Ten-Minute Methodology: How to Ask Good Research Questions

SpringBoard is pleased to present a guest post by Harold Henderson, CG, a Trustee of the Board for Certification of Genealogists. 

Genealogists are often confused. It comes with the territory. We can’t always avoid it, but we would prefer not to wallow in it. Have you ever had a conversation (with yourself or someone else) that began something like this made-up one?

I would like to find out if Joseph born about 1785 who married “unknown” is my fifth great-grandfather, or sixth. Who did Joseph marry? Who were the children? When and where were they born? Is Joseph the father or grandfather of my Jerome J. Jenkins, and if Joseph is not his father then who? And what about Jemima living in Joseph’s household in 1850? Is she his wife?

To escape this confusion, we need to slow down. In particular, we need to ask one question at a time—and make that question specific.

A good research question does two things. First, it identifies a unique individual. (Sometimes it might be a group or an event, but let’s keep it simple for now.) Second, it specifies what we want to learn about that individual.[1]

We can improve the question by working from the known to the unknown and leave the suppositious Joseph on the shelf for now.  But even “Who is Jerome’s father?” doesn’t identify a unique individual. There are too many Jeromes. Better would be, “Who is the father of Jerome J. Jenkins who was born about 1832 in Kentucky and lived in Hamilton County, Ohio, in 1850?” This question identifies a specific individual (Jerome J. Jenkins) and the specific information we’re looking for (his father). It’s specific enough that we can measure a proposed answer against the Genealogical Proof Standard.[2]

Of course, as genealogists we really want to know all about Jerome, so this procedure may strike us as awfully narrow. But a specific question is a magical thing. By concentrating our attention we focus on the right person and learn more than we expected about him. For instance, we might

  • follow Jerome through later censuses and check to see if a possible sibling or a candidate father shows up in or near his household;
  • locate Jerome’s death record (Does it name a father or a county of birth?);
  • seek out a late second marriage record (Does it name a father or a county of birth?);
  • check for any appearance in county history “mug books” where he settled;
  • check his or his children’s marriage records for a church affiliation that might yield information; and
  • find what role Jerome played in the Civil War (If he served, did his muster roll name a county of birth? Did he or any dependents apply for a pension?).

If we find a possible birthplace and a candidate for a father and/or siblings, we can seek out their records, especially the candidate father’s probate. If looking forward doesn’t help, we might cast a wider net and

  • review the 1840 census for heads of Jenkins households in Kentucky containing boys aged 5–10; and
  • look for Jerome in Hamilton County land records and start building a list of friends, associates, and neighbors who might lead back to Kentucky.

And so forth, depending on what turns up in the process. Obviously any leads or contradictions (such as other census records saying he was born in Indiana instead of Kentucky) could redirect the search drastically.

Thorough research for Jerome’s father will inevitably pull in more than just that relationship. When we pursue the answer to this one question, we find ourselves gathering information that answers other questions and even starting to fulfill our original naïve wish to learn “everything” about Joseph.

And we’re sticking with our one question. We don’t start by googling a name and staring at dozens of probably irrelevant results. We follow a strategy that could provide a firm foundation for the next question we ask.

 Seek specifically, and ye shall find abundantly!

 



[1] Genealogy Standards: 50th Anniversary Edition (Nashville, Tenn.: Turner Publishing, 2014), Standard 10, “Effective Research Questions,” 11–12.
[2] “The Genealogical Proof Standard,” Board for Certification of Genealogists (http://www.bcgcertification.org/resources/standard.html : viewed 22 January 2016).

The words Certified Genealogist are a registered certification mark, and the designations CG, CGL and Certified Genealogical Lecturer are service marks of the Board for Certification of Genealogists®, used under license by board certificants after periodic evaluation.

Ten-Minute Methodology: Documentation and the Research Report

Some genealogists have been confused about whether to include source citations in research report introductions and summaries. SpringBoard is pleased to offer clarification of this question by the expert on genealogical source citations, Elizabeth Shown Mills, CG, CGL, FASG, FNGS, FUGA.

Documentation and the Research Report
Elizabeth Shown Mills, Certified Genealogist®, CGL

It’s a fact of life. In a world governed by laws and standards, rules for one aspect of life often collide with rules for another. So it seems with the research report by which we genealogists chronicle each block of research we do.

A research report has one basic goal: to provide written documentation of the research process we executed, the findings we developed, and the conclusions we reached. As a work product, a research report is expected to achieve an appropriate balance of both writing skills and documentation skills. Each of these two skills is governed by one fundamental concept.

For documentation, the basic rule is this:

Each “fact” or assertion we make, if it is not “common knowledge,” should be supported by credible evidence. That evidence will be presented in one of two ways:

  • For simple, direct evidence, that “fact” or assertion can be supported by a simple source citation.
  • For more-complicated evidence, our conclusion must be supported by a proof argument or a proof summary.[1]

For expository writing, the basic rule drilled in us since middle school is this:

Introduce your subject, then

  • say what you’re going to say,
  • say it,
  • then say what you’ve said.

A good research report will embody both concepts. However, report writers sometimes perceive a conflict.  As a step toward understanding the issue, let’s outline the two types of work products.

