Census Techniques and Strategies for Finding Elusive Ancestors

"Census records are the most used—yet most under used of all genealogical resources." Elizabeth Shown Mills

Preparation
A. Learn about the nature of censuses.


 * Use the Wiki at wiki.FamilySearch.org to learn about federal, state, and local censuses on (a) United States Census page, and (b) (State) Census pages, for example, Illinois Census
 * Explain a census is in the order of visitation, and what happened if no one was home.
 * Discuss why some information on a census may vary from year to year.
 * Discuss what is primary, what is secondary, and what is usually reliable information.

B. Mental preparation


 * Avoid the "census taker skipped my family" attitude trap
 * Develop a "track 'em down!" attitude – plan on finding each family member in every federal, state, local, and school census during their lifetime.
 * Be thorough – plan to look up each city directory, tax record, voter list, and plat map.

C. A well-documented family group record is important for providing search clues. Add a "new event" line for each census.

D. Use a research log for both successful and failed searches.

Index searching techniques
A. Most census indexes will help you find your ancestors, but don't trust an index that doesn't work.

B. For your 1st computer search “less” is good - start with name and state of residence only.


 * If your 1st search fails—


 * Never assume your ancestor was skipped in the census.
 * Never assume your ancestor's name appears exactly as you expect.
 * Always assume you can find an elusive ancestor on the census with more research.

C. If too many hits—one-by-one begin adding fields doing the search over with the added info.

D. If not enough hits—


 * Try alternate competing indexes for the same census
 * Search neighboring counties, states, or nations for the family (change jurisdiction).
 * Use Soundex searches (a phonetic index).
 * Try wildcard searches (*, $, and ?).
 * Keep the same surname, but search for the given name as an initial only.
 * Keep the same surname, but search for abbreviations of a given name (Wm/William, Jno/John).
 * Keep the same surname, but search nicknames of the given name (Polly for Mary, Bill for Wm). See Wiki at Traditional Nicknames in Old Documents - A Wiki List.
 * Keep the surname, drop the first name, search on the middle name or middle initial.
 * Keep the same surname, search without any given names at all (narrow to state and/or county), and go through the hits one by one.
 * Narrow down the locality to a state and/or county, then search on first name only (or age only) and go through the hits one by one. Watch for sibling’s first names in case the surname changed.
 * Search the surname with different vowels (GIL = GAL, GEL, GOL, GUL, GYL).
 * Search for the surname with double letters added or deleted (BAKKER, FULER).
 * Search the surname with the first 3 letters transposed (WIGHT = IWGHT, WGIHT, WIHGT).
 * Substitute frequently misread letters in searches - Printed Sources page 336.
 * Phonetic substitutes in searches - Printed Sources page 331.
 * Search for relatives (children, parents, siblings, uncles, aunts, cousins).
 * Search for next door neighbors from the previous or following census year.
 * If your ancestor had an unusual occupation, search without any names for that occupation.
 * Search the original census in the area where the person lived (without indexes).
 * Search census substitutes like tax records, directories, and plat maps.

When you do find an ancestor in the census—
A. Make note of neighbors – search 3 pages before/after, write nearby surnames on your research log

B. Nine things you must do before you lay your head on the pillow (to bring out clues, and prepare for better analysis later):


 * Capture a photocopy for later analysis (flash drive or paper copy).
 * Type the footnote on the front of the copy to start you thinking about the source.
 * Assign a file number electronically or on back of paper copy.
 * Summarize your findings on all appropriate research logs–often more than 1 family.
 * Transfer all parts of the data to the family group for each family member. At least add a new event (census) for each census year for each family member.
 * Add a footnote for each piece of data added to the family group.
 * Give a preliminary assessment of the data in each footnote's comments field.
 * Print an updated family group (discard the old).
 * File the new family group (and paper photocopy)

C. Move on to the next census year for the same family's members.

Census follow through
A. Track down all the schedules: population, veterans, slave, mortality, pensioners, manufacturers, agricultural, Indians, and defective schedules.

B. Track down various census copies: local, state, and federal, 2nd filmings, and 2nd enumerations.

C. Track down all census years (federal, state, and substitutes like tax records and directories)

D. Track down other documents for the family, evaluate and correlate the information in them.

Census analysis
A. Most census data are fairly accurate, but. ..

B. Do NOT trust censuses fully. Limitations and how to overcome them.


 * Over counts: IF you find a person more than once, check EACH locality for more records.
 * Under counts: compare the tax lists, directories, plat maps, and consult other record types.
 * Misspellings: be flexible, search for alternate spellings, see Printed Sources page 331.
 * Name changes: search all available records, search on given names or for neighbors.
 * Misunderstandings (hearing, language, or alphabet problems): see misspellings.
 * False Information: compare a variety of sources to recognize misleading information.
 * Missing years and records (between Asnapshots@ or lost records): use all available sources.
 * Boundary confusion: Thorndale &amp; Dollarhide, Map Guide to the U.S. Federal Censuses, 1790-1920.

C. For 1790-1840 censuses with tally marks list which family members the tally marks most likely represent based on corresponding gender and age.

D. Analyze and interpret changes over time. Ponder the nuances and implications. Ask probing questions about the data, and what further records they imply exist.

E. Correlate every census in each member of the family's life with data from all the other sources in their lives. Use logic, inference, and corroboration to assess the reliability of the data.

F. Analyze your family in community context, especially in census and land records.


 * Extract full information on individuals in the area with the same surname.
 * Extract full information on individuals by a different surname living in the same household.
 * Identify the neighbors, at least a dozen before and after. Notice who the land owners were.
 * If the family lived near a county or state line, study individuals of the same surname in the adjacent counties or states.
 * Comb the neighborhood for families with similar naming patterns, origins, or occupations.

G. Correlate a variety records of neighbors and infer relationships.


 * Use census, directories, tax records, court records, plat maps, &amp; land to identify neighbors.
 * Use proximity, naming patterns, occupations, document witnesses, analysis, and deduction to infer probable relationships.
 * Correlate other sources about neighbors and use logic to confirm or reject those inferences.

H. Draw a map of the route the census taker went in order to learn the whereabouts of neighbors.

J. Check to be sure all conclusions are reasonable and fit well with the evidence.

K. Explain contradictory evidence.

L. Share your well-documented findings (as part of the analysis, vetting, and research process).

Homework challenge: Find and footnote a family group for EVERY United States census entry for EVERY year for EVERY American ancestor on your pedigree.

Case Studies and Methodology

 * Giroux, Amy, “A Classic Census Problem—Identifying Fathers Before 1850: John Bulson of Orange County, New York,” NGSQ 89 (December 2001): 259-74.
 * Hatten, Ruth Land, “Finding ‘Missing Men’ on Early Census Records: The Example of Thomas Russell,” NGSQ 81 (March 1993): 46-50.