Counting method
The Weave connects allegation categories selected together in the same report. Its geometry is preserved from the recovered drawing; its counts now come directly from the retained source.
Choose the observation
Latest report per LWDA case selects one row for each exact identifier in column B: the greatest report-create date in column K, with later worksheet rows breaking same-date ties. This produces 68,708 observations. No identifiers are normalized, and flags from different reports are never combined.
All report rows retains all 73,094 observations, including amendments. A case can therefore contribute more than once. A source flag describes the selected report; the latest-row rule does not establish that it is a final legal account of the matter.
There are 3,893 identifiers with multiple rows and 126 identifiers tied at their latest report date. Every inspected identifier, employer, date and boolean flag was valid. No report was excluded or imputed. Source dates are day-level values; their original strings are retained alongside ISO dates.
Map the source fields
The following mapping is applied within each report. Wage statements combine two source flags with a logical “or,” so a report selecting both counts once. Other source categories remain outside this drawing.
| Displayed category | Source column | Original field label |
|---|---|---|
| Meal & rest breaks | Y | Meal and Rest Breaks |
| Wage statements | AD or AE | No Wage Statements; Improper or Incomplete Wage Statements |
| Wages due on termination | S | Not Paid Wages Due on Termination |
| Overtime | Q | Overtime |
| Minimum wage | P | Minimum Wage |
| Pay for all hours worked | R | Not paid for all hours worked |
| Expense reimbursement | V | PMNT Reimbursement of Empl Expenses |
| Other unpaid wages | T | Other Unpaid Wages |
The field labels do not establish off-the-clock work, a regular-rate theory, rounding, waiting-time penalties, or a derivative legal relationship. Those conclusions are not inferred from the flags.
Read the drawing
A ribbon counts observations selecting both of its categories. Each qualifying report contributes one to that pair. A report selecting three categories contributes to three distinct pairs; it remains one observation in the denominator.
An arc is sized by the sum of its pair counts with the other enabled categories. Its percentage divides that sum by the sum for all enabled arcs. Each pair has two endpoints and is counted at each one. Arc percentages are shares of pair incidences, not shares of unique reports.
The percentages beside the category controls divide the number selecting that category by all observations in the chosen counting unit. Selected-pair percentages use the same denominator. Reports with none of the eight categories remain in that denominator.
Disabling categories changes the ring’s pair geometry. The threshold hides ribbons below the chosen count without changing arc proportions, source counts or the observation denominator. The pairing list provides the same selection by keyboard.
Reproduce a result
Open a pairing’s source rows, select the same counting unit, and count rows where both categories’ source flags are true. The table links each observation to its worksheet row and original boolean cells. Download its filtered CSV or the complete source workbook.
The drawing exports counts for all 28 possible pairs, including whether each is enabled and visible under the current threshold. Its PNG retains the native canvas backing resolution and records the source hash and active controls. The calculation module, manifest and derived records are available for independent inspection.