MCHMORE One Name Study

METHODOLOGY

The purpose of this page is to explain how the family trees have been constructed from the available source data.

The two basic problems that have to be solved are:

to decide if different records refer to the same person
to link each person to their M*CH*MORE parent.

The general procedure has been to look for consistent evidence from at least two sources. When this has not provided a solution, we have followed one of two rules:

If there is some evidence of a link, include the person in a family tree but indicate uncertainty.
Otherwise, put the person in a collection of unlinked families and hope that further evidence turns up later.

Finding evidence for a link is easier in some cases than others. We look at the following cases:

19th century families in England and Wales
17-18th century families in England and Wales
Earlier families in England and Wales
20th century families in England and Wales
Families in other countries

19th century families in England and Wales

Readily available records of persons born in the 19th century in England or Wales normally provide enough information to solve both of the above problems:

Census records for the years 1851, 1861, 1871, 1881, 1891 and 1901 establish many links between the generations. Persons can usually be matched between censuses by their age and place of birth and the composition of their family. The 1841 census contains less information, but can also be valuable.
Civil registration indexes can be used to confirm dates and places of birth and marriage suggested by the census records, but they do not give any further information. Registration certificates are more helpful (providing names and occupations of parents), but these are not freely available. Death indexes and certificates are sometimes useful, mostly in eliminating alternative possibilities.
Dates of birth can be confirmed from monumental inscriptions such as gravestones, which usually state the age at death and often show family relationships.
Spouse names can often be inferred from a combination of the above sources.

Of course, there are many cases where the above do not provide convincing evidence of a link. This happens, for example, when a person was not at home at census time, the person died before 1851, a birth was not registered, two persons of the same name were born at about the same time in the same place,or no gravestone has been found. Additionally, records cannot be treated as 100% reliable because many people were illiterate (and so had no written record of their personal details) and many transcriptions contain errors (due to the difficulty of reading the handwriting).

17-18th century families in England and Wales

Prior to 1837, the only systematic records are church baptism, marriage and burial registers. Some registers go back to the beginning of the 17th century, but the early records are difficult to read. Very few parish registers have been transcribed.

These records contain very little information: Baptismal registers give only the names of the father and mother (from 1813, also the occupation and abode of the father), marriage registers only give the names and parishes of the bride and groom and the names of the witnesses, and death registers often only give the name (from 1813, also the age at death).

The task of linking different records together is made even more difficult by the following factors:

Many parishes kept no records during the Commonwealth period between September 1642 and October 1653.
17th century baptismal registers only recorded fathers� names.
There was only a small number of forenames in use at the time. For example, between 1600 and 1750, nearly 80% of the male children baptised at Buckfastleigh parish church were either John, Philip, Robert, Thomas or William.
No 17th century M*CH*MORE and less than 5% of 18th century M*CH*MOREs had a middle name.

On the other hand, ordinary people were much less mobile than they are today and rarely moved further than a neighbouring parish. See E A Wrigley (Ed.), An introduction to English historical demography, pub. Weidenfeld & Nicholson, 1966: 'Generally speaking, marriages between persons resident in the same parish, and those involving a partner from an adjoining parish or one within a five-mile radius, account for 75-80 percent of all marriages, and if we extend the radius to fifteen miles, we are likely to include all except an insignificant fraction of places of origin of partners' (p. 22).

In order to allow any links to be made, however conjectural they may be, the following assumptions have been made.

Children born to the same-named parents in the same or neighbouring parishes are siblings, and the parents married in the same or a neighbouring parish.
Couples married between 18 and 35 years old, and had children when the wife was between 18 and 40 years old.
Where a burial record states a person's parents, that person died as an infant.

These assumptions, which are based on links where little doubt exists, still do not always allow a definite link to be made. This is especially the case in parishes like Stokenham and Buckfastleigh, where there was a relatively high density of M*CH*MOREs and many had the same first name. On the other hand, they may also have introduced false links.

Earlier families in England and Wales

Before about 1600, only the wealthier people left records�mainly in the form of wills, leases and other legal documents. Not many M*CH*MOREs fell into this category, and linking them is virtually impossible.

20th century families in England and Wales

Linking people in 20th century families is made easier by certain factors:

There are many more first names in use, and middle names are almost universal. As a result, any particular combination of forenames identifies very few people or even a unique person.
From 1912, the civil registration birth index gives the mother's maiden name and the marriage index gives the spouse's surname.
Contributors know the names and dates of birth and marriage of their children and parents (and often grandparents) and may possess the corresponding certificates.

On the other hand, there are no census data available, many M*CH*MOREs have emigrated and privacy considerations restrict the sharing of data.

Other countries

The availability of the information needed to make links varies widely between countries:

Decennial censuses are available for the USA from 1790 to 1930 and for Canada from 1851 to 1901, but Australian census returns were destroyed after analysis.
Civil registration indexes are available for some US and Australian states, but are limited to certain dates and do not exist in any consolidated, easily searchable form.
Some individual church and cemetery records are available, but not easily accessible.
Other valuable records (such as the US Social Security Death Index or Australian electoral rolls) may be available on a country by country basis.