Task Force)
Jodi-Anne Brzozowski (Statistics Canada)
Karen Mihorean (Statistics Canada)
Maria Giuseppina Muratore (Istat, Italy)
Frits Huls (Statistics Netherlands)
Beata Gruszczynska (Institute of Justice of Poland)
Alison Walker (Home Office, ...

With the panel design, sample attrition is a concern.
184.

The information was used
to create an inventory of victim surveys in the region, to be
used as a tool to analyze differences and communalities in
the methods and definitions adopted.

Additionally, objectives relating to these detailed data should be set in a way that is realistically
achievable.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Depending on the budget availability a range of questions can be answered from a survey of this
type such as:
• How much crime is there and what are its characteristics?

Data collected cover the period 1970-to date.2
11.

It addresses both
the protection of respondents by means of informed consent and protection of privacy and confidentiality,
as well as data dissemination standards.

Since April 2001,
the BCS has become an annual fixture in England and Wales providing information on trends in victimization
as well as detailed information on the nature of those crimes.
19.

In the example
above, it is possible that some of the missing sub-groups (e.g., those in prison and the homeless) will
have higher levels of victimization so their exclusion may affect the overall estimates.

Sampling errors
10
Manual on victimiZation surveys
represent the difference between an estimate derived from a sample of a population and the result that
would have been obtained if the entire population had been asked the same question.

Multiple or repeat victimizations can be captured, although
there will normally be a limit to the number of such incidents
recorded in order to prevent outliers overly affecting the
estimates
Often includes attempts within the main category Includes attempts, but usually able to separately identify these
Sources of error:
Reporting by the public Sampling error
Processing error, edit failure, non-responding police
department
Non-sampling error related to: coverage, respondent or reca...

The accuracy of statistics is influenced by the ability of people to recall past victimizations.

Analyze and explain
• Undertake special analysis
• Compile clearance documentation
• Analyze and write up key findings
• Measure and explain comparability with other results
• Produce relative standard errors and other reliability
measures
7.

Coverage defined by age incurs additional costs because not all persons approached are eligible,
while coverage defined by other attributes such as race incurs extra costs and is less likely to result in a
fully representative sample.

There are several ways to select statistically viable
samples.

As the surveys matured and started to produce long-run time series, they became
increasingly useful in the measurement of trends and influential both inside and outside government
as a measure of success or failure in reducing crime.
1 United Nations Crime and Justice Statistics http://www.uncjin.org/Special/history.html
2 UNODC website: http://www.unodc.org/unodc/en/data-and-analysis/United-Nations-Surveys-on-Crime-Trends-and-the-Operations-of-
Criminal-Justice-Systems.html
3 Kauko Ar...

Finally, it proposes a draft list of key topics which could be
considered as elements for international comparability of victim surveys.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Challenges and possible approaches 132
Table 14.

The Manual provides a comprehensive source of information for developing
national victimization surveys and will be particularly useful for those countries who may be
embarking on a survey of this type for the first time.

Interviewers should reassure respondents of the confidentiality
of the responses, and remain empathetic and non-judgemental.

And unless those who choose to participate differ in no important
ways with respect to the survey’s aims from those who choose not to participate, the information gathered
cannot be generalized to the general population.

Counting rules vary from jurisdiction to jurisdiction.

Australia conducted its first victimization survey in 1975 and further surveys were carried out in
1983, 1993, 1998, 2002 and 2005 with the intention now being to run a survey each year to provide
‘headline measures’ about a select range of personal and household crimes.

Increases in evidence-based policy-making and holistic approaches to social problems have driven
demand for social indicators that support the measurement of progress, well being and social inclusion.

Not only do these studies now provide comparative data, but
thanks to the longitudinal aspect they also provide some data on trends in crime for a number of participating
nations.
22.

Assemble and disseminate
• Assemble statistical products (reports, web content,
supplementary tables, etc.)

• Any victimization estimate has to be evaluated together with the corresponding sampling error or
confidence interval level in order to really appreciate differences and similarities with the corresponding
statistics from police data.

