Industrial Distribution of Representation Elections Held in Cases Closed, Fiscal Year 2000

p. 178. The crop shows only the first page of the table; the body continues past the bottom edge (the last transcribed row, "Printing and Related Support Activities", is the last fully printed row on this image) and no footnote text appears on this crop, so `footnotes` is empty even though the stub heading carries a raised "1" marker ("Industrial Group^1") whose text must be printed elsewhere. Column structure: 15 columns. "Number of elections in which representation rights were won by unions" spans the four columns Total / AFL-CIO unions / Other national unions / Other local unions; "Valid votes cast for unions" spans a second, identically labelled group of four. The remaining headings ("Total elections", "Number of elections in which no representative was chosen", "Number of employees eligible to vote", "Total valid votes cast", "Total votes for no union", "Eligible employees in units choosing representation") are tall single cells spanning both heading levels and are repeated in each level, as is the stub heading. The last heading is broken across lines as "Eligible employ- ees in units choos- ing rep- resentati on" and has been rejoined to "Eligible employees in units choosing representation". Rows that are indented industry-group subtotals and printed between heavier rules: "Agriculture, Forestry, Fishing, and Hunting", "Mining", "Utilities", "Construction", "31-Manufacturing". "Utilities" is both a detail row and its own group (it sits alone between rules). The subtotal label "31-Manufacturing" is printed with that leading sector number, unlike the other group labels; transcribed as printed. Suspected misprint (transcribed as printed, not corrected): in the "Oil and Gas Extraction" row the "Total valid votes cast" figure (480) exceeds the "Number of employees eligible to vote" figure (405). The 480 is internally consistent with the rest of that row (219 + 59 + 0 = 278 union votes, 278 + 202 = 480) and with the "Mining" subtotal's 1,090, so the questionable figure appears to be the 405 eligible, which is nevertheless consistent with the subtotal 1098 (405 + 452 + 241). Both are as printed. Low but genuine values confirmed by cross-footing: "Support Activities for Agriculture and Forestry" and "Support Activities for Mining" each show 4 in the final column ("Eligible employees in units choosing representation"); these sum correctly into the group subtotals 1180 and 284 respectively, so they are not misreadings. The digit-grouping comma is applied inconsistently by the printer: the "Number of employees eligible to vote" and "Eligible employees in units choosing representation" columns print four- and five-digit values without commas (e.g. 1596, 11690, 20720, 9750), while the adjacent vote columns use commas. Both are transcribed exactly as printed. All component-to-total relationships were checked and foot correctly throughout (union-vote components to their Total, Total union votes plus no-union votes to Total valid votes cast, won-election components to their Total, Total won plus no-representative to Total elections, and every detail row to its group subtotal across all 14 numeric columns), which corroborates the column assignments.

p. 179. 15 columns. The two spanner headings ("Number of elections in which representation rights were won by unions" over Total/AFL-CIO/Other national/Other local, and "Valid votes cast for unions" over the same four) are repeated across their columns; all other headings are tall single headings repeated in both levels. This is a "--Continued" page, so the table is cut off at both ends: the "32-Manufacturing" subtotal line (row 5) covers component industries that begin on the preceding page, so its components do not sum within this crop (the four printed detail rows above it sum to 53 elections, not 88). The "33-Manufacturing" and "Wholesale Trade" subtotals do sum exactly from the detail rows printed above them. The body continues past the bottom edge of the crop below "Clothing and Clothing Accessories Stores". The stub heading carries a superscript 1, but no footnote text is printed within this crop, so `footnotes` is empty. Typesetting note: in the "Number of employees eligible to vote" and "Eligible employees in units choosing representation" columns, four- and five-digit values are printed without digit-grouping commas (e.g. 17636, 10395); they are transcribed as printed. The adjacent "Total valid votes cast" and vote columns do use commas. All rows were cross-checked arithmetically and are internally consistent: total elections = won + no representative chosen; union won total = AFL-CIO + other national + other local; union votes total = its three components; total valid votes cast = union votes + votes for no union. No misprints detected and no doubtful or illegible cells.

