ANU Centre for Social Research & Methods

Permanent URI for this collection

The ANU Centre for Social Research and Methods (CSRM) research focuses on: - The development of social research methods - Analysis of social issues and policy - Training in social science methods - Providing access to social scientific data


Recent Submissions

Now showing 1 - 20 of 70
  • ItemOpen Access
    The role of informal wage negotiations in explaining the gender wage gap
    (Centre for Social Research & Methods, Australian National University, 2017) Shankar, Sriram; Gray, Matthew
    It is well known in the gender wage gap literature that wage differentials between men and women can be explained by a range of factors including differences in productivity, occupational segregation, and wage discrimination. In this paper we use the Fair Work Commission’s 2014 Australian Workplace Relation study (AWRS) linked Employer- Employee data set to estimate the contribution that individual level negotiation of wage/salary between the employee and employer makes to the gender wage gap. Preliminary analysis of the 2014 AWRS reveals that men are more likely to attempt to gain a better wage/salary through negotiation with their manager and are more likely to be successful if they attempt to negotiate than are women. The paper uses regression based Blinder-Oaxaca (BO) decomposition we measure (both in terms of sign and magnitude) how much each of the variables capturing how bargaining about wage/salary contributes (both in dollar and percentage terms) towards the overall gender wage gap. Our results indicate that once we control for bargaining the gender wage gap is significantly reduced. In other words gender wage gap is found to be much less among those who bargain relative to those who don’t.
  • ItemOpen Access
    Longitudinal Study of Indigenous Children (LSIC) technical report: education
    (The Australian National University, 2019) Biddle, Nicholas; Edwards, Ben; Lovett, Raymond; Radoll, Peter; Sollis, Katherine; Thurber, Katerine
    This report evaluates the education measures in the Longitudinal Study of Indigenous Children (LSIC). Education measures in the LSIC were found to be internally valid and perform as expected. The LSIC is a robust dataset that, if used carefully, can improve our understanding of the development of Indigenous children, and help design good public policy. For analysts, we recommend using the data with confidence, while remaining aware that some variables perform better than others and that models using the education measures (especially those specific to the LSIC) tend to have low explanatory power. We also recommend taking advantage of the longitudinal data rather than the cross-sectional data. For reviewers of papers based on LSIC data, we recommend taking into account the unique circumstances of the survey, and that models will be estimated with low precision and with variables that differ from those collected in other datasets. Finally, for policy makers, we recommend making decisions using longitudinal research and considering funding a top-up sample.
  • ItemOpen Access
    Panel mixed-mode effects: does switching modes in probability-based online panels influence measurement error?
    (The Australian National University, 2020) Kocar, Sebastian; Biddle, Nicholas
    Online probability-based panels often apply two or more data collection modes to cover online and offline populations, and to collect data from onliners who do not respond online in time to contribute to a given wave. As a result, offline/ online status can change during the life of the panel for some individuals, which can improve response rates and representativeness, but may cause increased measurement error. In this study, we use Life in Australia™ survey data and online panel paradata to identify respondents who switched modes; almost 4% of the whole panel was interviewed using both online and offline modes in the first 2-years, and almost one-third of those 4% switched mode more than once. We selected all repeated substantive survey items, identified any relevant changes in responses that could be explained with mode effects, and determined the effect of mode switching on changes to answers, controlling for panel conditioning, panel fatigue and sociodemographic characteristics of panellists. This study identified a limited number of panel mode effects from panellists switching modes of data collection over time. We found evidence of recency and some social desirability, and established that measurement error may be more common when the proportion of mode switchers is higher. Moreover, panel conditioning had an effect on the frequency of changing answers; respondents provided more stable answers if they were more conditioned. We conclude that combining mode effects with panel conditioning, as well as an increasing representation bias over time, may lead to less accurate estimations in longitudinal surveys.