An essay based on a research topic has three main parts:

  1. INTRODUCTION, where we
    1. Give readers the most-basic information needed to understand the subject.
    2. Tell the reader what our research will prove.
  2. BODY OF PAPER, where we
    1. Present our findings and analyses that, all together, make the case for what we said we would prove.
    2. Support each “fact” or assertion with a citation to a credible source.
  3. CONCLUSION, where we
    1. Reiterate our main points.
    2. Issue a call-to-action if appropriate.

A report based on a segment of research also has three main parts:

  1. INTRODUCTION, which consists of
    1. Background: the who, what, when, and where that we will be researching—i.e., the essentials about the problem that readers need to know.
    2. Executive Summary: an easy-to-find and quick-to-grasp overview of the results of the research—positive and negative; often presented as a bulleted list of conclusions.
  2. BODY OF REPORT (“RESEARCH & FINDINGS”), where we
    1. Present an item-by-item account of what we have searched, what we did or did not find, whatever significant problem or anomalies we encountered, and the conclusions we have drawn from this body of evidence.
    2. Support each abstract or transcript with a citation to the record; document each contextual “fact” we add from general study; and provide a proof argument for all conclusions we reached from indirect, complex, or conflicting evidence. We may or may not attach image copies of records, with citations on each image and a cross-reference in the report.
  3. CONCLUSION, where we
    1. Reiterate our main points.
    2. Make suggestions for future research based upon our latest findings and conclusions.

Both types of writing have the same essential needs. Both follow the same pattern. It’s a pattern we see in a variety of educational venues, from journal articles that often begin with extracts or abstracts, to textbook chapters and conference syllabi that often begin with bulleted lists of key points.

The perception of conflict arises when we overthink the documentation rule and assume it must override the basic rule for expository writing. The result is a new assumption with a logistical impossibility:

Assumption:
If every statement of fact must have a citation of source, then every fact asserted in the introductory background and executive summary must also carry documentation.

Logistical Impossibility:
Given that the purpose of the background and the summation is to provide an easily digestible recap of main points—and given that research conclusions are based on the whole body of evidence—then providing citations for any assertion in the introduction is possible only when a point is based on simple direct evidence.[2] Even then, in the background or the summation, the totality of the citations could easily overwhelm what is supposed to be a simple recap at the start of the report.

When our summary points are based on extensive and complex evidence, “documentation” of each point would often require a lengthy discussion of how that conclusion was reached. That discourse and all the citations necessary to support it would strip the introduction of its core function: a quick summation of the main points the paper will develop.

As genealogists who strive to meet all standards, do we violate the documentation rule when we summarize background facts or briefly recap conclusions in our introduction?

No.  Standard 2, the one most relevant, instructs us to attach citations for

  • each statement … that is someone else’s observation, deduction, or opinion;
  • each fact that is not common knowledge;
  • each image the genealogist shows of someone else’s creation; and
  • each conclusion the genealogist establishes.[3]

Standard 2 does not state that each fact or conclusion must repeat its citation or its supporting proof argument each time the point is mentioned.

In sum:
Each part of our research report has a specific function. The presentation of documented evidence is the function of the body of the report—the section typically labeled “Research & Findings.” The function of the introduction is to provide a directory or a road map of what’s to come, so our readers will not be lost in the maze of evidence that the body of the report presents.[4]

 


[1] Summarized from Board for Certification of Genealogists, Genealogy Standards, 50th Anniversary Edition (Nashville and New York: Ancestry Imprint, Turner Publishing, 2014), Standards 1–3, 53.

[2] For example, in the introduction, the set of “facts” we assert for a problem person might be simply cited to the client’s file or letter or to a prior report.

[3] Genealogy Standards, p. 6, Standard 2: Specificity.

[4] For additional guidance on the creation of research reports, see Elizabeth Shown Mills, “QuickLesson 20: Research Reports for Research Success,” Evidence Explained: Historical Analysis, Citation & Source Usage (https://www.evidenceexplained.com/content/quicklesson-20-research-reports-research-success : posted 23 May 2015).

The words Certified Genealogist are a registered certification mark, and the designations CG, CGL and Certified Genealogical Lecturer are service marks of the Board for Certification of Genealogists®, used under license by board certificants after periodic evaluation.

Ten-Minute Methodology: Beyond the Index—or Not

Our goal, supported by genealogy standards, is to use whenever possible original records and primary information.[1] That’s the gold standard. When we find an index or other derivative source, we set about finding the original from which it was created. That was the gist of our last post on indexes.

When we know what we want, and we can’t get at it because of access restrictions or record loss or destruction, we are challenged to use our creativity and knowledge of sources to provide substitutes. When no substitutes surface after reasonably exhaustive research, we use the index as our best source. This is, however, a last resort.

Access limitations

The Social Security Death Index (SSDI) is the go-to resource for U.S. deaths after 1936.[2] However, it does not point us to original death records. We have to find these ourselves. Sometimes original death records are protected by access restrictions, and we have to seek other sources that provide the same or nearly the same type of information.

