Final Outcome of Representation Elections in Cases Closed, Fiscal Year 1971

p. 184. CROP: the table is cut off at the bottom of the page. The 3-union-election block ends mid-group at "AFL-CIO v. Local v. Local" with no subtotal row; the table (and sections B, C, etc., if any) continues on the following page. A partially sliced line of type at the very bottom edge reads "See footnotes at end of table." — it is only half-visible and is not a numbered footnote, so `footnotes` is empty. The two footnote markers that are referenced (superscript 1 on the caption, superscript 2 on "Total elections") are defined at the end of the table, not on this page. PERCENT COLUMN: the decimal points in "Percent won" are broken/faint in this 1-bit scan, so several cells image as e.g. "50 3", "50 5", "90 7", "80 0", "79 4", "100 0". All are one-decimal percentages and every one of them reproduces won/total to the printed rounding, so they are transcribed with the decimal point (50.3, 50.5, 90.7, 80.0, 79.4, 100.0). 5-vs-6 GLYPHS: this typeface's 5 and 6 are hard to separate at scan resolution. Each ambiguous cell was settled by the table's own column and row arithmetic, all of which closes exactly: "Teamsters v. Teamsters" Total elections = 5 (4 won + 1 no rep.); "Teamsters v. Natl." Other national unions = 5 (17 + 5 = 22 total won, and the 2-union subtotal 63 = 45+5+7+6); "AFL-CIO v. AFL-CIO v. Natl." employees AFL-CIO unions = 57 (57 + 9 = 66); "AFL-CIO v. AFL-CIO v. Local" employees Other local unions = 605 (754 + 605 = 1,359); "AFL-CIO v. Teamsters v. Natl." employees Other national unions = 250 (135 + 98 + 250 = 483). By contrast "Natl v. Local" Other national unions really is 6 (6 + 13 = 19). SUSPECTED MISPRINT (transcribed as printed): in the 2-union block the "Elections in which no representative chosen" entries for "Natl. v. Natl." and "Natl v. Local" appear to be transposed. As printed, "Natl. v. Natl." shows 7 total elections, 100.0 percent won, 7 won, but 1 election with no representative chosen (7 + 1 = 8, not 7); "Natl v. Local" shows 20 total, 95.0 percent, 19 won, but 0 with no representative chosen (19 + 0 = 19, not 20). Swapping the 1 and the 0 makes both rows internally consistent, and the column subtotal (147) is unaffected either way. Every other row and every column subtotal in the table adds correctly. ROW LABELS: leader dots run straight into the abbreviation "Natl", so a terminal period cannot be distinguished from the first leader dot. Labels ending in that abbreviation are given as "Natl." on the normal abbreviation convention. The row transcribed "Natl v. Local" images with no period after the first "Natl", unlike "Natl. v. Natl." directly above it; that may be a dropped period in the original. STRUCTURE: 16 printed columns. "Percent won" through "Other local unions" are tall headings that occupy both of the lower heading levels, so they are repeated there; "In units won by" sits only at the middle level over the last four employee columns. "A. All representation elections" is a centred section line spanning the table with no values.

p. 185. Continuation page ("—Con."): the column headings and the section line "A. All representation elections" are reprinted here, but the body rows are the tail of the list of participating-union combinations begun on the previous page. The footnote texts for markers 1 (in the caption) and 2 (on "Total elections") are not printed in this crop. Doubtful / damaged cells: - Row "AFL-CIO v. AFL-CIO v. AFL-CIO v. Teamsters", column "Employees eligible to vote – Total": printed glyphs read 83, while the same row's "In elections won" and "In units won by – AFL-CIO unions" both read 33. Transcribed as printed (83); it is very likely 33 (broken/filled 3, or a misprint), since the single election was won, so Total should equal "In elections won". - Row "Teamsters v. Teamsters v. Teamsters", column "Percent won": the decimal point has not printed ("100 0" on the page); transcribed 100.0 to match the other rows of that column. - Row "3(or more)-union elections", column "In units won by – Other national unions": the thousands separator prints as a period ("9.460"); transcribed 9,460, which the column arithmetic confirms (2,053 + 1,103 + 9,460 + 1,136 = 13,752). - Row "AFL-CIO v. AFL-CIO v. AFL-CIO v. Teamsters": the period after the first "v" is not visible in the label; written "v." for consistency with the rest of the stub. Internal checks that do pass, supporting the digit readings: total row 4,445 + 3,917 = 8,362; 2,453 + 1,349 + 418 + 225 = 4,445; 4,445/8,362 = 53.2 percent; 274,700 + 311,455 = 586,155; 164,498 + 39,867 + 36,798 + 33,537 = 274,700. Subtotal row: 17 + 9 + 5 + 5 = 36; 36 + 5 = 41; 36/41 = 87.8 percent; 13,752 + 1,742 = 15,494. Cells shown on the page as a dash rule (inapplicable combinations) are transcribed as empty strings; printed zeros are kept as 0.

