pintos-os.org Git - pspp/blob - tests/language/stats/examine.at

   1 dnl PSPP - a program for statistical analysis.
   2 dnl Copyright (C) 2017, 2019 Free Software Foundation, Inc.
   3 dnl
   4 dnl This program is free software: you can redistribute it and/or modify
   5 dnl it under the terms of the GNU General Public License as published by
   6 dnl the Free Software Foundation, either version 3 of the License, or
   7 dnl (at your option) any later version.
   8 dnl
   9 dnl This program is distributed in the hope that it will be useful,
  10 dnl but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12 dnl GNU General Public License for more details.
  13 dnl
  14 dnl You should have received a copy of the GNU General Public License
  15 dnl along with this program.  If not, see <http://www.gnu.org/licenses/>.
  16 dnl
  17 AT_BANNER([EXAMINE])
  18
  19 AT_SETUP([EXAMINE])
  20 AT_KEYWORDS([categorical categoricals])
  21 AT_DATA([examine.sps], [
  22 DATA LIST LIST /QUALITY * W * BRAND * .
  23 BEGIN DATA
  24 3  1  1
  25 2  2  1
  26 1  2  1
  27 1  1  1
  28 4  1  1
  29 4  1  1
  30 5  1  2
  31 2  1  2
  32 4  4  2
  33 2  1  2
  34 3  1  2
  35 7  1  3
  36 4  2  3
  37 5  3  3
  38 3  1  3
  39 6  1  3
  40 END DATA
  41
  42 WEIGHT BY w.
  43
  44 VARIABLE LABELS brand   'Manufacturer'.
  45 VARIABLE LABELS quality 'Breaking Strain'.
  46
  47 VALUE LABELS /brand 1 'Aspeger' 2 'Bloggs' 3 'Charlies'.
  48
  49 LIST /FORMAT=NUMBERED.
  50
  51 EXAMINE
  52         quality BY brand
  53         /STATISTICS descriptives extreme(3)
  54         .
  55 ])
  56
  57
  58 dnl In the following data, only the extreme values have been checked.
  59 dnl The descriptives have been blindly pasted.
  60 AT_CHECK([pspp -O format=csv examine.sps], [0], [dnl
  61 Table: Reading free-form data from INLINE.
  62 Variable,Format
  63 QUALITY,F8.0
  64 W,F8.0
  65 BRAND,F8.0
  66
  67 Table: Data List
  68 Case Number,QUALITY,W,BRAND
  69 1,3.00,1.00,1.00
  70 2,2.00,2.00,1.00
  71 3,1.00,2.00,1.00
  72 4,1.00,1.00,1.00
  73 5,4.00,1.00,1.00
  74 6,4.00,1.00,1.00
  75 7,5.00,1.00,2.00
  76 8,2.00,1.00,2.00
  77 9,4.00,4.00,2.00
  78 10,2.00,1.00,2.00
  79 11,3.00,1.00,2.00
  80 12,7.00,1.00,3.00
  81 13,4.00,2.00,3.00
  82 14,5.00,3.00,3.00
  83 15,3.00,1.00,3.00
  84 16,6.00,1.00,3.00
  85
  86 Table: Case Processing Summary
  87 ,Cases,,,,,
  88 ,Valid,,Missing,,Total,
  89 ,N,Percent,N,Percent,N,Percent
  90 Breaking Strain,24.00,100.0%,.00,.0%,24.00,100.0%
  91
  92 Table: Extreme Values
  93 ,,,Case Number,Value
  94 Breaking Strain,Highest,1,12,7.00
  95 ,,2,16,6.00
  96 ,,3,14,5.00
  97 ,Lowest,1,3,1.00
  98 ,,2,4,1.00
  99 ,,3,2,2.00
 100
 101 Table: Descriptives
 102 ,,,Statistic,Std. Error
