Final Outcome of Representation Elections in Cases Closed, Fiscal Year 1965

p. 210. 16 columns: stub, "Total elections^2", six under "Elections won by unions", "Elections in which no representative chosen", six under "Employees eligible to vote" (Total, In elections won, and four under the second-level spanner "In units won by"), and a final tall column "In elections where no representative chosen" that stands outside the "Employees eligible to vote" spanner (separate rule on both sides). DECIMAL POINTS: the percent column is printed with one decimal (e.g. "60 2", "100 0" in this 1-bit scan, where the period has dropped out). I have restored the decimal point (60.2, 100.0, ...), the same repair as the space-after-comma. No value was computed or rounded. ABBREVIATION PERIODS: the scan drops light punctuation unevenly, so some row labels appear as "AFL-CIO v AFL-CIO" / "Natl" and others as "AFL-CIO v. AFL-CIO" / "Natl."; I have written "v." and "Natl." throughout, which is what the printed page carries. Likewise the panel head is printed letterspaced as "A ALL REPRESENTATION ELECTIONS"; I read the period after "A" as present but faint. DOUBTFUL CELLS — two rows do not reconcile, and these are the only arithmetic failures in the whole grid: (1) Row "AFL-CIO v. Natl.": I read Total won = 140 and Percent won = 83.3, but AFL-CIO 82 + Other national 59 = 141, and Total elections 169 - no representative chosen 28 = 141; 141/169 = 83.4. Both of those cells may in fact read 141 and 83.4. (2) Row "AFL-CIO v. Local": I read Total won = 130, but AFL-CIO 68 + Other local 61 = 129, Total elections 134 - no representative chosen 5 = 129, and the printed Percent won 96.3 equals 129/134 (130/134 would be 97.0). That cell may in fact read 129. Note that the subtotal 689 for "2-union elections" is satisfied either by the pair (140, 130) as I read them or by the pair (141, 129) implied by the row arithmetic, so the column sum does not settle it. Values are transcribed as I read them, uncorrected; these three cells are worth re-checking against a higher-resolution scan. Everything else reconciles: all four union columns sum to "Total won", "Total won" + "no representative chosen" = "Total elections", the four "In units won by" columns sum to "In elections won", and that plus the last column equals "Total" employees, in every row; the 1-union, 2-union, 3-(or more)-union blocks each sum to their subtotal line, and the three subtotals sum to "Total representation elections". Dash rules ("------") are transcribed as empty cells; printed zeros are kept as 0. Leader dots after row labels are dropped. The crop ends just below the last body row: the line "See footnotes at end of table." is clipped at the bottom edge but legible, and the numbered footnotes (marker 1 on the caption, marker 2 on "Total elections") are printed elsewhere, as is panel B of the table, which continues past this page.

p. 211. Continuation page of Table 13 ("—Con."); it carries the full column headings and only panel B (section A is on the preceding page). No footnotes are printed below the table on this crop; the raised "1" visible in the caption belongs to the caption and is therefore not transcribed. Decimal points: in the "Percent won" column the decimal point is nearly lost in this 300-ppi bitonal scan, so the values read as "61 8", "100 0", etc. They are unambiguously one-decimal percentages and are transcribed with the point (61.8, 100.0, ...). Possible misprint: "AFL-CIO v. Natl" (2-union block) shows Percent won 82.7, but 135 of 163 elections won is 82.8. The other three digits of that cell (163, 135) are clear and every other percent in the table matches its own quotient to one decimal, so either the source rounded down or the final digit is a faint 8 misread as 7. Transcribed as printed. The "v." in the versus-style row labels prints inconsistently in the scan — some lines appear to lack the period (e.g. "AFL-CIO v AFL-CIO", "AFL-CIO v Natl", "Teamsters v Local"). This is scan dropout of a period in worn type; the labels are normalized to "v." throughout. Cells filled with a leader/dash rule ("--------") are transcribed as "" (no value), and printed zeros are kept as 0; the two are distinguished throughout. Every row and column of the panel cross-foots: the three block subtotals (1-union, 2-union, 3-or-more-union) sum to "Total RC elections" in all 15 numeric columns, each block's detail lines sum to its subtotal, and within each row the union-by-union breakdowns sum to "Total won" and to "In elections won", with Total minus In-elections-won equal to the last column. This arithmetic consistency is the main basis for the column assignments.

p. 212. COLUMNS: No column headings are printed on this crop (it continues a table from an earlier page), so header_rows is empty. The body has 15 numeric columns plus the stub, counted from the column rules: (1) total elections, (2) percent won by unions, (3-7) elections won, total and by AFL-CIO / Teamsters / other national / other local, (8) elections lost, (9) total eligible voters, (10-14) employees in units choosing a representative, total and by the same four affiliations, (15) employees in units choosing no representative. Column placement was verified arithmetically: in panel C every column footing and every internal sum checks (e.g. 9,623 = 7,263 + 649 + 232 + 1,479; 173 = 109 + 49 + 9 + 6; 400 = 173 + 227), and in panel D all sums check except the one cell noted below. SUSPECTED MISPRINT: Panel D, row "1-union elections", column 12 (employees in units choosing Teamsters) is transcribed as printed, 578. It is inconsistent with the rest of the table: the row's own total requires 518 (3,189 + 518 + 546 + 0 = 4,253), the "Teamsters" detail row directly beneath prints 518, and the column footing requires 518 (518 + 174 + 750 = 1,442 in "Total RD elections"). The glyph is ambiguous at scan resolution, so this may be a 518 whose "1" has broken into what looks like a "7" rather than a typesetting error. DECIMAL POINTS: Column 2 is a one-decimal percentage. The decimal point survives the scan in only two cells ("41.3" for C 1-union elections and "100.0" for D AFL-CIO v Teamsters); elsewhere it prints as a gap ("43 3", "100 0", "0 0"). I have written the point in every cell of that column, since each value matches the quotient of columns 3 and 1 to one decimal (173/400 = 43.3, 0/2 = 0.0, etc.). No value was rounded or otherwise altered. STRAY MARKS: Panel C, "Total RM elections", column 12 shows "649," with a trailing comma and nothing after it; the arithmetic confirms 649, so the comma is treated as an artifact and the cell is given as 649. Panel C, "3 (or more)-union elections", column 6 shows a small apostrophe-like speck before the 0; transcribed as 0. LABELS: The centred panel titles "C. ELECTIONS IN RM CASES" and "D. ELECTIONS IN RD CASES" are given as label-only rows. Row labels read "AFL-CIO v Teamsters" etc. with no period after "v" in every case but one, panel C's "AFL-CIO v. Natl", which is transcribed as it appears; the period is probably present in all of them and simply lost in most. "Natl" may be printed "Natl." but any period is indistinguishable from the leader dots that follow. Empty cells in the image are dash rules ("------"), rendered as "". Footnote markers 1 and 2 belong to headings on the preceding page and appear nowhere on this crop.