Genealogical Biases: What Are They and How to Minimize Them

Our biases can adversely affect how we collect, analyze, and interpret our genealogical research. While confirmation bias may be the most frequently mentioned bias in genealogy, many more exist. Simply being aware of these biases and how they influence our research can minimize their harmful effects.

In this blog post, I expand genealogy’s list of possible biases to what I believe are the 10 most relevant biases, situate them within genealogical research, and discuss how to diminish their influence. At the end of this post, I provide a self-auditing check list to assist in minimizing these biases. I have also developed a companion YouTube Learning Module that can easily be shared with others.

What Are Biases?

In the simplest terms, biases – or cognitive biases as they are referred to within the academic literature – are subjective decision rules that distort objective reality.[1] In decision making, we often rely on judgement rules (‘rules of thumb’ or heuristics) to help us simplify complex situations. Consequently, biases are the negative consequences of adopting these judgement rules.[2]

Genealogy research relies on our interpretation of incomplete evidence with conflicting information from multiple sources that may possess their own set of biases. Our biased judgement rules develop to cope with these challenges and are driven by our motivations for interpreting evidence and where we seek validation for our conclusions.[3]

Our motivations for interpreting information are often to reduce uncertainty, anxiety, or the risk of being wrong. Sometimes our motivations are to cope with an abundance of information by preserving current beliefs or relying on easily available information. At other times, our motivations are to reduce uncertainty by increasing the perceived reliability of our conclusions by making them appear more defensible or credible. Information reduction behaviors are concerned more with feeling ok about derived conclusions rather than the appearance of being correct through uncertainty reduction by increasing the perception of accuracy.

At other times our biases are derived from where we seek reassurance for our conclusions. When the reassurance originates internally, confidence in our decisions is derived from our own beliefs, preferences, or familiarity, which is typically more concerned with self-consistency and our self-preservation. When reassurance originates externally, confidence is derived from the visibility, authority, quantity, or consensus of information implying reliability without requiring full analysis.

Taken together, our motivations and sources of validation can be transformed into a four-quadrant matrix helping us identify the genealogical biases affecting our decision-making processes (see below). I refer to it as the Sources of Genealogical Biases Matrix.

Genealogical Biases

As a marketing professor, I perform academic research as part of my job responsibilities. We are trained to recognize and minimize our biases. I’ve taken my academic knowledge and more than 30 years of family history experience to adapt cognitive biases to genealogical research.

Using the Sources of Genealogical Biases Matrix as an organizing framework, I present what I believe are the 10 most relevant biases in genealogy and categorize them into its four quadrants:

• Belief Reinforcement
  • Confirmation Bias
  • Commitment Bias
• Information Availability
  • Anchoring Bias
  • Accessibility Bias
  • Printed Source Bias
• Status Quo
  • Endowment Bias
  • Familiarity Bias
  • Ambiguity Bias
• Power in Numbers
  • Extension Bias
  • Bandwagon Bias

Belief Reinforcement: Defending the Narrative

Biases here are influenced by our desire to reduce the anxiety with information overload, and we search for validation among our own beliefs and preferences.

Confirmation Bias is the tendency to focus on information in a way that confirms our expectations. When we know what we’re looking (or hoping) to see in records or DNA matches and find it, we accept our observations as confirming our theories without much additional evaluation or validation. For example, we find a DNA match whose family tree has an ancestor of interest, and we immediately conclude this how we are related without analyzing the match’s tree further for other possible shared ancestors or reviewing the shared matches to determine if the match is a member of a relevant genetic network.

Commitment Bias favors our existing beliefs especially those exhibited publicly. The more shared our research conclusions are, the more hesitant we may be to seek or discount other alternative sources that conflict our existing narrative. Family stories, whether oral or written down, are particularly vulnerable to commitment bias.

Information Availability: Wearing Rose-colored Glasses

Biases here are again influenced by our desire to cope with information overload, but we search for validation in the sources we use to draw conclusions. We often become too optimistic about our sources and overweigh information that is easy to access, prominent, or presented first.

Anchoring Bias is the tendency to overrate initial information ignoring high quality evidence that can be more difficult to obtain. Information encountered first is often what we remember the most. Whether accurate or not, it can become the reference point from which all future research is compared and its importance overstated as a result.

Accessibility Bias is the tendency to rely on readily available information. Many of us remember when records were mostly accessed at the local courthouse or depository. They weren’t online. While Ancestry, FamilySearch, Newspapers.com, and countless archives and other sources have curated accessed to many records online, there remain many that are only found offline. Reasonably exhaustive research is more than clicking and hours of screen time, and not all records are indexed, e.g., FamilySearch’s Full-text Search.

