Conference will be reviewing May 2011 results but is only asking 50% of questions it should be: There is more to learn if we ask the right questions

At the Lib Dem Conference next week there will be a review of the May 2011 election (see document here) which asks some interesting questions but it only goes so far in asking the right questions to get all that we could learn. The review will be asking:

1. Was your candidate in place early enough? Did he or she campaign for long enough?

2. Were the methods of campaigning varied enough?

3. Did you produce enough literature? Was it of a good enough quality?

4. Do you feel there was enough emphasis on community politics?

5. Did we target correctly in your area? What lessons can we learn for the future?

6. Was support from Party Headquarters adequate? If not, how could it be improved?

7. How did you deal on the doorstep with the Liberal Democrats being in the coalitiongovernment?

8. Did you campaign purely on local issues or also on national issues? What worked best?

9. What more can be done to recruit extra activists and keep existing ones motivated?

10. Knowing what you know now, how will you fight the next set of elections differently?

11. How did holding the referendum on the same day affect:

                                      i: Turnout?

                                      ii. Activist priorities?

                                      iii. The result?

While these questions are important there are some equally important questions that we should be asking. You can learn a lot from asking about why went wrong, what we learnt and what we are going to do differently next time. This gives us what didn’t work. However, this is only half the story. We can also ask where what we did worked well (such as question 8) and come up with ‘what worked’.

In May 2011 what worked would be any local party who scored better than the average result.

These are the results for Scotland as they are available. Any local party who did better than a -14.23% from them didn’t do too badly given the context. There is some evidence that where the Lib Dems have an MP we did better, which is a generalisation that seems to have gained traction with many including those in the Tories as advertised in ConservativeHome:

The Liberal Democrats lost 40% of the council seats they were defending but did much better in places where they had an incumbent MP

But this is not strictly true. Taking Scotland as an example we did ‘well’ in areas where we don’t have an MP and poorly in areas we do e.g. Orkney. So this is not strictly true and so there is a lot to learn about what local parties have done to attract support/keep support in their areas which goes beyond the received wisdom. If we don’t ask these questions we lose important lessons.

I have been contacting local parties where I see a good result and ask them what they thought they did which attracted the support which was bucking the national trend. The main message I am getting is that it is due to the strength of the candidate but I feel there is more to it than that. We need to be asking more specific questions about what they are doing, how do they attract support, supporters, specifically. So I would like the review to ask:

Did your party beat the average national poll for the Lib Dems? / Did you do better than expectations? If so what did you do specifically which contributed to this result? What campaigning issues worked in your area? How did you choose them? What material did you use and how much do you think this contributed? What worked on the doorstep? How did you get this result?

Do negative campaign ads work?

Image by gorfor via Flickr

Are voters turned off by negative campaign ads and mudslinging? Some say they are yet it doesn’t stop political candidates or parties from using them. Do we know the evidence about whether negative ads work or do we use them hoping they will work, or because we all know a good negative ad which did work? The results from the research may actually help inform and improve our campaigning.

Research on negative campaigning and negative campaign advertising has produced some conflicting results. Some studies suggest that negative campaign ads are more easily remembered and therefore have a greater influence on voters’ attitudes and vote decisions. Other research, however, provides evidence that the opposite is true. Moreover, while some research suggests that candidates who run negative ads are more likely to win, other research suggests that running negative ads makes a candidate more likely (or at least equally likely) to lose. There are also conflicting conclusions about the effect of negative advertising on voter turnout–some research concludes that negative campaigning depresses turnout while other findings suggest that intense competition (often characterized by negative campaigning) enhances voter turnout. For an excellent review of this research see The Effectiveness of Negative Political Advertisements: A Meta-analytic Review.

While the research on negative advertising in political campaigns is not clear the conventional wisdom among campaign professionals is that negative ads do work. That is, while voters might not like negative ads, their perceptions of candidates attacked in negative ads are tarnished by the information they are exposed to. But that is not to say that they should be used wholesale as there are significant downsides and risks to using them. So what can we learn from the research to improve our campaigning?

