Development of standardised inventories of legal capability

Grant: £45,000
July 1, 2015

The project aims to arrive at a standardized tool for evaluating legal confidence which can be used to measure the legal confidence of individuals. In order to arrive at this standardized tool, a technique called factor analysis is used.

Factor analysis allows researchers to investigate concepts that are not easily measured directly by collapsing a large number of variables into a few interpretable underlying factors. Factor analysis has been used to arrive at a standardised test for complex concepts such as @ldquo;socio-economic status” or differential diagnosis of psychological conditions. Researchers collect a series of questions that relate to concepts that they believe to be associated with the phenomenon they are attempting to measure (for example, in the case of “socio-economic status” these might include questions about “income”, “education”, “house value”, “number of crimes per year in neighbourhood”). These questions are put to a large sample of individuals, the results collected, and the researcher examines the data to see if there are any patterns in the data which would indicate that the answers given to the questions that have been asked are related to an underlying phenomenon that is not being directly measured by the questions. For example, in the case of socio-economic status, researchers observed patterns in responses given to questions about income, employment, and level of education. Statistical techniques are then applied in order to work out which of the question responses are most likely to account for changes in the phenomenon that is not being directly measured (e.g. in the case of “individual socio-economic status”, (the underlying non directly measured phenomenon), difference in responses to questions about income level account for more change in socio-economic status than level of education). In order to be sure that the questions posed accurately predict the presence of the underlying concept across the population as a whole (and not just within a particular sub-group) they should be put to a group of people who are as representative of the wider population as possible. The best way to guarantee this is to pose these questions to a representative, random probability sample. In order to arrive at a tool for that is not overly long or onerous, the exercise must be repeated, this enables the researcher to refine and reduce the list of questions that are relevant to assessing the underlying concept.

It must be noted that this is an experimental project. Standardised tools for measuring concepts such as “legal confidence” have not been developed before. It may be that the questions that are posed do not reveal patterns in the data that indicate an underlying phenomenon that we might define as “legal confidence”. This is an unlikely but possible outcome that should be borne in mind when considering the level of funding that should be dedicated to this project.