Conversely Mixed Mode in the Swedish Crime Survey Sanna Wallin, researcher The Swedish National Council for Crime Prevention Thank you! It’s very nice.

Slides:



Advertisements
Similar presentations
Sampling Frames for Establishment Surveys International Workshop on Industrial Statistics Beijing, China, 8-10 July 2013.
Advertisements

STATISTICS FOR MANAGERS LECTURE 2: SURVEY DESIGN.
1 CDBG Income Survey Requirements For Grant Administrators.
Customer : contractor : December, 2012 Sociologic Research on Awareness of Industrial Property Protection Possibilities.
Quality in the Swedish Business Database The Quality Survey 2004 Round Table Beijing 2004 Swedish presentation, session 5, 18 th Round Table, Beijing –
Adding a Mode as a Final Step in the Follow-up of a Panel Survey Seminar on New Frontiers for Statistical Data Collection Geneva, Switzerland, 31 October.
The Challenge of Non- Response in Surveys. The Overall Response Rate The number of complete interviews divided by the number of eligible units in the.
Introduction to NCHS Rob Weinzimer, Special Assistant for Outreach Centers for Disease Control and Prevention National Center for Health Statistics.
Data to be released after chamber annual meeting and Methodology Statement.
Data to be released after chamber annual meeting and Methodology Statement.
Methodology Sample of 402 Interviews conducted between August 19-30, 2015 Screening qualifications included West Virginia residency, current voter registration,
National Center for Health Statistics (NCHS) Centers for Disease Control and Prevention.
Implementing cawi into the data collection process Kees van Berkel Mariëtte Vosmer Jerusalem, July 2013.
Miami, Florida, 9 – 12 November 2016
WORKSOURCE INTEGRATED FRONT-END SERVICES
Casey Eggleston, Jennifer Hunter Childs, & Gerson Morales
Rachel Vis-Visschers & Vivian Meertens QDET2, 11 November 2016
COMPLEMENTARY TEACHING MATERIALS
City of Huber Heights Public Opinion Survey
Session 3: Innovative solutions
Context for the experiment?
Part III – Gathering Data
Anna Długosz Central Statistical Office of Poland
SNAP – Retirement Planning Fact Finder
Planning efficient recruitment
Session: 12 Integrating civil registration, vital statistics, population registers and identity management, 15 November 2017 Workshop on the Operation.
How to get students from Europe to Asia?
Implementing the Istanbul Convention
Striving to be a Vibrant Club
Assessing Community Opinions, Strengths, & Needs
Cell phone problem and alternative approaches
MINI-SPEC REFERENDUM April 2010 Report
The second wave of the new design of the Dutch EU-SILC: Possibilities and challenges Judit Arends.
The importance of administrative data in the era of SDGs
Survey Design Steps in Conducting a survey
Chapter 2: The nonresponse problem
The European Statistical Training Programme (ESTP)
Case study: The effect of text message reminder on survey nonresponse
13th Workshop on Labour Force Survey Methodology
Surveys of Consumers: Mixed Mode Experiments
Statistics on crime and criminal justice, Task Force meeting on victimisation, 4–5 June 2009, Luxemburg Review of the current situation in respect of the.
The European Statistical Training Programme (ESTP)
Background to the development of a European Victimisation Survey
Modernising the Swedish National Travel Survey to be launched in RVU Sverige Luxembourg
Outbound Tourism Statistics in Turkey
National needs for AES Purpose - describe participation in learning during a 12 months period. The main parameters are; Participation rates in different.
1 Panel 2, Position 3 Jack D. Ripper.
Activities of the UNECE-UNODC Task Force on Victimization Surveys
UNODC-UNECE Manual on Victimization Surveys: Content
Response to Instruction/Intervention (RtI) for Parents and Community
PUBLIC OPINION POLL ON DRAFT CONSTITUTION
Response to Instruction/Intervention (RtI) for Parents and Community
Sampling Designs and Sampling Procedures
Session 4 – From pilot to regular surveys: the costs aspects Introduction: cost impact of possible design strategies Malta, January 2008 Transition.
Metro ACEs Data 2018 Community Health Needs Assessment
Project on translating and testing a victimisation survey module
Measuring Audience and Impact
Vice President, Health Care Coverage and Access
Plans for the 2021 Population and Housing Census
How the Affordable Care Act Has Improved Americans’ Ability to Buy Health Insurance on Their Own Findings from the Commonwealth Fund Biennial Health Insurance.
P3.
Changes in the Canadian Census of Population Program
MSP Regional Travel Behavior Inventory Program
International Crime Victimisation Survey (ICVS_2)
Chapter 2: The nonresponse problem
Chapter 5: The analysis of nonresponse
New Mexico Census outreach message/messenger testing survey
Statistics is... a collection of techniques for planning experiments, obtaining data, and then organizing, summarizing, presenting, analyzing, interpreting,
“THIS WILL TAKE JUST A MINUTE OF YOUR TIME”
New Mexico Census outreach message/messenger testing survey
Presentation transcript:

Conversely Mixed Mode in the Swedish Crime Survey Sanna Wallin, researcher The Swedish National Council for Crime Prevention Thank you! It’s very nice to be here. My name is Sanna Wallin and I’m here to talk about Conversely Mixed Mode in the Swedish Crime Survey. I’m a researcher at the Swedish National Council for Crime Prevention which is a public agency under the Ministry of Justice and a knowledge centre for the justice system. The council contribute to knowledge in the area of criminal policy and promote crime prevention efforts. We do this by working up and producing official crime statistics and other statistics on instruction from the Government. conducting applied research as instructed by the Government, in cooperation with justice system agencies, and on our own initiative.

