Reasons for Unit Non-Response
|
|
- Brooke Bruce
- 6 years ago
- Views:
Transcription
1 Reasons for Unit Non-Response Failure of the data collector to locate/identify the sample unit; Failure to make contact with the sample unit; Refusal of the sample unit to participate; Inability of the sample unit to participate (e.g. ill health, absence, etc); Inability of the data collector and sample unit to communicate (e.g. language barriers); Accidental loss of the data/ questionnaire. Unit non-response is defined relative to the eligible sample. In other words, if the sampling frame contains ineligible units, these do not contribute towards response/ non-response. Unit non-response is often divided into three components: non-contact, inability to respond, lack of co-operation (refusal). Response rates can usefully be divided into these (or other) components.
2 Item non-response A sample unit participates but data for some survey items are not available for analysis. Reasons could include: Refusal to provide an answer Inability to provide an answer Other failure to answer (e.g. by accident) Provided answer being of inadequate quality (e.g. incomplete, implausible, failing an edit/consistency check, etc.) Item non-response can be caused by: the action of the sample member (e.g. refusal to answer); the action of an interviewer (e.g. failure to ask a question that should have been asked, or failure to record the answer adequately); the survey design (e.g. poor routeing instruction). In practice, these factors interact.
3 Response Pattern We can define the survey response pattern by a matrix R = [r jk ], where r jk = 1 if item j is observed for unit k, and r jk = 0 otherwise. Possible response patterns include: Unit (k) y 1k y 2k y 3k y qk full response item non-response item non-response unit non-response Note that, due to item non-response, the set of units available for analysis depends on the item or items required.
4 Non-Response Error The effect of non-response on a survey estimate y r can be defined: y r = y n + nr n ( yr ynr ) where y r = statistic for the r responding units, y n = statistic for all n sample units, y nr = statistic for the nr non-responding units. (Groves, 1989, p.133). nr n ( y r ynr ) is the estimate (the realisation in one implementation of the survey) of total non-response error.
5 Components of Non-Response Error The non-response error has two elements: the non-response rate; the difference between responding and non-responding units in terms of y. Both of these elements are important. Note that the second element will result from both non-response bias and variance. In practice, studies suggest that bias is usually the main component.
6 Components of Non-Response Error ctd. Non-response errors can be divided into two components: Errors due to unit non-response Errors due to item non-response Each of these can in turn be decomposed into subsources, for example: Unit non-response due to: Non-contact Refusal to respond Inability to respond Item non-response due to: Routing (instrument) error Routing (interviewer) error Refusal to respond Inability to respond Etc.
7 Non-Response in Different Types of Surveys Survey structure: Cross-sectional Longitudinal Rotating Panel Flow sample Respondent type: Business Individual Household Individual within household
8 Cross Sectional Surveys Components of non-response: Refusals Non-contact Inability to respond Constraints on non-response: Limited field time Limited budget Survey population Survey task (burden)
9 Longitudinal Surveys Components of NR same as for cross-sectional surveys. But structure of NR different: Complete NR Wave NR (gaps) Attrition NR patterns depend partly on survey policy and partly on field efforts (especially "tracing").
10 Patterns of response to a 4-wave panel survey Policy 1: Issue all eligible cases at every wave. 16 possible patterns: Wave
11 Policy 2: Issue only wave 1 responding cases at each subsequent wave. 9 possible patterns: Wave Policy 3: At each wave, issue only cases responding to previous wave. 5 possible patterns: Wave
12 Example: England and Wales Youth Cohort Study Panel with 3 waves; Implements policy 1 (issue all eligible cases at every wave). 8 possible response patterns: Wave: No. of cases % of cases 1 8, , , , (From Lynn, Purdon, Hedges and McAleese, 1994)
13 Panel Attrition Specific to longitudinal studies. Can result from non-traceable or refusal cases Impacts can be quite different from cross sectional surveys Differences should be studied (using data from earlier waves). Design stage: strategy for dealing with attrition (both non-response and population exits ) Response rates as products of wave response rates
14 Panel Attrition ctd. Non-response due to mobility (e.g. sample unit has moved and cannot be traced): not unique to longitudinal surveys but can be especially problematic particularly if long time gap between waves Experience of the first wave may alter the response behaviour on subsequent waves: Almost unique to longitudinal studies Interview experience therefore very important
15 Survey Types and Data Collection Modes Interviewer surveys Face-to-face surveys Telephone surveys Self completion Mail surveys Web surveys Diary keeping Mixed mode strategies Supervised self-completion Mail questionnaire with interviewer follow-up Interview and diary keeping period Interview and self-completion Different modes for different waves in panel surveys
16 Survey Types and Data Collection Modes ctd. Face-to-face surveys typically yield higher response rates than telephone surveys. Response for mail surveys usually much lower than other modes of data collection. Differences are mainly due to level of respondent motivation possible in different modes. In mixed mode strategies: motivation of respondents vs. time and costs. Interview surveys: role of the interviewer is crucial. Telephone surveys: scope for interviewer influence is more limited. Mail surveys: only communication is via document design and content (q re, advance or covering letter, reminders).
17 Measuring and Presenting Response Rates Response rates measure the proportion of eligible sample units that successfully provided data, i.e. the reduction due to non-response of cases available for analysis. Response rates are more likely to be reported than almost any other survey process quality indicator. Response rates are often mistakenly used as a measure of quality of survey statistics. Remember, response rate is just one component of non-response error.
