The politics of numbers: Crime statistics in China

Abstract

This study treats crime statistics as a subject, not as a means for statistical exercise. We use the historical development of statistics in the West as a backdrop to elucidate the current status of Chinese crime statistics. Examples of the use and misuse of crime statistics are introduced to contextualize this discussion. We highlight the feasibility of conducting investigator initiated criminological research that could yield alternative (and more scientific) Chinese crime and delinquency data. It is concluded that Chinese official crime statistics remain severely limited and untrustworthy as a means of true crime measurement. The substantive improvement in the availability, credibility and future use of Chinese crime statistics is likely to depend on a normative shift in the appreciation of the value of social science and the usefulness of statistics.

Full text

1
Chapter 11
The Politics of Numbers: Crime Statistics in China
(Phil He, Revised on 9/10/12)
Abstract
This study treats crime statistics as a subject, not as a means for statistical exercise. We use the
historical development of statistics in the West as a backdrop to elucidate the current status of
Chinese crime statistics. Examples of the use and misuse of crime statistics are introduced to
contextualize this discussion. We highlight the feasibility of conducting investigator initiated
criminological research that could yield alternative (and more scientific) Chinese crime and
delinquency data. It is concluded that Chinese official crime statistics remain severely limited
and untrustworthy as a means of true crime measurement. The substantive improvement in the
availability, credibility and future use of Chinese crime statistics is likely to depend on a
normative shift in the appreciation of the value of social science and the usefulness of statistics.
Introduction
Crime statistics remain essential in the studies of criminology and criminal justice policy. The
production, use, misuse, abuse and usefulness of crime statistics have long been topics of critical
discourse. Although rare, two previous studies have been published in late 1990s that focused on
making sense of Chinese official crime data (i.e., He and Marshall 1997; Yu and Zhang 1999).
Unfortunately, the conclusions were so dire that the authors successfully prevented themselves
from using Chinese official crime statistics over the past fifteen years. More optimistically, the
questionable Chinese official crime statistics are no longer the only data source for
criminological studies. This chapter is an attempt to update earlier studies on Chinese crime
statistics with newer developments from China. It offers a comprehensive critique in order to
better understand the changing status of Chinese crime statistics and their uses for future
criminological studies. Juxtaposed with historical reviews and recent scholarships from the
Western nations, this study places a China specific inquiry on the roles of statistics in general
and crime statistics in particular in a broader context. We go beyond the limitations of Chinese
official crime statistics to explore promising alternatives that are based on recent development in
investigator initiated criminological research in China. We discuss how Chinese criminology can
benefit from scientific research including collection and utilization of original crime and justice
data.
Era of Statistics
In Western Europe, the era of statistical enthusiasm is said to have begun in the 1820s and 1830s
(Landes 1972). Almost two centuries later, however, there is no indication that the age of
statistics is nearing its end. To the contrary, in the past few decades the evolutions and
revolutions in digital information, data storage and computational capacities and computer
literacy seem to have permeated the world. Numerical data are now ubiquitous in everyday lives.
Yale law school professor and econometrician Ian Ayre’s (2007) bestselling book is aptly titled
Super Crunchers: Why Thinking-by-the Numbers is the New Way to be Smart. The Economist
(2007) forewarns “Super Crunchers presents a convincing and disturbing vision of a future in

2
which everyday decision-making is increasingly automated, and the role of human judgment
restricted to providing input to formulae.”
There are multiple reflections on the historical origin of modern statistics. The birth and spread
of quantification in early 19th century Europe (Starr 1987) and America (Cohen 1982) was seen
as an integral component of state building. It’s been argued that the very science of a state is to
govern based on statistics (see Rose and Miller 1992; Hand 2007; Lampland and Star 2009;
Miller 2001). Schumpeter (1954) posits that capitalism could proudly take credit for increasingly
quantitative habits of mind. Others suggest that the modernist “lust for precision” is to counter
disorder by first counting it properly (Hacking 1990). If the latter statement is agreeable, a
curious student of Confucianism would readily detect the philosophical contrast between the
numerical (i.e., counting things) and the normative (e.g., rectifying terms, or Zhengming, per
Analects) preferences explicated in the rules of governance. Whether, when, to what extent and
why the West and the East converge on the statistical fervor are interesting research questions.
At least in the West, an abundance of scholarly research is currently available with focus on
myriad aspects of producing and using statistics.
The statistical obsession bears close association with Western social scientific and industrial
development in the last two hundred years. The realization of learning through numbers is said to
be a major ingredient of western modernity (Urla 1993). ‘Quantifacts’ or the statistical
representations that make the world factual were seen as the foundation of ‘euromodernity’
(Comaroff and Comaroff 2006). Furthermore, the field of statistics emerged in the service of the
state with politics as its birthmark (Desrosieres 1998). Traversing through the history of statistics,
Porter (1995: 8) observes that:
The appeal of numbers is especially compelling to bureaucratic officials who lack the
mandate of a popular election, or divine right. Arbitrariness and bias are the most usual
grounds upon which such officials are criticized. A decision made by the numbers has at
least the appearance of being fair and impersonal. Scientific objectivity provides an
answer to a moral demand for impartiality and fairness. Quantification is a way of
making decisions without seeming to decide. Objectivity lends authority to officials who
have very little of their own.
There are incessant debates on the powers and perils of statistical creations. Without a doubt,
there is utilitarianism behind the creation of statistics: we count or measure things in order to
change them or change our own behavior in response to them (Stone 1997). Statistics are
politically consequential thus highly contentious. Andreas and Greenhill (2010) emphasize that,
it is the trust and faith in counting that gave quantification its political potency. Alonso and Starr
(1987) reiterate that political judgments control the choices of both the process (e.g., what to
measure, how to measure it, and how often to measure it) and the presentation of the results
based on these measures.
Statistics thus serve two purposes, says Joel Best (2001: 13): “one public, the other often
hidden.” Some critics qualify that political manipulation of statistics is typically more subtle than
outright fabrication, and that the common techniques include the deceptive use of classifications
and tolerance for methodological inadequacies that yield data with useful political effect (Starr