They were also interested in elements of electronic security employed, behaviour change over time, detailed
geographic locations, etc.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Although police statistics and
victim surveys normally cover comparable geographic areas, if appropriately nationally representative,
victim surveys may exclude some categories of victims, such as very young children or persons residing
in institutions such as a prison, hospital, care centre or military barracks.

It looks at the use of screener questions, incident questions, answer modalities and question
response patterns.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

This sort of information can be useful in stratifying the sample (see
section below), identifying sub-groups, data processing, imputation, estimation, record linkage,
quality assessment and analysis.

As the surveys matured and started to produce long-run time series, they became
increasingly useful in the measurement of trends and influential both inside and outside government
as a measure of success or failure in reducing crime.
1 United Nations Crime and Justice Statistics http://www.uncjin.org/Special/history.html
2 UNODC website: http://www.unodc.org/unodc/en/data-and-analysis/United-Nations-Surveys-on-Crime-Trends-and-the-Operations-of-
Criminal-Justice-Systems.html
3 Kauko Ar...

These should form part of the goal setting process determined at the
beginning of the survey wherever possible, enabling subsequent methodologies employed to anticipate
constraints and maximise the data that can be generated within those constraints.

Administrative sources (such as police or judicial statistics) cannot provide a sufficiently reliable
and comprehensive analysis of crime on their own.

There is a strong relationship between frequency, timelines, survey goals and available resources
when conducting a victim survey.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Surveys of minorities may require the translation of the fieldwork documents or recruitment
and training of multilingual fieldwork staff.
108.

To become ‘known to the police’ a crime must survive a succession of victim and police
decisions, including recognition by the victim that a crime has occurred, police notification, and,
entering the occurrence into official police records.

For example, when comparing the two national crime sources in the United States,
homicide would be removed from Uniform Crime Reporting Program (police data), and simple
assault would be removed from National Crime Victimization Survey data.

Government measurement of crime initially focused on administrative sources, such as police and
court statistics.

Snowball sampling
begins by identifying a person who meets the criteria for inclusion in the study.

Chapter II. Planning a crime victimization survey
Chapter II of this Manual describes the relationship between information obtained from conducting
crime victimization surveys and that obtained from police-recorded crime statistics.

A sub-population may
be defined by characteristics such as age, gender, citizenship status or ethnicity.

Substitution in cases of replication can
be considered.

Perfect comparability between these different statistics is not possible because victim
surveys reflect the experiences of victimization as perceived by the victims, whereas the other sources
are the product of different administrative systems and operational processes which will vary due to
differences in legal codes, operational systems and the differing cultural contexts.
6.

The ICVS plays an important role in
providing more harmonized victimization data allowing comparisons between countries while national
and some sub-national victim surveys will often have a larger scope and sample size to allow a closer
focus on topics of particular relevance to a specific country.
7.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Police-reported statistics obtain data from police administrative records either at the aggregate
or micro-data level.

In the first digit a random number between 1 and 2 is inserted,
in the second a random number among 1,2 and 3 is inserted, and so on, in the lth digit a random
number among the l+1 integers 1,2,…, l+1, is inserted.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim survey...

At these meetings a factor emerged which has continued
to remain in the foreground through all subsequent efforts, namely, the problem of comparability of
definitions.1
10.

Criminal offences that are not as prevalent
in the community will require large enough samples to be representative of the population, however,
data for low prevalence offence types can still be subject to higher error rates than offences that are
more common, and sometimes will not be reliably measured by survey methods.

It considers the concepts of prevalence and incidence and discusses the issues of multiple and
repeat victimizations.

This additional information was thought essential
to support the development of crime prevention strategies.

Many other countries have explored the feasibility of victimization surveys
through pilot studies or the inclusion of questions on crime in the questionnaires used for already
existing household surveys on other social issues.
20.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Sixth, identify duplication
if multiple frames are used.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Historical development
The oldest survey included in the inventory is the US
National Crime Victimization Survey, started on 1972 and
still ongoing.

These include maintaining tight control
over the survey, building expertise and staffing capability over time, which can minimize costs.
98.

Alternatively,
if a statistical agency is commissioned to run the survey but does not have access to persons
with specialist substantive experience in criminology or justice-related backgrounds, then this expertise
may also have to be brought in from outside the organization.

