European conference on Qualitywww.stat.gov.lt

1
REAL AND IMAGINATIVE RESPONSE BURDEN DIFFERENT
WAYS OF TREATMENT, MEASUREMENT AND REDUCTION
Daliute Kavaliauskiene Bronislava
Kaminskiene Statistics Lithuania
2
SUBJECT-MATTER

1. MEANS USED TO REDUCE THE BURDEN ON RESPONDENTS
2. MEASUREMENT OF RESPONSE BURDEN

3
ORGANISATION OF WORK

• A PERMANENT WORKING GROUP ON RESPONSE BURDEN
  SINCE 2003
• AN ACTION PLAN RENEWED ANNUALLY

4
MEANS USED

1. REDUCTION OF THE NUMBER OF STATISTICAL SURVEYS
2. REVISION OF STATISTICAL QUESTIONNAIRES
3. IMPROVEMENT OF SAMPLING PROCEDURES
4. APPLICATION OF ADMINISTRATIVE DATA IN STATISTICAL
   SURVEYS
5. NEW IT TECHNOLOGIES FOR STATISTICAL DATA
   COLLECTION

5
REDUCTION OF STATISTICAL SURVEYS

• JOINING TOGETHER SEVERAL SURVEYS
• ABOLITION OF SURVEYS

6
REVISION OF STATISTICAL QUESTIONNAIRES

• ABOLITION OF IRRELEVANT INDICATORS
• IMPROVEMENT OF METHODOLOGICAL GUIDELINES FOR
  FILLING IN STATISTICAL QUESTIONNAIRES

7
IMPROVEMENT OF SAMPLING PROCEDURES

• REDUCING THE SAMPLE SIZES
• TRYING TO DISTRIBUTE THE RESPONSE BURDEN MORE
  EVENLY AMONG BUSINESSES

8
APPLICATION OF ADMINISTRATIVE DATA IN STATISTICAL
SURVEYS

• 2 STS INDICATORS BASED ON ADMINSTRATIVE DATA OF
  THE STATE SOCIAL INSURANCE FUND BOARD
• NUMBER OF EMPLOYEES
• WAGES AND SALARIES
• SBS INDICATORS FOR ENTERPRISES WITH UP TO 10
  EMPLOYEES BASED ON ANNUAL FINANCIAL ACCOUNTS FROM
  STATE TAX INSPECTORATE
• ABOUT 16 THOUS. RESPONDENTS

9
NEW IT TECHNOLOGIES FOR STATISTICAL DATA
COLLECTION

• INTRODUCTION OF E-QUESTIONNAIRES
• DRAFT PROJECT ON THE POSSIBILITY TO EXTRACT DATA
  DIRECTLY FROM BUSINESS ACCOUNTANCY SYSTEMS

10
PARTICIPATION OF ENTERPRISES IN STATISTICAL
SURVEYS
  2005 2006 2007 2008
Number of active enterprises 72 330 73 334 76 516 81 376
Number of enterprises participating in at least one statistical survey 46 717 45 025 45 384 45 742
Participation rate in percent 65 61 59 56
Average number of surveys per enterprise 3,7 3,6 3,5 3,4
Number of questionnaires for enterprises 187 175 157 157
Average number of questionnaires per enterprise 6,8 6,9 6,5 6,1
11
REASONS TO START MEASUREMENT OF THE STATISTICAL
RESPONSE BURDEN

• POLITICAL INTEREST IN THE SUBJECT
• RESPONSE BURDEN ONE OF THE QUALITY DIMENSIONS
  IN TQM SYSTEM
• IMPLEMENTATION OF IDEAS TO REDUCE THE BURDEN

12
THE MAIN RESPONSE BURDEN INDICATOR
TIME USED TO FILL IN QUESTIONNAIRES

• ADVENTAGES
• EASY TO COMPARE AMONG DIFFERENT TYPE OF
  ACTIVITIES, SIZE OF AN ENTERPRISES, E.T.C.
• EASY TO COMPARE AMONG COUNTRIES
• EASY TO CALCULATE FOR DIFFERENT TIME PERIODS
  (IRRESPECTIVE OF LABOUR COSTS)
• EASY FOR RESPONDENTS TO REPORT DATA

13
EVALUATION OF THE RESPONSE BURDEN

• ALL QUESTIONNAIRES SUPPLEMENTED BY A QUESTION ON
  THE RESPONSE BURDEN
• TIME AVERAGE AND MEDIAN CALCULATED FOR
  QUESTIONNAIRES
• THE TOTAL RESPONSE BURDEN CALCULATED FOR ALL
  BUSINESSES IN THE SAMPLE
• THE PURPOSE OF THE SURVEY IS TO MEASURE THE
  BURDEN IMPOSED ON BUSINESSES, NOT THE ACTUAL
  RESPONSE BURDEN

14
RESULTS OF EVALUATION OF RESPONSE BURDEN
15
ANALYSIS OF DATA

• TENDENCY OF THE RESPONDENTS TO OVERESTIMATE THE
  STATISTICAL BURDEN IN COMPARISON WITH OTHER
  ADMINISTRATIVE OBLIGATIONS
• HUGE VARIABILITY OF DATA SHOWS THE IMPACT OF THE
  HUMAN FACTOR ON THE RESPONSE BURDEN ESTIMATION
• REAL AND IMAGINATIVE (SUBJECTIVE) RESPONSE BURDEN
  TWO COMPONENTS OF THE SAME PHENOMENON

16
FACTORS INFLUENCING THE SUBJECTIVE RESPONSE BURDEN

• TOO HEAVY REAL RESPONSE BURDEN
• GAP IN INFORMATION ON RESPONSE BURDEN OF OTHER
  BUSINESSES
• LACK OF KNOWLEDGE ABOUT THE IMPORTANCE OF THE
  DATA PROVIDED
• INSUFFICIENT DIALOG OF STATISTICIANS WITH
  RESPONDENTS

17
WAYS TO REDUCE RESPONSE BURDEN

• ALLOCATION OF SAMPLES AS EVENLY AS POSSIBLE OVER
  THE POPULATION
• MORE INFORMATION FOR THE RESPONDENTS ON THE
  RESPONSE BURDEN
• INCREASED DIALOG WITH RESPONDENTS
• FEEDBACK FOR THE RESPONDENTS

18
Thank you.Questions are welcome