Chapter 2 - Working without fear: Results of the Sexual Harassment National Telephone Survey (2012)

Sex Discrimination

Working without fear:

Results of the Sexual Harassment National Telephone Survey

Chapter 2: Sexual harassment national telephone survey 2012

Chapter 2 provides a brief introduction to the 2012 National Survey. It outlines the objectives of and background to the 2012 National Survey. It also explains the methodology of the 2012 National Survey and identifies key methodological differences with previous sexual harassment national telephone surveys conducted in 2003 and 2008.

2.1 Objectives

The objectives of the sexual harassment national telephone survey were two-fold. The first objective was to identify the prevalence, nature and reporting of sexual harassment in Australian workplaces over the past five years. The second objective was to analyse trends in the prevalence, nature and reporting of sexual harassment in Australian workplaces across each wave of the national telephone survey.³

The Commission’s reports on the sexual harassment national telephone survey provide the only national data on sexual harassment in Australian workplaces. It is hoped that this data will help to improve awareness of sexual harassment in Australian workplaces and identify trends in the prevalence, nature and reporting of sexual harassment during the period covered by the survey. It is also hoped that this data will help to strengthen the effectiveness of efforts to ensure that all employees enjoy safe work environments free of such harassment.

2.2 Background

In 2002, the Commission undertook a review of complaints it had received under the Sex Discrimination Act 1984 (Cth) concerning sexual harassment in employment.⁴ The review found that complaints about sexual harassment in the workplace accounted for a significant proportion of complaints under the Act. It also found that sexual harassment presented particular challenges for women in paid employment.

The 2002 review provided important information about the number, nature and outcomes of complaints about sexual harassment in the workplace. However, there remained a significant gap in information concerning those individuals who had experienced sexual harassment in the workplace but who had not submitted a complaint under the Sex Discrimination Act.

In 2003, the Commission decided to conduct a national telephone survey on the prevalence, nature and reporting of sexual harassment in Australian workplaces to address this gap. The 2003 National Survey was the first such survey conducted in Australia and involved telephone interviews with 1,006 individuals between 18 and
64 years of age, who were representative of the Australian population in terms of sex, age and area of residence. Key findings of the survey included that sexual harassment was widespread in Australian workplaces and only a small proportion of individuals who had been sexually harassed had made a formal report or complaint about that harassment.

Following the 2003 National Survey, concerning evidence continued to emerge about the pervasiveness and harmful effects of sexual harassment in Australian workplaces. For instance, during the Sex Discrimination Commissioner’s Listening Tour, conducted in 2007 and 2008, women and men from across Australia identified sexual harassment in the workplace as a serious and ongoing problem.⁵ Many women and men also highlighted concerns regarding the general lack of awareness about sexual harassment and employers’ limited understanding about how to respond effectively to allegations of sexual harassment.

The Sex Discrimination Commissioner subsequently identified the prevention of sexual harassment as a key priority for her term.⁶ She also committed to repeating the national telephone survey, with a view to tracking trends in the prevalence, nature and reporting of sexual harassment.

The second sexual harassment national telephone survey was conducted in 2008. The survey, which was based on a slightly modified version of the 2003 National Survey questionnaire, involved interviews with 2,005 individuals, who were aged between 18 and 64 years and representative of the Australian population in terms of sex, age and area of residence. The 2008 National Survey found that sexual harassment continued to be a problem in Australian workplaces, despite some improvements since 2003. It also found that understanding about sexual harassment remained limited and targets infrequently reported sexual harassment to employers or other bodies.

In June 2011, the Commonwealth Government enacted the Sex and Age Discrimination Legislation Amendment Act 2011 (Cth). The Act made a number of amendments to the Sex Discrimination Act, including strengthening protections against sexual harassment in workplaces, schools, and in the provision of goods, services or facilities. It also strengthened protections against sexual harassment through the use of new technologies. Many of these changes were recommended in the Commission’s report on the 2008 National Survey.

In 2012, the Commission engaged Roy Morgan Research to conduct its third sexual harassment national telephone survey. As with previous surveys, the 2012 National Survey investigated the prevalence, nature and reporting of sexual harassment in Australian workplaces during the past five years. It also analysed related trends across the different waves of the survey.

The 2012 National Survey, which was based on a slightly amended version of the 2008 National Survey questionnaire, involved interviews with 2,002 individuals aged 15 years and over, who are representative of the Australian population in terms of sex, age and area of residence.