 103 Breaking Strain,Mean,,3.54,.32
 104 ,95% Confidence Interval for Mean,Lower Bound,2.87,
 105 ,,Upper Bound,4.21,
 106 ,5% Trimmed Mean,,3.50,
 107 ,Median,,4.00,
 108 ,Variance,,2.52,
 109 ,Std. Deviation,,1.59,
 110 ,Minimum,,1.00,
 111 ,Maximum,,7.00,
 112 ,Range,,6.00,
 113 ,Interquartile Range,,2.75,
 114 ,Skewness,,.06,.47
 115 ,Kurtosis,,-.36,.92
 116
 117 Table: Case Processing Summary
 118 ,Manufacturer,Cases,,,,,
 119 ,,Valid,,Missing,,Total,
 120 ,,N,Percent,N,Percent,N,Percent
 121 Breaking Strain,Aspeger,8.00,100.0%,.00,.0%,8.00,100.0%
 122 ,Bloggs,8.00,100.0%,.00,.0%,8.00,100.0%
 123 ,Charlies,8.00,100.0%,.00,.0%,8.00,100.0%
 124
 125 Table: Extreme Values
 126 ,Manufacturer,,,Case Number,Value
 127 Breaking Strain,Aspeger,Highest,1,6,4.00
 128 ,,,2,5,4.00
 129 ,,,3,1,3.00
 130 ,,Lowest,1,3,1.00
 131 ,,,2,4,1.00
 132 ,,,3,2,2.00
 133 ,Bloggs,Highest,1,7,5.00
 134 ,,,2,9,4.00
 135 ,,,3,11,3.00
 136 ,,Lowest,1,8,2.00
 137 ,,,2,10,2.00
 138 ,,,3,11,3.00
 139 ,Charlies,Highest,1,12,7.00
 140 ,,,2,16,6.00
 141 ,,,3,14,5.00
 142 ,,Lowest,1,15,3.00
 143 ,,,2,13,4.00
 144 ,,,3,14,5.00
 145
 146 Table: Descriptives
 147 ,Manufacturer,,,Statistic,Std. Error
 148 Breaking Strain,Aspeger,Mean,,2.25,.45
 149 ,,95% Confidence Interval for Mean,Lower Bound,1.18,
 150 ,,,Upper Bound,3.32,
 151 ,,5% Trimmed Mean,,2.22,
 152 ,,Median,,2.00,
 153 ,,Variance,,1.64,
 154 ,,Std. Deviation,,1.28,
 155 ,,Minimum,,1.00,
 156 ,,Maximum,,4.00,
 157 ,,Range,,3.00,
 158 ,,Interquartile Range,,2.75,
 159 ,,Skewness,,.47,.75
 160 ,,Kurtosis,,-1.55,1.48
 161 ,Bloggs,Mean,,3.50,.38
 162 ,,95% Confidence Interval for Mean,Lower Bound,2.61,
 163 ,,,Upper Bound,4.39,
 164 ,,5% Trimmed Mean,,3.50,
 165 ,,Median,,4.00,
 166 ,,Variance,,1.14,
 167 ,,Std. Deviation,,1.07,
 168 ,,Minimum,,2.00,
 169 ,,Maximum,,5.00,
 170 ,,Range,,3.00,
 171 ,,Interquartile Range,,1.75,
 172 ,,Skewness,,-.47,.75
 173 ,,Kurtosis,,-.83,1.48
 174 ,Charlies,Mean,,4.88,.44
 175 ,,95% Confidence Interval for Mean,Lower Bound,3.83,
 176 ,,,Upper Bound,5.92,
 177 ,,5% Trimmed Mean,,4.86,
 178 ,,Median,,5.00,
 179 ,,Variance,,1.55,
 180 ,,Std. Deviation,,1.25,
 181 ,,Minimum,,3.00,
 182 ,,Maximum,,7.00,
 183 ,,Range,,4.00,
 184 ,,Interquartile Range,,1.75,
 185 ,,Skewness,,.30,.75
 186 ,,Kurtosis,,.15,1.48
 187 ])
 188
 189 AT_CLEANUP
 190
 191 AT_SETUP([EXAMINE -- extremes])
 192 AT_KEYWORDS([categorical categoricals])
 193 AT_DATA([examine.sps], [dnl
 194 data list free /V1 W
 195 begin data.
 196 1  1
 197 2  1
 198 3  2
 199 3  1
 200 4  1
 201 5  1
 202 6  1
 203 7  1
 204 8  1
 205 9  1
 206 10 1
 207 11 1
 208 12 1
 209 13 1
 210 14 1
 211 15 1
 212 16 1
 213 17 1
 214 18 2
 215 19 1
 216 20 1
 217 end data.
 218
 219 weight by w.