3
1987). Yet, others observe that statistics are not only artificial but also inadequate. Porter (1995:
86) comments that “quantification has the virtues of its vices. The remarkable ability of numbers
and calculations to defy disciplinary and even national boundaries and link academic to political
discourse owes much to this ability to bypass deep issues.” Is it not fair to notice the
contradictions? Statistics travel far because they travel light, and statistical knowledge can be
loaded yet empty at the same time.
‘Knowledge is powerful’, thus statistics as quantified knowledge are powerful too, Sir Francis
Bacon would probably agree with the logic. What might be less agreeable is the source for such
power. Truth seems to be the common denominator for Weber (1949) and Foucault (1980) who
share the view of knowledge as an instrument of power. Foucault (1980: 133) comments
specifically that “truth is to be understood as a system of ordered procedures for the production,
regulation, distribution, circulation and operation of statements,” and “truth is linked in a circular
relation with systems of power which produce and sustain it.” Nietzsche (1977) is more
parsimonious: ‘there are no facts, only interpretations.’ Objectivity underwriting the truth, adds
Porter (1995: 3), is “required for basic justice, honest government, and true knowledge.”
Reflecting truth or not, the applications of statistics, however, are error prone due to both the
producers and the consumers of the statistical information. Statistics, disregard of their validity,
are often uncritically accepted and reproduced because the consumers entrust the experts who
produce and present the numbers (Andreas and Greenhill 2010). There is a fairly long list of
books with shared focus on the use, misuse and abuse of statistics (e.g., Best, 2001, 2004, 2008;
Blastland and Dilnot 2009; Eberstadt 1995; Spirer et al. 1998; Kimble 1978; Fairley and
Mosteller 1977; Huff 1993; Mauro 1992; Maier 1995). These researchers focus on three major
themes: (1) the generation of problematic statistics; (2) why consumers of statistics believe them;
and (3) danger signs of which consumers of potentially bad statistics should be aware. All of
these books prove to be fascinating read (who can resist the chapter titles such as “Percentage:
Machiavellian Misleaders” or “The Fine Art of fooling”). The message is never clearer: use
statistics with caution!
There is a recurring complaint in the field of social sciences that many number crunchers, from
students to seasoned scholars, do not care to know enough about the data they use (Maier 1995).
This is a stern warning especially applicable to those self-claimed quantitative methodologists.
Writing on the contradictions he sees in the use and misuse of crime statistics, Manning (2009:
451-52) said it best:
Officially reported crime statistics are now more available, more easily stored,
systematized and retrieved, and can be visually displayed in maps, diagrams, tables, and
charts. They remain in place as representations of self-audited practices. Perhaps it should
be noted initially that there is no body of data that is more consistently and brilliantly
critiqued than officially gathered and processed crime data, and yet it is repeatedly used
without apology in every major journal that publishes work on crime and crime control.
The Politics of Official Crime Statistics

4
Much of the same critiques on general statistics are applicable to the discussions on official
crime statistics. In the West, the validity and reliability of official crime statistics are more often
under scrutiny than elsewhere. Governments of developed countries are also more likely to
invest in maintaining lengthy series of official crime data. It is understood that collecting,
distributing, interpreting and comparing official crime statistics is always a fluid process. Highly
aggregated official national crime statistics, especially when seen (and used) as the main source
of crime data, are met with strong suspicion from disciplined criminologists and well informed
public. There is a significant movement in recent decades in Western European and North
American countries to establish alternative crime measures to cross-check the validity of official
crime statistics and, to facilitate regional and international comparisons of crime and crime
control policies. The U.S. National Crime Victimization Survey, the British Crime Survey, the
International Crime Victimization Survey and the International Self-report Delinquency Study
are prime examples.
Countries outside the developed world often have neither interest (expressed as government
mandate) nor financial and technical capacities to create and maintain viable (large-scale,
especially longitudinal) national alternative crime database. Moreover, in countries characterized
as authoritarian regimes, police recording processes are likely to remain opaque to the outside
world for quite some time. There is often no way to gauge the citizen reporting practices in these
countries either. Behind a veil of secrecy, not only the fidelity of official crime data is
questionable for any credible scientific purposes, but politicians and pundits of crime (and public)
policy can and do manipulate crime statistics without being challenged. Unfortunately,
developed world is not immune from playing the politics of numbers. The degree of misuse and
abuse of crime statistics and the reasons for the absence of pluralistic crime data deserve close
examination. We offer two major strands of such discussions with examples below.
The Normative Dimension
Official statistics remain a vital part of academic inquiry, political discourse and governmental
policy in Western societies (e.g., Kitsuse and Cicoourel 1963). There is a normative dimension
involved in the creation of crime statistics. In the context of criminal justice and social control, to
make crime and crime control count often involves transforming crime and the criminal justice
system into something that can be measured and evaluated (Haggerty 2001a, 2001b). The
invention of official statistics such as crime rates in the 1830s and unemployment rates around
1900s was to serve a more important function, i.e., the transformation of the condition of a
society by shifting what was traditionally an individual matter (e.g., suicide, criminal acts,
unemployment, disability, and etc.) to a collective social responsibility (Porter 1995). How
public measures recreate social worlds has been a topic interested by many sociologists (see
Espeland and Sauder 2007).
Researchers have now spent considerable time in studying the sociological aspects of official
crime statistics. Changing definitions of criminal behaviors, selective law enforcement, nuances
in citizen reporting and police recording practices and manipulation of the public images of
crime have all been prime themes for theoretical exercises based on labeling, conflict and social
constructionist perspectives (see Kitsuse and Cicourel 1963; Hindness 1973; Taylor et al. 1975;

