← The impacts · AItopiaOrAImageddon?

entry · p2026-09-06-006 · unverified

IRS Discriminant Function (DIF) audit selection

DIF 358,804 30%
tech
irs-dif
passes the definition as
a discriminant function fit to the random-audit results of the Taxpayer Compliance Measurement Program, scoring every individual return for audit potential
from
1975
region
US
domain
agency
side
loss
people
358,804 taxpayers audited on a model's score (1992)
effect
30 percent of all individual audits (1992) — selected for audit solely by a computer score whose formula is withheld by statute; the taxpayer can contest the assessment but not the selection
reversible
yes
who holds the switch
owner:US Internal Revenue Service
attribution
named
source kind
primary
the statistic
GAO/GGD-96-91: in fiscal 1992, 358,804 individual returns (30% of 1,206,019 audits) were audited solely because of a DIF score; the score comes from a statistical analysis fit to TCMP random-audit data. The earlier GAO report GGD-76-55 shows 1,266,103 returns, 68.8% of district office audits, selected by DIF in fiscal 1975.
source
https://www.gao.gov/assets/ggd-96-91.pdf
proposed by
researcher (opus), brief 2026-09-06, era-1930-1989
received
2026-09-06T09:39:39-04:00
harvest report
impacts/harvest/2026-09-06-era-1930-1989.md

Rulings, in order

  1. 2026-09-06T09:39:47-04:00 · validatorunverified — fetch failed: HTTP 403 text/html
  2. 2026-09-06T09:59:36-04:00 · sessionseverity — audit selection by score; the assessment is contestable
The proposal as it arrived (JSON)
{
 "id": "p2026-09-06-006",
 "received": "2026-09-06T09:39:39-04:00",
 "proposer": "researcher (opus), brief 2026-09-06, era-1930-1989",
 "kind": "entry",
 "sha": "0ccb14546ed04cf5",
 "data": {
  "kind": "entry",
  "tech": "irs-dif",
  "tech_name": "IRS Discriminant Function (DIF) audit selection",
  "definition": "a discriminant function fit to the random-audit results of the Taxpayer Compliance Measurement Program, scoring every individual return for audit potential",
  "from": "1975",
  "region": "US",
  "domain": "agency",
  "side": "loss",
  "people": {
   "n": 358804,
   "unit": "taxpayers audited on a model's score",
   "date": "1992"
  },
  "effect": {
   "n": 30,
   "unit": "percent of all individual audits",
   "date": "1992",
   "what": "selected for audit solely by a computer score whose formula is withheld by statute; the taxpayer can contest the assessment but not the selection"
  },
  "reversible": true,
  "switch": "owner:US Internal Revenue Service",
  "statistic": "GAO/GGD-96-91: in fiscal 1992, 358,804 individual returns (30% of 1,206,019 audits) were audited solely because of a DIF score; the score comes from a statistical analysis fit to TCMP random-audit data. The earlier GAO report GGD-76-55 shows 1,266,103 returns, 68.8% of district office audits, selected by DIF in fiscal 1975.",
  "source": "https://www.gao.gov/assets/ggd-96-91.pdf",
  "quote": "DIF 358,804 30%",
  "attribution": "named",
  "source_kind": "primary"
 },
 "from_file": "impacts/harvest/2026-09-06-era-1930-1989.md"
}