This can have major impacts on the size
of the sample and the resources needed to produce the survey outputs required.
18
Manual on victimiZation surveys
BOX 4: Survey objective examples
• Prevalence of a particular phenomenon (i.e., persons, households, or businesses experiencing victimization)
• Estimate the number of crimes occurring during a specified period
• Establishing baseline crime victimization rates
• Exploring movements in crime victimization rates
• Replicate past surv...

Analyze and explain
• Undertake special analysis
• Compile clearance documentation
• Analyze and write up key findings
• Measure and explain comparability with other results
• Produce relative standard errors and other reliability
measures
7.

Crime also
can result in significant economic costs such as the provision of law enforcement services and corrective
services as well as costs to businesses and households either as a consequence of crime or implementation
of preventative measures.
24.

On the one hand, this may refer to the financial budget associated with the
enterprise, but can also include the amount of personnel resources that can be dedicated to a project
through to the technological resources available through the data design, collection, processing and
dissemination phases.

The value of stolen items will also affect the willingness of a victim to report
an incident.
43.

One of the benefits of understanding the information needs of key stakeholders 73. is that it enables
an appreciation of the analytic requirements of users.

B. Goals/objectives/purposes of survey/collection history
Stakeholder and user needs and priorities
71.

In the surveys in the 2005 UNODC-UNECE inventory that utilized stratified samples, the
most common stratifiers used are geographical areas, degrees of urbanization, and respondent age and
gender.
168.

Victimization surveys have found that one of the primary reasons that an incident does not come
to the attention of the police relates to the fact that the victim felt that the incident was not important
enough.

Perfect comparability between these different statistics is not possible because victim
surveys reflect the experiences of victimization as perceived by the victims, whereas the other sources
are the product of different administrative systems and operational processes which will vary due to
differences in legal codes, operational systems and the differing cultural contexts.
6.

Decision support
• Manage stakeholder and user requests for information
from the survey
• Undertake customised consultancies to meet specific user
needs
• Produce special articles
• Maintain links with key stakeholders and users
• Provide insight into the meaning of the numbers for key
stakeholders and users
4.

On the other hand, outsourcing may offer a faster and less expensive alternative.

For example:
• Scoping studies, including desktop research, may extend to focus groups or roundtable
exercises with panels of experts to determine key items and definitions
• Design of questionnaires and questions
• Cognitive testing, or other forms of questionnaire testing, which can be very beneficial but labour
intensive
• Preparation of materials required to go into the field (coding for use with technology; preparation
of printed materials)
• Creation of training required...

• Any victimization estimate has to be evaluated together with the corresponding sampling error or
confidence interval level in order to really appreciate differences and similarities with the corresponding
statistics from police data.

For
agencies which do not maintain panels of interviewers or other forms of infrastructure necessary to
conduct population-based surveys, outsourcing can be an important option to cost and consider.
96.

These concepts are introduced
later in this introductory section.

Non-probability sampling methods can be divided into two broad categories: purposive
and accidental/haphazard/convenience.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim survey...

In addition, a ‘crime’
may disappear or be redefined at any point in the process, for example, as a result of operational rules,
classification decisions, recordkeeping failures or a decision by the victim or the police to drop charges.

Transform inputs into statistics
• Produce aggregate estimates
• Impute data
• Identify and correct deviation anomalies
• Identify and treat significant aggregate level anomalies
• Identify and resolve outliers
• Weight data
• Validate weighted data
• Confidentialise data
2.

• Develop collection instrument
• Determine testing and QA strategies
• Test and evaluate concepts, definitions, questions, procedures,
training, documentation, instrument and
methodologies
• Finalise data items, questions, collection instrument and
collection procedures; specify derivations
• Develop and test input and output systems/other systems
or databases
6.

Then there will be subsequent requirements for large enough samples of relevant
groups to be identified in analysis and possibly special arrangements for interviews; for example, in the case of surveys on
minority groups there may be the need to translate the original questionnaire into different languages and to provide bi-lingual
interviewers, including the provision of minority interviewers from the same background as the groups being surveyed.
41
Chapter III. METHODOLOGICAL I...

