Sources

Where the numbers come from

Every figure here is taken from one of these publications as its authors reported it. Findings reported only in words are described in words, not plotted.

Historical accounts

Single accounts, sacred texts, epics and early collections. Each says on the map what kind of thing it is, because a poem, a funerary spell and a case note are not the same sort of evidence.

Studies on the map

Further reading used for the breakdowns

Greyson B. The Near-Death Experience Scale: construction, reliability, and validity. Journal of Nervous and Mental Disease 1983;171:369-75. The 16-item scale and its 7-point threshold.
Lange R, Greyson B, Houran J. A Rasch scaling validation of a "core" near-death experience. British Journal of Psychology 2004;95:161-77. Shows the scale measures one underlying dimension, with depth not tied to sex or age.
Athappilly GK, Greyson B, Stevenson I. Do prevailing societal models influence reports of near-death experiences? A comparison of accounts reported before and after 1975. Journal of Nervous and Mental Disease 2006;194:218-22. Tunnel 16% vs 64%, 24 accounts per group.
Cassol H, Martial C, Annen J, et al. A systematic analysis of distressing near-death experience accounts. Memory 2019;27:1122-29. 17 of 123 narratives (14%) distressing.
Cassol H, Pétré B, Degrange S, et al. Qualitative thematic analysis of the phenomenology of near-death experiences. PLoS ONE 2018;13:e0193001. Open access. Themes in 34 narratives, the basis for text-coding approaches.
Kellehear A. Experiences Near Death: Beyond Medicine and Religion. Oxford University Press, 1996. Cross-cultural review; the tunnel as a largely Western motif.
Journal of Near-Death Studies, 1982-2025, free at the UNT Digital Library. The field's own journal, 1,081 items.
Near-Death Experience Research Foundation, nderf.org. The largest public archive of first-person accounts. This site links to it and does not copy from it.
International Association for Near-Death Studies, iands.org. Research, support groups and an experience registry.

Rules this site follows

Three kinds of data are kept apart and never averaged together: published research, external archives (linked only), and experiences people share here.
Every proportion carries its sample size and a 95% confidence interval.
Community figures are hidden below 10 submissions and greyed below 30; comparisons between groups need 30 or more in each.
A pattern is called significant only when it survives multiple-comparison correction, differs by at least 10 percentage points, and does not contradict published research.
The 3D world reads only a parameter file in which every value is cited.