Hungarian Quality Report

Hungarian quality report

1. Response rates

The sample design consisted of 1096 respondents, 646 answersarrived, subsequently, theresponse rate was 59%. The response-non-response distribution isas follows:

Codes / Number interviews / Distribution, %
Successful / 646 / 59.0
Address isunidentifiable / 4 / 0.4
Address does not exist / 7 / 0.6
Empty dwelling / 15 / 1.4
Not dwelling / 5 / 0.5
Target person is not available / 149 / 13.6
Target person has died/moved / 117 / 10.7
Refused answer / 135 / 12.3
Inability to answer / 16 / 1.5
Total / 1094 / 100.0

The response rate by counties was different,in the Hungarian capital it wasconsiderablylowerthan in the others (39.-,2%),it wasrelative highin counties: Tolna 72.3%, Szabolcs-Szatmár72.0%, Győr-Sopron 70.5%, Hajdú-Bihar 65.5%, Pest 64.8%.

Comparing the data of the original sample design and the respondents’ file, the original design was only a little distorted. Younger persons answer less frequently than the older ones.

Distribution of the sample design and the respondents by sex ad age

Sample design / Respondents
Age / Men / Women / Together / Men / Women / Together
18-24 / 13.3 / 11.3 / 12.2 / 13.7 / 9.3 / 11.3
25-29 / 9.5 / 6.2 / 7.8 / 7.6 / 4.8 / 6.0
30-34 / 11.7 / 7.6 / 9.5 / 10.0 / 7.6 / 8.7
35-39 / 8.7 / 8.1 / 8.4 / 9.3 / 7.6 / 8.4
40-44 / 8.2 / 8.6 / 8.4 / 10.3 / 8.5 / 9.3
45-49 / 7.6 / 6.7 / 7.1 / 9.3 / 7.9 / 8.5
50-54 / 6.4 / 8.8 / 7.7 / 6.2 / 9.3 / 7.9
55-59 / 6.8 / 7.3 / 7.0 / 5.8 / 8.7 / 7.4
60-x / 27.8 / 35.4 / 31.9 / 27.8 / 36.3 / 32.5
Total / 100.0 / 100.0 / 100.0 / 100.0 / 100.0 / 100.0

1. Organisational aspects of the testing

The testing of the victimization survey was carried out by the Hungarian Central Statistical Office, based on its own resources.The implementationwas coordinated by the Social Services Statistics Department, the operational work was organized by the Regional Directorate Debrecen at local level in the selected counties Győr-Sopron, Hajdú-Bihar, Tolna, Borsod-Abaúj-Zemplén, Pest, and in the capital Budapest. The most experienced and practiced interviewers were selected to carry out the field work. Training was organized for the interviewers at the end of April 2009in each county. The trainers were the experts of the Social Services Statistics Department and the employees of the Counties’ agencies. The duration of the trainingwas approximately 2 hours. The instructionscontained all aspects of the survey: main aims of the testing, parts of the questionnaire packet, problematic issues of the questionnaire and the financial aspects of the testing. At the end of the traininginterviewers filled out a “test” questionnaire to checkup whether theymake sense of the instructions andhow they can apply fundamentals in practice.

The questioning period was 11-24 May. The interviewer couldask only that person who was chosen in the sample design. If the chosen person was available, the interviewer had three possibilities to try to contact the respondent. When the third attempt to contact was also unsuccessful, the interviewer was asked to mark the reason.

The questionnaire was divided into two main parts. The first part(A Respondent and household characteristics, B Feeling safe and worries about crime, C Screening questions, D Victim form, E Other victimisation issues, F Other safety issues) was aface-to-face interwiev by paper-pencil method, the second part (G Violence) was filled out by the respondent later without the interviewer or if he/she needed help they could fill in it together. This part of the questionnaire could be sent to the Counties’agencies or collected later by the interviewer in a closed envelop.