  • ItemOpen Access
    Consent to data linkage in a child cohort study, Growing Up in Australia: The Longitudinal Study of Australian Children
    (The Australian National University, 2019) Bandara, Dinusha; Edwards, Ben; Mohal, Jatender; Daraganova, Galina
    In this paper, we provide new evidence on the factors associated with consent to data linkage in young people from a sample of 16-17-yearolds (born in 2004) participating in Growing Up in Australia: The Longitudinal Study of Australian Children. To our knowledge, this survey is the first time consent to data linkage has been studied in young people under 18-years of age. We extend the existing literature by examining economic record linkage, which is of particular concern in longitudinal surveys given the complexity of income support, benefits and pensions in many developed countries. The findings show that young people's consent to data linkage is above 80%. The findings also shed light on significant demographic and psychosocial correlates of consent to data linkage, the influence of prior commitment to the survey by the household, the influence of previous consent to other forms of data linkage and the influence of family members' decisions on consent from the young person.
  • ItemOpen Access
    A new method of estimating the number of Indigenous business owner-managers
    (The Australian National University, 2020) Shirodkar, Siddharth; Hunter, Boyd; Foley, Dennis
    Accurate estimates of the size of the Indigenous business sector are valuable for policy makers, practitioners and academics. Such estimates provide one measure of Indigenous economic advancement. A difficulty in developing these estimates has been a lack of suitable data for calculating key unknowns, such as changes in Indigenous identification by individuals. Using the three-wave Australian Census Longitudinal Dataset (ACLD), which links individuals between the past three (2006, 2011 and 2016) Censuses of Population and Housing, this paper addresses these data gaps and presents a transparent methodology for estimating the number of Indigenous business owner-managers. Using information about the greater rate of Indigenous self-identification over time from the ACLD, we estimate that around 19,400-Indigenous Australians were business owner-managers in 2016, almost double the number in 2006. We also estimate that Indigenous business ownership as a share of the Indigenous working-age population grew from around 3.3% in 2006 to around 3.7% in 2016. This increase occurred at a time when the rate of non-Indigenous business ownership decreased from 10.0% in 2006 to around 8.7% in 2016, reflecting the ongoing consolidation of the global economy since the global financial crisis. Although the Indigenous business ownership rate remains low compared with the non-Indigenous rate, the continued growth during challenging times is testament that greater numbers of Indigenous people are aspiring to the opportunity and ambition that business ownership affords. But barriers such as implicit or unconscious bias in society still exist and may limit opportunities for faster Indigenous economic advancement.
  • ItemOpen Access
    The Online Panels Benchmarking Study: a Total Survey Error comparison of fndings from probability-based surveys and nonprobability online panel surveys in Australia
    (The Australian National University, 2018) Pennay, Darren; Neiger, Dina; Lavrakas, Paul J.; Borg, Kim
    The pervasiveness of the internet has led online research, and particularly online research undertaken via nonprobability online panels, to become the dominant mode of sampling and data collection used by the Australian market and social research industry. There are broad-based concerns that the rapid increase in the use of nonprobability online panels in Australia has not been accompanied by an informed debate about the advantages and disadvantages of probability and nonprobability surveys. The 2015 Australian online Panels Benchmarking Study was undertaken to inform this debate, and report on the fndings from a single national questionnaire administered across three different probability samples and fve different nonprobability online panels. This study enables us to investigate whether Australian surveys using probability sampling methods produce results different from Australian online surveys relying on nonprobability sampling methods, where accuracy is measured relative to independent population benchmarks. In doing so, we build on similar international research in this area, and discuss our fndings as they relate to coverage error, nonresponse error, adjustment error and measurement error.
  • ItemOpen Access
    Text messages and reminder calls in student and alumni web surveys
    (The Australian National University, 2019) Phillips, Benjamin; Compton, Shane
    The impact of text messages (SMS) and reminder calls on response to web surveys was experimentally tested using a crossed design on two surveys in the Quality Indicators for Learning and Teaching suite of studies for the Australian Government Department of Education: the May 2018 Graduate Outcomes Survey and the 2018 Student Experience Survey. SMS and telephone reminders were associated with increased probability of response for both surveys. Telephone reminders were more effective than SMS on a per-contact basis. However, taking into account the higher cost of a reminder call than a reminder SMS, sending an SMS reminder was more cost-effective. This research adds to the limited literature on the effectiveness of reminder calls and SMS for surveys.
  • ItemOpen Access
    Measuring natural disasters through self-report: the case of a national child cohort study
    (The Australian National University, 2019) Edwards, Ben; Gray, Matthew; Borja, Judith B.