For example, New York state death certificates are closed for at least fifty years after the death.[3] A recent article by BCG associate Laura DeGrazia demonstrates use of alternate sources for an inaccessible death certificate for Theresa (Sabbatino) Rebaschio, who died 9 November 2009. Laura substituted correlated information from three sources, an obituary, a gravestone, and interviews with Theresa’s daughter. She also checked two other sources without success. Laura’s source citation details them all.[4]

152 Theresa C. Rebaschio obituary, Tributes.com (www.tributes.com/show/Theresa-C.-Rebaschio-877155151), which includes her dates of birth and death as well as a condolence message that mentions her late husband, Joe. Theresa C. Rebaschio marker, Block 5, Section 50, Range 32, Plot M, Grave 194, Pinelawn Memorial Park (Farmingdale, N.Y.), read by the author 20 Oct. 2013, which indicates her birth and death dates. Telephone interviews with Theresa’s daughter (name withheld for privacy) by the author, 2011–2013. Neither Theresa’s birth nor death record is in the public domain. No death notice or obituary was found in Newsday (Melville, N.Y.). No record of her estate was found in Nassau Co.

Pursuing leads in five sources was certainly time consuming. It gave Laura the confidence in this identity and date and place of death to publish her findings without the death certificate.

Sometimes it’s OK to cite the SSDI or another index. Sometimes it’s about all we have after reasonably exhaustive research. In research for the same article, Laura used information from the SSDI as a starting point to corroborate or disprove a date and place of death. She came up with an obituary that confirmed both but was unsuccessful in locating an estate record in two target counties. Her source citation includes the obituary, the SSDI, and her negative searches.[5] It’s important to stress that using the SSDI was a starting point, not the end of the research.

101 Anthony De Grazia obituary, Poughkeepsie Journal (Poughkeepsie, N.Y.), 24 Apr. 1985, p. 14, cols. 1–2. Anthony De Grazia entry, Social Security Administration, “Social Security Death Index,” database, Ancestry.com (www.ancestry.com). No estate record was found in Dutchess or Orange County.

Lost or destroyed originals 

Index to an unidentified account book, first page.

Sometimes indexes and extracts are all that remain after original sources have been lost or destroyed. Then we’re happy to have them at all! In these cases we must use them, cite record loss or destruction, and explain that what we used is as close as it is possible to get to originals.

For example, the New England Historic Genealogical Society (NEHGS) has several original indexes to account books, but not the account books themselves, and the names of the merchants are unknown. Still, much valuable family information can be gleaned from just the indexes. This image of an account book index page shows in just one section of A’s many Attwoods distinguished by first name, “Senr” and “Junr,” and “Capt.” Other people are designated by residence (“of plimton”), race (“Indian”), or occupation (“Sailor”). Index entries identify customers from towns in a cluster south of Boston (Kingston, Plymouth, Plympton, Middleborough) and a time period around 1758–1764.[6]

When this and other similar indexes are catalogued and digitized (hopefully early 2016), they will provide researchers with invaluable original source material.[7] Even though they are “only” indexes, they offer primary information (the merchant’s identification of his clients) that may be otherwise unavailable and that might solve someone’s genealogical mystery. They will figure legitimately in genealogical source citations.

As another example, the Vermont Vital Records Index resulted from a Vermont law that “required all town clerks to transcribe, in full, records of births, marriages, and deaths in the possession of the town and churches.”[8] The transcriptions were made on index cards now available online at FamilySearch and Ancestry. Researcher Catherine Desmarais, CG, found an index entry for a birth recorded at St. Paul’s Cathedral in Burlington, Vermont. Pursuing the original source, she found that a 1971 fire had destroyed St. Paul’s and 140 years of its records.[9] A copy of the church records had been made for the diocese, but it does not include the event in question. In this case, Catherine used the index entry, but she did not stop there. She corroborated the index information with census and death records, citing all three sources in her work product.[10]

Indexes sometimes point us to missing or unavailable records, challenging us to pursue reasonably exhaustive research in our quest for original records and primary information. Going beyond the index strengthens the foundation supporting our genealogical conclusions. It demonstrates our commitment to working to genealogy standards by providing evidence from the best sources we can possibly find. 

Quiz (optional): Do you know of an index that survives after originals have been lost or destroyed? What is its source citation?

Extra credit (also optional): What other source(s) could substitute for the missing records?

Photo courtesy of NEHGS, R. Stanton Avery Special Collections. The author gratefully acknowledges input from Timothy Salls and the following BCG associates: Jeanne L. Bloom, Ruy Cardoso, Laura Murphy DeGrazia, Catherine Desmarais, Joan Hunter, and Michael Leclerc.


[1] Genealogy Standards (Nashville, Tenn.: Turner Publishing, 2014), Standard 13, Source-Based Content, 12–13; and Standard 38, Source Preference, 23.

[2] “United States Social Security Death Index (FamilySearch Historical Records),” FamilySearch (https://familysearch.org/learn/wiki/en/United_States_Social_Security_Death_Index_ (FamilySearch_Historical_Records)) : accessed 17 November 2015), “Record Description.”

[3] “Genealogy Records & Resources,” New York State, Department of Health (http://www.health.ny.gov/vital_records/genealogy.htm : accessed 16 November 2015).

[4] Laura Murphy DeGrazia, CG, FGBS, “Con Spirito: Violinist Giuseppe De Grazia, 1855–1937, of Marsicovetere, Italy, and New York City,” The New York Genealogical and Biographical Record 146 (January 2015): 68. This is a good example of a proof in a footnote.