During the period of questioning we followed the procedure with attention,and asked for a weekly status report by phone. The local contact points were in touch continuouslywith the interviewers to solvethe problems. There wasanother form of the feedback as well: we prepared a specialquestionnaire for the interviewers about their experiences on the questionnaire and the field work. The main results of them were used in this quality report.

Thedata entry was madeon regional level. The regional data were forwarded to the HCSO Informatics Department, where the experts prepared the database.Finally, this database was forwarded to the Social Services Statistics Department. Wechecked the database concentratingon the main logical coherence (following instructions connected with each question, false answers, etc).

As the aim of this project was the testing of the victimization survey module and not the analysis of data, we have not analyzedthem yet but we plan to do this in the future.

1. Description of difficulties encountered

Based on answers of interviewers in the above mentioned special questionnaire, we summarize the experiences and recommendations:

-As we could not apply either CAPI or CATI methods forthe field work, there were some problems connected totransforming from the original logic to paper-pencil logic. There were editorial and technical problems too. These causedseveral difficulties in the field work.

-In general,the respondents’ opinion was that the questionnaire was too long, too detailed. It was the most difficult and the longest questionnaire in the last few years.

-In the opinion of the interviewers,the work needed more efforts to involve the chosen person and maintain his/her interest and attention to make a successful interview. Some interviewers presented gifts to the respondents in the hope that it would increase the inclination to answer.

-The repetition of some questions and answer categories,as it was necessary to specialize them by the interviewer,taxed the respondents’ patience.

-Many respondents reluctantly answered some questions such as date of birth, place of birth (they feared of identification), income, owning a car and/or asecond home (which are possible indicators of living conditions), tools for defending home (because of fear of burglary if an unauthorized person gets this information), violence (many respondent considered this issue as part of their privacy), bribery (because it is a crime, however gratuity paidt to doctors and nurses is not a crime in the respondents’ opinion.)

-The theft of the parts of bicycles is not a crime in many respondents’ opinion. Hence it was necessary to call theattention of the respondents to this.

-Many respondents consideredonly crimes which were reported to the police. The interviewer had to remind them of other possibilities.

-Many respondents tried to answer on other kinds of crimes which were not in the questionnaire. Hence we propose that it would be useful to introduce an “other crimes” category.

-Concerning the section Q of the questionnaire, the own completionwas not successful, but if the interviewer helped, the respondentsaccepted the co-operation.

1. Comparison of the results with the data from other crime surveys conducted in Hungary

In Hungary, the last crime survey at national level was carried out by the National Institute of Criminology (NIC)in July 2003. We do not have any database from this survey therefore we cannot compare data. But it is intended to do this work in the future and to publish a study about the main results. This kind of comparison is difficult, because the NIC and the Eurostat surveys were based on different sample designs,the NIC survey was representativeon national level but the Eurostat’s survey was conductedonly in five counties and Budapest. The main issues were different in the two surveys and the questions were different in the most instances as well (wording, different aspects of issues, level of detailing).

However, there are somecommon or very similar issues and concrete questions which can be the basis of a future comparison. These questions are as follows:

1. - Eurostat survey: (B2): “Howoften do you usually walk alone in this area after dark?”

- NEC survey: “How often do you go out after dark?”

1. - Eurostat survey: (B3) “How safe do you feel walking alone in the area where you live after dark?”

- NEC survey: “How safe do you feel or would you feel when walking alone near your home after dark?”

1. - Eurostat survey: (B6): “How safe do you feel in your home at night?”

- NEC survey: “How often do you fear of criminals in your home at night alone?”

1. - Eurostat survey: (D26): “In some countries, agencies have been set up to help victims of crime by giving information, practical or emotional support. Did you or anyone else in your household know about these services?”

- NEC survey: “Do you know about agencies, foundations which have been set up to help victims of crimes?”If yes, which one?

1. - Eurostat survey: (F4): “Do you or anyone else in your household own a handgun, shotgun or rifle?”

- NEC survey: “Would you like to own a handgun, shotgun or rifle for self protection?”

