Introduction

This book argues that an analysis of social cohesion needs to focus on whether people from different backgrounds have similar life opportunities as well as assessing interaction, interdependence and conflicts of interests of people from different backgrounds. Whatever the vision of a cohesive society is, there are important questions to answer about where we stand now, and how we can measure change. This chapter focuses on the measurability of social cohesion. We argue that use of flawed data, often analysed inadequately, misleads us as to where we are and where we are going. This is not inevitable, however, because better data and better analytical techniques are already out there.

We note that many existing measures of cohesion emphasise and critique perceptions and ‘cultural’ explanations of problems in cohesion, rather than providing material explanations of competition and exclusion and ideas of social cohesion (Phillips, 2005, 2006; Broadwood and Sugden, 2008; DCLG, 2008a; Ratcliffe et al, 2009). Administrative data not only help to correct that emphasis but, when coupled with survey data, can be used to look at attitudes, behaviour and status.

Building on the previous two chapters by Fuller and Simpson respectively, this chapter begins with a short discussion about the existing measures of cohesion and the limitations with current, and most frequently used, sources of data. The following section considers how the present state of the science could be greatly improved by using a system called ‘Neighbourhood Knowledge Management’ (branded as nkm) applied to normal local authority level administrative data and case studies. We then turn our attention to the practical and ethical issues involved in using such data and how they can be managed.

The problems with existing measures of cohesion

There are a number of problems with existing measures of cohesion. The problems are of three kinds:

• • what is being measured;

• • the sources of data; and

• • how the data are analysed.

What is being measured

Fuller has already argued that a negative approach to measuring cohesion fails to value or measure the benefits that stem from extensive interaction and positive relationships. Both Fuller and Simpson argue that physical proximity is neither a necessary nor a sufficient condition for positive interaction or wellbeing and both highlight problems of confusing statistical association with causality.