Attitudes Towards Crime and Punishment in the EU

1
Attitudes Towards Crime and Punishment in the EU
Results from the ECSS with Focus on Germany
ECSS Workshop Bernd Kuehnrich Gallup Europe,
Brussels Harald Kania November 21th, 2005
2
Overview

• Cross-country analysis with focus on Germany
• Main topics of our analysis
• Structure and methodology of analysis
• Results
• Further Analysis (Harald Kania)

3
Main Topics of our Analysis

• Attitude toward punishment
• Fear of crime
• Reporting crime
• Attitude towards police
• Seriousness of crime

4
Structure and Methodology of Analysis

• Descriptive data
• Cross-country comparison with focus on Germany
• Metric ANOVA gt Multiple comparisons (t-Test)
• Ordinal Kruskal-Wallis H-Test gt Mann-Whitney
  U-Test
• Regression Model for German sample
• General model Validation
• Specific models for subsamples
• Further Analysis (Harald Kania)

5
Overview

• Cross-country analysis with focus on Germany
• Main topics of our analysis
• Structure and methodology of analysis
• Results (Attitude toward punishment)
• Used database
• Used items
• Descriptive data
• Cross-country comparison (Prison Term)
• Regression models (Sanction Categories)

6
Used Database
7
Used Items (1) Sanction Category

• Wording q320
• People have different ideas about the sentences
  which should be given to offenders. Take for
  instance the case of a 21 year old man who is
  found guilty of burglary for the second time.
  This time hes taken a TV. Which of the following
  sentences do you consider the most appropriate
  for such a case?
• Categories
• Community service
• Fine
• Suspended sentence
• Prison

8
Descriptive Data (1) Sanction Category
9
Descriptive Data (1) Sanction Category
10
Used Items (2) Length of Prison Term

• Wording q321
• For how long do you think he should go to
  prison?
• Categories (8 / 14)
• 1 1 month or less
• 2 2 - 6 months
• 3 6 - 12 months
• 4 1 year
• 5 2 years
• 6 3 - 5 years
• 7 6 - 10 years
• 8 gt 10 years

11
Descriptive Data (2) Prison Term
12
Descriptive Data (2) Prison Term
13
Cross-Country Comparison Prison Term

• Ordinal scale level gt non-parametric tests
• Overall comparison gt Kruskal-Wallis H-test (p
  .000)
• Further pairwise comparisons gt Mann-Whitney
  U-test
• 18 Countries gt 153 pairwise comparison possible
• Germany vs. all other 17 countries

14
Cross-Country Comparison Prison Term
15
Overview

• Cross-country analysis with focus on Germany
• Main topics of our analysis
• Structure and methodology of analysis
• Results (Punitiveness)
• Used database
• Used items
• Descriptive data
• Cross-country comparison (Prison Term)
• Regression models (Sanction Categories)

16
Categorical Regression Procedure

• Simultaneous use of nominal, ordinal, and
  numerical variables in one model
• Quantification of categorical data
• Assigning numerical values to the categories
• Resulting in an optimal linear regression
  equation for the transformed variables
• Quantifications reflect characteristics of the
  original categories
• Treating quantified categorical variables in the
  same way as numerical variables

17
Regression Model Procedure (CatReg)

• 17 relevant predictors in model (Literature)

• Sex
• Age
• Victimization (3)
• Level of self security (4)
• Occupation (5)
• Income quartiles (4)
• People in household (6)
• General life satisfaction (4)
• Fear of crime Emotional (4)

• Immigrant in family
• Going out How often?
• Attending religious service How often?
• General health rating
• General happiness rating
• Fear of crime Cognitive component
• Fear of crime Projected burglary
• Satisfaction with police

• Successive removal of items when
• quantification of punitiveness was not
  interpretable
• Beta lt .05
• Probability of error p gt .01
• ?R2 was too small

18
General Model Germany

• Quantification of criteria (dV) Punitive
  attitudes Sanction category

19
General Model Germany
Model with 9 Predictors (N 1662)
20
General Model Germany
Cross-validation of Model (N 800)
21
Specific Models for Subsamples
General Model German Sample
Victim Status
Sex
Non Victims
Non-Contact Crime
Contact Crime
Men
Women
22
Overview

• Cross-country analysis with focus on Germany
• Main topics of our analysis
• Structure and methodology of analysis
• Results (Punitiveness)
• Used database
• Used items
• Descriptive data
• Cross-country comparison (Prison Term)
• Regression models (Sanction Categories)
• Further exploration and explanation

23
Further Exploration and Explanation

• Comparison of contrast groups with extreme values
  for the punitive attitudes items
• Most Lenient Group Community Service (n
  1023)
• Most Punitive Group Prison at least one year
  (n 143)

24
Further Exploration and Explanation
z-Scores Variables
-0.20
-0.15
-0.10
-0.05
0.00
0.05
0.10
0.15
0.20
Sex
Age
Victimization
Level of Self Security
Income Quartiles
Household Size
General Life Satisfaction
Fear of Crime Emotional
25
Overview

• Cross-country analysis with focus on Germany
• Main topics of our analysis
• Structure and methodology of analysis
• Results (Attitude toward punishment)
• Further Analysis (Harald Kania)

26
End of First Presentation

• Thank you for your attention

27
General Model Germany Sex Age

• Interaction AGE x SEX

28
General Model Germany Victimization
(n 841)
(n 469)
(n 352)
29
General Model Germany Self Security
(n 182)
(n 296)
(n 357)
(n 827)
30
General Model Germany Occupation
(n 886)
(n 90)
(n 118)
(n 425)
(n 143)
31
General Model Germany Income
(n 536)
(n 466)
(n 320)
(n 340)
32
General Model Germany People
(n 525)
(n 565)
(n 288)
(n 198)
(n 29)
(n 57)
33
General Model Germany Life Satisfaction
(n 42)
(n 135)
(n 1058)
(n 427)
34
General Model Germany Fear of Crime
(n 452)
(n 677)
(n 394)
(n 139)
35
Subsample Model Men
36
Subsample Model Women
37
Subsample Model Non-Victims
38
Subsample Model Victims Non Contact
39
Subsample Model Victims Contact Crime