There has been a relative dearth of published data on farmers and farming attitudes over recent years, and the introduction of a regular index of farming opinion by the Irish Examiner has been a welcome and worthwhile development, says Larry Ryan of Behaviour & Attitudes.
2015 Irish Examiner/ICMSA survey
As someone who has worked in market research for 25 years, I have seen the size of the farming community decline from more than 20%, to its current level of roughly 7% of the Irish population.
We are now at a stage whereby marketing professionals are suggesting that farmers be “re-integrated” into the population at large, no longer having their own separate social class.
Indeed, social class F had always been unique to us here in Ireland, and its demise arguably spells the end of very specific commercial interest in farmers as a group — and a dangerous assumption that the behaviour and attitudes of farmers now mirrors those of the wider population.
Whether such a change is appropriate, it is always good to work on a survey of farming attitudes and beliefs, and we have been delighted to partner with the Irish Examiner in developing this innovative survey.
What is unusual about the poll is that fieldwork is spread out over a series of agricultural shows, running through the summer months.
So, rather than being a snapshot of public opinion at a moment in time, it is more a blending of political and social opinion, running across the summer months.
One of the positive aspects of this approach is that it is likely to eliminate spikes or troughs in sentiment which can and do occur in response to a speech or an event, and which can make some polls seem unusual as a result.
Arguably, it gives a slightly truer impression of the standing state of the farming public on political issues, parties and themes.
A possible downside is that if there is very active debate about one of the issues being polled, this may be “blended out” where a poll is run across an extended time period.
We don’t see any evidence of this in our 2015 poll, however.
Dearth of data
There has been a relative dearth of published data on farmers and farming attitudes over recent years, and the introduction of a regular index of farming opinion by the Irish Examiner has been a welcome and worthwhile development.
Although the Central Statistics Office and others may periodically collect data on farming, farm sizes and broad agricultural trends, there is no study that overlays this data with farmer’s attitudes on real or substantial issues.
Indeed, as a non-farmer, I see little about farmers’ points of view in the mainstream media, and one needs to turn to the specialist farming press to get a truer sense of the mood among farmers.
Furthermore, most of the wider media attention directed to farming happens when there is a dispute or an issue.
That leaves city dwellers like me with the impression that farmers are broadly reactionary in many of their standpoints and opinions.
The reality, as the study shows, couldn’t be further from the truth.
We have had many queries on how we field this project, and I am happy to outline the broad methodology here.
First and foremost, we sit down with the Irish Examiner in the run-up to the summer, and debate the issues and themes that we want to cover off in the survey.
We endeavour to avoid any area within which we feel there is likely to be profound change between the start and the end of the summer.
Having agreed on the content, we draft questions which, like all responsible polls, aim for balance and objectivity in their phrasing.
Having agreed the questionnaire, we allocate fieldwork across a series of 8 to 10 agricultural shows right through the summer period.
We deploy slightly more interviewers at the larger shows, and fewer at some of the smaller events.
However, we believe it is important to cover off a good spread of shows, including ones which attract a less mainstream group, and indeed, ones in more remote areas.
Our interviewers are briefed and attend the shows, interviewing farm dwelling adults on a randomised basis.
In other words, they will utilise a simple randomisation procedure asking every “nth” passer-by whether they live on a farm or not.
We don’t impose demographic quotas, so we aren’t aiming for a set proportion of men and women, or of specific age groups or types of farmer.
This ensures that the result achieved (the numbers in dairying, the numbers in beef, etc.) are effectively ‘survey results’ rather than pre-determined quotas or amounts.
Despite this randomisation and lack of quotas, we see underlying demographic similarity from year-to-year, with similar numbers of men and women and of respondents in each age group.
As with all polls, we don’t look for volunteers to take part, but rather approach and screen potential respondents who have been selected on the basis of a randomisation procedure.
Accuracy and Error Margin
People often ask about the precision or reliability of opinion polls.
Like all properly conducted research exercises, whether utilising a quota or a random selection procedure, the data for a survey can be deemed as accurate to within a predictable margin of error.
One of the rules of statistics is that, if a sample is properly selected, adhering to pre-determined procedures, and sufficiently large, then the results achieved will lie within a predictable margin of error, which is principally determined by the size of the sample.
Utilising a sample of somewhere between 500 and 600 farmers gives us data with a predictable margin of error of plus or minus 4%.
In other words, 19 times out of 20, an achieved percentage of, say, 25% could be as high as 29% or as low as 21%, provided all survey rules have been observed. If we increase the sample size to about 1,000, we reduce the error margin down to about 2%.
Clearly it is difficult, when the constituency is farmers, to prove or disprove the accuracy of these results.
However, we would point out our success in the context of the most recent European election.
Behaviour & Attitudes ran a large exit poll for RTÉ so that they could predict the election result, and identify its drivers, the next morning.
This involved us interviewing just under 3,000 voters as they exited their polling stations, with interviewing being spread across every electoral unit, and with shifts divided up into morning, afternoon and evening.
Having completed our fieldwork at the close of the polls on election day, we delivered constituency by constituency results to RTÉ for their 7am news bulletin.
An enormous team of interviewers was involved, and indeed most of our office staff also went out and interviewed a quota of 20 or so voters at some stage during the day.
On that occasion, we were able to compare the claimed voting pattern, as volunteered to our interviewers, with the actual European Election result.
Our poll was deemed to be accurate, with greater than 99% precision, indeed very closely mirroring the order of election, the magnitude of candidate-to-candidate transfers, and the precise make-up in terms of seats, with an almost uncanny degree of precision.
While we can’t “prove” our farmer-related findings in the same way, we are confident that the procedures used in bringing you this Irish Examiner/ICMSA poll are of equal quality and reliability.
Behaviour & Attitudes are proud to have fielded the third Irish Examiner farmers’ poll, in association with the ICMSA, and to bring Irish Examiner readers an overview of these shifts, and we hope that this poll leads to greater discussion, debate and indeed recognition of the wider communities interest in farmers as a group.
* Larry Ryan is a Director of B&A, and jointed the company in 1997. Behaviour & Attitudes is Ireland’s leading independent market research agency, and celebrates 30 years in business this month.
© Irish Examiner Ltd. All rights reserved