1. Reading and interpreting documents
Working with original documents required the surmounting of three main obstacles: handwriting, language and content. There is a wide range of different hands and a careful seventeenth century hand can be more legible than a twentieth century scrawl. The building up of sets of upper and lower case alphabets for individual hands can help in the deciphering of individual words and care should be taken to establish whether a particular word might be an abbreviation or an unusual spelling - orthography was standardised much later in written hands than in printed books. There are three selections of local documented by the Devon Record Office which include facsimiles, transcripts and, in some cases, translations: Documents of local history: a selection of papers on Devon's past (1969), Oakum: strands drawn from the maritime history of Devon (1970) and Nos voisins d'Outre-Manche (1973). These can all give useful practice. Most modern records are in English, but Latin lingered on in some ecclesiastical records until the early eighteenth century. In some medieval documents French can also be found. Many early documents are largely formulaic and a detailed knowledge of the language need not necessarily stand in the way of getting the gist of the contents. It is important to know the nature of the document being examined, its administrative background and thus what is implied by the forms of words used. Such knowledge will also help with an awareness of what is included in a particular type of record, for example the scope of the heath tax or the information required by the Tithe Commissioners. The Historical Association's Short guide to records series is useful here.

Further reading:
Baxter, J.H. Medieval Latin word-list (1947)
Buck, W.S.B. Examples of handwriting 1550-1650 (1965)
Hector, L.C. The handwriting of English documents (2nd ed, 1980)
MacLaughlin, E. Reading old handwriting (2nd ed, 1987)
Munby, L. Reading Tudor and Stuart handwriting (1988)
Newton, K.C. Medieval local records: a reading aid (1986)
Stuart, D. Latin for local and family historians (1995)
Thoyts, E.C. How to read old documents (1980)

2. Indexing documents
Indexing is not as easy as it sounds and some of the computer programs simply provide undifferentiated strings of page numbers after each heading. Many people for this reason prefer to use manual card indexing systems, but with many word processing systems providing indexing facilities consideration should be given to using these aids while at the same time realising that manual editing will normally be necessary.

Indexers should look out for three main features in the documents they index: personal names, place and other proper names, and subject terms. Depending on the type of document being indexed these may be in the same or separate sequences and there may on occasion be other indexes where appropriate, for example a chronological index for such persons as office holders.

There are a number of factors to be considered:

The form of personal names. Normally these are inverted to bring the surname to the front, but for medieval times the forename may also be significant. Will the surnames be indexed as they stand or will the form of surname be standardised? Standardisation is helpful for the earlier period but should be used with caution from the 18th century onward. In either case "see" or "see also" references should be used. Would a brief identification be helpful, e.g. "Brice, Andrew (printer, Exeter)" or "fl. 1475"? This takes time but can make the index much more helpful and can repay dividends when prolific entries have to be split. Also, if a database is used to produce the index, it might be possible automatically to generate place indexes under Exeter and subject indexes under Printers for the said Andrew Brice.

The form of place-names. Again the question of standardisation must be raised and authorities such as Ordnance Survey can be adopted as a model. Where appropriate reference from the form of names found in the original documents should be given. As with personal names, some form of identification may be useful, e.g. "Portledge (house, Alwington)".

Subject entries: these are in many ways the most difficult and hence the most neglected type of indexing. The choice of terms is difficult, as is the level of specificity and form of subdivision. Generally there is a danger in being too general in the choice of terms e.g. "Trade" instead of "Wine trade". It is difficult for a person looking for something specific to know under what more general term the items may have been placed. Avoid inverted terms wherever possible, cross-references can always be given, e.g. "Transport, road. See Road transport". Prefer concrete to abstract terms, e.g. "Birds" rather than "Ornithology".