18 Refusals Factors influencing household survey participation (Groves & Couper, 1998): 1. Societal-level factors 2. Attributes of the survey design 3. Characteristics of the sample person 4. Attributes of interviewer 5. Respondent-interviewer interaction Social environment Survey design Household Interviewer Household-interviewer interaction Decision to co-operate or refuse
19 1. Societal-level factors The degree of social responsibility felt by sample persons Legitimacy of societal institutions The degree of social cohesion (Goyder, 1987). Interviewer respondent interaction: Particular persuasion strategies (on the part of the interviewer) Decision making strategies (on the part of the respondent) (Groves, Cialdini and Couper, 1992). 2. Attributes of the survey design The mode of the initial contact affects The number of channels of communication between interviewer and respondent (Groves, 1978). The selection of persuasion strategies to employ and the effectiveness of alternative strategies (Groves, Cialdini and Couper, 1992).
20 Other attributes: Perceived burden (e.g. length of the interview being requested, sensitivity of questions) Respondents level of interest and knowledge in the survey (topic, description) Incentives 3. Characteristics of the sample person Socio-economic characters Psychological characteristics Past experiences (Goyder, 1987). These factors tend to produce a set of pre-dispositions that affect the decision. They also affect the initial approach of the interviewer to the sampling unit. (Groves, Cialdini and Couper, 1992).
21 4. Attributes of interviewer Observable socio-demographic characteristics of the interviewer may affect the script evoked in the respondent s mind at the first contact with the interviewer. (Groves, Cialdini and Couper, 1992). No research has yet found strong links between stable interviewerpersonality characteristics and success in gaining co-operation. Reasons for this might be interviewers being relatively homogeneous group, tailoring, social skills, and other adoptive behaviours (Groves and Couper, 1998). Morton-Williams (1993) argues that social skills can be taught and offers an outline of such training. Those with greater interviewing experience tend to achieve higher rates of co-operation than those with less experience. It is still unclear, whether this is a selection effect (less successful interviewers terminate their employment earlier) or a training effect (due to the benefits of coping over time with diverse situations in recruiting respondents) or both (Groves and Couper, 1998).
22 There is some evidence that interviewers who, prior to the survey, are confident about their ability to elicit co-operation tend to achieve higher co-operation rates (Groves and Couper, 1998). 5. Respondent-interviewer interaction Persuasion strategies employed by the interviewer are determined not only by the interviewer s ability, expectations, and so on but also by features of the survey design and by characteristics of the immediate environment and broader society. Similarly, the responses that the sample person makes to the request are affected by a variety of factors, both internal and external to the respondent, and both intrinsic and extrinsic to the survey request. (Groves, Cialdini and Couper 1992).
23 Non-Contact A conceptual model for contacting sample households (Groves & Couper 1998): Social environmental attributes Physical impediments Socio-demographic attributes Accesible at-home patterns X Likelihood of contact Number of calls X Timing of calls Interviewer attributes Contact likelihood is a function of three factors: 1) Physical impediments that prevent visiting interviewers from alerting the household to their presence 2) When household members are at home 3) When and how many times the interviewer visits the household.
24 Reasons for Non-Contacts Association between when household members are at home and when interviewer calls: Life styles that lead to reduced time at home (patterns of work, related patterns of shopping and entertainment) Rural/urban differences Local crime rate (perceptions) Ethnic/cultural differences One-person households households with no children <5 or adults >70 Hidden refusals: Fear of crime/strangers/salesmen Too busy Answer machine screening
25 Example: The Survey Participation Process in the British Crime Survey It can be useful to identify the logical stages of the survey participation process. This was done for BCS (next page). It can be seen that there are many stages at which either a non-contact or a refusal could occur (see diagram on next page, from Laiho and Lynn, 2000). This illustrates the heterogeneity of the survey non-response phenomenon. Survey design and implementation features should be used to address each of these possible non-response outcomes.
26 Survey participation process in BCS (Laiho & Lynn 2000) Advance letter to the sampled address Contact attempt Office refusal Ineligible Dwelling Unit (DU) Multiple DU - Insufficent address - Not traced Single DU - Not yet built - Vacant/ derelicted / demolished - Empty - Business/industrial only - Other Listing of all adult (16+) members of the DU 1) List in systematic order by flat number 2) random selection of the dwelling unit Information about number of persons 16+ refused No contact with responsible/ any adult in selected DU (after 5+ attempts) Random selection of the respondent Contact attempt - respondent Respondent contacted No contact with selected person after 5+ attempts Completed interview Refusal Other reason for non-response Personal refusal by respondent Broken appointment, no recontact Proxy refusal on behalf of selected respondent Ill at home Away/in hospital during survey period Senile/incapacitated Inadequate English Other
27 Studying Non-Response Designs for Determining Characteristics of Nonrespondents: Special studies of non-respondents Using information on the sampling frame Asking others about non-respondents or having interviewer provide information about them (example on next page) Comparison of respondent characteristics by call number Comparison of respondent characteristics to census or other external information Studying persons who drop out of a panel survey after an initial interview