5
Bottomley and Coleman 1981). In a fascinating case study of the interplays of violence, crime
statistics, state politics and media frenzy in postcolonial South Africa, Comaroff and Comaroff
(2006) present three salient characteristics of crime statistics: (1) the paradox of dis/trust (i.e.,
crime statistics are widely cited but not necessarily trusted nor understood); (2) alienation and
intimacy (i.e., aggregated crime statistics displace visceral experience with numbers, in the
meanwhile, faceless crimes occurred elsewhere become the objects of first-person affect); and (3)
phenomenology of figures (i.e., crime statistics are seen not as a representation of reality, but as a
reality in themselves). According to the authors, crime statistics were turned into diagnostic
measures of a national health and “a discursive currency” by means of which government,
citizens, experts and the media communicate with each other, while quietly inserting their own
“inventions, inflections, inflations” in the dialogues (Comaroff and Comaroff 2006: 211; italics
added). Turning against the post-apartheid South African media, Comaroff and Comaroff (2006)
indict that the media dominated by the affluent Whites inflates mass fear by reporting incident,
not incidence, stresses on sensational acts of violence to feed the impression that massive
criminality has come to saturate the country.
Taking it beyond the border of a single country, the politics of numbers in global crime and
conflict are well illustrated in a volume edited by Andreas and Greenhill (2010). The authors
explore how war time death tolls, refugee flows, human trafficking numbers, and drug smuggling
estimates are inflated, deflated, or simply fabricated in the service of political goals. Their
volume offers an excellent read based on a series of international case studies. It produces lively
illustrations of the powers and perils of statistical maneuvering motivated by interest group
competition, organizational and bureaucratic imperatives, and the international politics and
diplomacy.
The Methodological Dimension
A very common misuse of official crime statistics is in the calculation of crime rate figures.
Crime rates are usually calculated against the size of population, often standardized by a unit of
100,000 to facilitate comparisons. As the denominator, population size could seriously impact
the crime rate calculation due to the presence of an unknown number of mobile/transient
/underground populations. This problem is not country specific. It is likely to be more acute in
countries going through rapid urbanization and/or immigration. Switching denominators are the
easiest trick to manipulate crime rates (up and down). Holding all else constant, the size of the
denominator has a direct inverse relationship with the crime rate.
Another practical problem is that alternative crime rates are not easily compared without proper
adjustments made to the denominators. The case in point is the comparison of the crime rates
from U.S. official crime statistics- Uniform Crime Reports (UCR) and the alternative crime data-
National Crime Victimization Survey (NCVS). The UCR crime rates are calculated as ‘crime per
100,000 U.S. population’ while NCVS rates are calculated as either ‘crime per 1,000 U.S.
households for property crimes’ or ‘crime per 1,000 persons age 12 and older in the U.S.’ for
personal crimes. It is recommended that researchers should apply adjustments to the denominator
of the rate computation formula for both UCR and NCVS crime series so the rates can be read as
“per 100,000 U.S. population” (see Ansari and He 2012).

6
Unlike other crime statistics, official crime statistics are the dual products of citizen reporting
and police recording. The former is found to be closely associated with public’s confidence in
the police and the overall legitimacy of the government. This makes it difficult to assign credit or
place blame with every rise and fall of crime rates. In addition, changes in law and media
generated moral panics can also artificially raise the number of recorded crimes without a true
corresponding change in criminal behavior (Meehan 2000). Manning (2009: 455) calls official
crime records “a misleading umbrella.” He argues that they reflect criminal behaviors on the
street through “a complex mixture of the reactions and judgments of agents of control.” There
are ample research evidences that support Manning’s (2009) statement. Ansari and He’s (2012)
recent study on the convergence between the UCR and NCVS crime series has considered the
potential impact based on factors such as the function of increased police productivity, changes
in citizen reporting practice, and changes in the process of measurement. For example, the
authors pointed out that the increase in the UCR crime rates in the 1980s and 1990s are attributed
to increased police productivity, e.g., improvements in police record keeping and
computerization, increase in the number of police officers per 100,000 population, increase in the
number of civilians in police special units, increase in participation of local police agencies in the
UCR and increase in the number of personnel involved in dispatching and record keeping (see
also, Catalano 2006; O’Brien 1999). Domestic violence legislation and mandatory arrest policies
for police departments, coupled with the creation of special domestic violence units, are also
factors considered by other researchers (e.g., Catalano 2006; Dugan et al. 2003). In addition, an
array of other factors have also been scrutinized in relation to changes in the UCR data, such as
attitudes of citizens toward crime and the police, changes in the measurement process of the
UCR, and the increased use of mobile phones.
The Status of Official Chinese Crime Statistics
Official crime statistics for the first four decades (since 1949) of the People’s Republic of China
(PRC) are difficult to find and even more difficult to read. He’s (1992) chapter on “Crime and
Control in China” is perhaps one of the earlier English publications that present aggregated
national crime incidence data from PRC with a time span from 1950 to 1988. Amazingly, not a
single source of data or academic reference was included in his chapter. The author, however,
did point out that all these data do not include economic cases and the compilation of crime
statistics was not possible during the decade from 1966 to 1976 due to Cultural Revolution.
Collection and reporting of crime statistics are done more regularly since the late 1980s. The
China Law Yearbook Editorial Commission, Ministry of Justice, has published reports of crime
statistics (together with other justice related statistics) on annual basis in the last two decades.
Tables of crime statistics are accessible through China Law Yearbook in both hard copies and
online versions. Nevertheless, very little published technical notes are available on how the
Chinese crime statistics are recorded, compiled and validated. Provincial, city/county and police
departmental level data are mostly left out of the official publications. The logic for Chinese
crime data reporting is not complicated, given the centralized policing system where the path of
data reporting goes from local police stations to the Bureaus of Public Security at counties/cities
and provincial levels before reaching the Ministry of Public Security in Beijing. The aggregated
Chinese official crime statistics can be quite confusing and should be used with extra care,
especially when applied in cross-national analysis.

