Analysis of Methods of Disposition of Unfair Labor Practice Cases Closed, Fiscal Year 1978

p. 267. 18 columns: stub + "All C cases" (Number, Percent of total closed, Percent of total method) + seven two-column panels (CA, CB, CC, CD, CE, CG, CP), each Number and Percent of total closed. Single panel, no side-by-side panels. Decimal points: in this 1-bit scan the decimal point in many percent cells did not survive printing/scanning, so cells appear as "100 0", "24 8", "66 5" etc., while other cells in the same columns clearly show the point ("33.3", "11.6", "1.9", "0.5"). Every percent in this table is typeset to one decimal place, so all percent cells are transcribed with the decimal point. No value was otherwise altered. CD cases / Number, row "Before issuance of complaint": the cell holds a parenthesized raised 2, "(2)", used in place of a figure (a footnote reference). Transcribed as "(2)". Column arithmetic is consistent with a value of 0 there (4,361 + 1,053 + 590 + 0 + 16 + 17 + 106 = 6,143). Apparent misprints, transcribed as printed, not corrected: - Row "Stipulated decision" (the second one, under "After hearing opened"): All C Number is 9, but the panel figures read CA 8 and CB/CC/CD/CE/CG/CP all 0, which sum to 8. - Related: row "After hearing opened" shows CC Number 1, but both of its sub-rows ("Stipulated decision" 0, "Consent decree" 0) read 0. Both discrepancies would resolve if the CC "Stipulated decision" cell were 1; I checked that cell twice and it reads 0 with a dash rule in its percent column. All other rows and panel columns cross-foot exactly. The crop ends at the table's closing double rule; no footnote text, caption or page number is shown, so the footnote keyed by "(2)" is not available here. Faint cells where a dash rule is very light (e.g. the percent cells in the "After hearing opened" and second "Stipulated decision" rows for CE, CG and CP) were read as dash rules and recorded as "".

p. 268. Grid is 18 columns: stub plus 17 printed value columns, counted from the column rules (a number/percent pair repeated across the page, preceded by a total number column, a percent-of-all-cases column and a percent-within-method column). No column headings appear on this crop: it is a continuation page, so `header_rows` is empty. The table also continues from the preceding page — rows above "Withdrawal" (e.g. the overall total and the agreement/settlement section) are not on this image. Scan quality: the period in many one-decimal percentages has dropped out or faded to a gap, so the print reads e.g. "34 7", "43 8", "0 0". These are written with the decimal point restored ("34.7", "43.8", "0.0"), since every cell in those columns is a one-decimal percent. No digits were altered. Column 10 of the two "Before issuance of complaint" rows holds a parenthesized raised 2, i.e. a pointer to footnote 2 rather than a value; transcribed as "(2)". In "By Board decision", the value 0.4 in the last column is printed slightly low, near the baseline of the following block, but it belongs to that row: the 2 cases in the preceding column equal the 2 shown for the "Adopting administrative law judge's decision" sub-row plus 0 for "Contested", and 2 of roughly 470 cases is 0.4 percent. Arithmetic checks pass as printed: the Withdrawal detail sums to 12,391 and its within-method percents to 100.0; the Dismissal detail sums to 14,125 and its within-method percents to 100.0; 122 + 190 = 312 for "By Board decision". One minor internal inconsistency in the fourth-from-last pair: "Dismissal" shows 7 cases at 12.4 percent where the column base implied by the other entries (about 57) would give 12.3; transcribed as printed. The dash rules in the "10(k) actions" row occupy every column except 1, 2, 10 and 11; they are rendered as empty cells per the convention.