 220
 221 examine v1
 222  /statistics=extreme(6)
 223  .
 224 ])
 225
 226 AT_CHECK([pspp -O format=csv examine.sps], [0],[dnl
 227 Table: Case Processing Summary
 228 ,Cases,,,,,
 229 ,Valid,,Missing,,Total,
 230 ,N,Percent,N,Percent,N,Percent
 231 V1,23.00,100.0%,.00,.0%,23.00,100.0%
 232
 233 Table: Extreme Values
 234 ,,,Case Number,Value
 235 V1,Highest,1,21,20.00
 236 ,,2,20,19.00
 237 ,,3,19,18.00
 238 ,,4,18,17.00
 239 ,,5,17,16.00
 240 ,,6,16,15.00
 241 ,Lowest,1,1,1.00
 242 ,,2,2,2.00
 243 ,,3,3,3.00
 244 ,,4,4,3.00
 245 ,,5,5,4.00
 246 ,,6,6,5.00
 247 ])
 248
 249 AT_CLEANUP
 250
 251
 252 AT_SETUP([EXAMINE -- extremes with fractional weights])
 253 AT_KEYWORDS([categorical categoricals])
 254 AT_DATA([extreme.sps], [dnl
 255 set format=F20.3.
 256 data list notable list /w * x *.
 257 begin data.
 258  0.88  300000
 259  0.86  320000
 260  0.98  480000
 261  0.93  960000
 262  1.35  960000
 263  1.31  960000
 264  0.88  960000
 265  0.88  1080000
 266  0.88  1080000
 267  0.95  1200000
 268  1.47  1200000
 269  0.93  1200000
 270  0.98  1320000
 271  1.31  1380000
 272  0.93  1440000
 273  0.88  1560000
 274  1.56  1560000
 275  1.47  1560000
 276 end data.
 277
 278 weight by w.
 279
 280
 281 EXAMINE
 282         x
 283         /STATISTICS = DESCRIPTIVES EXTREME (5)
 284         .
 285 ])
 286
 287 AT_CHECK([pspp -O format=csv  extreme.sps], [0], [dnl
 288 Table: Case Processing Summary
 289 ,Cases,,,,,
 290 ,Valid,,Missing,,Total,
 291 ,N,Percent,N,Percent,N,Percent
 292 x,19.430,100.0%,.000,.0%,19.430,100.0%
 293
 294 Table: Extreme Values
 295 ,,,Case Number,Value
 296 x,Highest,1,18,1560000.000
 297 ,,2,17,1560000.000
 298 ,,3,16,1560000.000
 299 ,,4,15,1440000.000
 300 ,,5,14,1380000.000
 301 ,Lowest,1,1,300000.000
 302 ,,2,2,320000.000
 303 ,,3,3,480000.000
 304 ,,4,4,960000.000
 305 ,,5,5,960000.000
 306
 307 Table: Descriptives
 308 ,,,Statistic,Std. Error
 309 x,Mean,,1120010.293,86222.178
 310 ,95% Confidence Interval for Mean,Lower Bound,939166.693,
 311 ,,Upper Bound,1300853.894,
 312 ,5% Trimmed Mean,,1141017.899,
 313 ,Median,,1200000.000,
 314 ,Variance,,144447748124.869,
 315 ,Std. Deviation,,380062.821,
 316 ,Minimum,,300000.000,
 317 ,Maximum,,1560000.000,
 318 ,Range,,1260000.000,
 319 ,Interquartile Range,,467258.065,
 320 ,Skewness,,-.887,.519
 321 ,Kurtosis,,.340,1.005
 322 ])
 323
 324 AT_CLEANUP
 325
 326 dnl Test the PERCENTILES subcommand of the EXAMINE command.
 327 dnl In particular test that it behaves properly when there are only
 328 dnl a few cases.
 329 AT_SETUP([EXAMINE -- percentiles])
 330 AT_KEYWORDS([categorical categoricals])
 331 AT_DATA([examine.sps], [dnl
 332 DATA LIST LIST /X *.
 333 BEGIN DATA.
 334 2.00
 335 8.00
 336 5.00
 337 END DATA.
 338
 339 EXAMINE /x
 340         /PERCENTILES=HAVERAGE.
 341
 342 EXAMINE /x
 343         /PERCENTILES=WAVERAGE.