Depending upon the specific expertise of the survey managers, it may be necessary to also pay for
external consultants to work on various elements of the project.

It discusses the
differences between victim surveys and police-reported data and suggests good practices in attempting
to compare data from the two sources.

It is important for survey questions to
be carefully worded and tested, and for survey interviewers to receive specialized training on how to
deal with this sensitive subject matter.

In some instances, crimes against businesses are covered in specialized surveys of businesses.

These dimensions of quality are overlapping and interrelated.

In this Manual, countries
with extensive experience in designing national victim surveys share their knowledge and experience
with the international community.
5.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of survey-rel...

In addition to seeking to understand crimes not well identified by police statistics, these surveys
were an important research tool to assist in identifying aspects of crime and victimization that could
not be easily captured by the administrative sources.

Challenges and possible approaches 132
Table 14.

Moreover,
the person selected for the sample may not be knowledgeable of such offences, whereas another household
member might be a better respondent for events that impact the entire household.
219.

• Prioritise information needs
• Determine broad survey parameters, sample and
technology
• Determine broad output requirements
• Identify data constraints and data quality requirements
• Identify issues and risks
• Obtain first stage approval
5.

The ease of public reporting
can consequently impact whether a relatively minor criminal incident is recorded by the police.

The Manual provides a comprehensive source of information for developing
national victimization surveys and will be particularly useful for those countries who may be
embarking on a survey of this type for the first time.

Crimes and
victimization Property crime Contact crime / violent
crime
No crime
specified
Household
burglary
Theft of
vehicles Other theft Robbery Physical
Assault
Sexual
offences
Measure of victimization
in the past 12
months
x x x x x x
Measure of repeat
victimization in the
past 12 months
x x x x x x
Reporting to the
police x x x x x x
Crimes involving
weapons x x x
Victims who
suffered physical
injury
x x x
Victim-offender
relationship x x x
Public con...

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Joint publication can contribute towards providing the public with a
BOX 2: Typical differences between police stati stics and victim surveys
Police data Victim survey
Data source
Administrative police records Personal reports from individual citizens
Data collection method
Comprehensive coverage of police agencies Sample survey
Data recorded in an operational logbook or computer system Data captured by a survey interviewer
National, according to the scope of the police service Natio...

The information was used
to create an inventory of victim surveys in the region, to be
used as a tool to analyze differences and communalities in
the methods and definitions adopted.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim survey...

It discusses the
differences between victim surveys and police-reported data and suggests good practices in attempting
to compare data from the two sources.

Quota 178. sampling is yet another approach.

London: Wiley.
7
Chapter II. PLANNING A CRIME VICTIMIZAT ION SURVE Y
chapter II. PLANNING A CRIME VICTIMIZATION SURV EY
35.

Government interest in looking beyond its own administrative
sources to surveys of the population increased during the Second World War. In Finland, a question
on property stolen was introduced into Gallup opinion surveys in the 1940s.3

Distribution of reference periods in one financial year of British Crime Survey 103
interviews (April-March)
Figure 7.

In these instances, it may be necessary to look for alignment with existing
collections or definitions, and keep these data sources in mind when considering survey goals and
subsequent methodologies.

These questions are particularly relevant for countries that are in the process of developing victim
survey programs for the first time and have limited experience in this field.

This is the first step toward the broader long-term objectives of improving the international comparability
of crime statistics in general and developing an integrated system of victim data collections
alongside police, prosecution, court and corrective services statistics.*

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

Chapter X. Closing the circle – evaluating completed surveys
Chapter X of this Manual demonstrates the importance of the evaluation process in the life cycle of
crime victimization surveys.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Survey mode for victimization surveys in the UNODC-UNECE inventory, 2005 91
Table 13.

Initially focusing on developed nations, with the involvement of UNICRI and UNODC
the surveys expanded to developing countries and Eastern Europe.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of sur...

The Chapter continues with an
examination of survey modes, covering face-to-face interviews, telephone interviews, self-administered
interviews, internet based questionnaires and mixed-mode interviewing.

Chapter III. Methodological issues
Chapter III of this Manual provides an overview of the survey process, the main differences in the type
of victimization survey that are available to the survey manager, the key goals and objectives which
could be identified for any victimization survey along with some of the methodological issues in relation
to achieving such objectives using a social survey.

