Life in Australia™ methods

Summary documentation

August 2024

Introduction

Established in November 2016, Life in Australia™ is Australia’s first and only national probability-based online panel. The panel is the most methodologically rigorous online panel in Australia and is one of only a small number of probability-based online panels worldwide. Members of Life in Australia™ are randomly recruited via traditional, high coverage sampling frames such as random digit dialling (RDD) or residential addresses and agree to provide their contact details to take part in surveys on a regular basis. Unlike other research panels, Life in Australia™ includes people with and without internet access, surveying offline panellists via telephone. Research conducted by the Social Research Centre shows Life in Australia™ produces survey estimates of comparable accuracy to other major approaches for conducting probability surveys and superior to estimates derived from opt-in online panels, the most frequently used method of undertaking surveys. Life in Australia™ hosts surveys every two weeks and comprises around 10,000 adults aged 18 years and over from across Australia. A comprehensive documentation of methods used by Life in Australia™ can be found here.

Sample design

Recruitment to Life in Australia™

The target population for Life in Australia™ is adults aged 18+ years resident in Australia. Life in Australia™ panellists have been recruited using a variety of probability sampling frames and survey modes. These have included random digit dialling (RDD) with computer-assisted telephone interviewing (CATI) in 2016 and 2018, address-based sampling with push-to-web where the primary mode of contact was mail together with supplementary use of CATI in 2019‒2021, SMS push-to-web where the sole mode of contact is text message using mobile RDD sample in 2021, 2023, and 2024, and interactive voice response (i.e. pre-recorded voice call) using mobile RDD sample in 2020 only.

Sample selection for surveys

Our standard approach in sample selection for general population surveys selects stratified random samples of Life in Australia™ panellists defined by age (18–34, 35–44, 45–54, 55–64, 65+), gender, education (less than a bachelor’s degree, bachelor’s degree or above) and speaking a language other than English at home. To come as close as possible to population norms on these variables, target numbers of completed surveys by stratum are set based on population proportions. Once stratum targets have been assigned, individual respondents are selected using a procedure designed to control burden across panel members.

A variety of approaches are used for sample selection for special populations based on the target population and client needs. Examples of surveys of special populations that have been fielded on Life in Australia™ include: state or territory specific samples, samples focusing on a particular demographic (e.g. people who were recorded female at birth), and longitudinal samples.

Field methods

Frequency and timing of waves

Typically, two Life in Australia™ waves are fielded per month, except for the second half of December and first half of January, where waves are not fielded due to the Christmas / summer holidays. Life in Australia™ is in field for 2 weeks. Waves are usually released to field on a Monday afternoon and closed out two Mondays later.

Surveys per wave

Waves may include multiple, independent surveys on behalf of different clients to different samples of panellists. Sample selected for surveys in a wave may overlap, although we use a burden control technique to ensure that invitations are spread as evenly as possible.

Omnibus surveys

Life in Australia™ occasionally runs brief so-called ‘omnibus’ surveys at the end of a primary survey. The introduction to the omnibus module clearly identifies the client for each question.

Contact methodology

The standard contact methodology adopted for online Life in Australia™ members is an initial survey invitation via email and SMS, followed by multiple email reminders and a reminder SMS. Up to 5 reminders in different modes (including email, SMS, and telephone) are administered within the fieldwork period. Telephone non-response of online panel members who have not yet competed the survey commences in the second week of fieldwork and consists of reminder calls encouraging completion of the online survey. Offline members with a valid mobile telephone number are also sent a short SMS invitation that contained a link to the survey as well as the reminder SMS halfway through fieldwork. We may deviate from this protocol in cases where the number of completed surveys midway through fieldwork appears likely to come over or under the budgeted amount.

Language of interview

Interviewing is conducted in English only.

Incentives

All members are offered an incentive to complete the survey. The incentives offered for completing the survey have a value of $10 for surveys up to 20 minutes in length and are incremented by $5 for every 5 minutes beyond that. Incentive options include Coles / Myer gift cards (offline panellists only), points redeemable as an electronic gift card from GiftPay, and charitable donations to a designated charity out of five selected charities offered. Panellists can also choose to opt out of receiving an incentive.

