Talk:Cluster genealogy

Apology to User:NextEdit
This article was created by User:NextEdit and deleted by User:Moriori who was patrolling new pages. User:Moriori has been involved with genealogy for more than 20 years and has collated a massive family tree going back and out for yonks. In that time he has looked beyond direct ancestors to include the collateral kin (brothers, sisters, aunts, uncles, and cousins), in-laws, friends, neighbors, and associates. He and his colleagues call this genealogy. But obviously, others call it cluster genealogy, and clearly the article should not have been deleted so it has been reinstated. It's an easy fix. What isn't an easy fix is that User:NextEdit was so disappointed at the deletion that he said he is leaving Wikipedia for good. User:Moriori regrets that very much and apologises to User:NextEdit, and hopes he somehow sees this message and reconsiders. Moriori 22:12, 3 June 2006 (UTC)
 * Thank you for the apology, and thank you for restoring this article. NextExit 03:19, 7 June 2006 (UTC)

My Attempt at Clarifying Genealogy vs. Cluster Genealogy
I have been using Cluster Genealogy for years, but just never knew it. I even use the term "cluster" when describing what I do to other people. I would define cluster as a record of a family or a subgroup of that family. I think Cluster Genealogy is a method of genealogical analysis, and thus falls within the realm of Genealogy as the broader discipline. It could be called cluster analysis.

I think one difference between the routine genealogical analysis and cluster analysis is in the use of the "cluster" as a link between a source and a family as well as a source and an individual. I use clusters of individuals as evidence to support who belongs to what family. Once a given cluster in a particular source citation is linked by analysis to a specific family (i.e., analysis being how many of the same individuals are in clusters already tied to the family), then I know I can use the detailed facts (direct and indirect) about the individuals in that cluster for the individuals of that family. This prevents errors in misassigning facts to individuals who have similar attributes (e.g., name, birth year) but are not really the same individuals.

Another difference between typical genealogical analysis and cluster analysis (and my second use of clusters) is in linking families together. For example, how do I connect married couple #1 with married couple #2 when I think that the husband of married couple #2 is the son of married couple #1? I do this by taking clusters of both families and comparing them to find similar family attributes, e.g. proximity (both clusters on same page of census, or subgroups of bother clusters in the same household), organization (both clusters attend same church), cemetery (both clusters are in same cemetery, and maybe even adjacent plots), occasion (both clusters at same event), and transaction (clusters engaging in business or legal matters with each other). The more that clusters have these attributes in common, the more likely they are linked by family relationship.

I think most professionally trained genealogists probably think in terms of clusters. However, the write-ups of their analyses fall short of identifying the cluster connections I identified above. This could be because the genealogical software that most now use to compile data, analyses, and conclusions and generate reports and publications does not provide any good place to record such analyses.

I think most researchers use the standard "family record and individual record" method because that is the way paper methods of collecting and compiling data have been for years and that is the way genealogy software is designed. Genealogy software has really brought ease to many of the tasks of genealogy, and has done so to such an extent that it is almost impossible to find a researcher that does not explain their research in terms of the paradigm of their chosen software. Note, this cluster analysis method is not the same thing as defining a source citation and then linking it to individuals and marriages, even if the linking to individuals allows the user to assign roles to the individuals regarding the given citation. Some genealogy software does the latter. However, until the software allows me to link source citations to a relationship between two families, the software will never quite measure up to this method of analysis. Leeirons (talk) 17:05, 4 October 2010 (UTC)