The need to avoid strings of undifferentiated numbers means that though should be given to a consistent way of subdividing larger headings. For example for a person the page numbers could be given in numerical order with a brief explanation e.g.:
Birth, 1; early years, 2-3; schooling 4-6 [etc]

An alternative is a form of subject grouping e.g.
Characteristics: generosity, 27; wit, 45.
Travels: Cevennes, 48; Pyrenees, 74

Further reading:
Collinson, R.L. Indexes and indexing (4th ed, 1972)
Mulvany, Nancy C. Indexing books (1994)
Wellisch, H. Indexing from A-Z (2nd ed, 1996)

3. Statistics
Statistics should be gathered into tables wherever possible, although selected statistics can be cited in the body of the text. Work with sources such as census returns and parish registers can soon build up considerable bodies of statistical information. Family reconstitution is a valuable exercise based on parish registers and other sources. From this the length of residency of families within a parish can be calculated as well as average ages at marriage, average family size, infant mortality rates and a range of other demographic data. Useful examples of family reconstitution have been undertaken for Devon parishes by R.R.Sellman, who has placed transcripts in the Westcourty Studies Library. The correct interpretation of the figures obtained is important. The researcher should be aware of the different types of averages: mean, median and mode and the danger of generalising from small or unrepresentative samples. While pie charts and bar graphs can be attractive and have immediate impact, all to often the precise statistics on which they are based are not included and readers may require these for aggregate or comparative purposes. Care should be taken too in drawing the correct type of graph. Linear graphs are only appropriate where there is a true progression along the line. Thus if a series of settlements were ranked in population size a bar chart could be drawn to compare these populations; a line drawn through the population figures might create an aesthetic curve but would prove nothing.

4. Computers in local history
The world of computers is a rapidly changing one but there are basically five ways in which the local historian may use computers:

4.1. Word processing of text: basically using the computer as a glorified typewriter with the ability to manipulate, move and edit chunks of text without retyping. Most standard word processing systems can accept text from other programs either direct or via an ASCII file, which removes all the special codes that individual programs use and so may lose special features such as tab columns or tables on transfer. Widely used programs include Microsoft Word, Microsoft Works, Word Perfect and AmiPro, and some of these have special features to compile tables or indexes. Most word processing packages now have the ability to incorporate scanned images and together with the ability to use a variety of typefaces and page layouts they are the equivalent of a desk-top publishing system.

4.2. Creation of databases: using the computer as a flexible filing system. Each record or row (equivalent to a card record) is broken down into fields or columns (equivalent to a line on an index card with a specific piece of information such a personal name, occupation, or age). It is important to define the structure of each record so that information can later be searched for and sorted as required. Thus names should normally be inverted to bring the surname to the front and if addresses are being stored in a series of fields, it is important that the town appears consistently in the same field in case sorting by town is required at a later date. There are two main types of fields, character and numeric. Numbers can be stored in character fields but it will not always be possible to perform calculations on them. Some database packages also have the ability to display scanned images, such as historic postcards, to which the database can then act as an index. Relational databases link information in different files through a common element in each record. Data can normally be moved from one database system to another, either directly or by converting the data into a special file where each field is delimited so that it can be recognised as a separate piece of data by the new program. It is important to ensure that any database system adopted can "speak" to other systems, otherwise it may not be possible to share or maintain information effectively.

4.3. Development of spreadsheets: this can be undertaken where there is a considerable quantity of arithmetical data to be analysed. A change in any one box (cell) automatically updates the totals in specified columns without extensive recalculations by the researcher. Most spreadsheets are also linked to facilities for producing pie charts or bar charts and spreadsheets can also accept textual data in a similar way to databases.

4.4. Searching for information on networks: with the development of the World Wide Web much more information has become available, including a considerable amount of information of use to the local historian. Among such sites is the Genuki Project, based at Newcastle upon Tyne, which has included much information of genealogical and historical interest on all counties of the British Isles. Devon County Council has its own web pages on Information Devon, including this information on the local studies service. National providers of relevance to local historians include the Institute of Historical Research and the National Archives.

4.5. Production of interactive resources: These can either be as web pages or as material issued on CD-ROM or in other formats. The current drive to provide access to historic images and documents through digitisation is an initiative in which local groups can participate and there can be funding available from lottery and other sources. There are many services which offer to undertake this on behalf of organisations, but much can be produced in-house, especially where a local history group can benefit from the expertise of a member who uses computers in their work. This is too wide and rapidly changing a field for detailed advice to be given here, but some guidance can be found on academic websites such as the Higher Education Digitisation service. An important factor is that data should be made available in a widely recognised format and not locked away in proprietary software systems. The Devon Core discussion paper may be of use here.

Further reading:
Bayley, N.J. CAGe: computer aided genealogy (1995)
Bradley, A. Family history on your pc (1996).
Gloucestershire Local History. Local History Societies and Computers. Guidelines No. 1 (2005).