Transcript Sociology 101

Institutional-Anomie, Political Corruption,
and Homicide Rates
Jerry K. Daday
Lisa M. Broidy
Dale Willits
Western Kentucky University
University of New Mexico
University of New Mexico
Institutional Anomie Theory (IAT)

Framework for examining the influence of cultural and
structural forces on violent crime rates between social
aggregates (Messner and Rosenfeld, 2006)
 Culture of the American Dream


Structural imbalance between the economy and noneconomic institutions



Individualism, achievement, universalism, fetishism of money
Education, Polity, Family, Health Care
Devaluation, Penetration, Accommodation
Weakened social support, social controls & pro-social
socialization
Prior Research

Core Hypothesis: Strong non-economic
institutions can reduce or mediate the anomic
effects of the economy


Inequality often serves as proxy measure for
economy
Non-economic institutions have a direct, negative
effect on violent crime rates
 % of GDP spent on education
 % of GDP spent on health care
 % of population that votes
Prior Research

Several cross-national and sub-national test
support IAT in whole or in part




Messner and Rosenfeld (1997)
Savolainen (2000)
Pratt and Godsey (2003)
Kim and Pridemore (2005)
Limitations of Prior Research

The polity remains an understudied area



Studies primarily use % voter turnout to assess the
strength/apathy of the polity
The effects of other non-economic institutions have
been studied extensively
The effects of political corruption on institutionalimbalance has been understudied


Significant implications for IAT
Corruption: economic principals influencing political
decision making
Current Study

Messner and Rosenfeld (2001) stress the need
for empirical tests of IAT that incorporate
corruption measures

“We have speculated that high levels of corruption
result under conditions of political dominance and that
the dominance of the so-called primordial institutions
such as the family and religion leads to excessive
social control and human rights abuses. No one to
our knowledge has yet tested these implications of
IAT.”
Current Study

We have attempted to address this need by

Examining the effects of corruption, as a measure
of the polity, on cross-national homicide rates



Directly (net of controls)
Net of non-economic institutions (educ and health care)
Interacting with non-economic institutions
Our study

Sample Size = 97 countries

Includes all countries for which data exist all on
variables for at least 1 year between 1998-2004




Average between 1998-2004
OLS Regression Models
Imputation: EM Algorithm
Dependent Variable: Logged Homicide Rate

United Nations Crime Statistics
Independent Variables

Polity: Political Corruption Index

Transparency International’s Corruptions
Perceptions Index (CPI)



From Heritage Foundation
Defines corruption as the abuse of public office for
private gain
Composite index draws on different polls and
surveys from independent institutions

Example: Global Competitiveness Report and World Bank
Business Environment and Enterprise Survey
Independent Variables

Non-Economic Institution Index



Education
 Teacher ratio and gross secondary enrollments in school
 World Bank Development Indicators (WBDI)
Health Care
 % of GDP spent on Health Care, infant mortality, and life
expectancy
 WBDI
It was necessary to index these two items because of
their high correlation

Loaded on a single component score, Eigenvalue = 3.54
Control Variables

% Female


WBDI
Inequality


Ratio of Top 20% of Income Earners to Bottom
20% of Income Earners
WBDI
OLS Regression Coefficients (n=97),
DV= log (homicide)
Variable
Model 1
Model 2
Model 3
Corruption Index
0.016**
0.010*
0.006
Non-economic Institutions
-
-0.251*
-0.871**
Interaction
(Corruption X NonEconomic Institutions)
-
-
0.011**
Female (%)
0.149**
0.168**
0.168**
Inequality
0.065**
0.066**
0.057**
Constant
-7.729**
-8.360**
-7.910**
R2
0.4047
0.4331
0.4734
* = significant at the 0.05 level, ** = significant at the 0.01 level
Results

Control variables: Inequality and % Female
have positive relationship with homicide rates

Corruption has a positive relationship with
homicide

Non-economic institutions have a negative
relationship with homicide

Mediating effects
Results and Conclusions

Interaction between Corruption and the Strength Noneconomic Institutions


Positive relationship
Non-economic institutions mediate the effects of corruption
up to a certain point



Before the corruption index reaches 79 (out of 100), noneconomic institutions continue to have an overall negative effect
on homicide
When the corruption index exceeds 79, the non-economic
institutions no longer insulate a country from the effects of
corruption
Conclusion


Pervasive effects of Corruption
Support for IAT: mediating influence of non-economic
institutions