7
A critical examination of Chinese crime statistics was conducted some fifteen years ago by He
and Marshall (1997). Using an organizational framework that highlights the impact of both
macro-level characteristics of Chinese society (i.e., legal philosophy, cultural values, political
ideology, and technological infrastructure) and factors affecting the quality and quantity of
Chinese crime statistics (i.e., legal definitions, citizen reporting and police recording practices),
the authors illustrate the problems of comparative use of Chinese crime statistics in
criminological studies, including nonstandard indicators, underestimation, and limited
availability and/or restricted access. Their conclusion was that Chinese crime statistics are
suspect as a measure of criminal misconduct, more so than in Western countries. Chinese official
crime statistics are quite useful, however, when treated as a subject of study. They can be viewed
as an index reflecting change in the Chinese government’s use of formal social control
mechanisms. Dutton and Lee’s (1993) study on policing strategies in the period of economic
reform in the 1980s offers one such example of how specialized, campaign style policing with
selective targeting (e.g., on the so-called social dregs-the new urban underclass and the transient
criminals in the more developed coastal provinces/cities) could artificially impact Chinese
official crime statistics. Moral decay associated with economic reform was targeted in the form
of “six evils”, i.e., prostitution; producing and spreading pornography; kidnapping of women and
children; drug using and trafficking; gambling; and defrauding people by superstitious means.
The campaign was declared a success that by 1990, more than 213,000 “six evils” related cases
involving 770,000 criminal suspects were recorded by the police (Dai and Jiang 1990).
Yu and Zhang’s (1999) case study examines a national survey of crime recording by the police
conducted in the late 1980s and early 1990s in China. The authors find that “local police failed to
report to their superior agencies about 70 to 80 percent of the crime known to them in the years
investigated” (Yu and Zhang 1999: 255). The under-recording is found to be directly related to
the seriousness of the offenses where property crimes have the lowest recording rate. In addition,
the authors warn that a large number of so called “public order violations”, offenses that would
otherwise be counted as crimes by international definitions, are not included in Chinese official
crime statistics. The Chinese police have the authority to sanction these relatively minor
offenses without any judicial review. The true volume of such violations would probably never
be known. Chinese legal scholars (e.g., Dai, 1994) argue that the real Chinese crime rate in 1990
should be 800 per 100,000 population rather than 200 per 100,000 as officially released. More
revealing is the authors’ explanation on why the Chinese police grossly under-record crime
statistics. Echoing He and Marshall’s (1997) earlier work on the impact of political ideology (i.e.,
the need to find diminishing crime in a socialist society and eventually a crime-free communist
society) but extending to include the actual organizational pressure (i.e., to achieve higher
clearance rate via lowering the denominator for crime rate calculation), and the loopholes in the
language of Chinese Criminal Law (e.g., Article 10 regarding ‘incidents of insignificant harm
may not be accounted crime’) in classifying certain criminal incidents, Yu and Zhang draw
(1999) a conclusion that more than adequately illustrates why the practice of under-recording
crime statistics by the Chinese police is both similar to and different from their contemporaries
of other nations.
The most comprehensive recent update on the availability of data in Chinese crime and criminal
justice research was provided by Liu (2008). The author is successful in presenting the
potentials for empirical criminological studies using quantitative crime and justice data

8
originated from China. Under official data sources, Liu (2008) includes Chinese Law Yearbook
(1978-current) and China Statistical Yearbook with the former as the focus of his in-depth
discussion. The author lists major reported statistical categories under official police,
procuratorate and court data and cites a limited number of empirical articles using these statistics.
The author includes (somewhat surprisingly in a section under official crime data) the largely
survey-based Wuhan birth cohort study, a collaborative international criminological project
funded by multiple sources, both domestic and international. The Wuhan study which was
carried out mostly in the 1990s is modeled after the research design of the famed Philadelphia
birth cohort studies pioneered by Melvin Wolfgang and colleagues in the U.S. in the 1970s.
Although potentially useful in analyses of juvenile delinquency and testing of western
criminological theories in Chinese context, the Wuhan birth cohort study data has yet to be fully
utilized. The data is available on the website of the Inter-university Consortium for Political and
Social Research (ICPSR #3751).
Having reviewed the works from both He and Marshall (1997) and Yu and Zhang (1999) on
Chinese official crime statistics, it is difficult to be optimistic about the utility of the Chinese
official crime data as a true measure of crime in China. In spite of the tremendous economic
developments that have occurred in China in the last two decades, very little substantive political
reforms have materialized. In particular, the Chinese criminal justice system and practices
remain largely untouched. There are, however, positive signs that empirical criminology and
criminal justice research are more tolerated and perhaps even appreciated by Chinese domestic
scholars and some practitioners. There is an increasing volume of published research literature
and alterative (albeit mostly city-based and limited) crime and delinquency data available. In the
next sections, we begin by introducing the major alternative Chinese crime data sources based on
Liu’s (2008) helpful recent summary, followed by demonstrations of the design features of a few
investigator initiated projects completed in China.
Alternatives to Chinese Official Crime Statistics
Liu’s (2008) introduction of victimization survey data mainly covers four key studies: the 1994
Beijing victimization survey was a part of the International Crime Victimization Survey; the
2004 Tianjin victimization survey was funded by the U.S. National Science Foundation and
jointly conducted by a group of U.S. criminologists, in collaboration with their Chinese
colleagues from Tianjin Academy of Social Science; the National Public Security Survey,
conducted by the Chinese national Statistics Bureau on annual basis since 2001; and the
International Crime Against Businesses Survey (collecting data on bribery, corruption, extortion,
fraud, and etc), which was carried out in four Chinese cities from 2005 to 2006. Judging by Liu’s
(2008) descriptions of the research designs of these victimization studies, the scientific values of
the data should be on a par with the established norm in the West. However, to our best
knowledge only the Tianjin victimization study has resulted in multiple well crafted research
articles in English language criminological journals. The Tianjin survey data is also accessible
via Inter-university Consortium for Political and Social Research (ICPSR #21740).
Coordinated multi-year self-report crime or delinquency survey is very rare in China. Liu (2008)
highlights the Tianjin prison survey which was conducted every three years since 1990 to all the
entering prisoners of the year. The survey itself is coordinated through Tianjin Academy of