Printed Source Bias is the tendency to believe published sources are accurate because they are in print, especially older published works. In an era when so much of what we do is online, seeing a physical book “feels” more authentic. Older publications similarly feel more genuine because they’ve “stood the test of time” or because the authors were generationally closer to our ancestor of interest and therefore more knowledgeable of the past. We should not give a printed source more authority simply because it is in print, but rather whether its conclusions are peer-reviewed, critically analyzed, and reasonably exhaustive.

Status Quo: Staying in the Comfort Zone

Biases here are influenced by our aspiration to reduce uncertainty and we look to ourselves and to what we know to make our conclusions appear more credible. To do so, we prioritize our familiarity with events or theories over certainty and change.

Ambiguity Bias is the preference for options for which we know the probably of success. Based on our own experience or understanding of a topic or method, we stick to what we know. Rather than attempt a new task or analysis technique, we rely on information gathered through conventional means for which we are familiar. For example, we may recognize that others have had success using DNA matches to draw conclusions, but we may doubt our own ability to analyze match data or may be unsure whether we inherited a DNA segment from an ancestor that might inform our research question. As such, we do nothing with our DNA data other than review our ethnicity percentages.

Endowment Bias is the tendency to overvalue what we already have compared to alternatives, as it more difficult to give up our conclusions than it is to acquire something new. This bias is related to the Anchoring and Accessibility Biases, but an Endowment Bias refers to overvaluing a collection of evidence rather than just the first sources (Anchoring), and it refers to the strength of our convictions rather than the ease in finding new information (Accessibility).

Familiarity Bias is the tendency to prefer events, theories, or methods merely because of our familiarity with them. Our propensity to stick with what we know might make us feel better, but it can limit our perspective on a research question and evaluation of information from a source. As uncomfortable as it may be, considering other viewpoints and methods affords better source interrogation and stronger derived conclusions. While similar to an Ambiguity Bias, a Familiarity Bias is more closely aligned with habits while an Ambiguity Bias is about risk aversion.

Power in Numbers: Distorting Reality

Biases in the last quadrant are also intended to reduce uncertainty, but we seek validation of our conclusions through overly simplified decision rules focusing on quantity as a substitute for quality and critical evaluation.

Bandwagon Bias is the tendency to support opinions and conclusions because they are popular. The use of a “bandwagon” comes from early references to circus culture where its band was transported by a large wagon, which was later used by politicians to imply a trend that people “jumped on” to show support. In genealogy, we might be more apt to accept a conclusion if it has been documented – whether correctly or not – in many online family trees. We may mistakenly believe it to be correct simply because so many people list it as such even though others may have thoughtlessly accepted a hint or copied it without evaluation.

Extension Bias is the tendency to think more is better than less, which is related to the conflation of quality and quantity. In life, we are trained to believe that more is better and sometimes, it is true. In genealogy, it’s more nuanced. What we need is reasonably exhaustive research and sound conclusions from an interrogated source. Original sources are more likely to be accurate than derivative sources, and many times one document can answer our research question without further evidence unless we suspect information may not be accurate or we are aware of conflicting evidence. Conclusions should be based on the quality of evidence rather than the quantity of evidence.

Overcoming Biases in Genealogy

The first step in overcoming genealogical biases is simply to be aware they exist. Awareness changes our perspective when evaluating evidence and prompts us to pause when forming conclusions. Beyond general awareness of biases, consulting the academic literature on decision making offers additional strategies,[4] which I adapt to genealogy.

Perspective-Taking

This strategy involves assuming the viewpoint of someone else who might have opposing beliefs or a different perspective. In genealogy, this might be another researcher who believes the parents of their ancestor are another couple and not the one you’ve concluded. It could also be a family historian, scholarly historian, or an archivist. How would they interpret the evidence? What information would they prioritize as more relevant or confirming?

Devil’s Advocate

Similar to perspective-taking, being a devil’s advocate involves intentionally arguing against the conclusions you’ve made to expose weaknesses and promote critical evaluation. Taken a step further, find someone to play the devil’s advocate rather than assume the role yourself.

Role Playing

One of my personal favorites is to present or pretend to present your conclusions to others, like your family (if they’ll listen) or other genealogists in a study group. I often pretend that I’m presenting at a genealogy conference or being interviewed by the hosts of a genealogy podcast where you feel greater pressure to get it right. I’ve been a guest on the Research Like A Pro Podcast, and so this is exercise is not so imaginary for me.

Proof Arguments

Another one of my personal favorites is to convert my research log and accumulated evidence into a genealogy proof argument, which synthesizes multiple, conflicting, and indirect pieces of evidence into a single, cohesive argument. Not only does the act of writing help you think more critically and uncover potential biases, but it also preserves your work in a published form that can easily be shared with others or attached to an ancestor’s profile in your online family tree.

Pros and Cons List

Rather than role playing, create a list of why you believe your genealogical conclusions are sound (pros) and why they are perhaps weak (cons). Creating a list reduces emotional arguments and replaces it with logical reasoning.