The 2012 National Survey involved interviews with an additional 1,000 members of the Australian Defence Force (ADF), as part of phase two of the Sex Discrimination Commissioner’s Review into the Treatment of Women in the Australian Defence Force Academy and Australian Defence Force (Defence Review). The simultaneous administration of the 2012 National Survey and the Australian Defence Force Survey (ADF Survey) allowed for comparisons to be made between sexual harassment in Australian workplaces in general and the ADF workplace in particular.

The results of the 2012 National Survey are outlined in detail in this report. The results of the ADF Survey, together with a comparative analysis of key findings of the 2012 National Survey, are available in the Defence Review report, which was published earlier this year.⁷

Importantly, the results of the 2012 National Survey identify trends in the prevalence, nature and reporting of sexual harassment across each wave of the survey, notwithstanding the minor modifications made to the survey questionnaire.

2.3 Methodology

Two thousand and two people aged 15 years and over were interviewed for the 2012 National Survey. This was the first survey to include young people (15 to 17 years) and older persons (65+ years). A quota sample was used to ensure that participants were representative of the Australian population in terms of age, sex and area of residence. The sample was not selected to reflect the Australian population by occupation, employer size, industry, or employment status (full-time / part-time).

A number of different approaches were used to recruit individual participants.

First, individuals with a fixed landline telephone at home (n=1,699) were recruited through Random Digit Dialling (RDD).

Second, individuals with a mobile telephone only (n=300) were recruited through Roy Morgan Research’s Single Source Database (SSD).⁸ Both RDD and SSD were used to maximise the participation of individuals in the increasing number of households without a landline connection.

Third, to ensure a representative sample of Aboriginal and Torres Strait Islander peoples, the Commission supplemented the above approaches by promoting the survey through the National Congress of Australia’s First People and the Commission’s own Aboriginal and Torres Strait Islander Social Justice networks.⁹
A total of 46 Aboriginal and Torres Strait Islander peoples participated in the 2012 National Survey, including three individuals who self-nominated through the third approach described above. This proportion is representative of the actual distribution of Aboriginal and Torres Strait Islander peoples in the general population, aged
15 years and over.¹⁰

The interviews were conducted using the Computer Assisted Telephone Interview (CATI) approach. This approach was used first in 2003 and again in 2008 as an efficient and effective way to interview a large sample of individuals about a sensitive topic. Using this survey methodology for the 2012 National Survey also had the advantage of ensuring that any changes in results can be attributed to factors such as changes in attitude, behaviour and understanding of sexual harassment, rather than non-sample error arising from different survey methodologies.

The interviews were conducted using the 2012 National Survey questionnaire (see Appendix 1), which was based on a modified version of the 2008 National Survey questionnaire. Amendments to the survey questionnaire were required in 2012 to accommodate changes to the definition and legal regulation of sexual harassment, since the last survey was conducted. Further changes were made to enhance the effectiveness of the survey questionnaire and increase the representation of a broad cross-section section of the community (eg young and older persons). A number of new questions were incorporated into the survey, including questions concerning the medium to long-term consequences of sexual harassment for targets, the presence of repeat and multiple harassers in the workplace (from the perspective of those who have been harassed), and the consequences for bystanders of taking action after witnessing or learning about sexual harassment.

The number and scope of changes to the survey questionnaire were limited, however, so as not to undermine the analysis of trend data across each of the different waves of the survey. In addition, a small pilot of the survey was conducted to test the amendments to the survey questionnaire.

The interviews for the 2012 National Survey were conducted by female interviewers because of the gendered nature of sexual harassment in Australian workplaces. Male interviewers were also made available in the event that an interviewee indicated that he or she did not feel comfortable being interviewed about sexual harassment by a woman.

As explained previously, the 2012 National Survey was the first survey to include young people (15 to 17 years). Young people were interviewed only if a parent or guardian granted them permission to participate in the survey. Parents and guardians were given the option to be present during the interview, but were asked not to influence the answers of the participant, either directly or indirectly. The need for parental permission may have resulted in a possible sampling bias.

The limitations of the CATI approach have been noted in the reports of the 2003 and 2008 National Surveys.¹¹ These include that the sample of people interviewed may not adequately represent individuals from culturally and linguistically diverse backgrounds, individuals who are unable to communicate via the telephone due to a disability, and individuals from lower socio-economic backgrounds.

Another potential limitation of the CATI approach is under-representation of individuals without a landline telephone in their home. The ongoing shift from (almost) universal access to landline telephones in the home to an increasing dependency on mobile telephones and the Internet amongst core sectors of the Australian workforce means that consideration needs to be given to the continued appropriateness of the CATI methodology for future waves of the sexual harassment national telephone survey.

