Size of Units in Representation Elections in Cases Closed, Fiscal Year 1988

p. 252. Crop is cut off at the bottom: the "1,000 to 1,999" row sits on the lower edge and anything below it (any further size classes, panel "B." for decertification elections, and the text of footnote 1 whose marker appears on the caption) is not on this image, so `footnotes` is empty. Panel A itself appears complete through "1,000 to 1,999": the rows I transcribed sum exactly to every printed total (number eligible 211,438; total elections 3,509; AFL-CIO 1,074; Teamsters 456; other national 63; other local 143; no representative 1,773), and the cumulative-percent column reaches 100.0 on that row. Dropped decimal points: the scan prints many percent cells with a faint or missing decimal point (e.g. the total row's AFL-CIO and Teamsters percents appear as "100 0", "Under 10" AFL-CIO as "25 5", "20 to 29" cumulative as "54 8", "30 to 39" percent of total as "8 1", "160 to 169" cumulative as "92 0", "400 to 499" cumulative as "98 9"). These are transcribed with the decimal point as typeset; the cumulative column reproduces the running sum of the "Percent of total" column exactly at every row, which confirms the placement. Zero percent cells: every cell whose count is 0 renders in this scan with doubled/overlapping glyphs, reading like "0.00.0" or "0 00 0" (e.g. "120 to 129" other national percent, "160 to 169" Teamsters percent, "180 to 189" other national and other local percents, "1,000 to 1,999" Teamsters, other national and other local percents). I transcribed each as 0.0, which is what the adjacent count of 0 and the column totals require, but the exact printed glyphs in those cells are not legible at this resolution. Headings: the fourth union pair reads "Other local unions / Number / Percent of size class" ("of", not "by") while the other three pairs and the no-representative pair read "Percent by size class"; transcribed as printed.

p. 253. Crop shows only part B of the table (part A is on the preceding page); the column headings are printed on this page and are transcribed. 15 columns. Zero-percent cells: wherever a count is 0 the matching percent cell prints a six-glyph string that I read as \"0.00.0\" (individual impressions vary between \"0.00.0\", \"0.000\", \"0 00.0\" and \"0.00 0\" because one or other period is faint). The cells are visibly wider than a 3-character \"0.0\", so I transcribed them uniformly as \"0.00.0\"; the intended value in every case is zero, and a reader should treat the exact glyph string as uncertain (it may be \"0.0\" or \"0.000\"). This affects every percent cell whose paired Number cell is 0. Total RD elections, \"Cumulative percent of total\": prints the same anomalous zero string (\"0.00.0\") rather than 100.0 — transcribed as printed; apparent misprint, since the last body row's cumulative percent is 100.0. Faint decimal points: the scan drops or thins the decimal point in several numbers (e.g. \"25 6\", \"3 6\", \"1 9\", \"89 6\", \"94 3\", \"99 4\", \"0 6\", \"120\", \"7 2\", \"1 7\", \"1 1\"). I restored the decimal point in each; every one is confirmed by arithmetic (the Percent-of-total column sums to 100.0 and reproduces the cumulative column exactly). Verification: Number eligible sums to 32,254; Total elections sums to 644; the five outcome Number columns sum to 125, 43, 5, 12 and 459 and each row's five outcome counts sum to its Total elections; each percent-by-size-class column sums to 100.0. All printed totals agree. Values that differ from a straight recomputation but are correct as printed (they are the figures that force each column to 100.0): Teamsters percent for \"10 to 19\" is 20.7 (9/43 = 20.9); Other local unions percents 16.8 for \"20 to 29\" (2/12 = 16.7) and 25.1 for \"40 to 49\" (3/12 = 25.0). \"100 to 109\", Elections in which no representative was chosen / Percent by size class: a small speck sits just left of \"0.9\"; I read it as a stray mark (a leftover leader dot), not a footnote marker or digit. A small stray mark sits above the stub heading at the top of the first column; it is not a footnote marker on any heading. The footnote line under the table is clipped by the bottom edge of the crop — only the upper half of the characters is visible; the reading \"1 See Glossary of terms for definitions.\" is confident on wording, less so on the final period.