Australian Institute of Criminology

Skip to content

Method

Research design

This research employed a quantitative, cross-sectional survey design, examining identity crime and misuse within the sample at one point in time. The operational definition of identity crime and misuse was the use of personal information without permission. This included obtaining or using personal information without permission to pretend to be someone else or to carry out a business in someone else’s name without their permission, or other types of activities or transactions. This definition excluded the use of personal information for direct marketing, even if this was done without permission. For this research, personal information was defined as:

Name, address, date of birth, place of birth, gender, driver’s licence information, passport information, Medicare information, biometric information (eg fingerprint), signature, bank account information, credit or debit card information, passport, personal identification number, tax file number, shareholder identification number, computer and/or other online usernames and passwords, student number and other types of personal information.

Survey questions

The survey contained a mixture of closed response and open-ended questions on the following topics:

  • perceptions of the seriousness of misuse of personal information and how risks will change over the next 12 months;
  • experience of misuse of personal information at any time in the past and over the preceding 12 months;
  • methods of victimisation in respect of the most serious occasion in the preceding 12 months;
  • actual financial losses, funds recovered and other consequences of victimisation;
  • awareness of the availability of court victimisation certificates;
  • reporting misuse of personal information;
  • behavioural changes arising from misuse of personal information; and
  • demographic and other characteristics of the sample including age, gender, place of normal residence, income, language spoken at home, Indigenous background and computer usage.

These questions were developed in consultation with the AGD and the survey was distributed to a number of stakeholders for input and suggestions prior to being deployed. Some of the questions were based on those contained in other similar surveys that have been conducted in Australia and overseas, in order to enable comparison.

The questions spanned a number of reference periods. These included participants’ current circumstances (eg place of normal residence, age and income), their lifetime experiences of identity crime and misuse, as well as identity crime and misuse they had experienced in the previous 12 months. The survey was delivered over a two week period in September 2013.

The survey, which included 23 questions in total, took approximately 10 minutes to complete. No identifying information was requested from respondents. A copy of the online questionnaire is attached at Appendix 1.

Sampling

The survey was administered to an online survey panel by i-Link Research Solutions, an external provider. The sample consisted of 5,000 Australians aged 15 years and over who had internet access and who had registered with the online survey panel provider. The sampling frame and survey hosting was undertaken by i-Link Research Solutions, with the de-identified data being provided to the AIC for analysis and reporting.

Potential respondents were randomly selected and invited to participate in the survey using quotas—namely, location, age and gender. Respondents were stratified across location, so that there was an oversampling in smaller states and territories, and under-sampling of the larger states compared with their representation in the Australian population aged 15 years and over. Age and gender were used as qualifying variables, so that the respondents were nationally representative according to ABS (2013) census data. Sampling was completed once the quotas had been met and a sample size of 5,000 participants had been obtained.

Participants received an incentive in exchange for completing the survey. Participants were able to select the type of reward they wished to receive from the range of incentives provided by the external provider. Examples of the incentives provided by the provider included:

  • instant member reward points (accumulated to redeem gifts—Caltex/Coles vouchers etc);
  • chance to win $50,000 prize draw quarterly;
  • donate rewards to an affiliated charity; and
  • monthly community member competitions/prizes and draws.

Analysis

Descriptive statistics were used to report the characteristics of the sample and experiences relating to the misuse of personal information. Further analyses were undertaken to examine the relationship between identity crime and misuse and the characteristics of the sample.

As the survey was designed to capture information relating to respondents residing in Australia, respondents who indicated they resided elsewhere were excluded from the sample. In some cases, outliers that did not fit within the range of possible responses were excluded from the analysis. In other cases, where participants provided responses in the free text ‘other’ response option that fit within the categories that were provided, their response was recoded as appropriate.

Weighting of data

Data were weighted by location to represent the spread of the population in Australia. ABS (2013) data that estimated the June 2012 resident population by greater capital city and by state and territory were used to develop the weighting matrix for the sample data. The process of weighting involved the application of a formula to data provided by each respondent to make each response proportionate in relation to the broader population from which the sample was derived. For example, respondents located in Sydney made up 11.01 percent of the sample; however, this location contains 20.55 percent of the Australian population (ABS 2013). The actual weighting for each location is shown in Table 1. All results refer to weighted data, unless otherwise specifically noted.

The results have not, however, been weighted to indicate national estimates of prevalence and financial loss that would have been experienced had the entire Australian population aged 15 years and over be surveyed, as the sampling frame was insufficiently robust to permit such estimations to be undertaken.

Ethical considerations

A number of ethical issues were taken into consideration when developing the research design. These included the need for anonymity of research participants, the requirement to provide informed consent, the ability of participants to withdraw from the research and the potential for the research questions to cause psychological discomfort, particularly as they related to victimisation experiences. Once these concerns were addressed, the project presented a low risk to participants and the research was approved by the AIC’s Human Research Ethics Committee.

In relation to the anonymity of the research participants, no information that could be used to identify the participants was collected. The results are presented in an aggregate format and as responses are anonymous, they cannot be matched to specific individuals.

In order to ensure that participants provided informed consent, a plain language statement was provided with the survey. This stated that by completing the survey, participants were consenting to participate in the research. As outlined in the plain language statement, at any stage of the survey participants had the option to opt out of completing the survey and by contacting the external provider, could have the responses they had already provided withdrawn from the dataset.

It was acknowledged that, while the risk of psychological distress associated with the research was minimal, there was the possibility that a participant may have felt discomfort answering questions about victimisation. As the participant chose to complete the survey and there was information explaining what the survey is about, it can be assumed that they were aware of the potential sensitivity of the survey content. Telephone and website details for Lifeline crisis support were also provided in the plain language statement.

Limitations of the research design

Limitations of the research design arose from the sampling procedure, as those who participated in the online panel may not have been representative of the Australian population. For example, those who subscribed to an online panel may have had a higher exposure to online fraud than people in the general population. The survey was also only available to those who had computer access.

It can be difficult to measure victimisation and misuse within a given timeframe as it is not always easy to determine when the offence took place due to the time lapse between when personal information is obtained and when it was misused, when it was identified by the victim, and when, if at all, it was reported.

Survey designs such as this also have problems with reliability (such as whether the same survey delivered to the same subjects would elicit the same responses) and validity (whether the survey is measuring what it was intended to measure).

The circumstances and complexity of identity crime may also make constructing a meaningful survey instrument difficult. Problems of telescoping information (ie including events outside the survey reference period), exaggerating facts or reporting selectively—all common problems with surveys and personal interviewing—can affect the accuracy of information gathered using conventional techniques. There may also be problems of veracity, as individuals may be reluctant to report victimisation where they believe that they personally contributed to the problem, such as by voluntarily providing their personal information.

Alongside the challenges of obtaining good-quality data, there are also problems that stem from the volume of ‘hidden’ or ‘undetected’ victimisation. Hidden victimisation may occur due to the level of deception that is involved in an incident, which results in it being undetected, or the full extent realised. The result is that calculations of incidence, financial loss and other impacts can, at best, only be estimates.

Despite these limitations, the results of the survey provide valuable information to inform policymakers and the public about the current extent and nature of identity crime and misuse in Australia.

 

Related links

Identity crime and misuse in Australia: Results of the 2013 online survey