 344
 345 EXAMINE /x
 346         /PERCENTILES=ROUND.
 347
 348 EXAMINE /x
 349         /PERCENTILES=EMPIRICAL.
 350
 351 EXAMINE /x
 352         /PERCENTILES=AEMPIRICAL.
 353 ])
 354 AT_CHECK([pspp -o pspp.csv -o pspp.txt examine.sps])
 355 AT_CHECK([cat pspp.csv], [0], [dnl
 356 Table: Reading free-form data from INLINE.
 357 Variable,Format
 358 X,F8.0
 359
 360 Table: Case Processing Summary
 361 ,Cases,,,,,
 362 ,Valid,,Missing,,Total,
 363 ,N,Percent,N,Percent,N,Percent
 364 X,3,100.0%,0,.0%,3,100.0%
 365
 366 Table: Percentiles
 367 ,,Percentiles,,,,,,
 368 ,,5,10,25,50,75,90,95
 369 X,Weighted Average,.40,.80,2.00,5.00,8.00,8.00,8.00
 370 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 371
 372 Table: Case Processing Summary
 373 ,Cases,,,,,
 374 ,Valid,,Missing,,Total,
 375 ,N,Percent,N,Percent,N,Percent
 376 X,3,100.0%,0,.0%,3,100.0%
 377
 378 Table: Percentiles
 379 ,,Percentiles,,,,,,
 380 ,,5,10,25,50,75,90,95
 381 X,Weighted Average,.30,.60,1.50,3.50,5.75,7.10,7.55
 382 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 383
 384 Table: Case Processing Summary
 385 ,Cases,,,,,
 386 ,Valid,,Missing,,Total,
 387 ,N,Percent,N,Percent,N,Percent
 388 X,3,100.0%,0,.0%,3,100.0%
 389
 390 Table: Percentiles
 391 ,,Percentiles,,,,,,
 392 ,,5,10,25,50,75,90,95
 393 X,Weighted Average,.00,.00,2.00,5.00,5.00,8.00,8.00
 394 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 395
 396 Table: Case Processing Summary
 397 ,Cases,,,,,
 398 ,Valid,,Missing,,Total,
 399 ,N,Percent,N,Percent,N,Percent
 400 X,3,100.0%,0,.0%,3,100.0%
 401
 402 Table: Percentiles
 403 ,,Percentiles,,,,,,
 404 ,,5,10,25,50,75,90,95
 405 X,Weighted Average,2.00,2.00,2.00,5.00,8.00,8.00,8.00
 406 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 407
 408 Table: Case Processing Summary
 409 ,Cases,,,,,
 410 ,Valid,,Missing,,Total,
 411 ,N,Percent,N,Percent,N,Percent
 412 X,3,100.0%,0,.0%,3,100.0%
 413
 414 Table: Percentiles
 415 ,,Percentiles,,,,,,
 416 ,,5,10,25,50,75,90,95
 417 X,Weighted Average,2.00,2.00,2.00,5.00,8.00,8.00,8.00
 418 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 419 ])
 420 AT_CLEANUP
 421
 422 AT_SETUP([EXAMINE -- missing values])
 423 AT_KEYWORDS([categorical categoricals])
 424 AT_DATA([examine.sps], [dnl
 425 DATA LIST LIST /x * y *.
 426 BEGIN DATA.
 427 1   1
 428 2   1
 429 3   1
 430 4   1
 431 5   2
 432 6   2
 433 .   2
 434 END DATA
 435
 436 EXAMINE /x by y
 437         /MISSING = PAIRWISE
 438         .
 439 ])
 440 AT_CHECK([pspp -o pspp.csv examine.sps])
 441 AT_CHECK([cat pspp.csv], [0], [dnl
 442 Table: Reading free-form data from INLINE.
 443 Variable,Format
 444 x,F8.0
 445 y,F8.0
 446
 447 Table: Case Processing Summary
 448 ,Cases,,,,,
 449 ,Valid,,Missing,,Total,
 450 ,N,Percent,N,Percent,N,Percent
 451 x,6,85.7%,1,14.3%,7,100.0%
 452
 453 Table: Case Processing Summary
 454 ,y,Cases,,,,,
 455 ,,Valid,,Missing,,Total,
 456 ,,N,Percent,N,Percent,N,Percent
 457 x,1.00,4,100.0%,0,.0%,4,100.0%
 458 ,2.00,2,66.7%,1,33.3%,3,100.0%
 459 ])
