Analysis of Methods of Disposition of Unfair Labor Practice Cases Closed, Fiscal Year 1983

p. 196. 18 columns: stub + All C cases (Number, Percent of total closed, Percent of total method) + Number/Percent-of-total-closed pairs for CA, CB, CC, CD, CE, CG, CP. Decimal points: the scan has lost the decimal point in essentially every percent value, which prints as e.g. "100 0", "27 9", "61 4". These are decimals (the digit groups align with a decimal column and the subtotals add up: 61.4 + 35.8 + 0.7 = 97.9; 97.9 + 2.1 = 100.0), so they are transcribed with a period. No value was otherwise altered. Doubtful cells: - "Total number of cases closed", All C cases / "Percent of total method": reads "0 0", transcribed as "0.0". This is odd for a total line (one would expect blank or 100.0); transcribed as printed, flagged as a possible misprint. - "Before issuance of complaint", CD cases / Number: the cell holds a single small, raised "2" with the CD percent cell blank. It may be a footnote marker rather than a value: the CD column arithmetic implies this cell should be 0 or blank (informal settlement CD = 6 = 0 + 6 + 0, and 6 + 2 formal = 8 = the "Agreement of the parties" CD total). Transcribed literally as "2". - "After hearing opened", All C cases / Number: a stray leader dot precedes "94"; dropped. - The second formal-settlement sub-row is printed "After Issuance of complaint, before opening of hearing" with an apparent capital I, unlike the matching informal-settlement row ("After issuance..."); kept as printed. - Many i's render dotless throughout the scan ("complaınt", "openıng"); normalized silently. Crop: the table is cut off immediately below the "Compliance with" row — the remaining body rows and all footnote text (the caption carries marker 1) fall outside this image, so `footnotes` is empty. All internal sums that can be checked are consistent as transcribed (e.g. 10,410 + 222 = 10,632; 6,533 + 3,803 + 74 = 10,410; 128 + 94 = 222 for All C cases, and likewise down each case-type column).

p. 197. Crop is a continuation page: the caption carries a footnote marker 1, but no footnote texts are printed on this image, and the table is cut off immediately below the "Dismissal" row (its stage breakdown continues on the following page). Likewise the rows at the top continue a method block begun on the previous page. Decimal points: in this 1-bit scan the decimal point in every percentage renders as a gap (e.g. "35 1", "100 0", "0 0"). These are one-decimal percentages and are transcribed with the printed decimal point (35.1, 100.0, 0.0); no value was computed or altered. "Before issuance of complaint", CD cases / Number: the cell contains only a small raised "2" (a footnote marker) with no digits, right-aligned in the CD Number column; transcribed as "^2". The column arithmetic supports an underlying value of 0 there (CD withdrawals: 0 + 1 + 0 + 0 + 0 = 1, matching the "Withdrawal" row's CD total of 1), so the raised 2 is a marker, not the number 2. "After board of court decision" is transcribed exactly as printed; it appears to be a misprint for "After board or court decision". Blank percent cells beside printed zeros (e.g. CD/CE/CG/CP in the "Administrative law judge's decision" row) are genuinely empty in the original and are given as "". Faint ink marks appear in the CE column of the "After board of court decision" row and in the CP column just above the "Dismissal" rule; both read as scan artefacts, not characters, and were not transcribed. Checks that passed: Board decision = Adopting + Contested in every column; Withdrawal = sum of its five stage rows in every column; percent-of-total-method for the withdrawal stages sums to 100.0.

p. 198. CONTINUATION CROP: no column headings are printed on this image, so header_rows is empty. The grid has 18 columns (stub + 17), counted from the vertical column rules; the structure is a count column, two percent columns, then seven count/percent pairs (the case-type columns, CA through CP). Column assignment was verified arithmetically: in every data row the seven count columns (5, 7, 9, 11, 13, 15, 17) sum exactly to the total in column 2 (e.g. 8,346+3,784+232+0+22+8+76 = 12,468; 115+27+1+0+0+1+2 = 146; 0+0+0+355+0+0+0 = 355). DECIMAL POINTS: the 1-bit scan has dropped nearly every period in the image. All percent cells print as e.g. "32 8", "98 0", "95 9", "0 0". These are plainly decimal fractions, so they are transcribed with the decimal point restored ("32.8", "98.0", "95.9", "0.0"). The same loss affects sentence periods in the two footnotes ("by stage See glossary", "under sec 10(k) of the act See table 7A"); periods have likewise been restored there. No digits were altered. Row 1, column 11 (the fourth case-type count column, the one that carries 355 in the 10(k) row) holds only a raised "2" with no number beneath it; it is transcribed as the bare marker "^2", consistent with footnote 2, which says CD cases closed at this stage are handled as 10(k) proceedings. Blank percent cells: throughout the table a count of 0 is printed with its adjacent percent cell left empty, so those percent cells are "". The blanks in columns 6, 8, 10, 14, 16 and 18 of the 10(k) row and in the trailing columns of the sparse rows follow that pattern. Rows 6-7 ("Adopting administrative law judge's decision (no exceptions filed)" and "Contested") are printed indented as subdivisions of "By board decision" and are set off by a horizontal rule; rows 8-9 and rows 10-11 are likewise separated by double rules. The indentation and rules are not representable in the grid, so all labels are given at one level. Possible oddity, transcribed as printed, not corrected: in column 4 (the second percent column) the "10(k) actions" row shows 0.0 against a count of 355, which is inconsistent with the other rows of that column (e.g. 146 -> 1.1, 93 -> 0.7). Either that column uses a different base for the final section or it is a misprint. Footnote marker 1 is not attached to anything visible on this crop; it belongs to the table caption or a heading on the earlier part of the table.