28 ALL RESIDENTIAL ADDRESSES (CONTACTS AND NON-CONTACTS INCLUDING VACANTS) 17. Does the address have an entryphone? Yes 1 No Which of the following are visible at the sampled address? CODE ALL THAT APPLY Burglar alarm 1 Security gate over front door 2 Bars/grilles on any windows 3 Other security device(s) 4 Estate/block security lodge/guards 5 None of these 0 INTERVIEWER ASSESSMENTS: 19. Are the houses/flats in this immediate area in a good or bad physical state? Mainly good 1 Mainly fair 2 Mainly bad 3 Mainly very bad Is the sampled house/flat in a better or worse condition outside than the others in this area? Better 1 Worse 2 About the same 3 Does not apply Do you know or think that the occupants are probably white 1 black 2 Asian 3 Other: 4 Don t know 5 22a. SAMPLED DWELLING IS: Whole house: detached 1 Semi-detached 2 IF NO DWELLING mid-terrace 3 SELECTED, CODE end terrace 4 FOR ADDRESS Maisonette 5 Flat: purpose-built 6 converted 7 Rooms, bedsitter 8 Unable to code 9 IF FLAT ETC (5-7 AT a.) ANSWER b-e. OTHERS - END b. CODE TYPE OF FLAT ETC: Self-contained 1 Not self-contained 2 Don t know 8 c. BUILDING HAS: Fewer than 5 floors 1 5 floors or more 2 Unable to code 3 d. FLOOR LEVEL OF MAIN ACCOMMODATION: Basement/semi-basement 1 Ground floor/street level 2 First floor 3 2 nd /3 rd floor 4 4 th 9 th floor 5 10 th floor or higher 6 e. BUILDING HAS: Common entrance: lockable 1 Common entrance: not lockable 2 No common entrance 3
29 Example: Using Interviewer Observation Data Data were collected by interviewers, using the form on the previous page, on the 1996 British Crime Survey. The following table uses the answers to question 22a. Response Selected Responding rate sample sample % % House: detached 82.6% semi-detached 79.6% end terrace 79.2% mid-terrace 77.7% Maisonette 74.9% Flat: converted 72.3% purpose-built 70.3% Rooms/bedsit 75.6% Unable to code 51.2% Base 13,117 10,059 Source: Lynn P (1996) Weighting for non-response, in Survey and Statistical Computing 1996, Chesham: Association for Statistical Computing
30 Example: Using Sample Frame Data These data are from the Scottish School Leavers Survey, a postal survey for which the sampling frame is a list of pupils and their school exam results. Highest Response Selected Responding Qualification rate sample sample % % 5+ Higher grades 91.1% Higher grades 85.1% Higher grades 81.7% Standard grades % Standard grades % Standard grades % Standard grades 4-7 only 62.6% No qualifications 59.6% Base 4,542 3,469 Source: Lynn P (1996) Weighting for non-response, in Survey and Statistical Computing 1996, Chesham: Association for Statistical Computing
31 Example: Using Geographical Data The following data are from the Italian Multipurpose Survey carried out by ISTAT. The figures presented are regression coefficients from a model to predict propensity to refuse. It can be seen that refusals are most likely in metropolitan areas and least like in small (rural) municipalities. Parameter s.e. Metropolitan area (0.147) Met. area ring (0.168) Large municipality (0.114) Small municipality 0 Source: Baldazzi et al (2002) Interviewer s effect on refusal risk in the Italian Multipurpose Survey: a multilevel approach, paper presented at the annual conference of the Italian Statistical Society, May 2002.
32 Example: Using Number of Calls These data are from the 1987 British General Election Survey. This kind of analysis can be done for any survey, even if there is no useful information from the sampling frame or from interviewer observation (e.g. postal or telephone survey). Interviewed at 1 st call Interviewed at 2 nd call Interviewed at 3 rd call Interviewed after 4+ calls Nonrespondents Social Class % % % % % Non-manual Unknown Manual Unknown Self-employed Unknown Unclassifiable Unknown Base 726 1, ,354 1,637 Source: Lynn P (1996) Weighting for non-response, in Survey and Statistical Computing 1996, Chesham: Association for Statistical Computing
33 Example: Comparing Non-Contacts, Refusals and Easy-to-get Households Data are from 1996 Health Survey for England. Lynn et al (2002) analysed calls record data and classified respondents as easy or difficult to contact, and willing or reluctant. It can be seen, for example, that persons who were difficult to contact were much younger than average and much more likely to be employed; but reluctant persons (potential refusals) were slightly less likely than others to be employed. Estimate Difficult to contact (6+ calls) A Reluctant (temporary refusal) B Hard-toget A+B Easy-to-get households C All responding households A+B+C Male (%) Age (mean) Owner-occupier (%) Employed (ILO definition) (%) White (%) Source: Lynn P, Clarke P, Martin J and Sturgis P (2002) The effects of extended interviewer efforts on nonresponse bias, in Survey Nonresponse (ed.s R M Groves, D A Dillman, J L Eltinge and R J A Little), New York: Wiley.
34 Common Non-Response Patterns 1. Many (most) household surveys find response rates lower in urban areas. 2. Many self-completion surveys find response rates higher amongst those with more education. 3. Interview surveys often find higher contact rates amongst many-person households than amongst 1- or 2- person households. 4. Many household surveys find contact rates higher amongst persons aged 65+, but cooperation rates lower. 5. In most (but not all) European countries, response rates are usually higher for women than men. 6. On most business surveys, response rates are higher amongst larger businesses. But These relationships may not be true for all countries, cultures and surveys.
Weighting for Non-Response
Survey and Statistical Computing 1996 Edited by R Banks et al. Compilation Association for Survey Computing Weighting for Non-Response Peter Lynn Abstract Non-response is a problem which pervades almost
More informationWhat can we learn from the Understanding Society Innovation Panel?
What can we learn from the Understanding Society Innovation Panel? Annette Jäckle Institute for Social and Economic Research (ISER) University of Essex An initiative by the Economic and Social Research
More informationAppendix D: Nonresponse Analysis
Nonresponse in the O*NET Data Collection Program can occur from any of three sources. Establishments can cause nonresponse when they fail to participate at the verification, screening, recruiting, or sampling
More informationMixing Survey Modes: Why do it and What are the Consequences?
