Telephone Surveys and the Mobile Phone Only Population Workshop 17, July, 2012.
The 2011 Workshop Day 1 Integrated Public Number Database (Michel Ryan) Is the exclusion of mobile phones from telephone surveys a problem: the U.S. experience? (Paul Lavrakas) Trends in Mobile Phone Only households (Anne Taylor, The S.A. Health Omnibus) Two separate presentations that used Morgan s Single Source data Mark Cooper-Stanbury: The declining proportion of households with a fixed-line telephone connection Dean Martin and Wendy Quinn from ACMA used landline RDD and mobile only households to explore issues relating to telecommunications usage 2
The 2011 Workshop Day 1 Results from the SRC/ ISSR Dual-frame demonstration survey (Darren Pennay) Total sample (n=700). Landline frame (n=400). Mobile frame (n=300). Mobile-only subgroup (n=83). An overview of the practical / methodological issues to consider when mounting telephone surveys which include mobile phone samples (Paul Lavrakas) 3
Address-based sampling The 2011 Workshop - Day 2 Geo-coded National Address File (G-NAF) (Richard Lindsay). John Henstridge from DAA also presented a paper on matching G-NAF to the IPND Using the electoral roll as a sampling frame for telephone surveys (Mark Cooper-Stanbury) Advocacy and next steps How did the US research community respond to the rapid growth in households without a fixed line telephone connection? (Paul Lavrakas) How can the Australian survey research community respond? (Mark Western) 4
Based on the very limited Australian evidence available at the time, the exclusion of persons living in mobile phone only households from telephone surveys seemed to be associated with similar biases as had been found in the U.S. The mobile phone only population: Younger 5
The mobile phone only population: Male Lower socioeconomic status H igher levels of participation in education 6
The mobile phone only population: Renters, lived in neighbourhood for less than 5 years, group households 7
The mobile phone only population: Geographically mobile Overseas born 8
The mobile phone only population: Aboriginal Australians 9
The mobile phone only population: Mental health problems Binge and heavy drinking 10
The mobile phone only population: Smoking Legal and illicit drug use 11
The mobile phone only population: Problem gambling and financial hardship 12
The mobile phone only population: F ewer social supports 13
We are playing catch up on this issue: The U.S. survey research community began reacting to the non-coverage issues raised by cell only households as far back as February 2003 (Cell Phone Sampling Summit 1) when the proportion of cell only households was around 3%. First Australian dual frame survey September 2010 First Australian workshop March 2011, when the proportion of households without a landline telephone connection was already 14% Three major issues to be overcome: 1. No geographic information is attached to a mobile phone number. 2. The need for more robust data for weighting purposes. 3. No leadership is being provided in survey research. 14
Increase in the proportion of adults without a landline telephone. 40 35 30 30.2 32.3 25 24.9 20 21.1 19 15 16.1 14 10 10 5 5 6 7 8 0 Jun 05 Jun 06 Jun 07 Jun 08 Jun 09 Jun 10 Jun 11 Dec-11 Source: National Health Interview Survey (July - Dec, 2011) : Australian Communications and Media Authority (Jan-Jun, 2011) 15
Agenda for today Opening session Social Research Centre Dual-frame Omnibus (Methodology) Social Research Centre Dual-frame Omnibus (Selected Findings) Transport and road safety (Michael Nieuwesteeg, TAC) Sexual activity and sexual experiences (Anthony Smith, ARCSHS) Gambling and Health (Alun Jackson, PGRTC) Morning session 2011 National Survey of Mental Health Literacy and Stigma (Bas Misson, SRC) Surveying Australia s communications users: Integrating the mobile only population into a single survey vehicle (Joseph Di Gregorio, ACMA) NSW Population Health Survey (Margo Barr, NSW Health) 16
Agenda for today Afternoon session Fieldwork and financial considerations in undertaking dual-frame surveys (Graham Challice, SRC) Optimal approaches to weighting (Michele Haynes, ISSR) Measures of data quality across the frames (Paul Lavrakas, Independent consultant) It s not just the mobile phone frame you need to worry about! (Graham Challice, SRC) Latest developments and emerging issues (Paul Lavrakas, Independent consultant) Final session Advocacy initiatives since last workshop (Darren Pennay, SRC) Summing up and next steps (Mark Western, ISSR) 17
Is this where we are at in Australia? There is a general consensus within the survey industry that cellphone only households need to be included in studies of the general public. The growing penetration of CPO households and the noted behavioural differences of individuals within these households make the systematic exclusion of such households from survey samples no longer a viable option (Link and Lai, POQ, Vol. 75, Number 4, Winter, 2011) 18