    Children are considered to be disproportionately affected by natural disasters related to climate change. The impacts on the development of children of being exposed to multiple natural disasters are not well understood. This paper reports on the development and validation of a cumulative measure of exposure to natural disasters (2013-17) at the area level, as well as an individual-level measure of the impact of these natural disasters using data from the Longitudinal Cohort Study on the Filipino Child and linked data from the International Disaster Database (EM-DAT). First, we show that a caregiver-reported measure of cumulative exposure to natural disasters had statistically significant associations with disasters reported by officials responsible for the geographic area and with disasters in EM-DAT. A substantial proportion of the variation in individual reports of exposure to natural disasters occurred at the area level (25%), supporting the idea that taking community averages reflects a consensus of the exposure to natural disasters. We then generated a community average measure of exposure to natural disasters, based on neighbours' reports but not individual self-reports - therefore providing an exogenous measure of disaster exposure in the local area for each household. Second, we show that this community measure was more strongly related to EM-DAT and barangay (small administrative unit) official reports than individual household reports. Third, many household factors (e.g. quality of housing) will mitigate the impact of a natural disaster. Even though exposure to a natural disaster may be a shared experience, we develop an individual-level measure of disaster impacts. Importantly, this measure of disaster impact was associated with measures of exposure (individual and community average), community ratings by officials and EM-DAT. However, the impact measure was only moderately associated with the community average exposure. Both the community average and disaster impacts measures were consistently related to household income and the adequacy of income in households. We discuss the implications of our study for more nuanced measures of disaster exposure and monitoring.
  • ItemOpen Access
    Building a probability-based online panel: Life in Australia™
    (The Australian National University, 2019) Kaczmirek, Lars; Phillips, Benjamin; Pennay, Darren; Lavrakas, Paul J.; Neiger, Dina
    Life in Australia™ was created to provide Australian researchers, policy makers, academics and businesses with access to a scientifically sampled cross-section of Australian resident adults at a lower cost than telephone surveys. Panellists were recruited using dual-frame landline and mobile random digit dialling. The majority of panellists choose to complete questionnaires online. Representation of the offline population is ensured by interviewing by telephone those panellists who cannot or will not complete questionnaires online. Surveys are conducted about once a month, covering a variety of topics, most with a public opinion or health focus. Full panel waves yield 2000 or more completed surveys. Panellists are offered a small incentive for completing surveys, which they can choose to donate to a charity instead. This paper describes how Life in Australia™ was built and maintained before the first panel refreshment in June 2018. We document the qualitative pretesting used to inform the development of recruitment and enrolment communications materials, and the pilot tests used to assess alternative recruitment approaches and the comparative effectiveness of these approaches. The methods used for the main recruitment effort are detailed, together with various outcome rates. The operation of the panel after recruitment is also described. We assess the performance of the panel compared with other probability surveys and nonprobability online access panels, and against benchmarks from high-quality sources. Finally, we assess Life in Australia™ from a total survey error perspective.
  • ItemOpen Access
    A universal global measure of univariate and bivariate data utility for anonymised microdata
    (The Australian National University, 2018) Kocar, Sebastian
    A universal global measure of univariate and bivariate data utility for anonymised microdata. This paper presents a new global data utility measure, based on a benchmarking approach. Data utility measures assess the utility of anonymised microdata by measuring changes in distributions and their impact on bias, variance and other statistics derived from the data. Most existing data utility measures have significant shortcomings – that is, they are limited to continuous variables, to univariate utility assessment, or to local information loss measurements. Several solutions are presented in the proposed global data utility model. It combines univariate and bivariate data utility measures, which calculate information loss using various statistical tests and association measures, such as two-sample Kolmogorov–Smirnov test, chi-squared test (Cramer’s V), ANOVA F test (eta squared), Kruskal-Wallis H test (epsilon squared), Spearman coefficient (rho) and Pearson correlation coefficient (r). The model is universal, since it also includes new local utility measures for global recoding and variable removal data reduction approaches, and it can be used for data protected with all common masking methods and techniques, from data reduction and data perturbation to generation of synthetic data and sampling. At the bivariate level, the model includes all required data analysis steps: assumptions for statistical tests, statistical significance of the association, direction of the association and strength of the association (size effect). Since the model should be executed automatically with statistical software code or a package, our aim was to allow all steps to be done with no additional user input. For this reason, we propose approaches to automatically establish the direction of the association between two variables using test-reported standardised residuals and sums of squares between groups. Although the model is a global data utility model, individual local univariate and bivariate utility can still be assessed for different types of variables, as well as for both normal and non-normal distributions. The next important step in global data utility assessment would be to develop either program code or an R statistical software package for measuring data utility, and to establish the relationship between univariate, bivariate and multivariate data utility of anonymised data.