The Swedish Crime Survey Annual survey by telephone interviews confidence in the justice system exposure to crime (victimization) fear of crime victims’ contacts with the justice system Conducted since 2006 A stratified random sample of 20,000 Swedish residents aged 16–79 First of all, what is the Swedish Crime Survey? It’s a annual survey of the general public in Sweden about confidence in the justice system, exposure to and fear of crime and also victims’ contacts with the justice system. The survey has been conducted since 2006 by a stratified random sample of 20 000 Swedish residents aged 16-79.

The Swedish population register Everyone who lives in Sweden is registered The population register contains information about : Name Spouse, children, parents, legal guardians and adoption Address Personal identity number Moves to and from Sweden Place of birth Addresses abroad Citizenship Death and burial site How do we get information about the residents in Sweden? The Swedish Tax Agency are responsible of the population registration in Sweden. In the process, details of those who are living in Sweden, and where, are recorded. Everyone who is registered is issued with a unique personal identity number (almost like social security number), which is used in contacts with government agencies and others. The survey population consists of everyone in the register aged 16-79 . In the Swedish Crime Survey we use name, address and personal identity number to be able to link data in other registers (like education, income and more).

The Swedish Crime Survey Information about the survey to residents home address Search for telephone numbers Screening and follow-up by telephone interviews Postal questionnaire to non-respondents Response rate lowered from 78 % in 2006 to 60 % in 2016 Everyone gets a letter with an invitation to and information about the survey. The interviewers search registers for phone numbers. The telephone interviews, conducted from January through April, starts with screening questions and for those victimized there are follow-up questions as well. Those who don’t answer by telephone (due to no contact) are sent a short postal questionnaire with screening questions. In the latest survey, conducted earlier this year, the response rate was 60 percent and 7 % participated through postal or web questionnaires. In the first survey, back in 2006, the response rate was 78 percent.

Non-response Here’s a graph illustrating the non-response from 2006 to 2016. As you can se, the overall non-response has gone from 22 % in 2006 to 39 % in 2016. The inability to answer is at the same level in 2016 as in 2006, about 3 %. There is an increase in the share of refusal to answer, from almost 12 to almost 16 percent. But the biggest difference in non-response is in the category no contact, from 8 to almost 21 percent. We think that this probably is due to new ways of communicating, fewer people have fixed telephony (landline) at home. In Sweden, only 51 % use fixed telephony, meanwhile 97 % use mobile phones and 94 % have internet at home. Also there is an reluctance to answer when number is not recognized (the caller is unknown). The interviewers spend more time searching for phone numbers (very expensive) than actual interviewing. It’s harder to find number for mobile phones. Compared to other surveys in Sweden who use postal questionnaires as single data collection method, the response rate is lower but almost constant over time.

Need for methodology change Fast increase in non-response rate Increasing costs Smaller geographical areas Both due the increasing non-response and the increasing costs there’s a need for methodology change to maintain high data quality in the Swedish Crime Survey. Also, there is a whish for results on smaller geographical areas than today which demands a larger sample.

New method – conversely mixed mode Screening by postal- or web questionnaire Follow-up of sample of victimized by telephone interviews Experiment with mixed-mode So, the new method consists of screening with web/postal questionnaire. To change method from telephone interviews to web/postal questionnaire you need to adapt the questions. A sample of victimized respondents are followed-up by telephone interviews. We are now working on a design for that. Last fall, Statistics Sweden (which is a government agency that produces statistics) made an experiment with mixed-mode within Swedish Crime Survey. Though the questions weren’t adapted to the new method the experiment gave the same estimated percentage of victims as the Swedish Crime Survey.

Split-half Split-half (screening) in 2017 parallel to Swedish Crime Survey Larger sample 200 000 residents in 95 local police districts Effects of method change The mixed-mode makes it possible (due to lower costs for data collection) to increase the sample size, from 20 000 to 200 000, in order to show results for every local police districts in Sweden. Overall, the new method makes the Swedish Crime Survey more cost-efficient and more flexible. The follow-up interviews can be conducted on a specific sample of the respondents (for example the youngest) or specific questions can be included (for example partner violence). The screening-part of the new method will be tested in a split-half in 2017 parallel to the Swedish Crime Survey. And the effects of the changes in method, the questions and the sample will be evaluated next year.

More… www.bra.se sanna.wallin@bra.se search for “Swedish Crime Survey” Do you want to know more about the Swedish Crime Survey, or about the Swedish National Council for Crime Prevention, visit our website. Please do not hesitate to email me if you have questions about the presentation. THANK YOU for your attention!!