This page serves as a form of health warning for historians using directories. Although very important as a source, directories may contain a number inconsistencies. Some of these inconsistencies are an interesting tale in themselves, however.
As you can see from other web pages in this section, directories are undoubtedly invaluable for many types of historical research. They contain a wealth of information relevant to studies at a local, regional and national level. Also, they have survived in sufficient numbers to allow widespread analysis of change over time. Nevertheless, it is important to be aware of the many possible pitfalls so that you can conduct a critical evaluation.
For the historian it is important to know whether or not the information in a directory is accurate:
It is relatively easy to spot inconsistencies between directories, but evaluating their overall reliability is more difficult.
You can compare mid-nineteenth century directory listings with census returns. There are not many comparative sources for earlier periods, though.
There are other ways of assessing reliability. Start by looking at how the directory was compiled and the points below in mind:
Larger firms
Larger firms such as Kelly’s employed full-time professional agents, whose job it was to systematically collect and collate information. Consequently they could produce more consistent and reliable directories than lone compilers undertaking speculative, money-making ventures.
Local firms
Some locally produced directories show evidence of plagiarism and hence can be discounted as reliable sources. For instance, Francis White published a directory of Norfolk in 1854 that was copied verbatim from William White’s directory of 1845.
Early compilation methods
Compilation methods varied according to the resources a publisher had available. The most reliable method was to interview every householder in the survey area. However, some publishers simply delivered circulars, or took details only from subscribers. Some of the earliest directories relied on the lists already held by registry offices, even when these were clearly out of date, or continued to publish old listings in revised editions.
Subsets
All directories cover only a subset of the population living within a particular area, but the size of this subset can vary significantly. The entry for the small market town of Ashby-de-la-Zouch in White’s 1861 directory contained just 17% of the working population, or 33% of householders.
Social exclusion
Women and the poor are particularly under-recorded. The former were excluded because they were rarely householders. The latter lay at the margins of society and were thus considered unworthy of inclusion. Indeed, many early directories claimed only to be listing the traders and merchants of a town, or its principal inhabitants.
Although the population coverage in each directory varied enormously, several general points can be made:
Town size has a significant impact upon the level of coverage, particularly in the eighteenth century. Compilers were most interested in tradesmen, manufacturers and merchants who tended to be concentrated in the larger centres.
The above bias towards listing tradesmen, professionals and manufacturers, also means that craftsmen tend to be under-recorded. Labourers and servants are rarely listed, especially in eighteenth-century directories.
In large towns the poor, transient and those living in multi-occupied houses were the most likely to be missed by directory compilers.
Spatial bias
Directory coverage can vary from town to town, between town and country and within different parts of a particular town.
All these forms of spatial bias are important to the historian, especially if they are trying to trace distribution patterns or compare different places. For instance, coverage of suburban households is often much lower than those residing in the city centre.
Internal consistency
Beware of inconsistencies within directory listings.
Duplicated entries are commonly found in directories, particularly for shopkeepers and manufacturers. These occur when a tradesman is listed under more than one occupational heading due to the diversity of his or her business. For instance, a grocer might also be listed as a tea dealer, provision dealer or general shopkeeper.
Names included in the alphabetical listing, may be omitted from other parts of the directory, such as the street section. Occupational information in often left out or abbreviated in this section too.
Limits to information
The amount of information about particular households can vary considerably between directories due to their layout and method of compilation.
Some directories list only names and occupations, whereas others include full street addresses. Some late nineteenth-century directories emphasise larger businesses by putting their entry in bold type, but in earlier editions there is no way of ascertaining the scale of a business from the directory listing.
Your research
Although this section highlights a range of problems associated with using directories, it is important to keep things in perspective. So long as you bear the above points in mind, you will still be able to appreciate the value and versatility of directories for historical research.