Late last year we were told that Canberra’s Parliament House is plagued by bullying, sexual harassment and sexual assault. A 456-page report by Sex Discrimination Commissioner, Kate Jenkins, claimed half of all people in Commonwealth Parliamentary Workplaces (CPW), that is Parliament House or electorate offices, have experienced at least one incident of bullying, sexual harassment or actual or attempted sexual assault, and one in three people working in federal Parliament have experienced some kind of sexual harassment there.
Prime Minister Scott Morrison described the statistics as appalling and disturbing. Brittany Higgins, whose claims of rape in Parliament House prompted the inquiry, called for ‘immediate action’. Along with most media, the Guardian said that Parliament had a ‘toxic workplace culture’.
Having spent five years in Parliament House, I decided to read the report in full. What I discovered was that while it sets out to show Parliament in a bad light, it reveals the opposite. As a workplace, Parliament is both ordinary and representative — neither sexual assault nor harassment occur any more frequently than in other workplaces, and there is no reason to believe bullying does either. The report is just a shoddy attempt to legitimise social engineering based on cherry-picked data.
The data in the report are derived from a survey of people who work in the Parliament or electorate offices. There are two types of surveys: those in which the data drive the conclusions, and those where the conclusions drive the data. This one is in the latter category.
There were 935 responses from 4,008 people invited to participate. The sample is self-selected, which means those who choose to participate are not necessarily the same as those who do not. The report says responses were weighted to ‘correct imbalances in the results due to any non-response bias’ but gives no details. It is not true.
Given the Higgins allegations, sexual assault is a priority. However, only nine respondents reported such assault, or around 1 per cent of the sample. A 2016 Australian Bureau of Statistics survey found 1.1% of Australian adults claimed to have experienced sexual assault in the past 12 months. That is, sexual assault is no more common in CPWs than in the community.
The report says 33 per cent of respondents (40 per cent of women, 26 per cent of men) have experienced sexual harassment in a CPW, with victims including both parliamentarians and staff. This is the same rate as in the broader population and is also unchanged from a similar survey in 2018. In other words, sexual harassment in the parliamentary environment is no greater than in the community and is not increasing.
The report also says 37 per cent of respondents claim to have experienced bullying in a CPW. Women are twice as likely as men to be bullies, women are more likely to be victims, and in three-quarters of cases, the perpetrator was more senior. The most common case is a junior woman claiming to be bullied by a senior woman. However, the report provides no comparisons with other workplaces and no basis for implying that CPWs are exceptional.
Although the survey contained 200 questions, potentially yielding a lot of useful information, the report provides very little. There are literally no responses to individual questions, and none of the tables and crosstabs normally found in opinion research reports. Despite 44 questions about sexual assault, no responses are reported – hence there is no information about location, timing, gender, age, relationships, employment, or about the perpetrators.
It is much the same with the other two issues. Despite 49 questions about sexual harassment, the analysis is brief and superficial. The report notes that reported rates of sexual harassment are higher when specific behaviours are mentioned rather than just a short legal definition. The questionnaire gives the legal definition in one question and mentions ‘Inappropriate staring or leering that made you feel intimidated’ and ‘Being followed, watched or someone loitering nearby’ as examples of sexual harassment in another question. But responses to either question are not provided.
There is little, also, about the responses to the 48 questions about bullying. As with sexual harassment, responses were probably influenced by how bullying was defined: examples in the questionnaire included ‘Others spreading misinformation, or malicious rumours’ and ‘Assigning meaningless tasks unrelated to the job’. Either way, we are not told.
The overall design of the questionnaire is flawed. Well-designed surveys ensure responses to key questions are not influenced by prior questions or information. Not in this survey. All the questions about sexual assault, harassment and bullying are preceded by questions like, ‘Is the workplace safe and respectful? Are sexual assault, sexual harassment or bullying tolerated? Are people treated fairly and equally regardless of age, race or cultural background, sexual orientation, disability or religious beliefs? Are there negative attitudes to women?’ By the time respondents get to questions about their own experience, their thinking might well have changed.