Mixing Survey Modes: Why do it and What are the Consequences? Prof dr Edith D. de Leeuw Share Survey Modes Workshop Mannheim, 8 July 2009 Copyright Edith D. de Leeuw Modes of Data Collection Interviewer-administered
More informationAppendix E: Nonresponse Analysis for Analysis Cycles 9 Through 12
Appendix E: Nonresponse Analysis for Analysis Cycles 9 Through 12 Appendix E: Nonresponse Analysis Establishments can cause nonresponse in the O*NET Data Collection Program at the verification, screening,
More informationEnergy Use in Homes. A series of reports on domestic energy use in England. Thermal Insulation
Energy Use in Homes A series of reports on domestic energy use in England Thermal Insulation Energy Use in Homes A series of reports on domestic energy use in England This is one of a series of three reports
More informationUse of a nonresponse follow-up survey to assess the validity of R-indicators as a measure of the risk of bias
Use of a nonresponse follow-up survey to assess the validity of R-indicators as a measure of the risk of bias Caroline Vandenplas 1, Caroline Roberts 1 and Michèle Ernst Staehli 2 1 University of Lausanne,
More informationKeeping Track of Panel Members: An Experimental Test of a Between-Wave Contact Strategy
Journal of Official Statistics, Vol. 27, No. 2, 2011, pp. 319 338 Keeping Track of Panel Members: An Experimental Test of a Between-Wave Contact Strategy Katherine A. McGonagle 1, Mick P. Couper 1, and
More informationNUMBERS, FACTS AND TRENDS SHAPING THE WORLD FOR RELEASE DECEMBER 8, 2014 FOR FURTHER INFORMATION ON THIS REPORT:
NUMBERS, FACTS AND TRENDS SHAPING THE WORLD FOR RELEASE DECEMBER 8, 2014 FOR FURTHER INFORMATION ON THIS REPORT: Kristen Purcell, Research Consultant Lee Rainie, Director, Internet, Science and Technology
More informationThe Application of Blaise III to the Israel Labour Force Survey
The Application of Blaise III to the Israel Labour Force Survey Edith Noy and Gad Nathan, Central Bureau of Statistics, Jerusalem, Israel 1. Introduction The Central Bureau of Statistics has only recently
More informationGlossary of Research Terms
Glossary of Research Terms January 2001 fact sheet I 2/11 Ad hoc Single surveys designed for a specific research purpose, as opposed to continuous, regularly repeated, or syndicated surveys. Advertising
More informationExperiments with Methods to Reduce Attrition in Longitudinal Surveys
8 Experiments with Methods to Reduce Attrition in Longitudinal Surveys Laura Fumagalli Heather Laurie Peter Lynn Institute for Social and Economic Research University of Essex No. 2010-04 February 2010
More informationGoing Online with a Face-to-Face Household Panel: Effects of a Mixed Mode Design on Item and Unit Non-Response
Survey Research Methods (2015) Vol. 9, No. 1, pp. 57-70 c European Survey Research Association ISSN 1864-3361 http://www.surveymethods.org Going Online with a Face-to-Face Household Panel: Effects of a
More informationUsing monetary incentives in face-to-face surveys. Are prepaid incentives more effective than promised incentives?
Using monetary incentives in face-to-face surveys. Are prepaid incentives more effective than promised incentives? Michael Blohm 1, Achim Koch 2 1 GESIS Leibniz Institute for the Social Sciences, Mannheim,
More informationThe National Survey of Fishing, Hunting, and Wildlife-Associated Recreation Cell Phone and Debit Card Test A Cost -Benefit Analysis
The National Survey of Fishing, Hunting, and Wildlife-Associated Recreation Cell Phone and Debit Card Test A Cost -Benefit Analysis Aniekan Okon Elke Mclaren Denise Pepe U.S. Census Bureau March 16, 2012
More informationREVIEW OF ATTITUDES AND PREFERENCES FOR WATER EFFICIENCY IN HOMES
REVIEW OF ATTITUDES AND PREFERENCES FOR WATER EFFICIENCY IN HOMES Dexter Robinson 1, Kemi Adeyeye 2, Della Madgwick 3, Andrew Church 4 1 D.P.Robinson@Brighton.ac.uk, 2 O.Adeyeye@brighton.ac.uk, 3 D.Madgwick@brighton.ac.uk,
More informationLIFE+ Up and Forward Project: Case Study
LIFE+ Up and Forward Project: Case Study B10 Bags and caddies Area: Manchester Date: January 2014 LIFE11 ENV/UK/000389 Contents Page 1. Executive Summary 2 2. Introduction 5 3 Campaign Area 6 4. Demographics
More informationCover Page. The handle holds various files of this Leiden University dissertation
Cover Page The handle http://hdl.handle.net/1887/48563 holds various files of this Leiden University dissertation Author: Groenenberg, Iris Title: CHECK D?! : determinants of participation in a two-stage
More informationKey Concept Overview
Management School Key Concept Overview Sampling Techniques KMGT 714 Data Driven Marketing Research and Analysis Week 4 2015 Laureate Education, Inc. Page 0 of 4 Key Concept Sampling Techniques In order
More informationEffects of Differential Branding on Survey Materials
Effects of Differential Branding on Survey Materials Nicole Bensky, Gretchen Grabowski, Justin Bailey, Chuck Shuttles, Michael Link, PhD. 1 1 The Nielsen Company, 501 Brooker Creek Blvd., Oldsmar, FL 34677
More informationDesign Review Report Highfields, Heath, Cardiff DCFW Ref: 104 Meeting of 19th May 2016
Design Review Report Highfields, Heath, Cardiff DCFW Ref: 104 Meeting of 19 th May 2016 Declarations of Interest Panel members, observers and other relevant parties are required to declare in advance any
More informationNONRESPONSE IN HOUSEHOLD TRAVEL SURVEYS
NONRESPONSE IN HOUSEHOLD TRAVEL SURVEYS Michele Zimowski, Roger Tourangeau, Rashna Ghadialy, and Steven Pedlow Prepared for: NORC Federal Highway Administration th 1155 60 Street October, 1997 Chicago,
More informationDATA MEMO. The average American internet user is not sure what podcasting is, what an RSS feed does, or what the term phishing means
DATA MEMO BY: PIP Director Lee Rainie (202-419-4500) DATE: July 2005 The average American internet user is not sure what podcasting is, what an RSS feed does, or what the term phishing means Large numbers
More informationFinScope Methodology
FinScope Methodology 1. FinScope Surveys The FinScope survey is a research tool developed by FinMark Trust. It is a nationally representative survey of how people source their income, and how they manage
More informationAbstract. About the Authors
Household Food Security in the United States, 2002. By Mark Nord, Margaret Andrews, and Steven Carlson. Food and Rural Economics Division, Economic Research Service, U.S. Department of Agriculture, Food
More informationPerformance Appraisal System in Medical College Libraries in Karnataka State - A study
