Development of Electronic Data Reporting (EDR) in Statistics Finland
Post on 29-Jan-2016
32 Views
Preview:
DESCRIPTION
Transcript
Development of Electronic Data Reporting (EDR) in Statistics Finland
13.9.2005 A 2Rami Peltola
Design goals
Framework for similar systems Multi-language support User authentication / authorisation Data security (collection and production databases) Simple method for transferring data between collection and
production databases “Mass emailer” for all kinds of collection systems
13.9.2005 A 3Rami Peltola
Three generations
Building cost index 2001sub-annual surveybuilt using Microsoft Windows DNA (Distributed iNternet
Application Architecture) 6 EDR solutions 2002-2004
both sub-annual and annual surveysVB.NET
4 EDR solutions 2005sub-annual surveysXCola
Data Provider relations and response rates: Business surveys
13.9.2005 A 5Rami Peltola
Background
Traditionally high response rates in annual and sub-annual business surveys
up to over 99%persistent staff
Good relations with data providersexperienced staffmany continuous personal contactsusually long-lasting relationshipshigh level of trust
13.9.2005 A 6Rami Peltola
Infrastructure and atmosphere for EDR in Finland
High level of using the Internetalmost every enterprise has access to internetbusiness surveys are typically made for the largest
enterprises Positive atmosphere for using internet for transactions with
the governmentrespondents are even enthusiastic about using Internet“It’s fun to fill in web forms instead of paper ones!”
13.9.2005 A 7Rami Peltola
Key issues when implementing EDR -solution
Simplifying data collection process Reducing need for human resources Reducing other data collection costs Improving the quality of collected data Decreasing non-response Speeding up the data accumulation Reducing response burden Enabling direct individual feedback for respondents Enabling browsing of previously submitted data Assuring high level data security
Timeliness
Cost-efficiency
Accuracy
Data providerrelations
13.9.2005 A 8Rami Peltola
Reducing response burden
Using e-mail for informing the survey and sending the reminderslink to the EDR -solution
Questionnaire is “always” available Good designing of the questionnaire
not just a copy of the paper form Helpful validity checks
reduce the need of additional inquiries Contextual on-line help Option to fill in the questionnaire in separate sessions Support for several languages
13.9.2005 A 9Rami Peltola
Enabling direct individual feedback for respondents (1)
Motivates respondent to use EDR -solution instead a paper form
For example enterprise’s own data compared with the data of it’s own industry
Respondents have found feedback usefulhas even led to inquiries for more information
13.9.2005 A 10Rami Peltola
Enabling direct individual feedback for respondents (2)
Monthly saleinquiry:
13.9.2005 A 11Rami Peltola
Enabling browsing of previously submitted data
Respondent have access to all the data it has previously submitted
simple short-term (sub-annual) surveys Pre-filling the form with data from previous surveys
useful especially in annual surveyshelps the respondent to remember how the figures were
compiled last year
13.9.2005 A 12Rami Peltola
Assuring high level data security
Data security audit by an outside consult All traffic on the Internet is SSL -encrypted An authentication / authorisation -process is always needed New user IDs and passwords every year User IDs and passwords are initially sent in a letter
only one of them can be sent by emailthe other one must always be sent in a letter or given over
by telephone Only a certain number of our staff have access to user IDs
and passwords (usually two persons per survey)
13.9.2005 A 13Rami Peltola
Decreasing non-response and speeding up the accumulation of the data
Response rates have remained on high level The average response time of monthly surveys has been
reducedin the best case by 7-8 days or 30%
The number of reminders sent has decreased substantiallyin the best case by 50%
The share of the respondents using EDR -solution has in most cases reached high level
sub-annual surveys > 60% (in the best case 85%)annual surveys ~ 30% (in the best case 50%)
13.9.2005 A 14Rami Peltola
