2018-2020 family tree and records reviews
In 2018 we embarked on a thorough review of all our family trees - over 3,350 of them at the latest count. What did this involve, and what have we achieved?
Newcomers to family history research are usually told never to rely on other people's family trees, but to research their own from primary records.
In some ways that is sound advice, because there are many very misleading trees available online and in various books and other publications. But there are also some very well-researched trees which can provide researchers new and old with a sound basis for their own trees. It is up to them to continue their research to prove or disprove what they have found.
The main problem with following online trees is that any subscribers are free to post to Ancestry and other commercial sites, and there is no check on the accuracy of their trees. Although some people who view these trees and spot mistakes will leave comments, to be helpful to both the tree owner and other researchers, this does not happen very much.
And unlike the French site Geneanet, whose trees are renowned for their reliability, Ancestry's format does not lend itself to supplying details of sources for tree details. The number of Ancestry trees which simply quote another tree as a source is staggering. Some other unknown person's tree is not a recognised source.
Jerrripedia was guilty in its early days of accepting users' trees without questioning them - partly because of the paucity of online sources to check the trees against at that time. Since the first online Jersey church records were made available in Jerripedia, and in our associated database, in 2012, users have been able to check their own trees before submitting them, and we have been able to check submitted trees.
Over the past eight years there have been countless examples of trees which have been refused, have had sections removed or amended because we have found errors.
Some of the earlier trees have been revisited over the years and changes and additions made, and the availability of the next batch of church records in 2016, plus the opportunity to view the original registers on Ancestry, have made the work of checking trees much more effective.
For a long time the Jerripedia editorial team concentrated on adding as much new content as possible to the site, but with the tenth anniversary of our launch approaching in February 2020, a decision was taken earlier to switch to a policy of consolidation and checking, with new content a secondary consideration.
This work started in the autumn of 2018 and continues to this day. A small team is carrying out an extensive review of the content of our database of family records. No longer do we need to rely on somebody else's transcription of register entries; we are able to view these and make our own decisions about what was actually written.
This work on the database will continue indefinitely. Records are being checked against the original registers, missing records and extra details added, and corrections made as necessary. We also continue to be advised, by users, of errors which need correcting, and act on their notifications as quickly as possible.
In 2019 a considerable amount of time was spent checking over 14,000 burial records for women identified as wives and widows. Many duplicate records were deleted, and maiden names were transferred to the correct field as necessary.
Our original intention to update our indexes of church records in Jerripedia was abandoned in favour of the complete integration of the database into family pages in the website. It is now possible to search baptism, marriage and burial records for individual families in our 1,200-plus Family pages, and every family tree now includes an A-Z search facility linking directly to the database. This ensures that any search at any time draws on the most up-to-date set of records available anywhere.
What errors remain?
Despite all the work which has gone into checking some half a million records, inevitably errors remain. We have concentrated our efforts on getting names correct, particularly family names, and then on ensuring that the year given in the records is the right one. Because of the way early records were processed, and the poor standards of legibility of some of the transcripts, some dates have been misread when adding them to our database. We think it is much more important to check that years are correct rather than to worry too much about months and days.
But, whenever we have reason to compare our records with the original registers, we try to ensure that every detail of the entry has been transcribed correctly. And our small editorial team is now comparing our records, line by line, with those in the original registers. This is a very time-consuming process and we doubt that we will ever be able to say that our records are 100% correct, because there are known to be mistakes in the registers themselves. But we are trying to spot as many errors, whatever their source, as we can, and also, most importantly, to identify missing records and add them to the database.
Family tree review
In the meantime work never ceases on the very important job of checking our family trees. During 2018 and 2019 we worked systematically down the alphabet updating trees to our standard format, checking content against church registers and making corrections and additions as necessary. Nearly all of our trees were subject to a detailed review.
Wherever possible we added dates and parishes for baptisms and marriages, and whatever background information the more recent registers contained about occupations of grooms, brides and fathers. In many cases trees used to stop around 1842, which was the latest date for which church records had been transcribed. Now many parish registers for marriages and burials are available as late as 1940 (earlier for baptisms) and we have been able to extend many trees closer to the present day.
Although we initially attempted to discover as much information as possible about the marriages of male members of the descendancy, we did not always do so for females, because they do not affect the main lineage. However, a further review of all trees was initiated in 2020 and full details of all marriages are progressively being added to trees. In the process we have been able to add substantial descendant lines previously missing from some trees, and entirely new trees are added when sufficient information becomes available. We continue to accept trees submitted by individual researchers.
The revised format for our trees introduced earlier in 2018 allowed much more detailed information to be contained in notes and references attached to the end of trees, and made our best researched trees much less cluttered and easier to read.
Our policy used to be to create new trees when a substantial section of an existing tree is submitted from a different source. This has led to some families having three, four or more trees covering essentially the same descendancy, making it difficult to decide which to update when more details emerge, and confusing our users.
We are now more likely to consolidate new information in the most appropriate existing tree. This is why the increase in the number of trees has slowed significantly. We are adding more data - it is just appearing in existing trees.
We have also added more branches to trees as we research them and find church records for individuals not previously shown, but time precludes our doing this for all the trees we check. For some we can do no more at this stage than satisfy ourselves about the accuracy of what is already included, and leave enlargement of the tree for another day.
Many trees now include comments indicating that they are no longer being reviewed and updated. This is usually because a more accurate and detailed tree exists for the same lineage. But this comment refers only to the Jerripedia editorial team. Nothing precludes our users from contributing to any of the trees in the site, and if you prefer to add to a particular tree, rather than others relating to the same family, that is your privilege.