2016 IJSRST Volume 2 Issue 3 Print ISSN: 2395-6011 Online ISSN: 2395-602X Themed Section: Science and Technology Performance Appraisal System in Medical College Libraries in Karnataka State - A study Pushpalatha
More informationIssues in Applying Adaptive Design to Establishment Surveys
Issues in Applying Adaptive Design to Establishment Surveys Jaki S. McCarthy US Department of Agriculture National Agricultural Statistics Service The 6th International Workshop on Internet Survey and
More informationPractical issues in the use of surveys in management research. George Balabanis
Practical issues in the use of surveys in management research George Balabanis contents When to use survey research Survey errors How to reduce survey errors Email and web surveys When survey research
More informationNQT Quality improvement study
5.5 Retention of NQTs 5.5.1 Overview Although the literature review for this study (see section 2 above) suggests that there is a problem with the retention of NQTs, the data from the survey of SLT members
More informationONS s Data Collection Programme: What does it mean for Labour Market data?
ONS s Data Collection Programme: What does it mean for Labour Market data? Ian O Sullivan, Ed Dunn & Andrew Phelps Office for National Statistics Social Survey Division Context Better Statistics, Better
More informationWORKSHOP ON RESPONDENT ISSUES: SAMPLING, WEIGHTING, AND NONRESPONSE
WORKSHOP ON RESPONDENT ISSUES: SAMPLING, WEIGHTING, AND NONRESPONSE Resource Paper Data Quality Problems in Travel Surveys An International Overview Wim F. de Heer and Ger Moritz Statistic Netherlands
More informationTreating Nonresponse in the Canadian National Longitudinal Survey of Children and Youth (NLSCY)
Treating Nonresponse in the Canadian National Longitudinal Survey of Children and Youth (NLSCY) An Evolution Over 6 Cycles Marcelle Tremblay ICCCS 2006, Oxford UK September 13, 2006 Talk Outline Treating
More informationData Collection Research using Paradata at Statistics Canada François Laflamme 1
Modernisation of Statistics Production (MSP2009) Conference in Stockholm 2 4 November 2009 Data Collection Research using Paradata at Statistics Canada François Laflamme 1 Over the past few years, Statistics
More informationPartner Telephone Survey Report May 5, 2014
Partner Telephone Survey Report May 5, 2014 EVALUATION SUMMARY Partner Telephone Survey Author: Shannon Robinson, Dorian Lunny Date: May 5, 2014 Approved by: Dr. Kit Young Hoon Page 1 of 20 Table of Contents
More informationAddress Based Sampling: Census Block Group Data Used to Define Incentive Structure
Address Based Sampling: Census Block Group Data Used to Define Structure Anh Thu Burks 1 & Michael W. Link 2 1 The Nielsen Company, 501 Brooker Creek, Oldsmar, FL 34677 2 The Nielsen Company, 1145 Sanctuary
More informationStatistical Design and Estimation taking account of Quality
Statistical Design and Estimation taking account of Quality Chris Skinner NTTS, 23 February 2011 Quality Quality as accuracy Aim: quality improvement 2 Quality Context despite increasing threats to quality
More informationPlans for the 1991 Redesign of the Canadian Labour Force Survey
Plans for the 1991 Redesign of the Canadian Labour Force Survey D. Drew, J. Gambino, E. Akyeampong and B. Williams, Statistics Canada J. Gambino, Social Survey Methods Division, 16-A, R.H. Coats Building,
More informationSample: n=2,252 people age 16 or older nationwide, including 1,125 cell phone interviews Interviewing dates:
Survey questions Library Services Survey Final Topline 11/14/2012 Data for October 15 November 10, 2012 Princeton Survey Research Associates International for the Pew Research Center s Internet & American
More informationSubset of constructed questionnaire items. q 2. Female. 4) Which one of these best describes your current situation? (please tick one box only)
Subset of constructed questionnaire items Socio- demographics items: 1) Are you male or female? Male Female 2) What is your ethnic group? (please tick one box only) White Mixed Asian or Asian British Black
More informationMODELING THE WEEKLY DATA COLLECTION EFFICIENCY OF FACE-TO-FACE SURVEYS: SIX ROUNDS OF THE EUROPEAN SOCIAL SURVEY
Journal of Survey Statistics and Methodology (217) 5, 212 232 MODELING THE WEEKLY DATA COLLECTION EFFICIENCY OF FACE-TO-FACE SURVEYS: SIX ROUNDS OF THE EUROPEAN SOCIAL SURVEY CAROLINE VANDENPLAS* GEERT
More informationRecent Developments in Assessing and Mitigating Nonresponse Bias
Recent Developments in Assessing and Mitigating Nonresponse Bias Joanna Fane Lineback and Eric B. Fink 1 U.S. Census Bureau, Washington, D.C. 20233 Abstract In this paper, we address recent developments
More informationPredictors of pro-environmental behaviour in 1993 and 2010 An international comparison. Janine Chapman
Predictors of pro-environmental behaviour in 1993 and 2010 An international comparison Janine Chapman December 2013 Published by the Centre for Work + Life University of South Australia http://www.unisa.edu.au/hawkeinstitute/cwl/default.asp
More informationResponsive design for household surveys: tools for actively controlling survey errors and costs
J. R. Statist. Soc. A (2006) 169, Part 3, pp. 439 457 Responsive design for household surveys: tools for actively controlling survey errors and costs Robert M. Groves and Steven G. Heeringa University
More informationResidential care sector: Working conditions and job quality
European Foundation for the Improvement of Living and Working Conditions sector: Working conditions and job quality Work plays a significant role in people s lives, in the functioning of companies and
More informationBenin Indicator Survey Data Set
Benin Indicator Survey Data Set 1. Introduction 1.1. This document provides additional information for the Indicator Survey data collected in Benin from 18 th May and 30 th September 2009 as part of the
More informationRacing and Thoroughbred Breeding Industry Recruitment, Skills and Retention Survey Report: April 2017
Racing and Thoroughbred Breeding Industry Recruitment, Skills and Retention Survey 2016-17 Report: April 2017 Contents Executive Summary... 1 Main Report... 3 Section 1: Introduction... 3 Introduction
More information1 PEW RESEARCH CENTER
1 Methodology This report is drawn from a survey conducted as part of the American Trends Panel (ATP), a nationally representative panel of randomly selected U.S. adults living in households recruited
More informationTrade Union Membership in the Labour Force Survey: Is it who you ask or how you ask them?