Frequently headlines appear in the news media that report on differing trends in crime depending
on the source of data from which the reports are derived – police-reported statistics, often referred to
as official crime statistics, or victim survey data.

Victimization surveys (or victim surveys, crime
victim surveys as they will be interchangeably referred to throughout the Manual) are now a recognized
tool that help governments and their public understand their crime problems and how better to address
them.

Should the survey be
run under any other circumstances than ‘in-house’, such
as via contract or external agency agreement, then the
funding required for different stages will depend upon the
experience of the agency with running surveys, how well
established the organization’s survey infrastructure and
interviewer workforce may be, as well as other commercial
issues such as competition, supply and demand.
109.

Victimization surveys (or victim surveys, crime
victim surveys as they will be interchangeably referred to throughout the Manual) are now a recognized
tool that help governments and their public understand their crime problems and how better to address
them.

With survey data collection, processing and output increasingly utilizing 102. new technologies, a
significant proportion of the budget may be required to purchase or maintain new computer systems,
networks or other tools if undertaken in-house and no other survey vehicle is being utilized.

Process used to translate user requirements into survey methodology 13
Figure 2.

The proportion of the population experiencing one or more criminal 168
victimizations in the past 12 months
Table 15.

Other issues identified
related to problems in defining some of the frauds, and overcoming confusion amongst respondents and their inability to
recognise some of these frauds.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

This can have major impacts on the size
of the sample and the resources needed to produce the survey outputs required.
18
Manual on victimiZation surveys
BOX 4: Survey objective examples
• Prevalence of a particular phenomenon (i.e., persons, households, or businesses experiencing victimization)
• Estimate the number of crimes occurring during a specified period
• Establishing baseline crime victimization rates
• Exploring movements in crime victimization rates
• Replicate pa...

On the one hand, this may refer to the financial budget associated with the
enterprise, but can also include the amount of personnel resources that can be dedicated to a project
through to the technological resources available through the data design, collection, processing and
dissemination phases.

Crime is no longer seen as a problem in isolation, and policy-makers, researchers and service providers
now view crime as relational to other social and economic conditions.

These dimensions of quality are overlapping and interrelated.

Timeliness of data is often related to the complexity of the data required.

Victims of robbery: type of weapon used in most recent incident 188
LIST of FIGURES
Figure 1.

Transform inputs into statistics
• Produce aggregate estimates
• Impute data
• Identify and correct deviation anomalies
• Identify and treat significant aggregate level anomalies
• Identify and resolve outliers
• Weight data
• Validate weighted data
• Confidentialise data
2.

This should not be confused with the achieved sample or final
sample, which represents the survey respondents actually contacted and enumerated for the survey.

The relationship between target population, sampled population and sample frame 27
Figure 5.

In addition, a ‘crime’
may disappear or be redefined at any point in the process, for example, as a result of operational rules,
classification decisions, recordkeeping failures or a decision by the victim or the police to drop charges.

However, because of the significant differences
in scope between these two sources, several ‘global’ adjustments are required before making such
comparisons.5

For highly skewed populations, some strata can be included in the survey with 166. certainty.

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of sur...

One of the benefits of understanding the information needs of key stakeholders 73. is that it enables
an appreciation of the analytic requirements of users.

The Manual on crime victimization surveys has been developed through a joint effort of the UNODCUNECE
Task Force on crime victim surveys which consisted of the following members:
Soula MacFarlane (Australian Bureau of Statistics, Chair of the Task Force)
Jodi-Anne Brzozowski (Statistics Canada)
Karen Mihorean (Statistics Canada)
Maria Giuseppina Muratore (Istat, Italy)
Frits Huls (Statistics Netherlands)
Beata Gruszczynska (Institute of Justice of Poland)
Alison Walker (Home Office, ...

In many countries, legislation covers the
activities of official statistical agencies or government departments, and it is vital that the planned
survey activities do not contravene any of the requirements set out in statute or policy.

Once the survey goals and objectives have been set through consideration of user requirements, a
detailed process is then required to design and implement the survey and analyze, disseminate and
evaluate its results.

