How to Lie with Statistics
Darrell Huff
Catalogues the standard tricks — the truncated graph, the biased sample, the unspecified average — by which real numbers are made to imply a false conclusion without anyone technically lying.
Evaluating claims — arguably the most valuable field in this entire index right now.
11 topics · 15 curated works
No prior grounding assumed.
How to Lie with Statistics
Darrell Huff · 1954
Catalogues the standard tricks — the truncated graph, the biased sample, the unspecified average — by which real numbers are made to imply a false…
+2 more at this level
Assumes you know the vocabulary.
Understanding History: A Primer of Historical Method
Louis Gottschalk · 1950
Argues that a source only becomes historical evidence once its authorship, date and motive are established, and lays out the tests for telling a…
+7 more at this level
Primary sources and full treatments.
What Drives Media Slant? Evidence from U.S. Daily Newspapers
Gentzkow & Shapiro · 2006
Measures newspaper slant by the partisan phrases it shares with politicians and finds it tracks what readers in a market want to hear more than who…
+3 more at this level
12 of 15 works
Darrell Huff
Catalogues the standard tricks — the truncated graph, the biased sample, the unspecified average — by which real numbers are made to imply a false conclusion without anyone technically lying.
Mike Caulfield
Teaches SIFT — stop, investigate the source, find better coverage, trace claims to the original — as a four-move alternative to close reading for judging claims found online.
link checked 17 Sept 2026Society of American Archivists & RBMS
Defines primary source literacy as a set of competencies distinct from general information literacy, built around the idea that a primary source is never neutral because the very acts of preserving, cataloguing and digitising it already mediate what a reader receives.
link checked 17 Sept 2026Louis Gottschalk
Argues that a source only becomes historical evidence once its authorship, date and motive are established, and lays out the tests for telling a credible account from an unreliable one.
Margaret Stieg Dalton & Laurie Charnigo
Surveys working historians to find they still locate primary sources mainly through informal networks, footnote-chasing and physical browsing rather than catalogues, even as they have come to rely on electronic databases for secondary literature.
link failing as of 17 Sept 2026Joel Best
Extends his own earlier argument to show that even sympathetic, well-intentioned advocates routinely misstate the statistics behind a cause, because a round, dramatic number spreads faster through the press than a hedged, accurate one.
Sumner et al.
Traces exaggerated claims in health news back to the university's own press release far more often than to the journalist who wrote the story.
link checked 17 Sept 2026Craig Silverman
Sets out a step-by-step method — provenance, source, date, location, motivation — for confirming an eyewitness photo or video is what it claims to be before a newsroom runs it.
link checked 17 Sept 2026Association of College & Research Libraries
Argues information literacy is a set of interrelated conceptual understandings, not a checklist of skills — authority is constructed and contextual, and information creation is itself a process worth interrogating.
link checked 17 Sept 2026Wineburg, McGrew, Breakstone & Ortega
Finds that history PhDs and even professional fact-checkers were readily fooled by professional-looking websites, and that only the fact-checkers reliably left the page to check who was behind it.
link checked 17 Sept 2026Wineburg & McGrew
Argues the fastest way to judge an unfamiliar website's credibility is to leave the page and read what the rest of the web says about the source, rather than scrutinising the page itself.
link checked 17 Sept 2026Gentzkow & Shapiro
Measures newspaper slant by the partisan phrases it shares with politicians and finds it tracks what readers in a market want to hear more than who owns the paper.
link checked 17 Sept 2026This subject genuinely sits in more than one domain. These fields approach the same ground with different methods.