Trade Union Membership in the Labour Force Survey: Is it who you ask or how you ask them? Rhys Davies Concerns regarding how the use of proxy respondents within the Labour Force Survey may affect the quality
More informationLocation Review Atlantic Park, Liverpool
Location Review Atlantic Park, Liverpool 2007 Introduction The following report provides a thorough assessment of the demographic profile, local labour market and extent of recruitment potential applicable
More informationReasons for unit and partial nonresponse in Websurveys
ISM Tokyo, November 7th 2002 Reasons for unit and partial nonresponse in Web Surveys Gašper Koren Faculty of Social Sciences University of Ljubljana gasper.koren@uni-lj.si Outline Survey research in Slovenia
More informationConsumer Knowledge of the Hazards of Carbon Monoxide Poisoning and Faulty Domestic Heating Systems. Mark Speed Jenny Dickson Sarah Birtles
Consumer Knowledge of the Hazards of Carbon Monoxide Poisoning and Faulty Domestic Heating Systems Mark Speed Jenny Dickson Sarah Birtles TABLE OF CONTENTS Page Number INTRODUCTION 1 KEY FINDINGS AND IMPLICATIONS
More informationIMPLEMENTATION GUIDELINE NO. 23
IMPLEMENTATION GUIDELINE NO. 23 Refuse Arrangements and Management for Multiple Residential Development Date of Resolution These guidelines were adopted by Council on the 22 February 2011 and takes effect
More informationWorking life research. How is new technology changing the work of employees?
Working life research How is new technology changing the work of employees? SAK working conditions survey 2018 2 TECHNOLOGY HAS A SIGNIFICANT PRESENCE in the world of work nowadays. This is a veritable
More informationWholesale sector: Working conditions and job quality
European Foundation for the Improvement of Living and Working Conditions sector: Working conditions and job quality Work plays a significant role in people s lives, in the functioning of companies and
More informationSurvey Report Community Newspaper Readership Survey [National Newspaper Association]
Survey Report 2012 Community Newspaper Readership Survey [National Newspaper Association] Center for Advanced Social Research The Donald W. Reynolds Journalism Institute School of Journalism University
More informationDoes giving people their preferred survey mode actually increase survey participation rates? An Experimental Examination
University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln Sociology Department, Faculty Publications Sociology, Department of 2012 Does giving people their preferred survey mode
More informationUsing Weights in the Analysis of Survey Data
Using Weights in the Analysis of Survey Data David R. Johnson Department of Sociology Population Research Institute The Pennsylvania State University November 2008 What is a Survey Weight? A value assigned
More informationSubject: Request for Quotations for: Nationwide household survey in Libya
Date: January 30, 2019 Ref.: RFQ/19/032 Subject: Request for Quotations for: Nationwide household survey in Libya The International Foundation for Electoral Systems (IFES) invites your firm to participate
More informationCitizens First 8 KEY INSIGHTS AND FINDINGS. Presented by: Dan Batista, Executive Director December 10, Citizens First 8
KEY INSIGHTS AND FINDINGS Presented by: Dan Batista, Executive Director December 10, 2018 1 Outline Background and Approach Key Findings How Are We Doing? Priorities for Improvement Service Expectations
More informationNational Statistics Omnibus Survey - Technical Report April 2004
National Statistics Omnibus Survey - Technical Report April 2004 1. The Sample Interviews are conducted with approximately 1,800 adult individuals (aged 16 or over) in private households in Great Britain
More informationCharacteristics of the Population of Internet Panel Members
Vol. 10, Issue 4, 2017 Characteristics of the Population of Internet Panel Members John M Boyle *, Ronaldo Iachan, Naomi Freedner- Maguire, Tala H Fakhouri ** * Institution: ICF Institution: ICF Institution:
More informationScottish Sector Profile
Scottish Sector Profile 2011 www.alliancescotland.org 1 Introduction 01 2 Summary of findings 02 3 Sector characteristics 03 4 Recruitment 05 5 Qualifications sought from candidates 06 6 Skill gaps 07
More informationMKT 450 Final Exam Review
MKT 450 Final Exam Review The final exam will be constructed as follows: 1. Approximately 10 multiple-choice questions on the global research issues and marketing ethics material covered since exam three.
