> Back to BCG Learning Center
> From OnBoard—Newsletter of BCG
Karen Stanbary, MA, LCSW, CG and the BCG DNA Committee, “The Truth About Pedigree Evaluation,” OnBoard 28 (September 2022): 17–18.
Today’s digital click-through age conditions us to expect immediate answers. Too often we buzz through match lists, clicking on trees and forming hasty conclusions. In this intoxicating excitement, we might forget a tried-and true genealogy tenet: Try as hard to disprove your hypothesis as you try to prove it. Most genealogists begrudgingly welcome this tenet after failed research attempts. The phrase echoes in our minds, but do we apply it when working with DNA test results?
Two standards express the principle:
- Standard 50, “Assembling conclusions from evidence,” says “Once a genealogist resolves conflicting evidence, all remaining relevant evidence items are compatible with a single answer to the research question.1
- Standard 17, “Extent,” says “thorough research gathers sufficient data to test— and to support or reject—hypotheses.”2
A common mistake in DNA analysis occurs when we rule in a conclusion instead of ruling out competing hypotheses. Such cherry-picking creates the dreaded confirmation bias. We might identify an autosomal DNA (atDNA) match, determine that the shared-DNA amount fits within an “expected range,” and discover that the test-takers share the same ancestor. Voila—easy-peasy—both inherited this DNA from that shared ancestor, right? WRONG! Whoa. Hold on. Pull back the reins. Don’t give in to the temptation to choose matches that support a preconceived or desired result. Do not blindly trust testing-company hints that flash supposed ancestors onto the screen. Slow and steady consideration of all potentially relevant evidence is the key to achieving accurate conclusions.
Standard 43, “Evidence integrity,” reminds us that “genealogists do not trim, tailor, slight, or ignore potentially relevant evidence to fit a bias or preconception, to harmonize with other evidence, or for any other reason.”3
Think about it: Anyone alive on planet earth at the time of a child’s conception could be the source of the child’s shared DNA. As ridiculous as this sounds, the underlying logic will help us achieve accurate results. Start big, and then eliminate candidates until one answer remains. Consider all ancestral lines as competing hypotheses until your evidence analysis, evidence correlation, and reasoning rule them out.
The Genealogy Standards help us avoid confirmation bias. The best defense against it is to conduct pedigree evaluation and mitigate any problems you might discover; for example, missing ancestors, undocumented parent-child relationships, unknown maiden name of a mother, and multiple common ancestors.
Consider standard 52, “Analyzing DNA test results.” This nuts-and-bolts standard guides us to proper interpretation. It lists eight variables that affect conclusions about genetic relationships. The first two remind us to focus on evidence from documentary sources:
- Accuracy, completeness, and depth of each pedigree included in the analysis
- The possibility of more than one common ancestor for each pair of DNA test takers.4
When two people share a valid atDNA segment, our task is to identify the common shared ancestor. We begin broadly by considering each ancestor in the compared pedigrees as a potential source of the shared
atDNA. Correlation, analysis, and inferential reasoning systematically eliminate parts of the pedigree from consideration.
We recognize and control assumptions about our research’s starting point. We assess the accuracy of a match’s pedigree by doing our own research to validate the match’s tree. We examine the compared pedigrees for completeness. Most of us have gaps in our pedigrees. A gap can hide the source of the shared DNA, despite known common ancestors in the compared pedigrees. We must mitigate gaps before making a conclusion about a biological relationship. This includes gaps in our pedigree but also any gaps in any match’s pedigree used in the analysis. The chain is only as strong as the weakest link.
Most of our ancestors lived in times and places with limited choices of marriage partners. That shortage can cause atDNA matches to be biologically related in multiple ways. Such situations affect genealogists’ decisions about how many generations of pedigree analysis are sufficiently deep to rule out multiple common ancestors. Recognition of this reality and a carefully targeted testing plan helps us avoid premature conclusions.
Mitigation is fundamental to interpreting DNA test results properly. Simply reporting the percentage of unknown ancestors for each pedigree is insufficient. Many mitigation strategies apply to incomplete pedigrees. The best strategy is to identify each parent-child relationship in the pedigree, with all findings supported by source citations. Autosomal DNA-analysis techniques like clustering, genetic networks, chromosome mapping, visual phasing, and segment triangulation can lead to expanded documentary research. A clearly reasoned argument may provide justification for discarding irrelevant pedigree branches. Correlation of genetic matches from both sides of an ancestral couple may provide a convincing argument to eliminate other ancestral lines from consideration (if the couple shares no common ancestors).
Another strategy for ruling out competing hypotheses is to incorporate evidence from additional DNA matches. Standard 52, “Analyzing test results,” asks us to consider the numbers and genetic relationships of people who provided the DNA samples in the study group.5 Standard 53, “Extent of DNA evidence,” provides parameters for gauging how many test takers we need to eliminate competing hypotheses.6
Just as no fixed number of sources is needed to demonstrate reasonably exhaustive research, no fixed number of matches is needed to prove a conclusion. Standard 53, “Extent of DNA evidence,” and Standard 50, “Assembling conclusions from evidence” both provide best practice guidance to help us answer the “how much is enough” question.7 We should include enough atDNA matches to eliminate competing hypotheses for an answer to a research question. Rather than focus on the number of matches, we consider what each match adds to a body of evidence eliminating potential ancestors as the source of shared atDNA. For example, X-DNA might help eliminate hypothetical ancestors.
Documentary research provides a basis for pedigree evaluation that also helps us select atDNA matches who provide the best evidence to rule out multiple common ancestral pathways.
DNA data drawn from test results provide direct evidence of a genetic relationship. The details of the genetic match compared with the details of other genetic matches provide indirect evidence of that relationship. We must correlate and integrate DNA data with evidence about the relationship from documentary sources. Correlation and logic are a proof’s best friends.
Sound DNA analysis methodology includes:
- Reasonably exhaustive research with many DNA test takers
- Recognition and analysis of underlying assumptions
- Systematic consideration and elimination of competing hypotheses
- Pedigree evaluation
- Mitigation of problems encountered in pedigree
Proper analysis of DNA test results extends skills we use in documentary genealogy research. We craft focused research questions and develop plans to answer the questions. We understand and appreciate the value of thorough research. We study and understand our sources. We use tests of analysis to assess the likely accuracy of each source we examine. We engage in tests of correlation as we seek corroboration. We employ inferential reasoning as we assemble a body of interrelated evidence to answer research questions. We resolve any conflicting evidence items we might encounter. We craft written reports detailing the evidence and reasoning supporting a relationship conclusion. We understand that we cannot prove a relationship until all the evidence, after conflicts are resolved, points to just one answer. DNA test results, properly interpreted, contribute to the aggregate body of evidence to answer a research question. Aren’t we lucky to have this new source of evidence to help solve our genealogical problems?
Endnotes
- Board for Certification of Genealogists, Genealogy Standards, 2nd , rev. (Nashville: Ancestry.com, 2021), p. 28. Emphasis added.
- Ibid., p. 14. Emphasis added.
- Ibid., pp. 25–26.
- Ibid., p. 30.
- Ibid., pp. 30–31.
- Ibid., p. 31.
- Ibid., pp. 28–29, 31.
This article was originally published in OnBoard, BCG’s educational newsletter and is protected by copyright. Individuals may download and print copies for their personal study. Educators are granted permission to provide copies to their students as long as BCG, OnBoard, and the appropriate author are credited as the source of the material. Republication elsewhere is not permitted.