DR-201b · Module 3

Common Source Bias Patterns

4 min read

Bias is not a character flaw — it is a structural feature of information production. Every source has incentives, perspectives, and constraints that shape what they report and how they report it. Understanding these structural biases is not about distrusting everything — it is about calibrating your interpretation to account for predictable distortions.

Five bias patterns recur across virtually every research domain. Commercial bias: the source profits from a particular narrative. Vendor white papers, sponsored research, and analyst reports funded by the companies they cover all carry commercial bias. The information may be accurate, but the framing systematically favors the sponsor. Survivorship bias: the source only sees successful examples. Case studies, customer testimonials, and industry awards all suffer from this — you see the winners but never the losers who tried the same approach.

Recency bias: the source overweights recent events relative to historical patterns. Quarterly earnings commentary often displays this — a single strong quarter becomes "accelerating momentum" and a single weak quarter becomes "headwinds." Confirmation bias: the source seeks and amplifies information that supports an existing thesis. Analyst reports that begin with a conclusion and then marshal supporting evidence are displaying confirmation bias, even when the evidence is real. Authority bias: the source defers to prestigious names regardless of domain expertise. A tech CEO commenting on macroeconomics gets cited because of their name recognition, not their economics expertise.

Do This

  • Ask "who benefits from this narrative?" for every source — the answer reveals commercial and positional bias
  • Actively seek disconfirming evidence for any conclusion that feels comfortable — if you only find support, you may be in a confirmation loop
  • Weight historical pattern data more heavily than single-period results to counteract recency bias

Avoid This

  • Dismiss biased sources entirely — biased sources often contain accurate information; you just need to adjust for the distortion
  • Assume your own analysis is unbiased because you are "being objective" — every analyst has blind spots
  • Treat authority figures as reliable outside their domain of expertise — prestige does not transfer across fields