Rediscovering the civic and achieving better outcomes in public policy
Our Research
Analysis of existing survey data to understand the civic-outcome link
The main interest of analysing existing survey data is to understand the link between dispositions & orientations and civic participation. Such analytical work is aiming to explore whether and what interventions can be undertaken to change behaviour, especially agents’ behaviour in situations of collective belief formation, action and decision-making.
We are re-analysing data from the Home Office and the Office of the Deputy Prime Minister to throw further light on the link between attitudes and civic behaviour and between civicness and policy outcomes. We are using the Home Office Citizenship Survey (HOCS), which ran in 2001, 2003, 2005 and 2007, which is a unique large dataset of about 10,000 core respondents per wave and a large ethnic minority boost. A core part of the research will analyse questions of the HOCS at the individual level to see if civic and civil activity may be associated with dispositions and orientations.
Randomised Control Trials (RCTs)
Randomised control trials are the gold standard for evaluating policy outcomes, superior to all other methods but they are hard to pull off and may not apply to all situations.
They rely on being able to randomise subjects, groups or communities into treatment or control groups. The intervention applies to the treatment group and the inference is made by comparing outcomes between the groups.
The basic statistics are quite simple, which is one of the attractions of experiments as they do not have the problems of complex statistical estimation of non-experimental evaluations. There is, of course, a lot of statistical theory behind them: see Shadish et al, Experimental and Quasi-Experimental Designs for Generalized Causal Inference (2002).
The main examples come from health, then employment, crime and education with only a few examples from citizen engagement.
- Get Out The Vote - Institution for Social and Policy Studies, Yale University
- Get Out The Vote Campaign - IPEG
- Magenta Book - Guidance Notes on Policy Evaluation
- Randomized controlled trial - Wikipedia article
- The University of York Trials Unit
Emerge Recycling Randomised Control Trial
The randomised controlled trial is delivered in partnership with Emerge, a not-for-profit organisation who deliver a doorstep recycling service to 6800 properties in Greater Manchester. Each household has a recycling bin which is collected weekly. The overall aim of the research is to provide evidence of how households can be successfully encouraged to recycle household waste.
We will start off by measuring recycling participation rates for all households who receive the Emerge service. Observations will be made over 3 weeks because some people may not recycle every week. We will then randomly select some streets where all the households will receive an intervention to encourage recycling. Other streets will be in a control group and receive no special attention. The randomisation will be undertaken independently of the research team, by the Trials in Public Policy Team at the University of York from who we have received advice on this design.
We will recruit a team of casual researchers to undertake the interventions. They will receive training and supervision from Emerge and IPEG. The canvassers will visit all households in the intervention group to: persuade householders of the benefits of recycling; encourage them to participate; ensure they understand the recycling scheme; and recruit volunteer community recycling champions. Householders will be provided with a leaflet about the scheme. Each household will receive up to two visits from the canvassing team to maximise contact. Shortly after the intervention: Emerge will repeat the visits ahead of the bin collection over a 3 week period to check which households are recycling. This will be repeated again several months later to observe longer term effects.
Deliberative Experiments
The deliberative experiment is driven by two developments in democratic practice. The first is the interest in mini-publics such as citizens’ juries, which whilst displaying an impressive level of commitment and reasoning by participants typically involve low numbers (usually 12-25). The second is the increasing interest in the role that ICT (information and communication technology) might play in engaging more citizens in political processes.
We are organising a citizens’ jury and an on-line deliberative space where the 1000 participants will be randomly-selected from Ipsos-MORI’s online panel. Two further on-line groups of 1000 will be exposed to the details and results of each deliberative intervention. A fifth group will serve as a control. Participants will undertake pre and post deliberation surveys and qualitative analyses of the deliberations will be conducted.
Our experiment addresses policy relevant questions such as: How do randomly-selected citizens behave when given the opportunity to participate in a large-scale online consultation event and how does their behaviour compare to that of participants in a citizens’ jury? What impact does involvement in deliberation have on the civic attitudes of those who take part? How does knowledge of the online event and citizens’ jury affect preferences and perspectives of non-participants.
Blackburn Design Experiment
This design experiment investigates effective ways to encourage civic behaviour, exploring how local authority staff can encourage callers to a contact centre to get involved in neighbourhood environmental activity. Our aim is to transform a passive one-way transactional relationship between consumer and provider into a more active two way co-production relationship.
Our experiment will include all callers from one neighbourhood who call the council’s contact centre about cleansing, environment or neighbourhood services. There are two steps to the experiment.
Firstly, officers in the contact centre will identify callers who might be willing to become more pro-active environmentally, provide some basic information on next steps and ask if they would like to get more involved. Our research will help identify how best to do this.
Secondly, the citizens who are identified as interested will be split into two groups. Half will join the intervention group. They will be encouraged - using a variety of approaches – to take further steps such as joining a local group, becoming volunteers or changing their environmental behaviour.
Our research will help identify which approaches work best. Half will join a control group: they will be sent an information pack about local activity they can get involved in.
Our design experiment combines quantitative and qualitative methods. An initial interview will seek baseline information on current civic behaviour and attitudes and ask about motives for getting more involved. A subsequent interview will be completed about 6 weeks after the initial referral to the neighbourhood officer, asking about civic activity and participants’ views on the this attempt to encourage neighbourhood activity.
Additional in-depth research interviews will be conducted with a sample of participants, staff involved in the intervention and community activists in the neighbourhood.
