Download
1 / 26
260 likes | 426 Views
Enumeration via internet – Estonian experience. Diana Beltadze , Ene-Margit Tiit Statistics Estonia UN Office in Geneva Meeting on Population and Housing Census 30.Sept – 4.Okt 2013. Content. Preparation period. Choice of methodology Using registers E-enumeration and CAPI
E N D
Enumeration via internet – Estonian experience Diana Beltadze, Ene-Margit Tiit Statistics Estonia UN Office in Geneva Meeting on Population and Housing Census 30.Sept – 4.Okt 2013
Content • Preparation period. Choice of methodology • Using registers • E-enumeration and CAPI • Electronic map • Census process. E-enumeration • Census moment and census time • E-enumeration process • Results of census • Tested environment for complex controlling surveys • High quality of results • Lessons of census • E-activities are successful in Estonia
Preparation period Statistics Estonia started the preparations for the 2010 round of censuses as early as in 2006. The first question was whether it was possible to organise the census without fieldwork, relying solely on the existing national registers. Estonian state registers were good but they did not cover the whole scale of information needed from census output: there was no information on occupations, educational level of elderly people etc. The Census Committee of the Government of the Republic approved a combined methodology as the method to be used in 2010 census round.
Combined methodology. Using registers Several registers were used on different stages of census. • Current learning was not asked from recipients, but information about it was taken from register; • Several data (citizenship, official family status etc) were taken from registers and prefilled, respondents had to check them. • Registers were used also • For preparing working-lists, • For filling random blanks in questionnaires • For assessing coverage
Combined methodology.Data-collection forms: CAWI and CAPI • Two new technical solutions were introduced: • self-completion of questionnaires on the internet • entry of answers directly in laptop computers during census interviews. • All data were automatically sent to census data-base, so that no data-entry was needed. • Data were on-line checked using a series of logical checks. • Additional possibilities for data-collection (in case of emergency or in unapproachable places as small islands) were paper questionnaires and telephone interviews (both used in less than in 1% cases).
Markingtheresidence’sspacecoordinates on electronicmap Peopleenumeratingthemselvesby internet hadtofindtheirhousing on e-map and checkit. Thesamewasdonebyenumeratorsduringtheperiodoffieldwork.
Web applications: E-respondent map application Statistics Estonia
Census moment and census time Census moment was the very last day allowed by Eurostat for member states: the 31st of December 2011 at 00:00 Time of the Census was divided into the following stages: e-Census – 31.12.11 – 1.02.12 data arrangement – 2.02 –19.02.12 Census interview – 20.02.12 – 31.03.12 The data arrangement started already from the beginning of e-census. It was necessary to avoid the cases that questioner enters the housing where all people have enumerated themselves by internet.
E-enumeration How many people will enumerate themselves by internet? The prognoses varied from 5% to 80%. After pilot census (31.12.2009—31.03.2010) quite conservative forecast by census team was – the share of e-enumeration will be 25%. In real life the result was much better: 67%. The probable reasons of the success: The long and intense campaign (why the census is necessary and useful for Estonia?) that ensured positive reception of census in society. Estonian peoples’ experience using internet (banking, voting etc) Transparency of census process: the results of census (by counties) were available on census homepage and were updated each hour. The idea to win the world record (to be better than Canada, whose e-enumeration score was about 55%) was spoken out by a popular writer and it started in society gambling for getting higher result.
Intensity of e-enumeration, No of persons enumerated themselves per day
Shareofe-enumerationbycounties sitlus või esitleja nimi
Almost uniform distribution of e-enumeration It was remarkable, that the difference between the share of e-enumeration of most active (Tartu) and most passive (Ida-Viru) counties was only about 60% – that means, in each village and town of Estonia there was a lot of people who were able to use internet for enumerating themselves.
Thefirstbonus – environment For census needs was created general-purpose systematic software (environment) that will be used in future for surveys, processing, analysing and combining data from different sources. The environment was during census thoroughly tested It can be used also for the forthcoming register-based censuses
Environment for monitoring the census process as whole The environment (software-system) created for census allowed: To create the questionnaires, implementing the help-texts and logical checks; To create the optimal interviewing areas (on map) and corresponding work-lists of addresses of dwellings; To gather the data (via internet or laptops) and create the data-base (of raw data); During the field-work stage tomonitor the work of each enumerator, check their success or lag; To define the roles of people working with data; to follow the security rules and restrictions to protect the data; To calculate the necessary output-variables using the census data.
Quality of census results Questionnaires of Estonian census were rather long: besides obligatory questions also ethnic nationality, mother tongue, command of foreign languages and local dialects, birthplace of parents and grandparents, number of children borne, also some questions about health, sources of subsistence etc. were asked. Estimated coverage rate was for persons about 98%, for housings – 99%, average for items (questions) about 99%. In general, quality of e-census was not worse than that of interviews. Only about 1% of questionnaires filled in by internet were improved by face-to-face interview.
Cost • Cost of Estonian census was considerably high – 10,4 € per capita. • One reason for this is that the population size is only 1,3 millions and the population density is low. • But: • Due to high e-census rate about 20% of planned resources were saved; • The environment useable for future surveys and data- analysis was prepared and tested; • The preparations for future register-based censuses were made.
We convinced that Estonian people are ready to use internet for different procedures, also when it is time-consuming and not very easy. To get good results in large-scale nationwide actions the long-lasting campaign and transparency of actions are necessary. The big problem of contemporary censuses is the mobility of people. In Estonia up to 20% of people do not live in their registered residence. Also the problem is the identity of addresses – it was greatly solved in Estonia by new address-standard. You cannot believe all technical solutions for 100% (e.g. existing internet-connection). It is reasonable to have plan B