The 2012 National Survey provided an important opportunity to investigate the merits of migrating from the CATI approach to an approach that utilises Computer Assisted Web Interview (CAWI). Significantly, it provided an opportunity to understand what, if any, impact a change in methodology might have on the results of the survey and the ability to track trends in the prevalence, nature and reporting of sexual harassment across the different waves of the survey.

A parallel study was therefore conducted in 2012 using the CAWI approach (CAWI Survey), which was in addition to the 2012 National Survey and the ADF Survey, described above. The CAWI Survey utilised the 2012 National Survey questionnaire, with slight modifications to ensure suitability for the online format. The CAWI Survey involved interviews with 480 individuals, who, as in the case of the 2012 National Survey, were representative of the Australian population in terms of sex, age and area of residence.

The Commission is currently reviewing the findings of the CAWI Survey, which are not reported here, and will make a determination regarding the most appropriate survey methodology before the next sexual harassment national survey is conducted in three to five years’ time.

2.4 Reading and interpreting the report

(a) Interpreting the data from different survey waves

When reading and interpreting this report, it is important to note that there are slight differences in the measures of prevalence of sexual harassment in the workplace in the past five years used across the different waves of the national telephone survey.

In the 2012 National Survey, the measure of prevalence of sexual harassment in the workplace in the past five years includes respondents who reported that they had:
(1) been sexually harassed based on the legal definition of sexual harassment (legal definition); and (2) experienced one or more behaviours that are likely to constitute sexual harassment under the Sex Discrimination Act (behavioural definition). This combined approach was expected to yield the most realistic incidence of sexual harassment.

In 2003 and 2008, the measure of prevalence of sexual harassment in the workplace in the past five years included respondents who reported sexual harassment based on the legal definition only. In the 2008 National Survey, respondents were also asked whether they had experienced one or more sexual harassment behaviours in the workplace that are likely to constitute sexual harassment under the Sex Discrimination Act. The prevalence of sexual harassment based on the behavioural definition was not combined with the prevalence of sexual harassment based on the legal definition to determine an overall prevalence. However, data gathered using the behavioural definition was combined with data gathered using the legal definition when analysing other parts of the 2008 data (eg nature and characteristics of sexual harassment).

Summary of definitions used according to wave and type of findings being reported
Findings reported	2003	2008	2012
Prevalence (in the workplace in past five years)	Legal definition only	Legal definition only (behaviours reported separately)	Legal and behavioural definitions
Nature and characteristics	Legal definition only	Legal and behavioural definitions	Legal and behavioural definitions
Prevention and response	Legal definition only	Legal and behavioural definitions	Legal and behavioural definitions

The different approaches to calculating the prevalence of sexual harassment in the workplace in the past five years limits the extent to which the prevalence data can be compared across each wave of the survey. Fuller comparisons of prevalence based on both legal and behavioural definitions will be available in future waves of the survey.

Any omission of comparisons in this report between the results of the 2012 National Survey and the 2003 and 2008 National Surveys may be because:

comparison across different waves of the survey is not meaningful due to the change in the calculation of prevalence
new questions were asked in the 2012 National Survey that were not asked in previous surveys
the data from previous surveys is inaccessible or incomplete.

(b) Interpreting the data with an expanded age range

As part of the 2012 National Survey, interviews were conducted with individuals aged 15 years and over. In contrast, the 2003 and 2008 National Surveys were limited to individuals aged between 18 to 64 years of age. To determine what, if any, impact the expanded age range had on the survey results, 2012 data for all age groups was compared with the same data for the 18 to 64 year old age group.

The comparative analysis showed that the expansion of the age range in the 2012 National Survey had virtually no impact on the data, except in the areas of employment and age. Figures related to employment are generally lower for the group that includes all age groups. For example, the figure for the total number of people who experienced sexual harassment at work is 68% for the group including all ages, and 71% for the group excluding the 15 to 17 and 65 and over age groups.

The differences in data concerning employment and age are to be expected because there are fewer individuals in the 15 to 17 and 65 and over age groups in the workplace. Where the inclusion of the two new age groups may have had an impact on comparison with previous waves of the survey, it is noted in the report.

(c) Reading the graphs

The values presented throughout the report have been rounded to zero decimal points (with the exception of those values between 0% and 1%). This means that some of the bars featured in the figures / graphs may not appear equal, even though they are shown as having the same value. For example, two bars on the same graph may be labelled as 5% but appear to be different lengths because one may represent an actual value of 4.5%, while the other may represent a value of 5.4%. Both have been legitimately rounded to 5%.

Unless otherwise stated, the results reported in the figures / graphs (and in other parts of the report) refer to the 2012 National Survey.

^Top