9
Social Sciences with approval and support of multiple local governmental agencies. Liu (2008)
points out that a good number of scholarly publications have used this survey data. Altogether,
data for 5 sweeps (1990, 1993, 1996, 1999, and 2002) of Tianjin Prison survey have been
collected.
The absence of Chinese governmental funding to support criminological research remains a
reality. Branches of social sciences are traditionally underdeveloped and under-funded in China
compared to their sister disciplines in hard sciences and commerce. However, researchers have
found creative ways to carry out original criminological research in China. One such example is
provided below.
The ISRD-Hangzhou study is a part of the (second-sweep) large scale International Self-Report
Delinquency Study (ISRD) conducted in more than thirty countries in 2006-2008 (Ren and Zhao
2011; Webb et al. 2011). The main purpose of the ISRD study is to examine prevalence,
incidence and correlates of youth crime. The questionnaire used in China was translated from the
standard instrument used in the ISRD survey. Before its administration the translated ISRD
instrument was pre-tested among Chinese exchange students studying at a local U.S. University
to gauge its fit with the Chinese social, cultural, and language contexts. Minor adjustments in
wording were made. The targeted student population was 7th-9th grade students aged from 12 to
15. Due to the large student population and complexity of its demographics in Hangzhou, a
multi-stage cluster sampling technique was employed for the sample selection. This type of
sampling approach is suitable when it is impractical to compile a complete sampling frame. As a
result, nine middle schools located in five core urban districts in the city were selected. In each
of the selected schools, one class was randomly selected from the 7th, 8th and 9th grades,
respectively. The final sample includes: 5 public schools (15 classes), 3 private schools (9 classes)
and 1 special school for migrant workers’ children (3 classes). The survey was conducted in late
December 2009 and early January 2010. The sample included a total of 1,082 students and 1,043
useable surveys were returned, resulting in a response rate of 96.4 percent.
The study introduced above demonstrates the feasibility of alternative crime and delinquency
data collection at city level in China. There are also indications of successful national level
studies, i.e., World Value Survey-China (WVS-China) and Chinese General Social Survey
(CGSS). WVS-China was coordinated by the Research Center for Contemporary China at
Peking University in 2005-2006 while CGSS was carried out by the Survey Research Center at
Hong Kong University of Science and Technology (HKUST) in 2003, 2004 and 2005. The
targeted sample size used in the WVS-China was 2,873 (Shen 2007) and for CGSS, 2,000
(Survey Research Center, HKUST, 2003). The design for WVS-China looks very promising.
According to Shen (2007), the survey has achieved a response rate of 78.6% resulting in 1,991
valid interviews. The student research associates were able to complete all field work within a
two-month time frame. The researchers used a GPS/GIS Assisted Area Sampling (spatial
sampling) technique that overcomes the inability to reach migrants in traditional household lists
based samples (Landry and Shen 2005). The spatial sampling technique is a particularly elegant
design feature, given the fact that the year 2000 Chinese Census classified 144.4 million people
(or 12% of total population) as migrants (National Bureau of Statistics of China 2002).
Recent Development in Chinese Empirical Legal Research