An example
of this type of objective may be for the survey to provide an estimate of the number of victims of
specific electronic or white collar crimes.
81.

How do interviewing methods
affect the comparability of results?

This Chapter also provides a guide to the typical structure of a crime victimization
survey and examines, in depth, the form of questions commonly asked, including those relating
to the respondent’s fear of crime, experience of household crime, personal victimization experience,
reporting of crimes to authorities and general attitudinal questions.

The Manual provides a minimum dataset of suggested
key topics for inclusion in national crime victimization surveys that will further facilitate international
comparability of crime survey results.

However, a number of other countries had
also begun to explore new and better ways to measure crime through interviewing members of the
public.

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

Based on client interaction, the survey goal was identified as ‘determine a benchmark of the level and characteristics of
consumer and mass marketed fraud victimization’.

It addresses both
the protection of respondents by means of informed consent and protection of privacy and confidentiality,
as well as data dissemination standards.

In addition to that, there are costs which increase or decrease depending upon the number of
persons, households or businesses to be sampled in the victimization survey.

We use the term crime to encapsulate the focus of a survey of crime victims but in some cases
the term can be understood to cover a wider range of events.

These
fundamental pieces of intelligence can be determined through understanding the potential stakehold-
6 Interested parties should consult: ABS Information Paper: Measuring Crime Victimization, Australia: The Impact of Different Collection
Methodologies, 2002, cat. no. 4522.0.55.001.
16
Manual on victimiZation surveys
ers in the victimization survey, and by conducting sufficient consultation to understand the information
needs that need to be addressed.

Chapter III. Methodological issues
Chapter III of this Manual provides an overview of the survey process, the main differences in the type
of victimization survey that are available to the survey manager, the key goals and objectives which
could be identified for any victimization survey along with some of the methodological issues in relation
to achieving such objectives using a social survey.

The Manual does not attempt to be prescriptive about which methods to use, but rather illustrates
examples of the types of methodologies available, including the pros and cons of the various methods.

Frequently headlines appear in the news media that report on differing trends in crime depending
on the source of data from which the reports are derived – police-reported statistics, often referred to
as official crime statistics, or victim survey data.

Victimization surveys (or victim surveys, crime
victim surveys as they will be interchangeably referred to throughout the Manual) are now a recognized
tool that help governments and their public understand their crime problems and how better to address
them.

Examples of this may be restrictions on the types of persons who can be questioned
(children or persons under or over a certain age), situations in which data can be collected (in
private settings or with other persons present), or subject-matter that can be dealt with through specific
methodologies (i.e. sensitive topics such as sexual assault).

Acquire data
• Prepare survey frame and sample
• Select sample
• Allocate interviewer workloads
• Conduct interviews and field editing
• Manage field operations
• Resolve field queries
• Capture responses and repair data
• Extract management information about the collection
process
8.

Counts of victims identified
through surveys may not appear in official police statistics, as the crime may not be reported and/or
recorded by police, therefore victimization estimates produced from surveys are likely to be higher than
data sourced from police records.
31.

• The clients were presented with a survey proposal, and memorandums of understanding were signed to secure funding for
the survey and outline the required outputs and delivery dates.

More in-depth measurement
of crime victimization in Australia is conducted on a less regular basis.

Other key factors to consider when embarking on a crime victimization survey include whether
an organization has the infrastructure available and capability to undertake a crime victimization
survey.

Joint publication can contribute towards providing the public with a
BOX 2: Typical differences between police stati stics and victim surveys
Police data Victim survey
Data source
Administrative police records Personal reports from individual citizens
Data collection method
Comprehensive coverage of police agencies Sample survey
Data recorded in an operational logbook or computer system Data captured by a survey interviewer
National, according to the scope of the police servic...

At these meetings a factor emerged which has continued
to remain in the foreground through all subsequent efforts, namely, the problem of comparability of
definitions.1
10.

The Manual does not attempt to be prescriptive about which methods to use, but rather illustrates
examples of the types of methodologies available, including the pros and cons of the various methods.

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

Based on client interaction, the survey goal was identified as ‘determine a benchmark of the level and characteristics of
consumer and mass marketed fraud victimization’.

