Size of Units in Representation Elections in Cases Closed, Fiscal Year 1986

p. 276. CROP: the image is cut off at the bottom immediately after the "3,000 to 9,999" row. The table continues below this crop (section A almost certainly has further size classes and/or a section "B." for decertification elections), and no footnote text is visible, so `footnotes` is empty even though the caption carries a raised "1" marker (the caption itself is excluded per instructions). DECIMAL POINTS: this scan systematically drops the small period in decimal numbers — every percent cell renders as e.g. "100 0", "22 7", "0 0". The decimals are unambiguous (one decimal place throughout; the cumulative-percent column accumulates correctly: 22.7 + 20.4 = 43.1, 43.1 + 11.7 = 54.8, etc.), so I have written them with decimal points. DROPPED COMMAS: for the same reason, two "Number eligible to vote" cells print without a visible thousands comma — "20 to 29" shows "10 402" and "110 to 119" shows "7 137". I have written them as 10,402 and 7,137; the column sum confirms these values (see below). SECTION LINE: the panel heading reads "A Certification elections (RC and RM)" in the image, with no visible period after the "A". Given that this scan loses periods elsewhere, I have transcribed it as "A. Certification elections (RC and RM)"; if a literal reading is wanted, the period is not legible. TOTAL RC AND RM ELECTIONS, "Cumulative percent of total": the cell holds a dash rule ("——"), transcribed as "". ARITHMETIC CHECK (all pass, as printed): Number eligible to vote 14,886…3,037 sums to 223,018; Total elections to 3,663; AFL-CIO Number to 1,073; Teamsters Number to 465; Other national Number to 59; Other local Number to 143; no-representative Number to 1,923. This independently confirms the digits and the column each value sits in. ONE APPARENT ROUNDING QUIRK IN THE ORIGINAL (transcribed as printed, not corrected): row "200 to 299", "Percent of total" is printed 3.0, though 114/3,663 = 3.11. The printed cumulative column is internally consistent with the printed 3.0 (94.3 → 97.3), so this is the source's own rounding, not a misreading. HEADING WORDING: the "Other local unions" sub-column is printed "Percent of size class", whereas the other four panels read "Percent by size class". Transcribed as printed. The scan also renders "in" as "ın" / "unions" as "unıons" (dotless i) in the headings; I have written the normal spellings.

p. 277. 15 columns: stub, four tall columns (Number eligible to vote / Total elections / Percent of total / Cumulative percent of total), four two-column union blocks under the spanner "Elections in which representation rights were won by", and a final two-column block "Elections in which no representative was chosen" that spans the top two heading levels. Punctuation in this 1-bit scan: decimal points and sentence periods are systematically dropped or nearly invisible at this resolution (percents render as "100 0", "31 4"), and the thousands comma is lost in a few numbers ("1 728", "1 585", "1 011", "1 798"). I have restored the decimal points and the commas. For the same reason the period after the panel letter in "B. Decertification elections (RD)" and the terminal period of footnote 1 are not legible; I have written both in the conventional form. The "Cumulative percent of total" cell of the "Total RD elections" row holds a dash rule ("- -"), transcribed as "". Doubtful cell: row "20 to 29", column "Other local unions / Percent of size class" reads "31 5" to me, i.e. 31.5, but 6 of 19 is 31.6; the final digit may be a 6 misread or a misprint. Transcribed as printed. Arithmetic checks run as a transcription check (not applied to the cells): the four count columns and the "Number eligible to vote" column all sum exactly to the printed Total RD elections line (36,221; 857; 131; 59; 2; 19; 646, and 131+59+2+19+646 = 857), and the "Percent of total" column accumulates exactly to the printed "Cumulative percent of total" column in every row, so the digits and their column placement are well corroborated. Two "Percent of total" entries differ from a direct division as printed (110 to 119 shows 0.5 where 5/857 is 0.58; 120 to 129 shows 0.4 where 4/857 is 0.47), but both agree with the printed cumulative column, so they are transcribed as printed. Footnote marker 1 is printed under the table but its anchor is not visible in the crop; it most likely attaches to the table caption, which is cropped off above. No footnote marker appears on any heading, row label or value in the body.