More informationSocial Impact Series
Social Impact Series WESTERN AUSTRALIA Issue #8 July 2017 Outcomes Measurement in the Community Sector: Are we Heading in the Right Direction? Authors: Zoe Callis, Paul Flatau and Ami Seivwright The University
More informationCharacteristics of the Population of Internet Panel Members
Vol. 10, Issue 4, 2017 Characteristics of the Population of Internet Panel Members John M Boyle 1, Ronaldo Iachan 2, Naomi Freedner- Maguire 3, Tala H Fakhouri 4 Survey Practice 10.29115/SP-2017-0025 Oct
More informationFinancial Services Authority. Equality and diversity workforce data report
Financial Services Authority Equality and diversity workforce data report 2009-2010 Contents Introduction 5 Scope 5 Reporting principles 7 Reporting periods 7 Reporting by gender 7 Reporting by race 12
More informationFear at Work in Britain
Fear at Work in Britain First Findings from the Skills and Employment Survey 2012 Duncan Gallie, Alan Felstead, Francis Green and Hande Inanc HEADLINES Fear at work can take several forms worry about loss
More informationPROFESSIONAL AND PERSONAL BENEFITS OF A DIRECT SELLING EXPERIENCE. ROBERT A. PETERSON, PHD The University of Texas at Austin
PROFESSIONAL AND PERSONAL BENEFITS OF A DIRECT SELLING EXPERIENCE ROBERT A. PETERSON, PHD The University of Texas at Austin EXECUTIVE SUMMARY Direct selling is simultaneously a channel of distribution
More informationWhat kind of jobs does the District have? The District is divided into three employment groups classified, academic, and unclassified.
EMPLOYMENT WITH THE DISTRICT What kind of jobs does the District have? The District is divided into three employment groups classified, academic, and unclassified. Classified service includes employees
More informationSmart energy outlook. February 2017
Smart energy outlook February 2017 Smart energy outlook February 2017 1 2 Contents Executive summary 4 Smart meters - 6 the verdict from those who already have one Case study - Eve Ogden 14 Understanding
More informationSURVEY DESIGN AND IMPLEMENTATION
NAMIBIA LABOUR FORCE SURVEY SURVEY DESIGN AND IMPLEMENTATION Introduction In Namibia Labour Force Surveys are carried out under the National Household Survey Programme, which had been launched after the
More informationDate: March 5, Ref.: RFQ Subject: Request for Quotations for Pre-Election Survey Firm in Nigeria
Date: March 5, 2018 Ref.: RFQ-18-022 Subject: Request for Quotations for Pre-Election Survey Firm in Nigeria The International Foundation for Electoral Systems (IFES), invites your firm to participate
More informationThe Fight Against Attrition National Population Health Survey (NPHS)
The Fight Against Attrition National Population Health Survey (NPHS) Methodology of Longitudinal Surveys University of Essex 12-14 July 2006 France Bilocq Chief, Health Statistics Division Statistics Canada
More information9-3. Learning Objectives
9-1 Surveys Learning Objectives 9-3 Understand... The process for selecting the appropriate and optimal communication approach. Factors affect participation in communication studies. Sources of error in
More information9-3. Learning Objectives
9-1 Surveys Learning Objectives 9-3 Understand... The process for selecting the appropriate and optimal communication approach. Factors affect participation in communication studies. Sources of error in
More informationNon-Response Reduction Methods in Establishment Surveys
Non-Response Reduction Methods in Establishment Surveys Jaki S. McCarthy, USDA s National Agricultural Statistics Service Morgan Earp, Bureau of Labor Statistics Presented at the Fourth International Conference
More informationAnimal Cloning. American Anti-Vivisection Society. Produced for. Prepared by. December 22, 2006
Animal Cloning Produced for American Anti-Vivisection Society Prepared by December 22, 2006 Copyright 2006. Opinion Research Corporation. All rights reserved. Table of Contents Page Methodology...2 Executive
More informationPurpose Remit Survey UK report Winter
Purpose Remit Survey UK report 2012-2013 Authors: Andy Scott and Gareth Morrell Date: June 2013 Prepared for: BBC Trust At NatCen Social Research we believe that social research has the power to make life
More informationMiddle Managers Outlook: Australia Overview of Findings October 2008
Middle Managers Outlook: Australia Overview of Findings October 2008 Accenture, its logo, and High Performance Delivered are trademarks of Accenture. Background / Methodology In October 2008, Accenture
More informationGetting the Bead of Hardship
Getting the Bead of Hardship Benjamin Messer, Jordan Folks, and Jane Peters, Research Into Action, Portland, OR Alex Dunn, Portland, OR Prapti Gautam, Southern California Edison, Rosemead, CA ABSTRACT
More informationCompeting Goals of Responsive Design in a Total Survey Error Framework: Minimization of Cost, Nonresponse Rates, Bias, and Variance
Competing Goals of Responsive Design in a Total Survey Error Framework: Minimization of Cost, Nonresponse Rates, Bias, and Variance Andy Peytchev International Total Survey Error Workshop August 2, 2012
More informationRESEARCH SUMMARY THE DIGITAL DIVIDE: COMPUTER USE, BASIC SKILLS AND EMPLOYMENT
RESEARCH SUMMARY THE DIGITAL DIVIDE: COMPUTER USE, A COMPARATIVE STUDY IN PORTLAND, USA AND LONDON, ENGLAND John Bynner, Steve Reder, Samantha Parsons and Clare Strawn OCTOBER 2008 2 RESEARCH SUMMARY INTRODUCTION
More informationZenith Model Recalibration and Validation Version Review of VISTA. February Public Transport Victoria
Zenith Model Recalibration and Validation Version 3.0.0 Review of VISTA February 2014 Public Transport Victoria Page Intentionally Left Blank Review of VISTA Draft Report Project No. ZML-VIC-Year4 COPYRIGHT:
More informationThe World Bank Listening to LAC (L2L) Pilot Project. Report on Attrition of Panel Participants in Peru and Honduras
The World Bank Listening to LAC (L2L) Pilot Project Report on Attrition of Panel Participants in Peru and Honduras September 2012 0 Contents Background... 2 Attrition Analysis... 3 Attrition in Peru...
