Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Internet-Based Collection of Statistical Data by 242e0MF

VIEWS: 0 PAGES: 42

									  Development of Electronic Data
Reporting (EDR) in Statistics Finland
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




                                  Rami Peltola     13.9.2005   A2
Three generations

 Building cost index 2001
    sub-annual survey

    built using Microsoft Windows DNA (Distributed iNternet

     Application Architecture)
 6 EDR solutions 2002-2004

    both sub-annual and annual surveys

    VB.NET

 4 EDR solutions 2005

    sub-annual surveys

    XCola

                                  Rami Peltola     13.9.2005   A3
Data Provider relations and response
      rates: Business surveys
Background

 Traditionally high response rates in annual and sub-annual
  business surveys
    up to over 99%

    persistent staff

 Good relations with data providers

    experienced staff

    many continuous personal contacts

    usually long-lasting relationships

    high level of trust



                                  Rami Peltola     13.9.2005   A5
Infrastructure and atmosphere for EDR in Finland

 High level of using the Internet
     almost every enterprise has access to internet

     business surveys are typically made for the largest

      enterprises
 Positive atmosphere for using internet for transactions with
  the government
     respondents are even enthusiastic about using Internet

     “It’s fun to fill in web forms instead of paper ones!”




                                   Rami Peltola      13.9.2005   A6
Key issues when implementing EDR -solution

 Simplifying data collection process
 Reducing need for human resources            Cost-efficiency

 Reducing other data collection costs

 Improving the quality of collected data
                                               Accuracy
 Decreasing non-response

 Speeding up the data accumulation            Timeliness

 Reducing response burden

 Enabling direct individual feedback for respondents             Data provider
                                                                    relations
 Enabling browsing of previously submitted data

 Assuring high level data security


                                          Rami Peltola           13.9.2005        A7
Reducing response burden
 Using e-mail for informing the survey and sending the
  reminders
     link 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
                                    Rami Peltola      13.9.2005   A8
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 useful

     has even led to inquiries for more information




                                 Rami Peltola     13.9.2005   A9
Enabling direct individual feedback for respondents (2)

   Monthly sale
    inquiry:




                                Rami Peltola   13.9.2005   A 10
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 surveys

    helps the respondent to remember how the figures were

     compiled last year




                                 Rami Peltola    13.9.2005   A 11
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 email

    the 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)
                                  Rami Peltola     13.9.2005    A 12
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

  reduced
     in the best case by 7-8 days or 30%

 The number of reminders sent has decreased substantially

     in 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%)

                                 Rami Peltola    13.9.2005   A 13
An example: Sale inquiry
Change in response media 10/2001 - 10/2004
    1600
    1400
    1200
    1000                                  10/2001
     800                                  10/2002
     600                                  10/2003
                                          10/2004
     400
     200
       0
           Fax     Mail   Internet

                           Rami Peltola         13.9.2005   A 14
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%


                                    Rami Peltola   13.9.2005   A 15
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




                                 Rami Peltola   13.9.2005   A 16
An example: Sale inquiry
Responses per day 10/2001 (until t+45 days)
    300

    250                                                     Due date
                                                      (accumulation 31%)

    200
                              Reminders
                           (accumulation 76%)
    150

    100

     50

      0

                             Rami Peltola       13.9.2005              A 17
An example: Sale inquiry
Responses per day 10/2002 (until t+45 days)
    450
    400                                                     Due date
                                                      (accumulation 49%)
    350
    300                       Reminders
                           (accumulation 81%)
    250
    200
    150
    100
     50
      0

                             Rami Peltola       13.9.2005              A 18
An example: Sale inquiry
Responses per day 10/2003 (until t+45 days)
    200
    180
    160
    140
                                                                       Due date
    120                                                        (accumulation 68%)

    100
     80                                    Reminders
                                      (accumulation 78%)
     60
     40
     20
      0

                            Rami Peltola                   13.9.2005              A 19
An example: Sale inquiry
Responses per day 10/2004 (until t+45 days)
    250


    200
                                                                Due date
                                                        (accumulation 76%)

    150


    100                            Reminders
                               (accumulation 86%)


     50


      0

                            Rami Peltola            13.9.2005              A 20
An example: Sale inquiry
Accumulation of data 10/2001 - 10/2004
     2000


     1500


     1000


     500


       0

            10/2001   10/2002   10/2003        10/2004

                                Rami Peltola             13.9.2005   A 21
Costs and Benefits of EDR solutions:
         Business surveys
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 5000

     In-house solution (XCola): 150 hours of work

 Maintenance costs of EDR solution per year (today)

