Source Analysis and Networking Solves Census Mystery
A 1900 census record in Wilson Township, Yell County, Arkansas, created a six-year research mystery. Using Ancestry’s index, John Devenport’s household included three daughters and one son: M. Devenport, Mary Vitali, Mike Vitali, and Josephine Vitali.[1] However, the image linked to the Vitali children shows they lived in Washington D.C., with their father Olfiren Vitale, not with John Devenport in Arkansas.[2] The image for John Devenport’s Arkansas household reveals additional analysis issues making it difficult to clarify the identity of his daughter “M.”
Several factors bring the Devenport household into question. The enumerator failed to complete the top section. With no state, county, township, or enumeration district, there is no obvious way to confirm that this entry was for a family in Wilson Township. In addition, the household information is faint. It seems that the enumerator wrote the names of two people and then erased them making the record even more difficult to read. Some parts of the document can be read. It appears that John Devenport had been married for 20 years and was born in Alabama. His daughter’s name is unreadable, but she may have been born in Alabama.[3] The FamilySearch and MyHeritage databases contain 27 images for Wilson Township but do not include this image.[4] Because the 1900 census records were microfilmed in the 1940s and the originals destroyed in 1949, no better image exists.[5] Did John Devenport live in Wilson Township or somewhere else and who was his daughter?
Figure 1: 1900 Census Image for “John Devenport” on Ancestry.com
Inverting the image colors and adjusting the contract or brightness did not help. No other Devenport households were enumerated in Wilson Township.[6] Performing an important analysis step, however, could have solved the problem. Every genealogist knows thorough analysis includes looking at the recordset as a whole. There are no images following this record, but there are 27 other images for Wilson Township. Collaborating with LaDonna Garner, M.A., R.V.T, owner of Leafseeker Consulting, presented a clue on the previous census page.[7]
The last entry on sheet 14A in Wilson Township reveals a partial resolution to the Devenport mystery. The image shows the original had deteriorated before being microfilmed. Sections at the top left corner and bottom right corner are missing. The last two entries on the page are barely readable due to a tear repaired with tape. Despite the legibility challenges, this is the entry for John Devenport who was born July 1857 in Alabama. He was widowed, and his household included his daughter who was born May 1883 in Alabama.[8] Unfortunately, his daughter’s name is obscured. It appears to start with a U or V and ends with an E.
Figure 2: 1900 Census Image for John Devenport Household on Sheet 14A
Unfortunately, attempts to invert the colors and adjust the contrast or brightness of this new image were no help. John’s daughter’s name continued to be elusive. A request for help from the Certification Discussing Group on Facebook elicited a suggestion from Karen Isaacson Leverich and provided an answer to this six-year mystery.
During the 1930s, the Works Project Administration included a project creating Soundex cards for the 1900 census. The indexers would have referred to the original census records that were easier to read as microfilming did not begin until the 1940s.[9] Genealogists who researched before the internet age are familiar with these cards. Lo and behold, FamilySearch published the images. The image on this derivative source is clear. John’s daughter is Unice![10]
Figure 3: Soundex Card for John Devenport 1900 Household
While it is still possible that the transcriber of the card made errors, there are lessons to be learned from this example. Words of wisdom to solve those research stumbling blocks…Never forget to analyze the source as a whole. Never forget to use the FAN principle to search for clues in the records of family, friends, associates and neighbors.[11] Never forget to search for related sources that could provide the answer.
My thanks to LaDonna Garner, M.A., R.V.T., Karen Isaacson Leverich, and Jill Morelli, CG, for their contributions to this post.
[1] “1900 United States Federal Census,” database, Ancestry (https://www.ancestry.com : accessed 15 May 2015), entry for John Devenport household, Wilson Township, Yell County, Arkansas.
[2] 1900 U.S. census, Washington County, District of Columbia, population schedule, Washington, p. 91–92 (stamped), enumeration district (ED) 125, sheet 5B–6A (penned), dwelling 86, family 90, Olfiren Vilate household; citing NARA microfilm publication T623, roll 164.
[3] 1900 U.S. census, Yell County, Arkansas, population schedule, Wilson Township, p. # [unavailable], enumeration district (ED) 160, sheet # [unavailable], dwelling # [unreadable], family # [unreadable], John Devenport household; “1900 United States Federal Census,” database and images Ancestry (https://www.ancestry.com/imageviewer/
collections/7602/images/4120034_01183 : accessed 15 May 2015); citing NARA microfilm publication T623, roll 80. This microfilm section includes 31 images.
[4] “United States Census, 1900,” database and images, FamilySearch (https://familysearch.org/ark:/61903/3:1:S3HY-677L-L5?cc=1325221&wc=9BW9-DPG%3A1030551501%2C1033766501%2C1033797301 : 15 May 2015), Arkansas > Yell > ED 160 Wilson Township > image 27 of 27; citing NARA microfilm publication T623, roll 80. This microfilm section includes 27 images and does not include the unlabeled entry for John Devenport. “1900 United States Federal Census,” database and images, MyHeritage (https://www.myheritage.com/research/record-10131-3429308/dotson-h-bently-in-1900-united-states-federal-census : accessed 27 December 2021), entry for Dotson H. Bently, Wilson Township, Yell County, Arkansas; citing NARA microfilm publication T623, roll 80. The Bently household is listed on the 27th image for this microfilm section.
[5] Loretto Dennis Szucs and Sandra Hargraves Leubking, editors, The Source: A Guidebook to American Genealogy, Third Edition (Provo, Utah: Ancestry, 2007), “Census Records: Legibility,” p. 164.
[6] 1900 United States Federal Census,” database and images, Ancestry (https://www.ancestry.com : accessed 15 May 2015), searched surname Davenport (or similar) living in Wilson, Yell County, Arkansas.
[7] Author thanks to LaDonna Garner, M.A., R.V.T, owner of Leafseeker Consulting, for her suggestion to review the previous page.
[8] 1900 U.S. census, Yell County, Arkansas, population schedule, Wilson Township, p. 344, enumeration district (ED) 160, sheet 14A, dwelling 258 or 259, family # [unreadable], John Devenport household; citing NARA microfilm publication T623, roll 80.
[9] Loretto Dennis Szucs and Sandra Hargraves Leubking, editors, The Source: A Guidebook to American Genealogy, Third Edition (Provo, Utah: Ancestry, 2007), “Census Records: Legibility,” p. 164. The author thanks Karen Isaacson Leverich for the suggestion to research the 1900 census Soundex cards.
[10] “Card Index (Soundex) to the 1900 Population Schedules,” images, FamilySearch (https://www.familysearch.org/ark:/61903/3:1:3QHK-H3TX-CDVL?cat=647008 : accessed 17 December 27, 2021), entry for John Devenport, Wilson Township, Yell County, Arkansas.
[11] Elizabeth Shown Mills, “QuickLesson 11: Identity Problems & The Fan Principle,” Evidence Explained https://evidenceexplained.com/content/quicklesson-11-identity-problems-fan-principle : accessed 28 December 2021). Drawn from Elizabeth Shown Mills, QuickSheet: The Historical Biographer’s Guide to Cluster Research (the FAN Principle) (Baltimore: GPC, 2012).
Great post! You did a terrific job. You bring the reader along from mystery to clarity. This is a lesson we often forget. Thanks for the reminder!