10
It’s worth mentioning of a unique recent development in empirical legal research in China. This
development is characterized by some observers as following the ‘Vera model of criminal justice
reform’ in the U.S. Funding for empirical legal research is strongly tied to experimentations
related to Chinese court reform, focusing typically on thorny issues related to criminal due
process. Ford Foundation Beijing Office is to be credited for playing a leading role in promoting
empirical legal researches in China in the recent decade. Other foundations and organizations
also funded similar projects (e.g., MacArthur Foundation and International Bridges for Justice).
From 2002 to 2011, 18 scholar-led demonstration projects on criminal procedure reform have
been funded by the Ford Foundation (Stutsman 2011). These empirical projects cover topics such
as police interrogation, pretrial discovery, bail and discretionary non-prosecution, live witnesses,
evidence collection, victim-offender mediation, criminal defense, plea bargaining and sentencing.
These projects and research teams were led by eight senior Chinese legal scholars in the field of
Chinese criminal procedure law. Researchers and scholars from Vera Institute of Justice,
Harvard Kennedy School of Government and Northeastern University in the U.S. played very
active roles in providing methodological training and assistance to the Ford Foundation criminal
justice project grantees in China. There is strong indication that the tradition of empirical legal
research is beginning to set roots in China. Most Chinese legal scholars (including all the
Principal Investigators of Ford demonstration projects in the last decade) we interacted with do
not consider themselves criminologists, although their research topics may overlap with the
subject matters studied by criminologists specializing in court and sentencing in the West.
Application of empirical methods, including experimental design methodology (with varying
levels of design sophistication and implementation fidelity), is a common feature of these
demonstration projects. These projects, however, should not be mistaken for traditional
sociological studies due to the obvious absence of any theoretical focus. Also, the statistical
analysis used to report the outcomes of these studies remain rudimentary, including mostly
simple counts, percentages and pie/bar graphs. We are not aware of any hypothesis testing being
used in any of these studies so far. Research design and statistics are taught in less than a handful
of Chinese law schools at the moment. Nevertheless, the awareness and interest have grown
substantially in recent years given the promotion of empirical legal research and a series of large
scale empirical methods training workshops co-sponsored by Ford Foundation and the Institute
of Procuratorate Theory, the research arm of the Supreme People’s Procuratorate.
Conclusion
Western scholars trying to decipher the historical Chinese nonchalance in scientific development
discover that “traditional China developed both a very rich tradition of empirical observation and
an active skeptical orientation but did not develop a scientific tradition as we know it, in part
because it lacked a third necessary ingredient to such a tradition, namely, a theoretical orientation,
an inclination to leave the world of practical application behind in an effort to construct and test
pure theoretical explanatory frameworks” (Bloom 1981: 34). This is an intriguing observation,
which appears to have face validity in describing the current status of Chinese criminology. The
difficult birth and arrested development of Chinese criminology in the 20th century is
understandable. China spent much of the last century in a survival mode, recovering from
dynasty transition, foreign invasions, civil wars, and ideological frenzy underlying socialist
revolutions. The 21st century Chinese criminology is very much a work in progress (a brief

11
review of the historical background and current status of Chinese criminology can be found in
Cao 2004 and Chapters 1 and 2 in this book). We argue that there is a need to establish an
independent Chinese criminological enterprise that breaks away from its status as a government
subsidiary. This will be an arduous journey where the process is just as important, if not more,
than its destination. Foucault (1980) once pointed out that the power circuits where (Western)
criminologists produce knowledge are threefold: the first sector is their class position (who do
they work for); the second sector includes the criminologist’s conditions of life and work as an
intellectual (the field of research, the intellectual’s place in the academy, the political and
economic demands in the university or other place of work); and the third sector is the politics of
truth in the wider society. Foucault’s (1980) statements are quite applicable to the Chinese
criminologists as well.
There is a normative dimension of discussion that often gets lost. Why has science and
philosophy generated greater interest from Western scholars than the Chinese? Why does
Chinese social science generally, and sociology and criminology in particular, struggle to grow
out of their infancy? Sciences, argues Karl Popper (1945), set free the critical powers of man.
John Dewey who taught in China in the early 20th century considers science an ally of
democracy. He treats scientific method as a means to subject beliefs to skeptical inquiry. He
even tried his hand at reforming the Chinese higher educational system. Democracy and science
were the popular vocabulary in the first two decades of the 20th century. How ironic that the
same two words are made popular again a century later.
One can argue that to understand the usefulness of Chinese criminology rests first in seeing its
potential value as a social science. Scientific knowledge depends on the taming of human
subjectivity. And, objectivity means the rule of law, not of men (Porter 1995). The rule of law
has never been a Chinese tradition. Unlearning one’s culture is a difficult task if not impossible.
Loader and Sparks (2011) hold a view that sees criminology as “democratic under-laboring.” In
their book interestingly titled Public Criminology? with a question mark at the end, Loader and
Sparks (2011) group the styles of criminological engagement in five categories: ‘scientific
expert’, ‘policy advisor’, ‘observer-turned-player’, ‘social movement theorist/activist’ and
‘lonely prophet’. The degree to which doing criminology (our emphasis) is an exercise of
learning, interpreting and practicing democracy is a good question to ask ourselves and our
colleagues in China. At this moment, we do not think Chinese criminologists are playing any of
the roles described by Loader and Sparks (2011).
Let’s now return to statistics. As we have discussed earlier, statistics do not exist naturally. They
are manmade and born with a political birthmark. Doing statistics is both a rationalized activity
and a learned skill. The use and usefulness of statistics have not been topics well discussed in
China until recently. Liu (2008: 131) is certainly correct in pointing out that studies of crime in
China tend to be “non-empirical” (he meant non-quantitative, perhaps) and that Chinese
criminologists are mostly law specialists with limited quantitative training. We also do not fully
disagree with Liu’s (2008:131) assessment that “a most difficult impediment for studying crime
and justice in China is accessing quantitative data.” We would argue rather that a true
appreciation of the usefulness of crime statistics, the value of scientific data collection and the
necessity for evidence-based policy must precede the actual use of crime data (for a similar
conclusion on the research-policy nexus in Taiwan, see the chapter on criminology in Taiwan in

