Graham Pike, Professor of Forensic Cognition, writes: Apparently the fact that I voted in the November 15th (2012) elections to fill the new role of Police Commissioners makes me part of a very small minority indeed. Overall turnout for the election was just 18% and as low as 10% in some areas. The poor turnout has hijacked the news cycle for the commissioners, and many people are asking why people did not turn out to vote – which is an interesting and important question. The Electoral Reform Society and 20 candidates for commissioner urged Theresa May (the Home Secretary) not to hold future elections in the winter, and the Electoral Commission unsuccessfully lobbied that letters be sent to all voters with information about who their candidates were. As well as cold weather and lack of information, a survey by The Royal United Services Institute and YouGov suggested that nearly 50% of voters believed that commissioners will make no difference to policing and crime, and only 10% said they understood what the new roles actually entailed. In addition, nearly two-thirds of respondents disagreed with candidates for commissioners being aligned with and supported by political parties.
That the poor turnout reflects very badly on what is a flagship policy in the Conservative’s manifesto on law and order does not seem in doubt, and to that end we really only need know how many people voted and who they voted for. However, it seems to me that not knowing why over 80% of the voting population stayed at home is a serious problem. Obviously we can explore this using follow-up surveys and polling analysis, but I can’t help thinking that our voting system is seriously flawed if it does not distinguish between voters who are apathetic and those who are against the entire point of the election. In designing a survey, certain types of question would be deemed badly constructed if they did not include options which said ‘none of the above’ or ‘I don’t know’, because otherwise the survey would be forcing the respondent to make a decision which did not reflect their attitude or opinion. For many decades in the UK the police have had to instruct eyewitnesses that the suspect ‘may or may not be present’ in identification procedures, as otherwise they would be leading the witness to believe that the suspect definitely was present. Psychologists refer to procedures that do not involve this explicit instruction as ‘bias lineups’. When interviewing, particularly when the police interview a witness, questions need to be carefully constructed to avoid being ‘leading’. For example, you should not ask ‘what colour coat was she wearing?’ as the question assumes the person was wearing a coat. In other words, the question makes a fundamental assumption that the witness is lead to believe, even if their own view was different. Surveys, identification procedures and interviewing avoid making assumptions about the views of those they are questioning and allow them to say ‘I don’t know’ or ‘None of the above’. All three of these, therefore, do not prejudice and are able to distinguish a genuinely positive choice from someone who doesn’t know the answer and from someone who doesn’t want to select any of the options presented.
I’m sure you can see where I’m going with this. Why doesn’t our voting system also allow these distinctions to be made? Where is the ‘I don’t know’ option for people that have had no information about any of the candidates, and more critically where is the option for people who want to indicate ‘None of the above’? By tomorrow we will have elected 41 Police Commissioners that more than 80% of the population did not vote for. If even half of the non-voters stayed at home because they fundamentally disagreed with the new roles, that will mean twice as many people were against the election than actually voted in it. The situation is likely to be worse, as many of those who voted may well have gone for ‘none of the above’ if given the chance, but the key point remains that our voting system has probably prevented a very large number of people from expressing their views, because it asks a leading question, has a biased construction and does not contain options standard on virtually every survey conducted.