[5] Laura Murphy DeGrazia, CG, FGBS, “Con Spirito: Violinist Giuseppe De Grazia, 1855–1937, of Marsicovetere, Italy, and New York City,” The New York Genealogical and Biographical Record 146 (January 2015): 64.

[6] Index to an unidentified account book, New England Historic Genealogical Society, R. Stanton Avery Special Collections, Mss A A33, uncatalogued, first page; digital image courtesy of Timothy Salls, Manager of Manuscript Collections, NEHGS. The index survives, but the remainder of the book does not.

[7] Timothy Salls (Boston, Mass.; tsalls@nehgs.org) to Judy Kellar Fox, email, 20 November 2015, “RE: unidentified account book index.”

[8] “State Registry History,” Vermont, Secretary of State (https://www.sec.state.vt.us/archives-records/vital-records/state-registry-history.aspx : accessed 1 December 2015).

[9] “Cathedral Church of St. Paul (Burlington, Vermont),” Wikipedia (https://en.wikipedia.org/wiki/Cathedral_Church_of_St._Paul_(Burlington,_Vermont)) : accessed 1 December 2015).

[10] Catherine Desmarais (Essex Jct., Vt.; stonehouseresearch@gmail.com) to Judy Kellar Fox, emails, 27 November and 2 December 2015, “Re: St. Paul’s question.”

CG, Certified Genealogist, CGL, and Certified Genealogical Lecturer, are service marks of the Board for Certification of Genealogists, used under license by Board-certified genealogists after periodic competency evaluation, and the board name is registered in the US Patent & Trademark Office.

Ten-Minute Methodology: When Index is a “Dirty Word”

Genealogical work supported by indexes alone can be unreliable. What? What’s wrong with the Social Security Death Index (SSDI)? It comes from a reliable source. Other indexes are good, too! Why not use them and cite them as sources?[1]

Standard 38, Source Preference, gives us the answer: “Whenever possible, genealogists prefer to reason from original records that reliable scribes carefully created soon after the reported events.”[2]

Standard 13, Source-Based Content, points out that our research plans may include indexes. However, when it’s possible, we “follow such materials to original records and primary information.”[3]

What is an index, anyway? We usually use the word to refer to an alphabetical list or guide that points us to more information elsewhere. By nature it is a derivative source, something created from something else.

Other so-called indexes offer much more than direction to find an item. Some include extracted information, helping us distinguish people of the same or a similar name.

Here’s an example of information gleaned from two online indexes. It gives a pretty good overview of one man. 

  SSDI[4] California death index[5]
Name John Kellar John M Kellar
Social Security Number 560-03-9682 560-03-9682
Death date Aug 1964 20 Aug 1964
Death place Sonoma [County], California
Birth date 14 Jun 1879 14 Jun 1879
Place where benefit was sent California
State (Year) SSN issued California (Before 1951)
Birth place Pennsylvania
Gender Male
Mother’s name Philippi

So what’s wrong with these indexes? Sure, they are derivative sources, but not all derivatives are bad. What can make index a “dirty word”?

There are two issues.

  • Extracting information involves recopying it, which introduces the possibility of error, no matter how carefully the extracts are proofread.
  • Creators of the indexes make choices about what information is extracted and what is left behind. Usually we are not given all the information from the record. This could lead to incorrect assumptions about what we do have. And what we’re missing could be clues to lead our research in productive directions.

The original death certificate for the above example reveals the extraction choices made by the California death index.[6]

If research stopped at the index level, the following information would be missed:

  • employment
  • father’s name and birthplace
  • mother’s given name and birthplace
  • spouse’s name and that she was the informant for the personal information
  • residence
  • length of residence in the state and county
  • medical and health data
  • burial information

Much of this information is primary, provided by knowledgeable informants (the spouse, the physician, the embalmer). This is one reason we are not satisfied with index entries alone. The original source may have errors, but they are not errors of transcription that could creep in when making an extract.

Indexes are useful in pointing the way. We note the information they offer and then set off to find an original source for the information we need.

Here’s the rub. First, we have to figure out if it is even possible to get the originals an index points to. The next step is learning how to get the originals. In the case of many online records, a digitized image of an original document is available with a few more mouse clicks. Other times we must determine the repository and then make contact in person or via telephone, email, or snail mail. Sometimes we are hamstrung by legal restrictions on records access, as with some recent vital records.

In the case of our death certificate example, Sonoma County, California, death records are available on Family History Library microfilm through 1919 only.[7] For a more recent death we can apply to the county or to the California Department of Public Health. (Ancestry even provides a link.) Oh yes, and we have to pay $21 and send an application through the mail. Or we can order online through a third-party site, which reduces the wait and increases the fee.

So here’s the question: Our n-generation genealogy includes many people in recent generations for whom death certificates are available, but not published, not on microfilm, and not online. It’s an expensive and time-consuming proposition to order death certificates for all the individuals. Yet genealogy standards urge us to use original sources. If we don’t, if we opt to use the “dirty word,” our list of source citations can include many derivative indexes. What do we do?