• Overall, negative political advertising produces negative evaluations of both the sponsor and the target.
• 87% (in one survey) of people are concerned about the level of personal attacks in today’s political campaigns
• Voters distinguish between what they feel are fair and unfair “attacks” in a political campaign.
• At least 57% (in one survey) believe negative information provided by one candidate about his or her opponent is relevant and useful when it relates to the following:

• Talking one way and voting another
• Not paying taxes
• Accepting campaign contributions from special interests
• Current drug or alcohol abuse
• His or her voting record as an elected official

• At least 63% (in one survey) indicated the following kinds of information should be considered out of bounds:

• Lack of military service
• Past personal financial problems
• Actions of a candidate’s family members
• Past drug or alcohol abuse

• Both younger and older people agreed that negative political advertising is not informative, but older people consider negative political advertising as less believable and have more negative attitudes toward the sponsor than younger people.
• Negative political advertising is more effective with lower income voters. They perceive negative political advertising as more informative and more believable and had more positive attitudes toward the sponsor than higher income level voters.

So voters do not treat all negative information equally and while negative ads have the capacity to weaken political support for a candidate’s opponent, being negative in a campaign can also diminish the attacking candidate’s stature among voters, producing a “backlash” effect, especially when that information is not perceived by voters as relevant to the campaign.

The rule is never use negative campaign tactics unless your have to. You only have to if you feel cannot win simply by presenting positive information about yourself.

Candidates most likely to use negative ads are challengers. Incumbents have generally spent years building positive images of themselves among voters. The longer an image of a candidate is maintained in the minds of voters, the more difficult it becomes to change that image. A challenger hoping to unseat an incumbent must provide evidence that the positive images voters have of their opponents are inaccurate.

The choice that candidates have to make is whether the negative information they want to emphasize in a campaign against an incumbent is important enough to voters to make them disavow their opponents and support them instead. When a negative ad aims at something outside of the bounds of what voters consider to be relevant and fair, the effects just might be opposite of what was intended.

Improving campaigning: What the research says about what works in political campaigning

What works in political campaigning? How many leaflets have been delivered, letters sent, doors knocked on? Remember the prerecorded message from Nick Clegg to 250,000 voters? Is this a reflection that the Lib Dems don’t know what works? Randomised experiments producing reliable evidence in politics are rare and so it is difficult to get the strategy as effective as possible for the resources you have. However, there are some experiments which give us some very specific results to improve our campaigning.

The get-out-the-vote campaign in the states has made extensive use of randomised experimentation since 1998 and have developed what they call the bottom line. While this is focused on increasing the number of people to vote it still offers parties some important lessons. So what does the research say?

Statistically significant results:

Door-to-door: This is considered the most effective method of influencing people to vote with one vote per 14 voters contacted (plus spill over effects).

Phone, volunteer: Getting enthusiastic volunteers and setting up a phone bank results in one vote per 38 contacts.

Commercial live calls: Getting a firm to make calls for voters produced some good results. For those who were trained it produced one vote per 35 contacts and for those who weren’t trained it produced one vote in 180 contacts

Election day festivals: This raised turnout in area by 1 – 2 % (but is based on only a few studies)

Not-statistically significant (in no particular order):

Leafleting: A high intensity job is an ineffective campaigning tool producing one vote per 189 voters reached.

Direct mail: direct mail advocating for a particular party had no detectable effect whereas non-partisan mail produced an increased voter turnout of boarderline significance.

E-mail: No detectable effect

Television: Raises postcode wide turnout by 0.5%

Radio: Raises city wide turnout by 0.8%

Robo calls: One vote per 900 called

The results show that what works is where some sort of rapport or relationship is built up between the voter and the campaigner. Where there is no relationship the results are insignificant and yet we spend a significant amount of time, money and effort on non-relationship based political activities.

Effective political campaigning is relationship based: it builds relationships between the party and the party members;  between the party members and party supporters; between the local party and the local community. Relationship based politics works by building a relationship between the party and the country.