An example: Sale inquiry Change in response media 10/2001 - 10/2004
0
200
400
600
800
1000
1200
1400
1600
Fax Mail Internet
10/2001
10/2002
10/2003
10/2004
13.9.2005 A 15Rami Peltola
An example: Sale inquiry
EDR users of all respondents:
after 1. month: 48%
after 2. month: 59%
after 3. month: 61%
after 4. month: 70%
Today: > 80%
13.9.2005 A 16Rami Peltola
An example: Sale inquiry
Reminders sent:
before EDR: ~ 1000
after 1. month: ~ 800
after 2. month: ~ 700
after 3. month: ~ 600
since 4. month: ~ 500
13.9.2005 A 17Rami Peltola
An example: Sale inquiry Responses per day 10/2001 (until t+45 days)
0
50
100
150
200
250
300
Due date(accumulation 31%)
Reminders(accumulation 76%)
13.9.2005 A 18Rami Peltola
An example: Sale inquiry Responses per day 10/2002 (until t+45 days)
0
50
100
150
200
250
300
350
400
450
Due date(accumulation 49%)
Reminders(accumulation 81%)
13.9.2005 A 19Rami Peltola
An example: Sale inquiry Responses per day 10/2003 (until t+45 days)
0
20
40
60
80
100
120
140
160
180
200
Due date(accumulation 68%)
Reminders(accumulation 78%)
13.9.2005 A 20Rami Peltola
An example: Sale inquiry Responses per day 10/2004 (until t+45 days)
0
50
100
150
200
250
Due date(accumulation 76%)
Reminders(accumulation 86%)
13.9.2005 A 21Rami Peltola
An example: Sale inquiry Accumulation of data 10/2001 - 10/2004
0
500
1000
1500
2000
10/2001 10/2002 10/2003 10/2004
Costs and Benefits of EDR solutions: Business surveys
13.9.2005 A 23Rami Peltola
Costs of EDR
The costs of developing web-based applications and running them has dropped by 60-70% during the last three years
Average investment cost per new EDR -solution (today)An outside service provider: EUR 5000In-house solution (XCola): 150 hours of work
Maintenance costs of EDR solution per year (today)An outside service provider: EUR 1000In-house solution (XCola): 50 hours of work
13.9.2005 A 24Rami Peltola
Development costs: In-house EDR solutions
The in-house solutions are already in the third generation phase During the first and second phases the total resource input was
about 2.5 person yearsmore or less “learning by doing”includes the development of a secure communication
environmentincludes the implementation of 7 solutions
XCola (third generation phase) development took about 1 person year
includes the implementation of 4 solutions
13.9.2005 A 25Rami Peltola
Benefits of EDR to Statistics Finland (1)
Four second generation solutions have been in production for more than year
3300 respondents per month and 800 per quarter Average percentage of work saved in the data collection phase is
over 40 (2 person years) The amount of ground mail has been reduced by 64 000 or 65%
(0.5 person years) The average response time has been reduced in the best case by
7-8 days or 30% Number of reminders sent has gone down by the half Investment has paid off in about a year
13.9.2005 A 26Rami Peltola
Benefits of EDR to Statistics Finland (2)
The data received are of better qualityboth annual and sub-annual surveyscommon estimate is “25% less errors”comprehensive study has not been made
As manual handling diminishes, it can be replaced by more rewarding tasks
13.9.2005 A 27Rami Peltola
Benefits of EDR to respondents
Questionnaire is possible to be completed more rapidly Pre-filling helps respondents to remember how they have
answered previously Validity checks prevent the sending of erroneous data
no additional inquiries Same piece of information needs always to be entered only
once Respondents like to use Internet Perceived response burden has gone down
13.9.2005 A 28Rami Peltola
Effects of the EDR on data collection process
Printing the questionnaires Transferring data to collection database
Mailing E-mail informing (mass emailer)
Receiving the questionnaires (mail, fax, e-mail) (Electronic data supply)
Validating and entering the data On-line validations + mass validation
Printing and mailing the reminders E-mail reminder (mass emailer)
Phone inquiry Phone inquiry
Non-individual delayed feedback Individual direct feedback
Limited access to previous own data Previous own data available
Manual exclusive treatment Electronic mass treatment
13.9.2005 A 29Rami Peltola
Data transfers