  • ItemOpen Access
    The social determinants of health and subjective wellbeing: a comparison of probability and nonprobability online panels
    (The Australian National University, 2018) Biddle, Nicholas; Sinibaldi, Jennifer; Sheppard, Jill
    As response rates to surveys decline all over the world, researchers are increasingly turning to sampling frames that are easier and cheaper to reach, and that have more predictable response rates. These include nonprobability web panels (NWPs) and probability web panels (PWPs). Although generally more expensive to construct, the latter have been shown in many instances to suffer from fewer biases and deviation from benchmarks. The literature comparing NWPs with PWPs is fedgling. We add to this research area by comparing measures of the social determinants of health that were estimated from a number of NWPs and PWP equivalents with a high-quality benchmark. The analysis finds that, when looking at the distributions of self- assessed health and life satisfaction, probability panels differ less from the gold standard than do nonprobability panels. This supports previous work, although we also show that this conclusion holds when a greater range of control variables is included in the model. However, some of the predictors of health are captured better using the nonprobability panels. In particular, the relationship between area-level disadvantage and health is better captured through a pooled nonprobability sample.
  • ItemOpen Access
    Measuring social attitudes with voter advice application data
    (The Australian National University, 2018) Sheppard, Jill
    This study directly compares survey data on social attitudes collected from an opt-in sample of voter advice application (VAA) users and a randomly recruited, probability-based online panel of respondents. Whereas much research to date has focused on the demographic representativeness of VAA-generated data, less is known about the attitudinal and other representativeness of these data. The findings from these Australian samples contribute to the emerging international literature. VAAs are proliferating as a source of 'big data' among public opinion and political science researchers, despite concerns about the representativeness of the opt-in samples. During July 2016, VAA developer Election Compass collected email address details for approximately 40,000 Australian users of its application in the weeks before the 2016 Australian federal election. In November 2016, this study surveyed the sample of VAA users on their attitudes towards a range of Australian social issues. In December 2016, the same questionnaire was administered to a probability-based sample, using an identical mode of administration and similar response maximisation techniques. The questionnaire contains a broad range of questions designed to identify dimensions of sociopolitical attitudes in Australian society. Comparing the composition of dimensions and relationships between variables within the data contributes to our understanding of incidental samples such as VAA users, and the extent to which we can and should make inferences from VAA-generated data. Comparison of point estimates of the unweighted and weighted datasets, and of estimates of the internal relationships between variables within both datasets suggests that VAA user data should not be considered externally valid sources of social attitude data, even after adjustment. Further, researchers and reviewers should use greater caution when adjusting VAA user data on the basis of observable measures (with wellestablished population parameter estimates) to explain unobservable measures without robust population parameter estimates (such as social attitudes).
  • ItemOpen Access
    Faith-based communities' responses to family and domestic violence
    (The Australian National University, 2020) Truong, Mandy; Sharif, Mienah; Pasalich, Dave; Olsen, Anna; Calabria, Bianca; Priest, Naomi
  • ItemOpen Access
    Evaluation of the Speak Out Against Racism (SOAR) program pilot
    (The Australian National University, 2020) Priest, Naomi; Oishee, Alam; Dunn, Kevin; Nelson, Jacqueline; Sharples, Rachel; Cronin, Darryl; Truong, Mandy; Francis, Kate; Paradies, Yin; Curry, Philip; Kavanagh, Anne
    The Speak Out Against Racism (SOAR) project is a major research study focused on understanding and addressing experiences and attitudes to racism and racial discrimination, and bystander responses to racism and racial discrimination in Australian schools.
