IN THIS ISSUE
This publication presents results from the Australian Bureau of Statistics (ABS) Personal Fraud Survey, conducted throughout Australia from July 2014 to June 2015 as part of the 2014-15 Multi-Purpose Household Survey (MPHS). This is the third national survey of personal fraud in Australia; the first Personal Fraud Survey was conducted in 2007 and the second in 2010-11.
Topics covered in the survey are card fraud, identity theft and scams (lottery, information request, pyramid scheme, relationship, up-front payment, financial advice, computer support, working from home, online trading or auction, and other types of scam).
The Survey also collects information on the socio-demographic details of persons who experienced personal fraud (such as age, sex and education). The data referred to within this commentary are available to download as data cubes from the Download tab within this publication.
WHAT IS NEW IN THE PERSONAL FRAUD SURVEY?
There have been a number of changes from the 2010-11 Personal Fraud Survey that have affected the availability or comparability of some data items in the 2014-15 Survey. Information about characteristics of incidents of personal fraud are not comparable across the two reference periods. The 2010-11 survey collected detailed characteristics about all incidents in the last 12 months for each fraud type. The 2014-15 survey collected detailed characteristics about the most recent incident only.
Additional questions were asked in the 2014-15 survey on the impact of fraud incidents. These included whether the respondent had discovered the most recent incident of fraud or someone else had informed them, the amount of time lost in the most recent incident, and whether their behaviour has changed as a result of the fraud incident. To accommodate these new questions, some information collected in the 2010-11 survey was not included in 2014-15: the number of cards fraudulently used, whether these were part of a joint account with another person, and the number of joint account holders.
Due to changes in the survey questionnaire wording regarding experience of identity theft, data from 2014-15 are not comparable with those from 2010-11.
The number of response categories for how details were obtained was increased in 2014-15 to reflect current common methods, such as social media. Additional questions were asked in the 2014-15 survey on the impact of identity theft incidents. These included the amount of time lost in the most recent incident, and whether behaviour has changed as a result of the fraud incident.
In 2014-15, information was collected on how the respondent became aware the incident was a scam. In addition, questions were asked on the impact of scam incidents, including the amount of time lost in the most recent incident, and whether behaviour has changed as a result of the fraud incident.
WHAT INFORMATION ABOUT DATA QUALITY IS INCLUDED IN THIS PUBLICATION?
The statistics included in this publication are based on survey data obtained from a sample of the Australian population. The figures contained in the tables are referred to as 'estimates,' and are calculated by adjusting results from a sample survey to produce an estimated result for the entire in-scope population. The difference between the result obtained from surveying a sample of persons rather than the entire in-scope population is referred to as sampling error. As a general rule, sampling error tends to decrease as the size of the sample increases. This means sampling error is likely to be higher for estimates relating to smaller subpopulations and low-prevalence experiences.
The relative standard error (RSE) is a standardised measure of sampling error. Estimates with a RSE of less than 25% are considered sufficiently reliable for most purposes, and these are referred to with a single estimate in the publication commentary. The amount of money lost due to identity theft and card fraud have RSEs between 25% and 50% and are considered less reliable. As these are key data items for users, these data have been presented in the publication commentary with upper and lower bounds for the 95% confidence interval, to assist users in interpreting these estimates. More information about significance testing can be found in the Technical Note.
Due to the relatively small numbers of persons experiencing certain types of crime, some of the estimates provided in the data cubes are subject to high sampling error; these are indicated by footnotes when presented in the publication commentary and through the use of cell comments in data cubes. Estimates with a RSE of between 25% and 49% are considered to be of lower accuracy, and users are advised to use these with caution. Where estimates have a RSE of 50% or more, the RSE value is not available for publication and users are advised that these estimates are considered too unreliable for general use.
All comparisons between populations mentioned in the publication commentary have been tested for statistical significance, with a 95% level of confidence that there is a real difference between the two populations being compared. More information about significance testing can be found in the Technical Note.
HOW IS RESPONDENT CONFIDENTIALITY PROTECTED IN THIS PUBLICATION?
To minimise the risk of identifying individuals in aggregate statistics, a technique called perturbation is used to randomly adjust cell values. Perturbation involves small random adjustment of the statistics and is considered the most satisfactory technique for avoiding the release of identifiable statistics while maximising the range of information that can be released. These adjustments have a negligible impact on the underlying pattern of the statistics. Perturbation has been applied to 2014-15 data presented in this publication.
Where the same statistic is published more than once in different tables, it is perturbed in the same way to ensure consistency across tables. However, cell values may not add up to totals within the same table as a result of perturbation.