Time and money. That’s often the cost of accessing original records. However they almost always give us far more than the index, and we can judge for ourselves the quality of the source and perhaps the reliability of the information in it. We don’t have to worry about introduced errors. We usually have new hints to work with, too. And we have the satisfaction of working to standards by getting closer to original records and primary information.

Watch for the next SpringBoard post for ways to meet standards when the original records indexes point to are lost, destroyed, or otherwise inaccessible.

 

 

 


[1] See Anita Anderson Lustenberger, “Skillbuilding: Using Indexes,” OnBoard: Newsletter of the Board for Certification of Genealogists 3 (September 1997): 24.
[2] Genealogy Standards (Nashville, Tenn.: Turner Publishing, 2014), 23.
[3] Ibid., 12–13.
[4] “U.S., Social Security Death Index, 1935–2014,” database, Ancestry.com (http://search.ancestry.com/search/db.aspx?dbid=3693) : accessed 28 October 2015), search for John Martin Kellar, d. 1964 California.
[5] “California Death Index, 1940–1997,” database, FamilySearch (https://familysearch.org/ark:/61903/1:1:VP4D-BWH : accessed 28 October 2015), John M Kellar, 20 Aug 1964; Department of Public Health Services, Sacramento.
[6] Sonoma County, California, Certificate of Death no. 4900-1176, John Martin Kellar, filed 26 August 1964; Register and Recorder, Santa Rosa. The Registrar’s Certification Statement is pasted on the left side of the certificate. It has been folded back for purposes of this post.
[7] “Death records, 1873–1919,” catalog entry, FamilySearch (https://familysearch.org/search/catalog/302916?availability=Family%20History%20Library : accessed 6 November 2015); citing Sonoma County Recorder.

Ten-Minute Methodology: “Reasonably Exhaustive”—How Do We Know We’re There?

Evaluating what we’ve done

There comes a time when we have to stop researching. Sometimes it’s when we feel we’ve answered our research question. Sometimes we’re unsure, fearing we missed something. We can evaluate our research to see if it’s reasonably exhaustive by asking questions about its breadth and strength.[1]

1. Do the research results answer our question? Have we used them to make a convincing case?

2. Does the research address all potentially relevant available sources? This is where we listen to that small voice that reminds us about that repository or informant we’ve been reluctant to visit or contact. Genealogy Standards asks if it covers “sources competent genealogists would examine to answer the same research question.”[2]

Hmm. How do we know, aside from the big names in the field, who’s a “competent genealogist”? How do we know what sources they would examine to answer our research question? This seems like a very subjective measure. It helps to go back to the last post and look at it with the mindset that competent genealogists wrote the standards. That’s better. Then we can see the lists of factors to consider when selecting our sources and use them as our guide.

3. Does the research cover a wide variety of sources and does it “replace, where possible, relevant authored narratives, derivative records, and information that is secondary or undetermined”?[3] We have to dig deeper and find out where those derivative sources got their facts. We can’t just trust someone who collected and published names a century ago.

4. Have we tested our collected data by analysis and correlation to establish its accuracy and consistency? Put another way, do “at least two sources of independent information items agree directly or indirectly on a research question’s answer”?[4]

5. Does the research provide “at least some primary information and direct, indirect, or negative evidence from at least one original record”?[5] This is the really big question that urges us to seek out the primary information from original sources beyond where the internet can take us.

6. Have all conflicts been resolved? How well?

When we’re ready to stop researching we acknowledge that when we make our conclusion public (in whatever format we choose), there is a possibility that someone somewhere will turn up a source with new evidence. These questions are intended to minimize the likelihood of that happening. We may ask one last overall question:

7. What is the risk of new evidence coming to light that will overturn this conclusion?

What? There’s more research to do?

 


[1] The definition of “reasonably exhaustive research” in Genealogy Standards informs most of these questions. See Genealogy Standards (Nashville, Tenn.: Ancestry.com, 2014), Glossary, 74–75. The criteria are treated more fully in Thomas W. Jones, Mastering Genealogical Proof (Arlington, Va.: National Genealogical Society, 2013). See especially Chapter 3, “GPS Element 1: Thorough Research,” 23–29. Also Thomas W. Jones, “When Enough Is Enough: How Much Searching is ‘Reasonably Exhaustive’?” Association of Professional Genealogists Quarterly 25 (March 2010): 25–33. The author thanks Alison Hare, CG, for invaluable input.

[2] Genealogy Standards (Nashville, Tenn.: Ancestry.com, 2014), 74.

[3] Ibid.

[4] Ibid.

[5] Ibid.

 

CG, Certified Genealogist, CGL, and Certified Genealogical Lecturer, are service marks of the Board for Certification of Genealogists, used under license by Board-certified genealogists after periodic competency evaluation, and the board name is registered in the US Patent & Trademark Office.

10-Minute Methodology: What Is “Reasonably Exhaustive” Research?

Let’s look at the searching part of research. As in the last post on research, we’ll see that it’s more involved than just looking.

Genealogy Standards offers this goal of reasonably exhaustive data-collection: “Genealogists attempt to collect all information potentially relevant to the questions they investigate.”[1] Got that? All information. All potentially relevant information. Potentially relevant to the questions under investigation.

