Implementing cawi into the data collection process Kees

Implementing cawi into the data collection process Kees van Berkel Mariëtte Vosmer Jerusalem, 21 -24 July 2013

Why • • Meet customers’ demands; Meet respondents’ expectations; Budget cuts. It’s all about the money; Increase response rates, reach more groups. Be flexible Implementing cawi into data collection process 1

History • 2005: start web data collection research; • 2005 – 2007: experiments ü ICT-survey ü Crime survey ü Survey measuring Underground Economy; • 2008 – now: surveys ü Dutch Housing and Living survey ü Crime survey ü National Mobility survey ü Health Survey ü Labour Force Survey; Implementing cawi into data collection process 2

Sampling frame Municipal basic registration of population data. This registration contains information of all people who are registered at a municipality. The municipalities own the data, but the Ministry of the Interior and Kingdom Relations is responsible for the regulation. Statistics Netherlands is allowed to use it, under strict conditions. Implementing cawi into data collection process 3

Sampling design First stage: stratified systematic pps sample of municipalities. Second stage: random sample of addresses. Stratification: the country is split into 40 corop areas. The sample is self-weighting: Every address has the same inclusion probability. Implementing cawi into data collection process 4

Sampling design In each stratum h : the number of addresses to be drawn nh is proportional to the number of addresses Nh , m : desired number of addresses to be selected in each selected municipality cluster size. gh : number of municipalities to be selected in stratum h. Implementing cawi into data collection process 5

Sampling design Small m many municipalities large travelling expences Large m few municipalities cluster effect large standard errors Face to face: cluster size m = 12 Mixed-mode: cluster size m = 1 Implementing cawi into data collection process 6

Fixed amounts Capi • Cawi respons Cawi non-respons Cati Increase of variance Implementing cawi into data collection process 7

Weighting procedure Mode included as weighting term Increase of inclusion weights due to partially approaching cawi non respondents by telephone or face to face Implementing cawi into data collection process 8

Approach strategy Approach letter + Reminder 1 + Reminder 2 + Telephone Face to face Implementing cawi into data collection process 9

LFS adaptations 1) 2) > 2 or > 3 in the age of 15 years or more to capi? % unknown sample unit / telephone number closed 18. 0 16. 0 14. 0 Quality of phone numbers 12. 0 10. 0 % 8. 0 Mobile 6. 0 Landline 4. 0 2. 0 0. 0 Sept Okt Nov Month Dec Implementing cawi into data collection process Jan 10

Distribution of responses in time Implementing cawi into data collection process 11

Response rates Implementing cawi into data collection process 12

Response results LFS (1) Implementing cawi into data collection process 13

Response results LFS (3) Implementing cawi into data collection process 14