  • ItemOpen Access
    Model of the Australian tax and transfer-system: a flexible open-source approach to tax-transfer-modelling
    (The Australian National University, 2018) Taylor, Matthew
    This paper introduces a new model of Australia's tax and transfer system: Model of the Australian Tax and Transfer System (MATTS). MATTS is a suite of Stata commands that provide researchers with the ability to model individual tax and transfer policies. The MATTS suite can be applied to a range of taxtransfer modelling problems and methodologies, and is freely available to anyone with an interest in tax-transfer research. This paper presents a specific application in which MATTS is used to illustrate how Australia's tax-transfer system augments the disposable incomes and effective marginal tax rates of single income support recipients. These simple examples illuminate some of the trade-offs involved in the design of means-tested tax-transfer systems.
  • ItemOpen Access
    Support for policy trials in Australia: level and predictors
    (The Australian National University, 2018) Biddle, Nicholas; Gray, Matthew
    An emerging view from a broad array of fields is that there should be greater use of evaluations of public policies in general, and the use of randomised controlled trials (RCTs), where feasible, in particular to test the effectiveness of new policies for which there is limited or no evidence about their likely impact. This is because of the potential for RCTs to provide reliable estimates of the causal impacts of the policy being trialled and considered for wider application. There is less evidence, however, on the level of support for such trials as a tool for policy among the general population. In this paper, we provide a summary of an online survey experiment that tested the level of support, and factors associated with support, for policy trials, and RCTs in particular. We found that about half the population supported a trial for a (hypothetical) policy intervention as opposed to introducing the policy to everyone at once. However, only around one-fifth of the population supported implementation of that trial through random assignment. We also found that (randomly assigned) policy area, support from experts for the policy, and party background of the policy instigator had large and significant associations with the level of public support for trials. We conclude that experts and policy makers who support trials in general and RCTs in particular need to engage with the community to explain the benefits, and to learn from community concerns.
  • ItemOpen Access
    A basic income for Australia? Exploring rationale, design, distribution and cost
    (The Australian National University, 2019) Ingles, David; Phillips, Ben; Stewart, Miranda
    This paper considers the potential for a basic income (BI) or guaranteed minimum income (GMI) scheme for Australia. We examine the proposal for a GMI advocated by the Henderson Poverty Inquiry in 1975. We briefly discusses the rationale for a BI and then focus on work incentive effects, design and financing of a BI in the Australian context. The paper describes and models four options that would move the current Australian system towards a partial or categorical BI, with an innovative approach of financing the BI by a wealth tax to keep the tax rate on earned income relatively low. Such a BI could help ease the effective marginal tax rates that affect families and welfare recipients, and would provide extra support to those with low or fluctuating incomes. For each option, the paper explores the required tax rate and the distributional outcomes for different family types and incomes.
  • ItemOpen Access
    Findings from the 2017 Speak Out Against Racism (SOAR) student and staff surveys
    (The Australian National University, 2019) Priest, Naomi; Chong, Shiau; Truong, Mandy; Sharif, Mienah; Dunn, Kevin; Paradies, Yin; Nelson, Jacqueline; Oishee, Alam; Ward, Andrew; Kavanagh, Anne
  • ItemOpen Access
    The best of times, the worst of times, or indifferent times: views of Australians on job security and the future of work
    (The Australian National University, 2019) Biddle, Nicholas; Gray, Matthew; Sheppard, Jillian; Taylor, Matthew
    The Australian labour market is changing, with new occupations being created, as others become more precarious. Some of this change is being driven by technological advances; some is due to external factors. Although there has been a considerable amount of research documenting this change and predicting what the labour market will look like in the future, there has been far less research looking at the attitudes of current workers towards their own jobs now and in the future. In this paper, we summarise and analyse a specially targeted survey (the 25th in the ANUPoll series) that looks at the attitudes of a representative sample of the Australian population, and test how these attitudes vary by important demographic, geographic and socioeconomic characteristics. We document different aspects of job security: perceptions of job security in Australia, including by industry; categorisation of different threats to job security; and perceived longer-term changes in work and society. We also undertake a randomised survey experiment that tests for the role of sex, ethnicity and qualifications in how people perceive medium-term labour market change, and compare these results with the views of 'experts'. We show that Australians are relatively relaxed about their current job and are somewhat concerned about finding a new job if they lost theirs, and that the general public has somewhat different views on what jobs are most at risk compared with experts.