Chris Huhne gets the SFP award in ‘Reinventing the State’

Image via Wikipedia

Chris Huhne makes a very interesting point in his Chapter in ‘Reinventing the State: Social Liberalism for the 21st Century’ – ‘The Case for Localism: The Liberal Narrative’. While talking about national policies and how they can actually provide negative consequences he says

Contrast this system with the private sector, where it is natural to experiment, to try something else. If they fail in some effort, they stop and try something else. If they succeed, then try more. There is a constant process if change and improvement – or the business dies – p.246

What he is describing is essentially the solution focused approach: Find what works, do more of it. If it doesn’t work, stop doing it and do something else. While it may seem so very basic and simple, this is because it is, and because everyone knows that this works because they have probably done it at some stage. However, we don’t always do this, otherwise Chris Huhne wouldn’t have to point this out in the first place. The problem often comes with focusing elsewhere other than on ‘what works’.

An example of how doing what works is not followed can be seen in the work of a mathematical biologist from Sydney whose study explained why ineffective medical treatments persist in the face of better proven methods. He found that quack treatments could spread more quickly than proven treatments. The explanation of this counterintuitive finding is that people pick up treatments by observing what other people use and then follow that, whether that is an effective treatment or not. This mechanism creates an unexpected advantage for ineffective treatments. The fact that ineffective treatments don’t work means that people who use them will be around longer and effective treatments take away the problem and so there are fewer people using them for longer. This means that there are more people left to be observed using ineffective treatments than effective treatments which will lead to more people copying them. This means that ineffectiveness of a treatment may be the very cause of its popularity.

Doing What Works: Obama Administration Using a Solution Focused Approach

Game theory attempts to mathematically capture behaviour in strategic situations, in which an individual’s success in making choices depends on the choices of others. The prisoner’s dilemma is a fundamental problem in game theory that demonstrates why two people might not cooperate even if it is in both their best interests to do so.

Over the years there have been competitions to find the most successful strategy. A strategy called ‘tit-for-tat’ consistently came out as the winner, which was a strategy of starting positive and then doing what the other party did in their previous move. That was until the strategy of Pavlov was developed. This was a strategy where you follow the same strategy as in the previous move if it was successful but change if it was not successful. Essentially a Do What Works strategy.

However, Pavlov did not beat the ‘always deceive’ strategy whereas tit-for-tat did. Therefore the Do What Works Strategy thrives once the always deceive strategy has been beaten. While this is an academic endeavour its real life application to politics is important.

Obama has started on the strategy of Doing What Works and it aims to deliver greater value and better results. This is exactly what a solution focused approach would advocate: a focus on outcomes with clear goals using tools such as scaling to measure progress. All of which is discussed on this blog in previous posts (e.g. here or here) and is shown in Obama’s administration’s video:

What Game Theory teaches us is that while this is the best strategy to produce the best results for the people, it is vulnerable and open to attack. Obama has come under pressure and heavy attack from his opposition. Maybe they are using the always deceive strategy? Not everyone wants you to do what is best and what is right.

Liberal Democrats – Doing What Works?

The argument goes that the Labour Party believes in a big state as this would help government perform its basic functions better. While the Conservative Party believes we should have a smaller state as this would increase economic efficiency and would therefore be better for long term economic performance. A government may have any size but if it makes unwise policy decisions based on erroneous information or thinking, it won’t be effective.

“For all Liberal Democrats, the aim is to reinvent, not to reduce, the state”, but what happens when the state is reduced in size by the Coalition Government? Do we then say we need to get back to the size it was in 2006/7/8? Or what happens when it gets bigger again in the future, do we then say it needs to be cut back again? Or are we happy with whatever size it is? The Lib Dems need to assert their position in a more distinct fashion than the current perceived piggy-backing onto the Tories small-state-is-best-narrative, which is frightening potential voters into believing the Lib Dems are now something to fear (mainly from the political left).

The Solution Focused Change website posted a video of Barak Obama on the subject which fits in with the solution focused approach and would fit well with the Lib Dems.

The question we ask today is not whether our government is too big or too small, but whether it works.

~Barack Obama

There is more discussion on this subject on this blog here. The issue with the debate is that Labour make an assumption that a big state is both good and necessary to protect people while the Tories assume a big state is wasteful and hampers private enterprise (which is bad for the people). This focuses the debate on the process of how to make the state bigger or smaller. The focus on the process is the problem as there is no debate on what works for people on the ground.

The simple message of ‘doing what works’ allows for experimentation to find what works which means the size of the state can be changed to find what is best for the time and for that particular area. It focused the debate on the outcome rather than the process i.e. the effect for the people and not a change in size of government due to preconceptions of what is best.