On 31st May, 2016 British Academy Post Doctoral Fellow Francesca Menichelli shared her expertise on ‘Comparative Research in Criminology’ based on her extensive experience in this area. Francesca is a British Academy Postdoctoral Fellow at the Centre for Criminology and obtained her PhD in Urban Studies from the University of Milano-Bicocca in May 2012 on ‘Police run open-street’ CCTV systems in Italy.

Francesca started her presentation with a historical insight into the comparative research in social sciences, as the same was crucial to the development of social science in the late 19th century. Founding figures in sociology like Emile Durkheim started a study to explain the difference suicide rates of Catholic and Protestants, whereas Weber analysed the relationship between religion and economic behaviour. If conducted today, these projects would constitute comparative research. But despite these exemplary research projects, comparative research became less relevant, especially prior to World War II mainly because the scholars became concerned about the standing of social research due its inability to confirm to the higher methodological standards, particularly in relation to hard sciences and natural science. However, after World War II, due to increased internationalisation and global contribution, methodological tools were modified to overcome previous problems. This fostered comparative research further.

Comparative research may be useful in various circumstances, but particularly in challenging or confirming existing theories, contributing to old theories, developing new theories or for of policy making.

However, one of the most important concerns in this research area is ‘comparability’. It’s really important to note that the judgement of the comparability or non-comparability, although cannot be made on absolute grounds but must necessarily be rested on specific properties and characteristics that are relevant in light of the research questions. She adds, ‘it may be important to ask yourself at this stage, what is to be compared, according to what’. For example, studying the impact of health care provisions on life expectancy in UK and US, as the both have provisions pertaining to privatisation.  However, if the research question pertains to unionisation of workforce in liberal market economies, one cannot compare UK and US as they both have liberal market economies. In this case, one might be better off to compare US with Germany due to distinct nature of market economies.

But a quite common mistake or a risk in comparative research is ‘ecological fallacy and individual fallacy’. The former refers to drawing assumptions about individuals based on data that is tailored at group level whereas individual fallacy is the opposite, i.e. drawing conclusions from groups according to individual data. For example, if there is a positive correlation between depravation and illiteracy between the group studied, it does not mean that all poor people will be unable able to read and write. Therefore one has to be clear, what the unit of analysis is and the conclusions being drawn are applied at same level. But there are two strategies to address these fallacies.

First is the maximization of difference. In this, a researcher may have a unit that differs in many ways but have one thing common. The strategy here is to highlight common causes. For example, in a study of political unrest and violence, it may be apt to study Barbados and Beirut. Although different in culture, geographical location, but have faced political unrest violence. Therefore one may uncover some of the common causes that might have led to political violence. Second is the maximisation of similarity. In this case the unit of analysis will have many common feature but are different in one aspect. Here we may compare religious affiliation and role of women in the labour market. In this case one may study Austria and Germany because they have similar culture and language but one is predominantly catholic the other is predominantly protestant.

The last step is the collection and analysis of data. Here it is pivotal to note that comparative research is not the same as comparative method. But one can have a comparative research that employs qualitative methods, quantitative method or mixed methods and all other the methods that apply to research where there is no comparative element but will also apply to research that is comparative in nature. Additionally, there are many other challenges which are required to note and appreciate.

First is the issue of comparability. It is essential to make sure that like is being compared to like. This can be referred to as the oranges and apples problem. The second is the issue of standardisation. Third, is the data collection method standardised and comparable? For instance the European Social Survey is cross-national and conducted every two years across Europe since 2001. In fact, methodological search for optimal comparability has led a standardised methodology.

Thereafter, the major concern is the problem of translation. In this regard it is essential to make sure that the questions are equally understandable in the same context. For example, asking public about civic servant’s mentality. It is possible that in one country it may invoke an image of someone who believes in laws and respect the procedure, while in another it may mean and imply corruption, favouritism etc. Furthermore, at times literal translation may appear to be a better option than syntactical translation. Assistance of an interpreter may also be used in this regard.

Lastly and most importantly, she highlighted the reasons for relying on comparative criminology. Increasing relevance of trans-national and cross border crime, policy making; investigating another country’s criminal justice system to import/adapt it into one’s own system. E.g. people in a different country have found a way to better way to fight robbery, and to check criminological theories in a different context than the one where they originated. it can employ cross-national data to test casual claims: is the link between age and criminal behaviour valid in different countries. Thereafter Francesca ended by giving a brief introduction of her own research on the local governance of crime prevention in three EU countries, i.e. England and Wales, France and Italy.