1. - Eurostat survey: (F1): “How often do you think of the possibility of becoming a victim of crime and how to avoid it?”

- NEC survey: “How often do you think of the possibility of becoming a victim of crime?”

1. - Eurostat survey: (F2): “Is your home protected by…”

- NEC survey: “Do you protect your home against burglary by anything?”

1. - Eurostat survey (F3): “To improve your personal safety…”

- NEC Survey: “Do you or did you do anything to improve your personal safety?”

1. In NEC survey the same 10 main crime types wereset upas in the Eurostat survey (theft of car, theft from car, car damage, theft of motor cycle-scooter-moped, bycicle theft, home burglary, other burglaries, property damage, robbery, theft). The “How many times did this happen?” and the “Did the police come to know about the incident?” questions are founded in both surveys. The E block questions (about consumer fraud) were in a less detailed issue in NEC survey. Sexual attacks, abuses werealso part of the NEC survey.

1. - Eurostat survey: (for example D25): “Overall, how satisfied or dissatisfied were you with the way the police handled this matter?”

- NEC survey: “How satisfied or dissatisfied are you with the work of the police?”

1. - Eurostat survey: (for example D24): “Did the police find out who did it?”

- NEC survey: “Had it been found out who the perpetratorwas?”

1. - Eurostat survey: (for example D33): “As a result of what happened have you or your household taken any action to try to prevent it happening again?”

- NEC survey: “Have you taken any action to prevent it to happen again?”

1. Comparison of the results with data from national administrative sources

In most cases the data from victimization survey and of administrative sources are not really comparable. There are several reasons for it:

-There are some differences between definitions of crimes indifferent approaches: first,the crimes which are considered as crime and questioned in thevictimization survey; second, crimes which are crimes in the respondent’s opinion; and the third, crimes which are crimes according to the penal code. The official administrative crime statistics is based on legal background. These three approachesmay bedifferent from each other and the comparability is very difficult.

-The Hungarian crimes statistics contain some information about victims of crimes. If we had had a national sample, not only data of five counties and Budapest, comparison could have been possible..

-We have to take into account the “dark number” in the case of official crime statistics, and – in our point of view – in the case of victimization surveys, too.

-One of the main difficulties in comparison of the administrative and survey data is that there is only one victim in administrative sources and because of the questions like “you or anyone else in you household” there may be more victims in the victimization surveys.

-In theory, it is possible to compare by crime types, but for instance in case of thefts we have to take into account the amount of loss. It is important because some not too serious thefts are considered by law not crimes but offences.

1. The issue of the reference period

We have already mentioned this problem in the second Hungarian report. The different intervals applied in the original questionnaire were specified in the Hungarian version.The 2006, 2007, 2008 years remained. The expression “last five years” was replaced by the one “since 2004” and the “last year” expression by the “in 2008”. As the field work period was May 2009 inHungary – we cannot get information about crimes which happened in the first months of 2009 (i.e. on the most current crimes). In our opinion, this problem would be solved if there was an identified reference period and aunified period of field work in each country (e.g. in the first three months of a given year).

Regarding analytical respects, it would be practical to apply the calendar year instead of “last year” categories. Many people usually remember what happened in a given year instead of “in the last 12 months” or “in the last five years”.

1. Breakdown of responses for each question

See Annex.

1. Estimate of the cost of implementing such a module on national level

The budget calculated for our application is sufficient to carry out a victimization survey in a 1000-persons-sample. Certainly, for a larger samplethis budget would not be enough. We would like to use the CAPI and CATI methods but we do not have any technical background to use them. We will take into consideration some “gifts” to the respondents of a next survey.

1. Conclusions and recommendations for further actions

In case of victims of crimeswhich were not listed in this questionnaire, in our point of view, it would be fruitful to use the “other crime” category to get more detailed information on the crime phenomena.

The problems connected with the repetitive questions and answers should be solved.

It would be important to define each crime type more detailed, because of the different details of the penal code and the public opinion.

1