Variable pages · IRS 990 Master Concordance
SA_03_PUB_ADD_L7AB_CY_M3
Public support from disqualified persons or in excess during the current tax year minus three years
- Description
- Current tax year minus three years
- Table
SA-P03-T00-SUPPORT_SCHEDULE_509(one)- Part
- Part III - Support Schedule for Organizations Described in Section 509(a)(2)
- Location
SCHED-A-PART-03-SECTION-A-LINE-07C-COL-B- Type
- numeric
- Scope
- PZ – 990 and 990-EZ
- Database
- F990
- Forms
- SZ
2 / 3
xpaths observed / mapped
317k
filings with a value
2009–2024
tax years observed
0 + 1 reviewed
open flags
Checks
Value checks
| Result | Value check | Detail |
|---|---|---|
| ✓ pass | Values parse as numbers | 100.0% of values are numeric |
| ▲ warn | No sudden change in the median between adjacent years | largest change 6.7x (IRS990ScheduleA/SubstAndDsqlfyPrsnsTotGrp/CurrentTaxYearMinus3YearsAmt, 990EZ, TY2017) |
| ✓ pass | Pooled xpaths are on the same scale (same return type) | largest ratio of medians between xpaths: 2.3x |
| ⓘ info | Sign convention | 0.0% of values are negative (fine for net amounts) |
Flags
| Flag | Severity | Xpath | Detail | Review |
|---|---|---|---|---|
SCALE_BREAK | ▲ warn | IRS990ScheduleA/SupportFromDQPsEtc/CurrentTaxYearMinus3Years | 990EZ medians differ 28x across xpaths (CurrentTaxYearMinus3Years vs CurrentTaxYearMinus3YearsAmt) | ✓ acceptednoteSame form line: the pre-2013 xpath (TY2009-12) and its post-2013 successor never overlap in a year. The check compares medians of per-year/schema-version q50 with zeros dropped, so zero-heavy eras and the small, larger-org TY2009-10 e-file population drive the ratio (old-xpath medians fall 2009->2012 toward the new-era level; yearly medians within one xpath swing as much). Same quantity. Evidence (990EZ): IRS990ScheduleA/SubstAndDsqlfyPrsnsTotGrp/CurrentTaxYearMinus3YearsAmt [Section A Line 07c Column (B)] TY13-24 n=116,245 yearly medians 0 to 5,000; IRS990ScheduleA/SupportFromDQPsEtc/CurrentTaxYearMinus3Years [Section A Line 07c Column (B)] TY9-12 n= 12,021 yearly medians 0 to 3,275 |
Xpaths
| Xpath (under /Return/ReturnData or /Return/ReturnHeader) | Form | Years | Filings | % filers | Repeats in | |
|---|---|---|---|---|---|---|
| x1 | IRS990ScheduleA/SupportFromDQPsEtc/CurrentTaxYearMinus3Years | SZ | 2009–2012 | 34,385 | 4.27% | |
| x2 | IRS990ScheduleA/SubstAndDsqlfyPrsnsTotGrp/CurrentTaxYearMinus3YearsAmt | SZ | 2013–2024 | 282,620 | 11.5% | |
| x3 | IRS990ScheduleA/Form990ScheduleAPartIII/TotalOf7aAnd7b/CurrentTaxYearMinus3Years | SZ | not observed | 0 |
% filers: percent of in-scope filers whose return has the xpath, in its latest year. Repeats in: share of filings where the element occurs more than once.
Coverage by tax year
| Filings | ’09 | ’10 | ’11 | ’12 | ’13 | ’14 | ’15 | ’16 | ’17 | ’18 | ’19 | ’20 | ’21 | ’22 | ’23 | ’24 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| x1 | 2.6k | 8.9k | 11k | 12k | ||||||||||||
| x2 | 12k | 14k | 15k | 15k | 16k | 18k | 19k | 23k | 29k | 33k | 36k | 53k | ||||
| Fill rate: % of all filings of the return type (largest xpath) | ||||||||||||||||
| 990 | 5 | 5 | 5 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 5 | 5 | 5 | 5 | 5 | 9 |
| 990-EZ | 6 | 5 | 5 | 4 | 4 | 4 | 4 | 3 | 4 | 4 | 4 | 5 | 7 | 8 | 8 | 16 |
Filings with a value, by xpath (rows, numbered as in the xpath table) and tax year. A change of row from one year to the next is usually the IRS renaming the element (most were renamed in 2013). Hover a cell for the exact count.
Values
| Return | Filings | % numeric | % zero | % negative | P25 | Median | P75 | P99 |
|---|---|---|---|---|---|---|---|---|
| 990 | 188,739 | 100.0% | 22.5% | 0.0% | 10,045 | 54,774 | 207,575 | 7,196,124 |
| 990-EZ | 128,266 | 100.0% | 67.2% | 0.0% | 0 | 975 | 13,000 | 126,328 |
Percentiles are the median over tax years of each year's percentile.