Methodology: Putting Out the “Coo-ee!”
In essence, the modern-day equivalent to the old “bush telegraph” (ie “word-of-mouth”) was utilised for the recruitment of participants, through use of social networking tools such as Facebook and by email and word of mouth.
The Facebook tool enabled Saunders to make contact with potential interviewees who were referred through the “private message” box function. Rurally located friends, families, colleagues and old school acquaintances all helped in “spreading the word” about this research.Eighty-four women employees and 23 employers (two males) were interviewed. These took place between July and December 2011 following approval by the University of Canberra’s Committee for Ethics in Human Research. Protecting the anonymity of participants was the main ethical consideration in this project. Before responding to the survey, respondents were asked to read the participant information form and sign a consent form, both of which confirmed that all responses would be de-identified. Initially all respondents were interviewed face-to-face with Saunders travelling to rural and remote destinations in Western Australia, the Northern Territory, South Australia and New South Wales. Not wishing to lose the “voices” of the most remotely based participants though, it became apparent that telephone interviews would be a good complement. It was preferred too by some participants who required an evening interview time. Interviews were recorded and transcribed and took from 30 minutes to 90 minutes with an average of an hour, depending on the experiences of each participant. One researcher, Saunders, completed a recording form for each interviewee.
Data analysis
A single individual translated the recording forms into numerical variables; this ensured that there would be no inter-coder variation.[460] For analysis we merged certain occupations to allow for meaningful comparison: teachers and nurses are grouped as “professional”; agriculture and horticulture are one category and we have aggregated those working in retail, hospitality or fast food.
We also designated our workplaces/locations by degree of rurality using the following population cut offs: high = 0-5000; average-high = 5001-15 000; average = 15 00130 000; average-low = 30 001-50 000; and low = >50 000. In addition the responses to two questions (“How does sexual harassment make you feel?” and “Why do you think people sexually harass others in the workplace?”) were aggregated into two categories to allow for some analysis, albeit limited by the low response rate. All serious effects (“angry”, “hurt”, “anxious”, “hopeless”, “humiliated and powerless”, “degraded”) were merged into one category and labelled as “serious”. Other responses such as “can help you in your job”, “depends on person”, “didn’t worry me”, “you get used to it”, were considered by us as “minimising”. The answers given to why people sexually harass were grouped similarly. Examples of minimising responses included: “blokiness”, “boys being pig-headed”, “harmless”, “joking, usually”.“Serious” included: “Exert power”, “gain power”, “power plays”, “power thing”, and “power trip”.
Using PASW Statistics 19, the total data set (n=104) was crosstabulated looking at variation in responses to relevant questions with sociodemographic variables such as age, seniority, education, rurality, occupation and ratio of men to women in workplace. In addition, employers (n=23) and employees’ responses to some questions were compared. A Pearson’s chi-square test of contingencies was used to evaluate statistical significance. Where appropriate, a series of unplanned 2x2 post hoc analyses were conducted to determine where statistical difference lay. The size of the cells in a number of cross-tabulations (eg more than one cell with less than five) precluded generation of statistically meaningful tests of variation.
Caveats
The number of rural workplaces with staff included in this study is not large enough to be representative of rural Australian workplaces. When disaggregated by occupation for instance, 11 per cent of respondents worked in the mines.
It is unlikely that 11 per cent of females employed in rural areas are employed in the mining industry.Further, people who participated in this research were self-selected as described above. It is possible that they were more likely to have experienced sexual harassment and were keen to discuss their experiences. Therefore the results reported next should be regarded as preliminary — providing indications but not necessarily absolutes.
Note too that whilst the definition of “rural” is a term that is intuitively ingrained into the psyche of most Australian people, most academics agree that the term “rural” is an elusive one[461] and that it is “constructed differently by different people”.[462] Our sample is reflective of our definition and therefore our findings are limited by those boundaries. One issue with the categorisation of “rural” by number of people is that “some towns may be classified as rural based on population size, density or location, but still have ready access to many facilities owing to their proximity to major population centres”.[463] In light of these obvious complexities in capturing the definition of rural, some academics prefer to define it as being anything that does not fall within the definition of metropolitan.[464] Whilst this provides a relatively simple springboard from which to examine rural trends, we recognise that the danger is that if the term “rural” is used to classify most regions outside of the coastal metropolitan areas, “it may cease to have much in the way of descriptive veracity at all, despite its continuing cultural resonances”.[465] While we had to impose population cutoffs for the degree of rurality, it is important, though, to also consider the predominant view of local residents in places such as Orange (which is a proud “rural community”) and this involves the identification of “elements of rural place/landscape/society/existence which together provide an approximation of the overarching concept of rurality”[466].