More informationMaximizing Home Energy Report Savings: Who Saves the Most, the Least and Why?
Maximizing Home Energy Report Savings: Who Saves the Most, the Least and Why? ABSTRACT Hannah Arnold, Opinion Dynamics, Oakland, CA Olivia Patterson, Opinion Dynamics, Oakland, CA Seth Wayland, Opinion
More informationTerms of Reference for i2i Credit Pilot Study in Zimbabwe. June 2017
Terms of Reference for i2i Credit Pilot Study in Zimbabwe June 2017 1 About i2i Insights2Impact (i2i www.i2ifacility.org) is a resource centre aimed at catalysing the use of data to improve financial inclusion.
More informationAssessing risk of nonresponse bias and dataset representativeness during survey data collection
Assessing risk of nonresponse bias and dataset representativeness during survey data collection Gabriele Durrant Joint work with Jamie Moore, Solange Correa and Peter W.F. Smith University of Southampton
More informationApplying the Tailored Design Method in a Randomized Control Trial Experiment Survey
Agenda Applying the Tailored Design Method in a Randomized Control Trial Experiment Survey Benjamin Messer Research Into Action, Portland, OR PAPOR Annual Conference, San Francisco, CA December 14-15,
More informationMode-Switch Protocols: How a Seemingly Small Design Difference can affect Attrition Rates and Attrition Bias
Understanding Society Working Paper Series No. 2012 07 December 2012 Mode-Switch Protocols: How a Seemingly Small Design Difference can affect Attrition Rates and Attrition Bias Peter Lynn Institute for
More informationOREGON ELECTRICITY SURVEY
OREGON ELECTRICITY SURVEY by Stephen M. Johnson, Ph.D., Associate Director with the assistance of Kimberlee Langolf January 1999 OREGON SURVEY RESEARCH LABORATORY UNIVERSITY OF OREGON EUGENE OR 97403-5245
More informationA plan for ensuring that our workforce reflects the communities we serve August 2017
A plan for ensuring that our workforce reflects the communities we serve August 2017 Page1 Foreword from Assistant Chief Constable Liane James As a Force, we recognise the benefits of employing a diverse
More informationThe impact of using web in the Danish LFS
The impact of using web in the Danish LFS Background breaks in time series The Danish core part of the Labour Force Survey (LFS) consists of approximately 19000 interviews each quarter. From January 2016
More informationWGO s response to the Waste Reduction by Waste Charging How to Implement document
22 nd January, 2013 TO: Council for Sustainable Development WGO s response to the Waste Reduction by Waste Charging How to Implement document 1. Summary a) The active discussion in our community over the
More informationPART THREE Guidelines for Survey Fieldwork and Panel Maintenance. Andrej Kveder
PART THREE Guidelines for Survey Fieldwork and Panel Maintenance Andrej Kveder Guidelines for Survey Fieldwork and Panel Maintenance 47 1. Fieldwork Guidelines The purpose of these guidelines is to provide
More informationRisk Perceptions of Urban Italian and United States Consumers for Genetically Modified Foods. Related Literature
AgBioForum, 7(4): 195-201. 2004 AgBioForum. Risk Perceptions of Urban Italian and United States Consumers for Genetically Modified Foods R. Wes Harrison Louisiana State University Stefano Boccaletti Istituto
More informationContact: Version: 2.0 Date: March 2018
Survey Sampling Contact: andrew.ballingall@fife.gov.uk Version: 2.0 Date: March 2018 Sampling allows you to draw conclusions about a particular population by examining a part of it. When carrying out a
More informationAgency Information Collection Activities: Request for Comments for Periodic Information Collection
DEPARTMENT OF TRANSPORTATION Federal Highway Administration Docket No. FHWA-2015-0004 This document is scheduled to be published in the Federal Register on 02/19/2015 and available online at http://federalregister.gov/a/2015-03462,
More informationUSING THE INTERNET AS A SURVEY MEDIUM: LESSONS LEARNT FROM A MOBILITY SURVEY OF YOUNG DRIVERS
USING THE INTERNET AS A SURVEY MEDIUM: LESSONS LEARNT FROM A MOBILITY SURVEY OF YOUNG DRIVERS Warren Harrison Eastern Professional Services Pty Ltd Ron Christie RCSC Services Pty Ltd ABSTRACT The internet
More informationIt s Déjà Vu All Over Again: More Revelations from a Lighting Panel Study
It s Déjà Vu All Over Again: More Revelations from a Lighting Panel Study Kiersten von Trapp, NMR Group, Inc., Somerville, MA Melissa Meek, NMR Group, Inc., Somerville, MA Scott Walker, NMR Group, Inc.,
More informationINFLUENCE OF SEX ROLE STEREOTYPES ON PERSONNEL DECISIONS
Journal of Applied Psychology 1974, Vol. 59, No. 1, 9-14 INFLUENCE OF SEX ROLE STEREOTYPES ON PERSONNEL DECISIONS BENSON ROSEN 1 AND THOMAS H. JERDEE Graduate School of Business Administration, University
More information