Data transfers between collection and production databases are handled with an external application
Data from collection database is first transferred to the temporary tables in the
production databaseand then synchronized with the actual tables
Solution is quite customizableEasy to customize for new collection systemsEasy to add new databases
13.9.2005 A 30Rami Peltola
Mass emailer
An external application to send emails to the respondents Modular approach
New systems can be added using textual configuration files
Reply requests (list of e-mails) can be added by writing sql statements to the configuration files
Supports attachments Replaces traditional letters
13.9.2005 A 31Rami Peltola
An example: Sale inquiryBackground Monthly inquiry using paper forms (~2080 enterprises) Data collection process (7-8 persons: ~2.0 working years)
Printing and mailing the questionnairesReceiving the questionnaires (mail, fax, e-mail)Validating and entering the dataPrinting and mailing the remindersPhone inquiry
Quarterly non-individual feedback for respondents Previously submitted data preprinted to questionnaires for
the last 2 months
13.9.2005 A 32Rami Peltola
An example: Sale inquiry Purpose
To describe the economic situation in different industriesTrade: 30 branchesServices: 23 branchesConstruction: 3 branchesManufacturing: 15 branches
To respond to EU-legislationThe Regulation of the EU concerning short-term
statistics (1165/98)
13.9.2005 A 33Rami Peltola
An example: Sale inquiry / Source data Tax Administration’s register
Value added tax -dataDelay 2 months in turnoverThe data accumulates for 6 months
Direct data collection (sale inquiry ~2080 companies)The most largest companies from each industry
Trade: ~660 companiesServices: ~600 companiesConstruction: ~200 companiesManufacturing: ~620 companies
Business RegisterBasic information about companies: branch of industry, location, contact
information etc.
13.9.2005 A 34Rami Peltola
An example: Sale inquiryEnd products
Monthly indicators (newsletters and Internet releases)Trade: 30 (European sample), 45 (preliminary) and
75 days delayServices: 45 (European sample) and 75 days delayConstruction: 75 days delayManufacturing: 75 days delay
Time seriesStarts mostly from 1995 (trade from 1985)
13.9.2005 A 35Rami Peltola
Data collection process in sale inquiry before EDRCollecting the sales data concerning March
April
Due date
Printing and mailing the reminders
Phone inquiry for non-respondents
Printing and mailing the questionnaires
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 1 2 3 4
Manual data entering
May
13.9.2005 A 36Rami Peltola
Data collection process in sale inquiry todayCollecting the sales data concerning March
April
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Transferring data to collection database
E-mail informing
Due date
E-mail reminders
1. phone inquiry for non-respondents on European sample
2. phone inquiry for non-respondents on European sample
Phone inquiry for the rest non-respondents
Electronic data transfers
13.9.2005 A 37Rami Peltola
EDR solution / Sale inquiry
13.9.2005 A 38Rami Peltola
Change in working hours used in data collection and validation (hours per year)
0
500
1000
1500
2000
2500
2001
2002
2003
2004
13.9.2005 A 39Rami Peltola
Hours used in development and maintenance of EDR solution (sale inquiry)
0
100
200
300
400
500
600
700
800
900
Hours
2002
2003
2004
Includes hours used in development of
infrastrucre
13.9.2005 A 40Rami Peltola
Experiences (1)
Feedback from respondents has been very positive:
Response burden has redused remarkably
Enthusiasm of persons involved in data collection
Manual data treatment has redused (at least by 50%)
Quality of data has improved: On-line validations, additional
information if data is not comparable etc.
13.9.2005 A 41Rami Peltola
Experiences (2)
Number of enquires made by respondents concerning EDR
solution:
First two months: ~100 / month (mainly questions
concerning base settings)
Since third month: ~30 / month (mainly forgotten
passwords)
13.9.2005 A 42Rami Peltola
Effects of EDR on data quality (ESS quality dimensions)
AccuracyAutomatic validations
TimelinessMail E -mail and webManual data entering Electronic data transfers
Comparability
top related