The most basic difference between the two types of crime measurement is the method of data collection.

This will also affect any comparisons between victimization survey results and police report statistics, since the latter are also
likely to include cases that have happened to non-residents.
27
Chapter III. METHODOLOGICAL ISSUES
• driving licence lists may be used as a sampling frame for persons old enough to have such a
licence.
• phone number listings may be used as a frame for homes with phones.
• a list of businesses or organizations in a business directory may be used as ...

Probability for selection in the
sample increases as the sampling fraction decreases.

Depending on the purposes to which the data
will be put, different levels of error may be acceptable for some data.

• How are duplicates treated?

Victimization surveys (or victim surveys, crime
victim surveys as they will be interchangeably referred to throughout the Manual) are now a recognized
tool that help governments and their public understand their crime problems and how better to address
them.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Statistical agencies may also have access to the records of government agencies responsible
for paying subsidies or financial support to particular groups e.g. for disability.

A reduction in sampling error yields from the fact that the samples are drawn from subsets
of the population with homogeneous, meaningful characteristics.

A reduction in sampling error yields from the fact that the samples are drawn from subsets
of the population with homogeneous, meaningful characteristics.

Victimization surveys as a statistical activity are generally quite innovative activities
given their reasonably short histories.

Quota sampling can be proportional
or non-proportional.

Excluded from the achieved sample (denoted by the cross-hatching) are the cases that could
not be contacted for a variety of reasons, for example because they were never at home (denoted as
access denied), or because there was insufficient time during the enumeration period (denoted as
physical barriers) as well as cases that were contacted but not interviewed because they refused or were
otherwise unable.

As the surveys matured and started to produce long-run time series, they became
increasingly useful in the measurement of trends and influential both inside and outside government
as a measure of success or failure in reducing crime.
1 United Nations Crime and Justice Statistics http://www.uncjin.org/Special/history.html
2 UNODC website: http://www.unodc.org/unodc/en/data-and-analysis/United-Nations-Surveys-on-Crime-Trends-and-the-Operations-of-
Criminal-Justice-Systems.html
3 K...

Crimes and
victimization Property crime Contact crime / violent
crime
No crime
specified
Household
burglary
Theft of
vehicles Other theft Robbery Physical
Assault
Sexual
offences
Measure of victimization
in the past 12
months
x x x x x x
Measure of repeat
victimization in the
past 12 months
x x x x x x
Reporting to the
police x x x x x x
Crimes involving
weapons x x x
Victims who
suffered physical
injury
x x x
Victim-offender
relationship x x x
Public con...

Generally,
sample size is a compromise between what the survey designer would really like to accomplish and
what the available level of funding will support.

However, the expectation that police-reported crime statistics and victim surveys
should produce similar figures originates from the belief that the two measure the same phenomena,
and are based on identical objectives, methodologies and populations when, in fact, they produce two
distinct sets of crime indicators.

Both actual and perceived risks of crime can be indicators of community well-being.
28.

During the 1960s criminologists began to understand the weaknesses in administrative sources and
sought alternative methods which might provide a more accurate and informative description of crime
problems.

The survey developer must evaluate the trade-offs between conducting long interviews (to maximize
the information collected by the survey) and the potential impact longer questionnaires may have
on survey costs and sample size.

If working with areas of new or emerging areas of victimization, or areas
which have not been surveyed before, methodologies may not be well developed at all.

These can be minimized through careful training and supervision, but can never be entirely
eliminated.

The Manual provides a minimum dataset of suggested
key topics for inclusion in national crime victimization surveys that will further facilitate international
comparability of crime survey results.

Some criminal incidents
are more difficult to measure as often they rely on a person realizing they have been a victim of
crime.

• What are the characteristics of victims and perpetrators?

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Logistical difficulties may need to be considered if the survey specifications include the collection
of data from persons living in geographically remote areas or areas without certain technologies available.

The type of survey seems to influence the respondents’ selection procedure to
a certain extent: for ad hoc victimization surveys the sample composed of households seems to be the prevailing practice (75
per cent of the surveys), while for multipurpose surveys with a victimization module the sample is based on households only
in 58 per cent of the surveys.