Response outcomes

The Social Research Centre uses standard industry definitions for calculating outcome rates. The completion rate (COMR) represents completed interviews as a proportion of all Life in Australia™ members invited to participate in each survey. The overall completion rate for a full panel survey is approximately 75% to 80%, resulting in a possible final sample size up to 8,000 (assuming all panel members are invited). We also report the cumulative response rate, which takes into account non-response to the invitation to join Life in Australia™, complete the panel profile, and attrition from the panel. Including all these points at which non-response can occur, cumulative response rates are around 4.5% as of August 2024.

Panel profile

Life in Australia™ collects extensive information on panellists as part of the recruitment profile survey. In addition, we refresh profile information about once a year.

Standard data file inclusions

The following panel variables are included as standard with full length surveys:

State/territory of residence
Resident in capital city or rest of state (Greater Capital City Statistical Areas)
State/territory of residence × capital city or rest of state
Socio-economic indexes for areas (Index of Relative Socio-economic Advantage and Disadvantage, national quintiles)
Gender
Age group (18‒24, 25‒34, 35‒44, 45‒54, 55‒64, 65‒74, 75+)
Country of birth group (Australia, main English-speaking countries [Canada, Ireland, New Zealand, South Africa, UK, USA], non-English-speaking countries [other countries])
Citizenship status
Uses a language other than English at home
Aboriginal or Torres Strait Islander status
Family household composition
Highest level of education attained

Additional profile items

Other data captured in the profile includes a number of other questions that can be added to Life in Australia™ datasets at a reasonable cost.

Data processing and outputs

Data quality checks for online completes

Data quality checks for online completes consist of checks for the following, where the specific checks made will depend on questionnaire content:

Logic checks
Proportion of ‘don’t know’ and ‘refused’ responses
Speeding
Straightlining
Verbatim responses to open-ended questions

We consider all these indicators when determining whether a respondent is removed for poor data quality. Data quality indicators other than verbatim responses are used to identify potentially problematic cases. Generally, verbatim responses are decisive, with those indicating thoughtful engagement with the survey being kept and others being removed (e.g. nonsense responses like ‘asdfgh,’ non sequiturs, profanities).

Data quality is tracked for panel members over time and those with repeated issues are retired from the panel.

After these checks, cases are removed due to poor data quality and are not counted toward the completion rate.

Weighting

Life in Australia™ weights are created in four steps:

Panel weights. Weights are created for the probability of selection into the panel and retention in the panel (the latter for existing panellists only) using a model-based approach.
Probability of selection weights. Weights adjust for the likelihood of selection from the panel to be invited to a specific survey.
Response propensity weights. For the sample invited to complete a specific survey, response propensity weights are created. A logistic regression model is used to predict the likelihood of each panel member completing the specific survey, conditional on characteristics available for both respondents and non-respondents. The model incorporates a wide range of demographic, attitudinal, and behavioural characteristics collected from all panel members. Weights are calculated for propensity classes.
Post-stratification weights. The weights are then adjusted to population benchmarks. For general population surveys, in most cases a ‘streamlined’ approach is used, which calibrates the sample to number of adults in the household, age by highest educational attainment, gender, use of a language other than English at home, region (capital city, rest of state), and state or territory.

Accreditation

All aspects of research conducted by the Social Research Centre are undertaken in accordance with ISO 20252:2019 Market, Opinion and Social Research Standard, the Research Society (formerly AMSRS) Code of Professional Behaviour, the Australian Privacy Principles, and the Privacy (Market and Social Research) Code 2021.

The Social Research Centre is an accredited Company Partner of The Research Society with all senior staff as full members and several senior staff QPR accredited. The Social Research Centre is also a member of the Australian Data and Insights Association and bound by the Privacy (Market and Social Research) Code 2021.

Menu