Final Outcome of Representation Elections in Cases Closed, Fiscal Year 1975

p. 245. The crop cuts the table off part-way through the 3-union block of section "A. All representation elections": there is no "3-union elections" subtotal line, no later sections, and no footnote block on this image, so `footnotes` is empty even though the caption carries a raised 1 and the "Total elections" heading a raised 2. Doubtful cells: - "AFL-CIO v. national", Percent won: I read the printed glyphs as 50.6, but 31/52 = 59.6, and every other percent in the panel agrees with its own row to one decimal. The middle digit is likely a 9 whose counter filled in during scanning; transcribed as seen. - "National v. local" (row label): the "v." has almost completely dropped out in the scan — only a faint mark remains between "National" and "local". Supplied from the pattern of the surrounding labels. - Last row "AFL-CIO v. national v. local", column "In elections where no representative chosen": the cell falls at the very bottom edge of the crop and no digit is legible, so it is left blank. Arithmetic on the row (Total 870 = In elections won 870) implies the printed value is 0. - "2-union elections", "Other national unions" under "Elections won by unions": the scan is heavy here; read as 39, which matches the column sum (16+8+10+5). All other rows cross-foot correctly (component columns sum to the printed totals, and "In elections won" + "In elections where no representative chosen" = "Total"), which supports the column assignments. A stray ink speck sits in the "Elections in which no representative chosen" column beside the "AFL-CIO v. national" row; it is not a value.

p. 246. Continuation page: no column headings are printed on this crop, so header_rows is empty. The first four rows complete a section begun on the preceding page (its earlier rows, including the 1-union and 2-union blocks of panel A, are not on this image); the section line "B. Elections in RC cases" then opens a new section that runs to the end of the crop. The table is a single panel, 16 columns wide (stub + 15 data columns). Column structure (inferred from arithmetic only, since no headings are printed, and stated here for the reader's orientation, not as transcribed text): col 1 total elections; col 2 percent; col 3 elections won, total; cols 4-7 elections won by AFL-CIO / Teamsters / other national / other local; col 8 elections lost; col 9 total valid votes; col 10 votes for unions, total; cols 11-14 votes for AFL-CIO / Teamsters / other national / other local; col 15 votes against. Every row verifies: cols 4-7 sum to col 3, col 3 + col 8 = col 1, cols 11-14 sum to col 10, col 10 + col 15 = col 9, and the 1-union / 2-union / 3-or-more subtotals sum down to "Total RC elections". Decimal points: the percent column (col 2) is printed with one decimal place, but the scan has dropped or faded most of the periods, so the cells image as "100 0", "48 2", "46 1" etc. I have written them with the decimal point ("100.0", "48.2", "46.1"); each value is confirmed by division (e.g. 1,020/2,214 = 46.1, 3,470/7,157 = 48.5, 11/12 = 91.7). Periods are likewise erratic in the row labels: some "v." render with the period and most "v" do not, and I have transcribed each as it images rather than regularising them. The period after "B" in the section line is faint and is supplied by analogy with the table's "A."/"B." section numbering. Doubtful cell — "Teamsters" row (panel B, col 12, votes for Teamsters): I read "23,416", but col 10 of the same row reads "23,461", and since Teamsters is the only union contributing votes in that row the two cells should be identical. The "1-union elections" subtotal directly below also prints 23,461 in col 12, and 23,461 + 53,032 = 76,493 (col 9), so the arithmetically correct value is 23,461. This is either a digit transposition in the original printing or a misreading of a degraded glyph; I have transcribed it as it images and left it uncorrected. This is the only cell in the crop that fails an arithmetic check. Row label as printed: "AFL-CIO v ALF-CIO v local" (third 3-or-more-union row, col 1 = 4) images with "ALF-CIO" for the second union; this is a typesetting error in the report for "AFL-CIO", transcribed as printed. Minor speckle, not transcribed: a stray mark precedes the "0" in col 15 of "National v national", and a stray comma-like mark precedes the "0" in col 15 of the second "All others" row; both cells are read as "0". Cells holding only a dash rule are rendered as "".