Our research question determines what “reasonably exhaustive” research looks like. We start with a problem, most commonly a question of identity or kinship. The individuals, what we want to know about them, the time and place in which they lived, and their life circumstances set the course of our research.

Time and place. Our research subjects lived in specific places in specific time periods. If we don’t know about the place or time period, we have to educate ourselves about history and geography at the least. Relevant considerations include

  • historical boundaries and their changes,
  • migration patterns and routes,
  • what sources are available in the relevant times and places.[2]

Life circumstances also determine the direction of our research. Aside from our subjects’ gender, race, and ethnicity, their life patterns suggest numerous research paths and types of sources. They interacted with legal and governmental entities and religious institutions. They may have served in the military or participated in social clubs. They had financial and trade dealings. They spoke a language and often left traces of their handwriting. They had many or few belongings of different kinds. They were wealthy or poor, free or enslaved. They left a legacy in their descendants’ DNA. Thinking of our subjects as people in families and communities suggests a wide variety of sources to discover and examine. Looking at all these factors in the relevant times and places will potentially lead us to a wide variety of sources that may name or bear on our research subjects.[3]

Quality information. At the outset the list of types of sources to be researched can be quite long. It includes “databases, finding aids, indexes, search engines . . . authored narratives, derivative sources, and documented and undocumented genealogies.”[4] However, we aim, wherever possible, to find the original records alluded to in other works and information provided by informants as close as possible to the events in question.[5] Sometimes this effort draws out our research time and effort considerably, all in the pursuit of “reasonably exhaustive” research.

The idea of testing our hypotheses addresses “reasonably exhaustive” research head-on.[6] Not only do we seek and gather data, but we also compare items to each other. Depending on how they correlate with other information, we accept them as evidence for or against our hypotheses. Some information may conflict, pushing us to research more until we can resolve discrepancies. Reasonably exhaustive research extends beyond searching to the mental effort of evaluating and processing our data to be sure we can test our theories.[7]

If we don’t have enough data to test, we broaden our search. We extend it to include our subject’s family, associates, and neighbors. We then seek out the same types of information indicated above, but for a wider circle of people.[8] This can be very time consuming and result in the accumulation of much data. Within it, however, we may well find the evidence we need to answer our research question.

So how do we know when it’s safe to stop? Watch for the next “10-Minute” post.

 


[1] Standard 19, Data-collection scope, Genealogy Standards (Nashville, Tenn.: Ancestry.com, 2014), 16.
[2] Standard 12, Broad context, Genealogy Standards, 12.
[3] Standards 12, Broad context, and 14, Topical breadth, Genealogy Standards, 12, 13.
[4] Standard 13, Source-based content, Genealogy Standards, 12–13.
[5] Ibid.
[6] Standard 17, Extent, Genealogy Standards, 14–15.
[7] Genealogical Proof Standard, bullets three and four, Genealogy Standards, 1–2.
[8] Standard 14, Topical Breadth, Genealogy Standards, 13.

Ten-Minute Methodology: Are You Searching or Researching?

Are you up to date? From the old Standards Manual to the new Genealogy Standards the first component of the Genealogical Proof Standard (GPS) is different. Have you noticed? We used to say, “We conduct a reasonably exhaustive search.”[1] That’s still true, but there’s more. We conduct reasonably exhaustive research.[2] Those two added letters bring much more to what others expect of us as genealogists and what we must expect of ourselves.

Research connotes more than search. It covers searching, of course—careful and thorough searching in a wide variety of sources. It also includes planning, critical thinking, and evaluation. And research includes strategies that go considerably beyond identifying relevant records and searching for a name of interest.[3]

Let’s look at how the concept is presented in reference works we use often.

The Encyclopedia Britannica dictionary includes in its definition of research

  • “careful or diligent search” and
  • “studious inquiry or examination, especially critical and exhaustive investigation . . . having for its aim
    • discovery of new facts and their correct interpretation;
    • revision of accepted conclusions . . . in the light of newly discovered facts;
    • practical applications of such new or revised conclusions.”[4]

The glossary in Genealogy Standards describes research as “an investigation designed to discover or interpret facts and thus to advance knowledge.”[5]

As genealogists, we examine sources and collect information, always subjecting both to critical evaluation. Standards 35 and 36 advise us to “appraise [the] likely accuracy, integrity, and completeness” of our sources and information.[6]

We also interpret the information we find. We think about it and decide if it becomes evidence to support our hypotheses. Standards in “Reasoning from Evidence” apply to the mental processes we perform on our collected data to turn it into evidence and to use that evidence to draw conclusions.[7]

Evidence Explained sheds more light on the concept of research:

  • “As history researchers, we do not speculate. We test. We critically observe and carefully record. Then we weigh the accumulated evidence, analyzing the individual parts as well as the whole, without favoring any theory.”[8]
  • “Research is much more than an accumulation of data. It is a process that requires continual comparison of new information against the old.”[9]

This first element of the GPS, even the word research alone, carries in it the sense and the value of the whole standard. The words of the GPS define us as not just lookup artists, no matter how skilled or experienced. We are more. As researchers we collect data, subject it to rigorous evaluation, compare and contrast it with other data and conclusions, and propose new information or conclusions. That’s a big responsibility. The GPS takes us there with the mindset of researchers, not just seekers.