p. 180. The crop is a continuation page: it begins with the "44-Retail Trade" subtotal line (its component detail rows are on the preceding page) and is cut off at the bottom immediately after "Credit Intermediation and Related Activities" — further body rows and any footnote text continue below the crop. The stub heading carries a raised 1 ("Industrial Group^1"), but the text of footnote 1 is not printed on this image. Two subtotal lines are both printed "48-Transportation" (row 15, following the Air/Rail/Water/Truck/Transit/Pipeline/Scenic/Support detail block, and row 19, following Postal Service / Couriers and Messengers / Warehousing and Storage Facilities). The second one appears to be a misprint — by NAICS it should be sector 49 (Transportation and Warehousing / Postal and Warehousing) — but both are transcribed exactly as printed. Subtotal and rule placement are unambiguous; the heavier rules group rows 2-5 under the "45-Retail Trade" subtotal, rows 7-14 under the first "48-Transportation" subtotal, rows 16-18 under the second, and rows 20-23 under "Information". All rows check out internally (valid votes for unions + votes for no union = total valid votes cast; detail elections sum to each subtotal), which supports the digit reading. Note the inconsistent printing convention in columns 7 and 14 (employees eligible to vote; eligible employees in units choosing representation), where four-and-five-digit values carry no digit-grouping comma (e.g. 8690, 10012, 17705, 3050, 4839, 8808, 2871, 2904, 1388, 2246); these are transcribed as printed, without added commas.

p. 181. The stub heading carries a superscript 1, but no footnote text is printed on this crop (the table is a "-Continued" page; the footnote presumably appears on the final page of the table). The table continues from the preceding page: the indented subtotal row "Finance and Insurance" (13 elections) exceeds the sum of the two Finance-and-Insurance detail rows shown here (5 + 1 = 6), because the earlier detail rows of that industry sector fall on the previous page. All other indented subtotal rows on this crop (Real Estate and Rental and Leasing; Administrative and Support, Waste Management and Remediation Services; Health Care and Social Assistance; Arts, Entertainment and Recreation) do foot exactly to the detail rows printed above them, in every column. Suspected misprint, transcribed as printed: for "Rental and Leasing Services" the "Total valid votes cast" (3,590) and "Total votes for no union" (3,152) far exceed the "Number of employees eligible to vote" (1426). The vote columns are internally consistent (438 + 3,152 = 3,590), so the error looks to be in one of the two figures as published; it is carried into the "Real Estate and Rental and Leasing" subtotal (eligible 1597 vs. total valid votes cast 3,736). Digit-grouping commas are printed inconsistently in the source: the "Number of employees eligible to vote" and "Eligible employees in units choosing representation" columns are set without commas (e.g. 49397, 26848) while the other numeric columns use them. Transcribed exactly as printed. Column heading 14 is hyphenated across lines as "Eligible employ-ees in units choos-ing rep-resentati on"; joined here to "Eligible employees in units choosing representation". All cells were legible; no column is cut off by the crop.

p. 182. The crop is a continuation page ("--Continued"), but the full column headings are reprinted on it, so header_rows is filled. 15 printed columns. Two spanner headings sit over four columns each ("Number of elections in which representation rights were won by unions" over Total / AFL-CIO unions / Other national unions / Other local unions, and "Valid votes cast for unions" over the same four sub-headings); the remaining headings are tall single headings repeated across both levels. In the "Total, all industrial groups" row the "Total votes for no union" cell is typeset wrapped onto two lines as "115,37" over "6"; it is transcribed as the single value 115,376 (which is consistent with 110,644 + 115,376 = 226,020, the printed total valid votes cast). Thousands separators are used inconsistently in the source: the "Number of employees eligible to vote" and "Eligible employees in units choosing representation" columns print values without commas (e.g. 2614, 262441, 120095, 1337) while neighbouring columns use them. Transcribed exactly as printed. The footnote reproduces the printed misspelling "Classificaiton" (for "Classification"). Two rows look internally odd but are transcribed as printed, not corrected: "Religious, Grantmaking, Civic, and Professional and Similar Organizations" shows 71 eligible employees in units choosing representation against 148 total valid votes cast and 156 eligible to vote; "Administration of Environmental Quality Programs" shows 0 in that same last column (consistent with 0 elections won). All column cross-footings checked out (unions total = AFL-CIO + other national + other local; total valid votes cast = votes for unions + votes for no union) in every row, which supports the column assignment.