Some researchers use genealogy software to record and track their research. Others use word processing software, writing as they go. I’m probably like many in using a combination of both. But in certain instances, neither of these options are as useful as we might like. Difficult research problems that require the collection and analysis of a large amount of data might well be solved by using spreadsheets.
I was attempting to identify the families of origin for my 3rd great-grandparents, John Gregory and Nancy (Shideler) Gregory, who had migrated from Greene County, Pennsylvania, to Delaware County, Indiana, in the early 1830s. A handful of probate records had helped to some degree, but not nearly as much as I had hoped. So I moved on to deeds to solve the problem.
When I searched for Gregory and Shideler grantors and grantees in those two counties, I discovered that there were more than 200 such deeds between 1792 and 1860. At first, I was thrilled there were so many records available to help me. After reading each one, however, I found that no single record cracked the case.
Perhaps a close analysis and comparison of the records would help, but how could I possibly analyze and compare so many records? After mulling the issue over for a while, I turned to Microsoft Excel.
I used one row for each deed and arranged them chronologically from top to bottom. Across the columns, I recorded the county where the deed was recorded, deed book and page number(s), grantor, residence of the grantor, grantee, residence of the grantee, amount of sale (where applicable), acreage (or town lot number), witnesses, and special notes.
Once I finished entering all the data for all the deeds, I sat back to see if this arrangement revealed anything that I had had difficulty seeing otherwise. And it did. While a number of Gregorys migrated out of Greene County, Pennsylvania, only one apparent family unit migrated to Delaware County, Indiana: one man of the right age to be John’s father and five people of the right ages to be John’s siblings. A similar group of Shidelers made the same migration in the same timeframe.
Further, the spreadsheet showed that those potential siblings, both Gregorys and Shidelers, had briefly settled in Miami County, Ohio, before moving on to Delaware County, Indiana. That fact led to yet more records, including some for John. For the Gregorys in particular, who rather insistently used the same few given names over and over, the spreadsheet also differentiated multiple people of the same name across three generations. It also showed ongoing interactions with Greene County more than a decade after migrating westward, prompting me to go back and look for records there in later years than I had previously considered.
In the end, John was firmly placed in his family of origin, and Nancy was aligned with her most likely family of origin. Research on the Shidelers continues.
In another case, I was looking for the families of origin for Andrew Allen and Maria (Lasher) Allen of New York City. An 1819 deed had given me hints as to the parents and siblings of Maria, but I had no such candidates for Andrew. Probate records had once again failed to help me. So I created an Excel spreadsheet to conduct a study of all Allens and Lashers in New York City directories and censuses from 1786 to 1820.
For this study, I arranged the spreadsheet differently. Each column represented a single city directory or census, chronologically from left to right. As I went through each source, every newly discovered person was given his/her own row. Where subsequent sources clearly referred to the same person, the data from that source was added to that existing row, forming a timeline for each person. But if a James or John or William Allen, for example, could not be easily differentiated from others of the same name, that person was given a new row.
In New York City, many people moved frequently and, to a lesser extent, changed occupations, complicating the problem of diffentiating people of same, common names. Arranging the rows of names alphabetically meant that all people of the same name were clumped together. That clumping of same names and their resultant timelines instantly highlighted two things: 1) significant timeline overlaps for similar occupations or addresses, indicating separate people with likely close relationships, and 2) complementary gaps in more than one timeline, indicating that those two or more timelines likely referred to the same person.
Further, by comparing Andrew Allen’s various addresses with the various addresses of other Allens, five candidates for close relations (out of nearly 300 Allens) were identified for further research. That research is ongoing.
As for Maria, however, the spreadsheet showed that there were few Lashers in New York City of the period to consider. The potential parents identified in the 1819 deed mentioned earlier were both covered extensively in the city directories and censuses studied. That data revealed that William and Margaret Lasher were almost certainly her parents; that William died about 1799 (in the absence of any known death, burial, or probate record); and that widow Margaret was likely the woman 65 or over in Andrew Allen’s 1820 household. It also connected Maria to a published Lasher genealogy that only briefly referred to William Lasher, but not his wife or any of his children. Far more was discovered about Maria’s family than I ever anticipated when I started this study.
If you ever find yourself juggling a large amount of data in your research, consider how a spreadsheet might help you analyze that data. And give thought to how you arrange the spreadsheet. The way data is arranged can reveal more than you might expect.