[1] The BCG Genealogical Standards Manual (Orem, Utah: Ancestry Publishing, 2000), 1.

[2] Genealogy Standards (Nashville, Tenn.: Ancestry.com, 2014), 1.

[3] The author is grateful to Elizabeth Shown Mills for input to this paragraph and encouragement overall.

[4] “Dictionary,” Encyclopædia Britannica (http://www.britannica.com/dictionary/research : accessed 28 June 2015), s.v. “research.”

[5] Genealogy Standards, 76.

[6] Ibid., 21, 22.

[7] Ibid., 23–29.

[8] Elizabeth Shown Mills, Evidence Explained: Citing History Sources from Artifacts to Cyberspace, 3rd ed. (Baltimore: Genealogical Publishing Co., 2015), 15.

[9] Ibid., 16.

Harold Henderson, CG: Analyze or Else!

SpringBoard is pleased to offer a post by guest blogger Harold Henderson, CG. Harold has been a professional writer since 1979, a professional genealogist since 2009, and a Board-certified genealogist since June 2012. He lives and works in northwest Indiana, serves as a trustee of the Board for Certification of Genealogists, and has published in National Genealogical Society Quarterly (NGSQ), New England Historical and Genealogical Register, New York Genealogical and Biographical Record, and elsewhere. His website is midwestroots.net.

Analyze or Else! by Harold Henderson, CG

Genealogists who meet standards do so in part by asking pointed, even impolite, questions about every document and piece of information they see. It’s called analysis. Genealogists who don’t meet standards do so by trusting everything they see, and not asking questions. And anyone who is still this trusting is not ready to apply for certification.

As ambitious genealogists, we need to know how to analyze sources. We need to do it all the time—and we need to feel queasy when we don’t. The example that follows involves no difficult problem, requires no unusual skill, and would not normally appear in a peer-reviewed genealogy journal. It shows the pitfalls of trusting the first source we find and the benefits of carefully analyzing all information from all sources.

Eliphas Thrall gravestone, courtesy of Jennifer Alford

Recently I wanted to document Eliphas Thrall’s birth date. His grave marker in Ohio gives his death date (19 March 1834) and his age at death (65 years, 8 months, 19 days).[1] This information does not determine a single definite birth date. Different methods of subtracting produce slightly different birth dates for Eliphas, between 28 and 30 June 1768.[2]

Case closed? No. I did not ask myself, “Is this true or false?” And if I had it wouldn’t have helped much. Sometimes grave markers and statements of age are mistaken, and sometimes they aren’t. I had to get down to a deeper level and consider the factors that would make the birth date more likely to be one or the other.

Whoever provided the information for the marker probably was not present for Eliphas’s birth in New England, and probably knew Eliphas’s birth date only by hearsay. That person might also have taken Eliphas’s supposed birth date and gone through a complex calculation, filled with chances of error, to figure his age at death. Could I find a way to get information about the birth that is closer in time to the event, more likely to come from an eyewitness?

I knew the family came from Connecticut and had lived in western Massachusetts. The published vital records of Granville, Massachusetts, give Eliphas’s birth date as 23 June 1767—more than a year earlier than the calculated dates from the grave marker.[3]

Case closed? Not. The book was published in 1914. Some conscientious twentieth-century person read through the Granville birth records (or a copy), and summarized them. Then they were typeset. That leaves plenty of chances for mistakes. So this handy, easily read published list is a derivative source. What do I do with a derivative source? Try to find what it’s derived from.

Do the original birth records for Granville survive from around 1767? Not only do they survive, they’re on line.[4]

The 1914 summarizer didn’t make any mistakes that I can see, but some information was lost in the process. The original lists the children in chronological order with a note in the middle making it easier to see that Eliphas was the last child born in Windsor, Connecticut, and his brother, two years later, was the first in the family to be born in Granville, Massachusetts. The handwriting also enables me to see that the entire list down to James was written in the same hand and with the same pen, no doubt at the same time—after the birth of Mary in 1775 and before the birth of James in 1778, which is in a different hand. So now I know that Eliphas’s birth date was written down, perhaps at the dictation of his father, between those two dates, before Eliphas had grown up.

Looking at the adjoining pages suggests that some of the top-of-page entries like the Thralls’ may have been made in sequence in the 1770s, with space left on the lower part of each page for additional children. Evidently in time a clerk went back and saved paper by filling the blank spaces with later entries.

What did my questions get me? Higher-quality evidence than I would have had if I had settled for the grave marker or the 1914 publication. The Granville record is not contemporaneous with Eliphas’s birth, but it is as close as I have been able to come so far. The Granville informant was much more likely to have been around when Eliphas was born than the tombstone informant. So unless and until new evidence appears to corroborate either 28–30 June 1768 or 23 June 1767 (and it may appear as I follow Eliphas to Vermont and Ohio), the chance of the 1768 date being wrong is greater.

This chase would have been worthwhile even if all three sources agreed right down to the day. The point is to look as hard as we can—and in genealogy that does not mean staring at the page until our eyes cross. It means recognizing that there are more questions to be asked and often more and better records to find.

Analysis is not a frill—and not always this straightforward. It is at the heart of what we do.