     An outside service provider: EUR 1000

     In-house solution (XCola): 50 hours of work




                                 Rami Peltola    13.9.2005   A 23
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 years
    more or less “learning by doing”

    includes the development of a secure communication

     environment
    includes the implementation of 7 solutions

 XCola (third generation phase) development took about 1
  person year
    includes the implementation of 4 solutions

                                   Rami Peltola     13.9.2005   A 24
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

                                 Rami Peltola    13.9.2005   A 25
Benefits of EDR to Statistics Finland (2)

 The data received are of better quality
     both annual and sub-annual surveys

     common estimate is “25% less errors”

     comprehensive study has not been made

 As manual handling diminishes, it can be replaced by more

  rewarding tasks




                                 Rami Peltola     13.9.2005   A 26
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



                                 Rami Peltola     13.9.2005   A 27
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


                                         Rami Peltola        13.9.2005        A 28
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 database
    and then synchronized with the actual tables

 Solution is quite customizable

    Easy to customize for new collection systems

    Easy to add new databases




                                     Rami Peltola      13.9.2005   A 29
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




                                   Rami Peltola      13.9.2005   A 30
An example: Sale inquiry
Background
 Monthly inquiry using paper forms (~2080 enterprises)
 Data collection process (7-8 persons: ~2.0 working years)

     Printing and mailing the questionnaires

     Receiving the questionnaires (mail, fax, e-mail)

     Validating and entering the data

     Printing and mailing the reminders

     Phone inquiry

 Quarterly non-individual feedback for respondents

 Previously submitted data preprinted to questionnaires for
  the last 2 months
                                  Rami Peltola      13.9.2005   A 31
An example: Sale inquiry
Purpose
 To describe the economic situation in different industries
    Trade: 30 branches

    Services: 23 branches

    Construction: 3 branches

    Manufacturing: 15 branches

 To respond to EU-legislation

    The Regulation of the EU concerning short-term

     statistics (1165/98)



                                   Rami Peltola      13.9.2005   A 32
An example: Sale inquiry / Source data
 Tax Administration’s register
    Value added tax -data

    Delay 2 months in turnover

    The data accumulates for 6 months

 Direct data collection (sale inquiry ~2080 companies)

    The most largest companies from each industry

        Trade: ~660 companies

        Services: ~600 companies

        Construction: ~200 companies

        Manufacturing: ~620 companies

 Business Register

    Basic information about companies: branch of industry,

     location, contact information etc.
                                     Rami Peltola      13.9.2005   A 33
An example: Sale inquiry
End products
 Monthly indicators (newsletters and Internet releases)
    Trade: 30 (European sample), 45 (preliminary) and

     75 days delay
    Services: 45 (European sample) and 75 days delay

    Construction: 75 days delay

    Manufacturing: 75 days delay

 Time series

    Starts mostly from 1995 (trade from 1985)




                                  Rami Peltola      13.9.2005   A 34
        Data collection process in sale inquiry before EDR
        Collecting the sales data concerning March
                                                      Manual data entering

April                                                                                May
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



                           Printing and mailing
                           the questionnaires

                                                     Due date


                                                                    Printing and mailing
                                                                    the reminders

                                                                      Phone inquiry
                                                                      for non-respondents

                                                     Rami Peltola             13.9.2005     A 35
        Data collection process in sale inquiry today
        Collecting the sales data concerning March
                                            Electronic data transfers
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                                         Due date
 collection database
                                                                   E-mail reminders
                             E-mail informing
                                                                                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

                                                                        Rami Peltola             13.9.2005       A 36
EDR solution / Sale inquiry




                              Rami Peltola   13.9.2005   A 37
Change in working hours used in data collection
and validation (hours per year)

    2500


    2000

                                            2001
    1500
                                            2002
                                            2003
    1000
                                            2004

     500


       0

                             Rami Peltola      13.9.2005   A 38
Hours used in development and maintenance of
EDR solution (sale inquiry)

     900
                                            Includes hours used
     800                                     in development of
                                                 infrastrucre
     700
     600
     500                                  2002
     400                                  2003
     300                                  2004

     200
     100
      0
                   Hours
                           Rami Peltola      13.9.2005            A 39
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.




                                      Rami Peltola      13.9.2005   A 40
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)




                                       Rami Peltola     13.9.2005   A 41
Effects of EDR on data quality (ESS quality
dimensions)
 Accuracy
    Automatic validations

 Timeliness

    Mail  E -mail and web

    Manual data entering  Electronic data transfers

 Comparability




                                  Rami Peltola     13.9.2005   A 42

								
To top