p. 247. 16 columns. The crop is a continuation page: it carries the full three-level column heading but only section C of the table (parts A and B are on earlier pages), and no footnote block is printed under it, so `footnotes` is empty (the markers ^1 in the caption and ^2 on "Total elections" are defined elsewhere). SUSPECTED MISPRINT: "Teamsters v. national", column "Total elections^2" is printed as 5, and the glyph looks like a 5 (same shape as the 5 in the "AFL-CIO v. local" row directly above). It is transcribed as printed, but it cannot be right: the row's own arithmetic requires 0 won + 1 no-representative-chosen = 1, the employees total for the row is only 4, and the "2-union elections" subtotal of 18 is reproduced only by 3+6+5+1+2+1 (as printed the six rows sum to 22). Every other cell in that column checks out, and 14/18 = 77.8 confirms the 18. A reader should treat this cell as 1, or verify against another copy. Percent-won column: the decimal point is very faint in this 1-bit scan and renders as a gap ("27 0"), but these are one-decimal percentages and are transcribed with the point (27.0, 26.6, ... 30.4). Each one is consistent with "Total won" / "Total elections", which supports the reading. Likewise the period is faint or lost after "C" in the section line and after "v" in several of the two-union row labels ("AFL-CIO v Teamsters", "AFL-CIO v local", "Teamster v local", "Local v local", "AFL-CIO v AFL-CIO v local"); it is clearly visible in "AFL-CIO v. AFL-CIO" and "Teamsters v. national". I have written "v." and "C." throughout on the assumption that this is scan loss, not a typesetting difference. The singular "Teamster" in "Teamster v. local" is as printed and is left alone. Cells holding only a dashed rule ("------"), used where a union did not take part, are "". Note the asymmetry in the "AFL-CIO v. AFL-CIO v. local" row: the local union's cell is blank under "Elections won by unions / Other local unions" but a printed 0 under "In units won by / Other local unions"; both are as printed. All other arithmetic verifies: the four 1-union rows sum to the "1-union elections" line, the six 2-union rows to "2-union elections", and the three subtotals to "Total RM elections", in all fifteen value columns; row-wise, Total won + no representative chosen = Total elections and employees in elections won + employees where no representative chosen = Total, everywhere except the flagged cell.

p. 248. Continuation page: no column headings are printed on this crop (they appear on the earlier page of the table), so header_rows is empty. The grid has 16 columns (stub + 15 data columns), counted from the vertical rules; the first row "D. Elections in RD cases" is the section line printed centred above the table's top rule and carries no values. Column assignment was verified by arithmetic throughout and every printed figure is internally consistent: cols 4-7 sum to col 3, col 3 + col 8 = col 1, cols 11-14 sum to col 10, col 10 + col 15 = col 9, and the "1-union", "2-union" and "Total RD elections" lines equal the sums of their member rows. The implied column meanings (not printed here) are: 1 total elections, 2 percent won by unions, 3 elections won, 4-7 won by AFL-CIO / Teamsters / other national / other local, 8 elections lost, 9 total valid votes, 10 valid votes for unions, 11-14 for AFL-CIO / Teamsters / other national / other local, 15 votes against participating unions. Typography: the decimal point in the percent column is faint or dropped in this 1-bit scan ("26 2", "17 1", "56 3", "100 0"); these are transcribed as 26.2, 17.1, 56.3, 100.0 etc., consistent with "33. 3" where the point is visible and with the column's arithmetic. Likewise the period after "v" is clearly printed only in "National v. local" and is faint in the other four comparison rows; it is rendered as "v." in all five, and the period after "D" in the section line is similarly reconstructed. Cells containing only leader dashes are rendered as "". Several small specks sit in the margins between columns (above the table at the top edge, and near the gap left of column 8 in the "AFL-CIO v. local" and "Teamsters v. local" rows); these read as scanner noise, not footnote markers, and no footnote markers are attached to any value on this crop. The markers 1 and 2 defined in the footnotes attach to the caption/headings on the preceding page.