[1] Find A Grave, database with images (http://www.findagrave.com : accessed 10 June 2015), memorial 19,659,389, Eliphas Thrall (d. 1834), Old Colony Burying Ground, Granville, Licking County, Ohio.

[2] RootsMagic and timeanddate.com give 28 June; calculator.net gives 30 June. The late lamented Master Genealogist program always indicated the result was approximate. A very full explanation is Barbara Levergood, “Calculating and Using Dates and Date Ranges,” NGSQ 102 (March 2014): 51–73. The inevitable variance is discussed on p. 52.

[3] Vital Records of Granville, Massachusetts, to the Year 1850 (Boston: New England Historic Genealogical Society, 1914), 85; Internet Archive (https://archive.org/details/cu31924028820609 : viewed 10 June 2015).

[4]  Town of Granville (Massachusetts), Town Records, Births, Marriages, Deaths 1751-1786, Samuel Thrall family, p. 114; digital image, FamilySearch  (https://familysearch.org/pal:/MM9.3.1/TH-267-11883-59149-92?cc : viewed 10 June 2015).

CG and Certified Genealogist are service marks of the Board for Certification of Genealogists, used under license by Board-certified genealogists after periodic competency evaluation, and the board name is registered in the US Patent & Trademark Office.

Skillbuilding: Eagleson on Conflicting Evidence

SpringBoard, an official blogger for the 2015 NGS Family History Conference, is pleased to offer a review of this Skillbuilding lecture, presented Wednesday, 13 May 2015:

W151: Pamela Stone Eagleson, CG, “Confronting Conflicting Evidence”, reviewed by Patricia Hobbs, CG.

Pam Eagleson advises that conflicts in our research must be resolved—we can’t just believe what we want about our ancestors. When we encounter a conflict, we conduct further research and carefully compare and analyze the sources used and information obtained. Although we are not always able to resolve the conflict, when we can, we describe our resolution in writing.

Pam’s talk begins with a short overview of sources, information, and evidence. She references Elizabeth Shown Mills’s research process map and a “Quick Lesson” from the Evidence Explained website. Pam also discusses the Genealogical Proof Standard.

Eagleson presents six examples of conflicting cases with wide-ranging outcomes. For each case, she identifies the types of sources, information, and evidence evaluated. One problem was resolved after compiling and correlating information from several sources. This resulted in the discovery of an error in a transcription, emphasizing the need to look beyond the easy-to-find indexes. Other cases were solved by consulting experts, by understanding the mindset of people in certain social situations, and by bringing a healthy level of skepticism to bear in identifications made by earlier generations. The most amazing solution was identifying a woman who at various times was referenced by four different surnames. Understanding the culture of the research locale was essential to solving this challenging problem.

We all face conflicting evidence in our research, and Pam Eagleson’s examples from her experiences help us to understand better the principles underlying the process towards resolution.

A recording of this lecture may be ordered from Jamb Tapes, Inc.

CG, Certified Genealogist, CGL, and Certified Genealogical Lecturer are service marks of the Board for Certification of Genealogists, used under license by Board-certified genealogists after periodic competency evaluation, and the board name is registered in the US Patent & Trademark Office.

Skillbuilding: Hait on Reasonably Exhaustive Research

SpringBoard, an official blogger for the 2015 NGS Family History Conference, is pleased to offer a review of this Skillbuilding lecture, presented Friday, 15 May 2015:

F351: Michael Hait, CG, “What Is a ‘Reasonably Exhaustive Search’?” reviewed by Nancy A. Peters, CG

Michael Hait, CG

Michael Hait, CG, began his lecture by reminding us that any single record can mislead or contain errors. Many genealogical researchers have heard about the Genealogical Proof Standard (GPS) and its role in preventing us from coming to wrong conclusions and chasing the wrong ancestors. Michael briefly reviewed the five elements of the GPS, the first of which tells us that “reasonably exhaustive research” is a prerequisite for sound conclusions. Yet some researchers might ask: What is reasonable? How do you conduct reasonably exhaustive research? In his lecture Michael answered those questions and showed an example.

In a humorous way, he first told us what is not the answer to the first question. He promptly discredited the myth that three is “the magic number” of sources needed to ensure an accurate conclusion. Michael then gave us some practical guidance for how to go about meeting this important first element of the GPS.

Michael gave another wise piece of advice—expand your horizons—and followed it with an example of what that means and why it’s important to our research. He illustrated his points with a case study example using direct, indirect, and negative evidence, which was taken from his own research and writing on the Hait family.

Any family historian who is serious about producing accurate work and determining sound kinship connections will find this lecture full of useful ideas and guidance.

If you missed this lecture at the 2015 NGS Conference, a recording is available from Jamb Tapes, Inc. In addition, Michael will repeat his talk for the upcoming BCG Lecture Series on 9 October 2015 at the Family History Library in Salt Lake City. This series of six lectures sponsored by BCG is free and open to the public. No prior registration is necessary.

 

CG, Certified Genealogist, CGL, and Certified Genealogical Lecturer are service marks of the Board for Certification of Genealogists, used under license by Board-certified genealogists after periodic competency evaluation, and the board name is registered in the US Patent & Trademark Office.