12
this Handbook) . In addition, we must reiterate an important point from an earlier discussion that
statistics tend to oversimplify complicated social facts. There are many things in life that could
not and should not be measured with simple statistics. It is the responsibility of those who
measure and present things quantitatively to resist the temptation to overstate the utility of
quantitative data, to fabricate explanations or to transmit statistical illusions.
As with other social statistics, crime statistics are only as meaningful as the skill (and we would
add, integrity) of the researcher who uses them (Maier, 1995). There remains a deficient critical
mass of well trained criminology specialists in China. Chinese criminologists, criminology
curricula and research are scattered in research institutes affiliated with government agencies,
law schools and police universities/colleges (see Liu and Yu 2010 for a detailed description of
the current status of Chinese criminology, including leading criminology centers, major
criminologists and the future development of the discipline). Some Chinese criminologists have
expressed frustration that they are afforded the status of second-class citizen, members of an
underdeveloped academic discipline. This experience is partly reflected in the lack of academic
prestige, promotion opportunities, adequate funding for research and travel and new courses to
teach in the curriculum. To a certain degree, this is not unlike the experiences many American
criminologists have had half a century ago and many criminologists elsewhere in the world are
having today. The encouraging news is that there is an increased amount of scholarly exchanges
between Chinese domestic criminologists and colleagues overseas. The Asian Criminological
Society (ACS) and the Association of Chinese Criminology and Criminal Justice in the U.S.
(ACCCJUS) are two professional organizations with developed networks to China. Researchers
inside and outside of China are creative in pursuing active criminological research with
increasing scientific rigor. The awakening of the criminological imagination and maturation of
criminology as a discipline in China will arrive in due course. Given what we have learned so far,
the substantive improvement in the availability, credibility and future use of Chinese crime
statistics requires a normative shift in the appreciation of the value of social science and the
usefulness of statistics. We hope this and other chapters included in the volume could serve as
harbingers for the growth and prosperity of Chinese criminological research in the decades to
come.

13
Reference
Alonso, W. and Starr, P. (eds) (1987) The Politics of Numbers, New York: Russell Sage
Foundation.
Andreas, P and Greenhill, K.M. (eds) (2010) Sex, Drugs, and Body Counts, Ithaca, NY: Cornell
University Press.
Andreas, P and Greenhill, K.M. (2010) ‘Introduction: The politics of numbers’, in P. Andreas
and K.M. Greenhill (eds) Sex, Drugs, and Body Counts, Ithaca, NY: Cornell University Press.
Ansari, S. and He, N (2012) (forthcoming) ‘Convergence revisited: A multi-definition, multi-
method analysis of the UCR and the NCVS crime series (1973-2008)’, Justice Quarterly.
Ayre, I. (2007) Super Crunchers: Why Thinking-by-the Number is the New Way to be Smart,
New York: Bantam.
Best, J. (2001) Damned Lies and Statistics: Untangling Numbers from the Media, Politicians,
and Activists, Berkeley: University of California Press.
Best, J. (2004) More Damned Lies and Statistics: How Numbers Confuse Public Issues, Berkeley:
University of California Press.
Best, J. (2008) Stat-Spotting: A Field Guide to Identifying Dubious Data, Berkeley: University of
California Press.
Blastland, M. and Dilnot, A. (2009) The Number Game: The Commonsense Guide to
Understanding Numbers in the News, in Politics, and in Life, New York: Gotham Books.
Bloom, A.H. (1981) ‘Language and theoretical vs. reality-centered morality’, pp. 21-37, in
Wilson, R.W., Greenblatt, S.L. and Wilson, A.A. (eds) (1981). Moral Behavior in Chinese
Society, New York: Praeger Publishers.
Bottomley, A.K. and Coleman, C. (1981) Understanding Crime Rates: Police and Public Roles
in the Production of Official Statistics, Westmead: Gower.
Cao, L. (2004) ‘The state of criminology in China’, The Criminologist, 29: 1-5.
Catalano, S. M. (2006) The measurement of crime: Victim reporting and police recording, New
York: LFB Scholarly Publishing.
Cohen, P.C. (1982) A Calculating People: The Spread of Numeracy in Early America, Chicago:
University of Chicago Press.
Comaroff, J. and Comaroff J.L. (2006) ‘Figuring crime: Quantifacts and the production of the
un/real’, Public Culture, 18: 209-46.
Dai, Y.S. (1994) ‘On the policy of public security’, Series on Studies of China’s Juvenile Crime
and Delinquency, Chongqing: Chongqing Publisher.
Dai, Y.S. and Jiang, B. (1990) ‘Mobilize and organize all possible social forces to strengthen the
public order’ （动员和组织各界社会力量加强社会治安）, People’s Police (人民公安) 4:
4-9.
Desrosieres, A. (1998) The Politics of Large Numbers: A History of Statistical Reasoning,
Cambridge: Harvard University Press.
Dugan, L., Nagin, D. S., and Rosenfeld, R. (2003) ‘Exposure reduction or retaliation? The
effects of domestic violence resources on intimate-partner homicide’, Law & Society Review,
37: 169-98.
Dutton, M. and Lee, T. (1993) ‘Missing the target? Policing strategies in the period of economic
reform’, Crime & Delinquency, 39: 316-36.
Eberstadt, N. (1995) The Tyranny of Numbers: Mismeasure and Misrule, Washington, D.C.:
American Enterprise Institute.