p. 186. Crop shows panel B of a table continued from an earlier page: no column headings are printed on this image, so header_rows is empty. The grid has 16 columns (stub + 15 value columns); from the rule structure and the arithmetic they are: number of elections; percent won by unions; elections won total; elections won by AFL-CIO / Teamsters / other national / other local; elections lost (no union); total employees eligible; employees in units won total; employees in units won by AFL-CIO / Teamsters / other national / other local; employees in units lost. These labels are NOT printed here and are stated only as a reading aid. "B. Elections in RC cases" is a centered section line inside the table rules with no values; transcribed as a label-only row. Decimal points: the 1-bit scan drops or faints the period in many percent cells, which print as e.g. "52 6", "63 0", "76 2", "89 6", "80 0", "77 8", "90 9", "94 7", "78 7", "100 0", "55 1". These are transcribed with the decimal point restored (52.6, 63.0, ... 100.0, 55.1), consistent with the cells where the point is clearly visible. The one exception is "Teamsters v. Local v Local", whose percent cell prints as a bare "0" with no decimal fraction; transcribed as "0". Row-label punctuation after "v"/"v." is unreliable at this scan quality; the period is transcribed where visible and omitted where it is not, so a few of these may differ from the print. "Other national unions" has faint stray marks after the label that could be scan noise or a small raised mark; no footnote marker is transcribed, and no footnotes are printed under this panel. The "5" in the elections column of "Teamsters v. Teamsters" has a stray mark to its left that reads like a comma or low period; taken as scan noise and transcribed as "5". "AFL-CIO v. Teamsters v. Natl" likewise has a stray tick just left of the "1" in the other-national column. The last two three-union rows wrap onto two printed lines ("AFL-CIO v. AFL-CIO v AFL-CIO v. / AFL-CIO" and "AFL-CIO v. AFL-CIO v. AFL-CIO v. / Teamsters") and are joined into one row each. All internal arithmetic checks pass: in every data row, elections won + elections lost = total elections, the four affiliation columns sum to elections won, the four employee-affiliation columns sum to employees in units won, and employees won + employees lost = total eligible. The 1-union, 2-union and 3-or-more subtotals equal the sums of their member rows in all 15 columns, and "Total RC elections" equals the three subtotals in all 15 columns. No misprint detected.

p. 187. This crop is a continuation page ("—Con.") carrying only panel C, "C. Elections in RM cases"; panels A and B are on earlier pages. The table is 16 columns wide (stub plus 15 numeric columns). No footnote texts are printed on this crop, so `footnotes` is empty, but two markers do appear: a raised 1 on the caption (omitted per instructions) and a raised 2 on the column heading "Total elections", recorded as "Total elections^2". Their definitions must be on an earlier page of the table. Row-label punctuation is uncertain where the label runs straight into the leader dots: "AFL-CIO v. Natl" and "Teamsters v. Natl" may each end with an abbreviating period ("Natl.") that is indistinguishable from the first leader dot — note that "Natl. v. Local" clearly has one. Likewise "AFL-CIO v AFL-CIO" and "Teamsters v Local" appear to lack the period after "v" that the other rows have; both are transcribed as they read. The bold "3" opening "3 (or more)-union elections" is part of the label, not a footnote marker. Arithmetic cross-checks all pass, which corroborates the column placement: the three subtotal rows sum their components exactly in all 15 numeric columns, "Total RM elections" equals the three subtotals in all 15, and every "Percent won" equals "Total won" / "Total elections" to the printed tenth (e.g. 68/216 = 31.5, 61/152 = 40.1, 166/418 = 39.7). No printed value appears to be a misprint. The scan renders 5 with an old-style glyph easily confused with 6; the affected cells (Teamsters "152" and "40.1"; the "5" under "In units won by — Teamsters" for "Teamsters v Local"; the "5" under "Other local unions" for "2-union elections"; "5,065") were each confirmed by the sum checks above.

p. 188. The crop carries no column headings — it is the final panel (D) of a table whose headings are printed on an earlier page — so header_rows is empty. Column count (15 data columns + stub) was taken from the vertical rules and confirmed by the three summary rows, each of which carries 15 values. "D. Elections in RD cases" is printed centred above the table's top rule; it is transcribed as a label-only section row. Dropped decimal points: the scan does not reproduce the decimal point in four percent cells — "43 2" (Other national unions), "50 0" (AFL-CIO v. AFL-CIO), "100 0" (AFL-CIO v. Teamsters) and "100 0" (Natl. v. Natl.). They are transcribed as 43.2, 50.0, 100.0 and 100.0; each is confirmed by the ratio of the two columns to its left (16/37 = 43.2, 1/2 = 50.0, 9/9 and 1/1 = 100.0). Label punctuation: the period after "v" is faint or lost in "AFL-CIO v. Teamsters" and "Teamsters v. Local", and the trailing period of "Natl." in "AFL-CIO v. Natl." and "Natl. v. Natl." merges into the leader dots. Periods are written in all of these for consistency with the rows where they are legible. Empty cells in the body are printed as short dash rules (------), transcribed as "". A printed 0 is distinguished from a dash rule and kept; note that "Teamsters v. Local" carries a printed 0 in the other-local column (7th data column) and in the following no-union column, and a printed 0 in the other-local employees column as well. A stray ink speck sits in the AFL-CIO row above the 5th data column, and another before the 3 in "AFL-CIO v. Natl."; both are treated as scan noise, not as characters or footnote markers. No arithmetic was altered. Every row adds across and every column adds down as printed: the 1-union, 2-union and Total RD elections lines all reconcile exactly with their component rows, and employees eligible equals units choosing a representative plus units choosing no representative in every row.