Combining Corpora and Statistics using Geographical Technologies: New Evidence on Nineteenth Century Infant Mortality Decline in England and Wales

The division between quantitative and qualitative approaches is fundamental to the study of the past. Quantitative approaches are predominantly used to study statistical sources in fields such as historical demography and economic history where large numerical databases are available. These approaches are well suited to situations where large volumes of digital statistics are available and are very good at finding relationships between variables. They can be criticised on two levels: firstly, while good at identifying relationships they are poor at establishing the causal mechanisms that cause them. Secondly, and more fundamentally, most information about the past is not in numerical form and thus cannot form part of a traditional quantitative analysis, thus many relevant factors cannot be included within the analysis. Qualitative sources, particularly texts, are richer in both the range of material available and in the amount of detail it provides about the conditions in which people lived. They are however much more complex to work with and traditionally required close reading which is slow and selective. Digital technologies offer the potential to overcome this divide and make use of the combined advantages of both types of source, making use of statistical sources to identify patterns and relationships, and textual sources to help explain the patterns found. This paper presents an example of this based on infant mortality in Victorian and Edwardian England and Wales.