Observational Learning

There’s nothing more humbling and inspiring than reading and/or seeing how other genealogists problem solve. Becoming aware of others’ methods and means for drawing conclusions can reframe our own problem-solving perspectives and judgment rules, which often influence our motivations for interpreting evidence and how we seek validation for our conclusions. As discussed previously in the Sources of Genealogy Biases Matrix, well-intentioned but misguided motivations and validation can lead to bias formation.

Instructional Learning

As a professor, I would be remiss if I didn’t mention formal education. Good instructional learning doesn’t just give us knowledge, it teaches us how to use that knowledge. That is, to critically think and avoid the type of decision rules where bias formation is more prevalent.

As genealogists, we are fortunate to have many wonderful opportunities for formal education including certification with the Board for Certification of Genealogists (BCG) or the International Commission for the Accreditation of Professional Genealogists (ICAPGen); university education programs like the University of Strathclyde or Brigham Young University; conferences like RootsTech, or organized classes like Research Like A Pro, Salt Lake Institute of Genealogy, Genealogy Research Institute of Pittsburgh (GRIP), and countless others.

AI as a Research Partner

Across all the above bias-minimizing strategies, artificial intelligence can be used to help us identify biases and potentially overcome them. When using AI for genealogy, there are two important items to note. First, your output is only as good as your input. As such, building the prompt for how you ask AI for assistance is important. Mark Thompson of The Family History AI Show outlines the necessary five components of an AI prompt: role, goal, task, format, and hallucination check.[5]

Second, be mindful of the data you share with AI platforms. You never want to share personal data for living people including DNA data, as this can become part of an AI model’s training data. I recommend watching Nicole Dyer’s RootsTech 2026 presentation for DNA Evidence Analysis with AI where she discusses how to manage AI data privacy.

Genealogy Bias Self-Audit Checklist

Below is a practical, self-audit checklist genealogists can use while researching, evaluating evidence, writing conclusions, or reviewing an existing family tree. It is designed to be quick to apply, non-judgmental, and aligned with good genealogical reasoning. I used Microsoft 365’s Copilot AI Web Chat to help me construct the checklist after identifying, describing, and categorizing the above 10 biases and associated quadrants.

Belief Reinforcement

• Am I favoring records that support my existing conclusion while discounting conflicting evidence?
• Have I fully analyzed all evidence, including records that complicate the story?
• Am I reluctant to revise this conclusion because it appears in a published tree, article, or family history?
• Would I reach the same conclusion if this were not my own ancestor?
• Am I continuing this line of research primarily because I’ve already invested significant time or money in it?
• Have I acknowledged contradictions rather than explaining them away too quickly?

Information Availability

• Am I relying heavily on the first record, online hint, or index result I encountered?
• Have I searched for less accessible sources (offline records, manuscripts, unindexed collections)?
• Am I treating a printed or published genealogy as reliable without independent verification?
• Am I assuming older sources are more accurate simply due to age?
• Do I continue rely on an early conclusion and fail to reassess it as new evidence emerged?

Status Quo

• Am I overvaluing my own research or conclusions compared to alternative interpretations?
• Do I prefer this conclusion because it fits familiar surnames, locations, or patterns?
• Am I avoiding plausible hypotheses because the evidence is incomplete or its conclusion unknown?
• Have I dismissed new interpretations too quickly because they feel unfamiliar or disruptive?
• Have I clearly distinguished what is known, what is probable, and what is speculative?

Power in Numbers

• Am I accepting this relationship because “everyone has it” in online trees?
• Have I evaluated the original sources, not just repeated claims?
• Am I assuming that more records automatically mean a more accurate conclusion?
• Do multiple sources actually represent independent evidence, or are they copied or derivatives of the same source?
• Have I weighed evidence quality over popularity?

---

---

Sources

[1] Haselton, M.G., D. Nettle, and D.R. Murray (2015). “The Evolution of Cognitive Bias,” in D.M. Buss (Ed.), The Handbook of Evolutionary Psychology, Hoboken, NY: John Wiley & Sons, p. 1-20.
[2] Das, T.K., and B. Teng (1999). “Cognitive Biases and Strategic Decision Processes: An Integrative Perspective,” Journal of Management Studies, 36(6), 757-778.
[3] Schwenk, C.R. (1984). “Cognitive Specification Processes in Strategic Decision-making,” Strategic Management Journal, 5, 111-128.
[4] Malicse, A. (n.d.). “Overcoming Bias in Analysis and Decision-Making: Effective Psychological Techniques,” PhilArchives.
[5] See for example, https://www.rootsusers.org/downloads/20241109-MarkThompson-ResearchDocumentsAndLettersWithAIHandout.pdf, and https://www.legacytree.com/blog/ai-in-genealogy-interview-with-expert-mark-thompson.