14
Espeland, W.N. and Sauder, M. (2007) ‘Ranking and reactivity: How public measures recreate
social worlds’, American Journal of Sociology, 113: 1-40.
Fairley, W. and Mosteller, F. (eds) (1977) Statistics and Public Policy, Reading: Addison-
Wesley.
Foucault, M. (1980) ‘Truth and power’, in C. Gordon (ed), Michel Foucault, Power/Knowledge-
Selected Interviews and Other Writings 1972-1977. Brighton: Harvester.
Hacking, I. (1990) The Taming of Chance, Cambridge: Cambridge University Press.
Haggerty, K.D. (2001a) ‘Negotiated measures: The institutional micropolitics of official criminal
justice statistics’, Studies in History and Philosophy of Science, 32: 705-22.
Haggerty, K.D. (2001b) Making Crime Count, Toronto: University of Toronto Press.
Hand, D. (2007) Information Generation: How Data Rule Our World, Oxford: Oneworld.
He, B. (1992) ‘Crime and control in China’, pp. 241-45, in Shelley L.I. and Katoh, H (eds),
Crime and Control in Comparative Perspectives, Berlin: Walter de Gyuyter.
He, N. and Marshall I. (1997) ‘Social production of crime data: A critical examination of
Chinese crime statistics’, International Criminal Justice Review, 7: 46-63.
Hindness, B. (1973) The Use of Official Statistics in Sociology: A Critique of Positivism and
Ethnomethodology, London: Macmillan.
Huff, D. (1993) How to Lie with Statistics, New York: W.W. Norton.
Karstedt, S. and LaFree, G. (2006) ‘Democracy, crime and justice’, Annals of the American
Academy of Political and Social Science, 605: 6-23.
Kimble, G.A. (1978) How to Use (and Misuse) Statistics, Englewood Cliffs: Prentice-Hall, Inc.
Kitsuse, J.I. and Cicourel, A.V. (1963) ‘A note on the uses of official statistics’, Social Problems,
11: 131-39.
Kitsuse, J.I. and Cicourel, A.V. (1963) The Educational Decision-makers, Indianapolis, IN:
Bobbs-Merrill.
Lampland, M. and Star, S.L. (eds) (2009) Standards and Their Stories: How Quantifying,
Classifying, and Formalizing Practices Shape Everyday Life, Ithaca: Cornell University Press.
Landes, D. (1972) ‘Statistics as a source for the history of economic development in Western
Europe: The Protostatistical era,’ in Val R. Lorwin and J.M. Price (eds), The Dimensions of
the Past: Materials, Problems, and Opportunities for Quantitative Work in History, New
Haven: Yale University Press.
Landry, P.F. and Shen, M. (2005) ‘Reaching migrants in survey research: The use of the global
positioning system to reduce coverage bias in China’, Political Analysis, 13: 1-22.
Lindesmith, A.R. (1965) The Addict and the Law, Bloomington, IN: Indiana University Press.
Liu, J. (2008) ‘Data sources in Chinese crime and criminal justice research’, Crime, Law, and
Social Change, 50: 131-47.
Liu, X. and Yu, Y. (2010) ‘Analysis of the status of Chinese criminology and its future
development’(犯罪学学科概况及发展趋势分析), Journal of Shangdong Police College（山
东警察学院学报), 4: 83-91.
Loader, I and Sparks, R. (2011) Public Criminology? London: Routledge.
Maier, M.H. (1995) The Data Game: Controversies in Social Science Statistics (2nd ed), Armonk:
M.E. Sharpe.
Manning, P.K. (2009) ‘Policing as self-audited practice,’ Police Practice and Research, 10: 451-
64.
Meehan, A.J. (2002) ‘The organizational career of gang statistics: The politics of policing gangs’,
The Sociological Quarterly 41: 337-70.

15
Miller, P. (2001) ‘Governing by numbers: Why calculative practices matter’, Social Research, 68:
379-96.
Mauro, J. (1992) Statistical Deception at Work, Hillsdale: Lawrence Erlbaum Associates,
Publishers.
Nietzsche, F. (1977) The Portable Nietzsche (translated by Kaufmann, W.), New York: Penguin
Books.
O'Brien, R. M. (1999) ‘Measuring the convergence/divergence of “serious crime” arrest rates for
males and females: 1960-1995’, Journal of Quantitative Criminology, 15: 97-114.
Popper, K. (1945) The Open Society and Its Enemies, London: Routledge.
Porter, T.M. (1995) Trust in Numbers: The Pursuit of Objectivity in Science and Public Life,
Princeton: Princeton University Press.
Rand, M. R., and Rennison, C. M. (2002) ‘True crime stories? Accounting for differences in our
national crime indicators’, Chance, 15: 47-51.
Ren, L. and Zhao, J. (2011) Executive Summary: A Pilot Study of Self-report Juvenile
Delinquency in Hangzhou, China. Huntsville, TX: College of Criminal Justice, Sam Houston
State University.
Rose, N.S. and Miller, P. (1992) ‘Political power beyond the state: Problematics of government’,
British Journal of Sociology 43: 173-206.
Schumpeter, J.A. (1954) History of Economic Analysis. New York: Oxford University Press.
Shen, M. (2007) World Value Survey-China, Technical Report.
Spirer, H.F., Spirer L. and Jaffe, A.J. (1998) Misused Statistics, 2nd ed., New York: Marcel
Dekker.
Starr, P. (1987) ‘The sociology of official statistics’, in W. Alonso and P. Starr (eds) The Politics
of Numbers, New York: Russell Sage Foundation.
Stone, D. (1997) Policy Paradox: The Art of Political Decision Making, New York: Norton.
Stutsman, T. (2011) ‘The use of demonstration projects to advance criminal procedure reform in
China’, Columbia Journal of Asian Law, 24: 335-66.
The Economist (2007) ‘Automated decision-making: The death of expertise’, The Economist,
Online. Available HTTP: http://www.economist.com/node/9795019.
Taylor, I., Walton, P. and Young, J. (1975) Critical Criminology, London: Routledge.
Urla, J. (1993) ‘Cultural politics in an age of statistics: Numbers, nations and the making of
basque identity’, American Ethnologist, 20: 818-43.
Webb, V., Ren, L., Zhao, J., He, N. & Marhsall, I. (2011) ‘A comparative study of youth gangs
in China and the United States: Definition, offending and victimization’, International
Criminal Justice Review 21: 225-42.
Weber, M. (1949) The Methodology of the Social Sciences. New York: Free Press.
Yu, O. and Zhang, L. (1999) ‘The under-recording of crime by police in China: A case study’,
Policing: An International Journal of Police Strategies & Management, 22: 252-63.