• Police data should be adjusted to account for any populations that are excluded from the victim
survey (i.e. children under the age of 15, crimes against corporate or business victims) and adjusted,
if possible, so that the data refer to the same reference period.

UNITED NATIONS
Geneva, 2010
MANUAL ON VICTIMIZATION
SURVEYS
UNITED NATIONS OFFICE ON DRUGS AND CRIME
UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE
NOTE
The designations employed and the presentation of the material in this publication do not imply the expression of
any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontier...

Chapter VII. Data processing, estimation and analysis
Chapter VII of this Manual examines data processing, including the transfer of data from the questionnaire,
coding processes, checking and editing.

For example, the survey objectives may be different if users wish to formulate or monitor policy, provide
input into operational processes, input into lobbying of industry or government agencies, or support
decision-making by businesses and organizations.
78.

There were initial fears
that government asking the public questions about their life and attitudes would be greeted with some
suspicion, but these fears proved to be unfounded.
13.

Aside from the issue of under-reporting of crime experienced by victims to interviewers, victim surveys
can only identify criminal incidents where there is an identifiable victim.

Distribution of reference periods in one financial year of British Crime Survey 103
interviews (April-March)
Figure 7.

These early surveys
focused on health and economic questions, along with certain public attitudes.

In 2005, this
survey and the Police Population Monitor were combined to form a new regular National Safety
Monitor.
16.

Although mobile phone companies keep
addresses for billing purposes, which might be utilized by survey researchers at some point in the future, in the case of European
countries, the possibility of locating mobile phone subscribers is currently limited for two reasons: First, the majority
of European countries are dominated by pre-paid mobile phone subscriptions that do not necessarily require subscribers to
give their addresses.

Examine user requirements
• Identify internal and external stakeholder and user needs
(what are the policy/research questions that need to be
answered?)

ETHICAL CONSIDERATIONS
A. Protecting respondents 171
B. Other ethical considerations 174
C. References 176
IX. DATA DISSEMINATION AND DOCUMENTATION
A. Scope and purpose 177
B. References 188
X. CLOSING THE CIRCLE – EVA LUATING COMPLETED SURV EYS
A. The importance of the evaluation process 189
B. Data quality 190
C. Survey goals 190
D. Utility of the survey 191
E. Review of survey protocols 191
F. Methodological research 192
G. References 192
APP ENDIX A
Glossary of survey-rel...

For example, timeliness may be crucial to the work of stakeholders
or major decision-making windows may be missed.

In the UK, the increasing
interest in the views of citizens was reflected in the launch of the Government Social Survey.
12.

* See the report of the Joint UNECE-UNODC Meeting on Crime Statistics, held on 3-5 November 2004 in Geneva (http://www.unece.
org/stats/documents/ces/2004/48.e.pdf ) and also the report of the Joint UNODC-UNECE Meeting on Crime Statistics held on 21-23
October 2008 in Vienna (http://unece.org/stats/documents/ece/ces/ge.14/2008/10.e.pdf ).
preface

v
CONTENTS
Summary of Chapters ix
I. Introduction to victimization surveys A. Purpose of the Manual 1
B. History of victim surveys: evol...

Government interest in looking beyond its own administrative
sources to surveys of the population increased during the Second World War. In Finland, a question
on property stolen was introduced into Gallup opinion surveys in the 1940s.3

Perfect comparability between these different statistics is not possible because victim
surveys reflect the experiences of victimization as perceived by the victims, whereas the other sources
are the product of different administrative systems and operational processes which will vary due to
differences in legal codes, operational systems and the differing cultural contexts.
6.

Probably the most important and influential innovation with respect to victimization surveys was
the launch in 1972 of the United States’ National Crime Survey, developed from work done by the
National Opinion Research Center and the earlier President’s Commission on Law Enforcement and
the Administration of Justice.

Whereas some crime victimization surveys accept the respondent’s view as to whether an incident was
a crime, others may use the assessment of the interviewer based on answers to a set of questions about
the occurrence of particular incidents.
32.

Acting without full appreciation of the needs and requirements for the survey will result in data
that are not relevant to the policy context, and potentially an inefficient use of resources.
65.