 460 AT_CLEANUP
 461
 462
 463 AT_SETUP([EXAMINE -- user missing values])
 464 AT_KEYWORDS([categorical categoricals])
 465 AT_DATA([examine-m.sps], [dnl
 466 DATA LIST notable LIST /x * y *.
 467 BEGIN DATA.
 468 1                   2
 469 9999999999          2
 470 9999999999          99
 471 END DATA.
 472
 473 MISSING VALUES x (9999999999).
 474 MISSING VALUES y (99).
 475
 476 EXAMINE
 477         /VARIABLES= x y
 478         /MISSING=PAIRWISE.
 479 ])
 480 AT_CHECK([pspp -O format=csv examine-m.sps], [0], [dnl
 481 Table: Case Processing Summary
 482 ,Cases,,,,,
 483 ,Valid,,Missing,,Total,
 484 ,N,Percent,N,Percent,N,Percent
 485 x,1,33.3%,2,66.7%,3,100.0%
 486 y,2,66.7%,1,33.3%,3,100.0%
 487 ])
 488 AT_CLEANUP
 489
 490 AT_SETUP([EXAMINE -- missing values and percentiles])
 491 AT_KEYWORDS([categorical categoricals])
 492 AT_DATA([examine.sps], [dnl
 493 DATA LIST LIST /X *.
 494 BEGIN DATA.
 495 99
 496 99
 497 5.00
 498 END DATA.
 499
 500 MISSING VALUE X (99).
 501
 502 EXAMINE /x
 503         /PERCENTILES=HAVERAGE.
 504 ])
 505 AT_CHECK([pspp -o pspp.csv examine.sps])
 506 dnl Ignore output -- this is just a no-crash check.
 507 AT_CLEANUP
 508
 509 dnl Tests the trimmed mean calculation in the case
 510 dnl where the data is weighted towards the centre.
 511 AT_SETUP([EXAMINE -- trimmed mean])
 512 AT_KEYWORDS([categorical categoricals])
 513 AT_DATA([examine.sps], [dnl
 514 DATA LIST LIST /X * C *.
 515 BEGIN DATA.
 516 1 1
 517 2 49
 518 3 2
 519 END DATA.
 520
 521 WEIGHT BY c.
 522
 523 EXAMINE
 524         x
 525         /STATISTICS=DESCRIPTIVES
 526         .
 527 ])
 528 AT_CHECK([pspp -o pspp.csv examine.sps])
 529 AT_CHECK([cat pspp.csv], [0], [dnl
 530 Table: Reading free-form data from INLINE.
 531 Variable,Format
 532 X,F8.0
 533 C,F8.0
 534
 535 Table: Case Processing Summary
 536 ,Cases,,,,,
 537 ,Valid,,Missing,,Total,
 538 ,N,Percent,N,Percent,N,Percent
 539 X,52.00,100.0%,.00,.0%,52.00,100.0%
 540
 541 Table: Descriptives
 542 ,,,Statistic,Std. Error
 543 X,Mean,,2.02,.03
 544 ,95% Confidence Interval for Mean,Lower Bound,1.95,
 545 ,,Upper Bound,2.09,
 546 ,5% Trimmed Mean,,2.00,
 547 ,Median,,2.00,
 548 ,Variance,,.06,
 549 ,Std. Deviation,,.24,
 550 ,Minimum,,1.00,
 551 ,Maximum,,3.00,
 552 ,Range,,2.00,
 553 ,Interquartile Range,,.00,
 554 ,Skewness,,1.19,.33
 555 ,Kurtosis,,15.73,.65
 556 ])
 557 AT_CLEANUP
 558
 559 AT_SETUP([EXAMINE -- crash bug])
 560 AT_KEYWORDS([categorical categoricals])
 561 AT_DATA([examine.sps], [dnl
 562 data list list /a * x * y *.
 563 begin data.
 564 3 1 3
 565 5 1 4
 566 7 2 3
 567 end data.
 568
 569 examine a by x by y
 570         /statistics=DESCRIPTIVES
 571         .
 572 ])
 573 AT_CHECK([pspp -o pspp.csv examine.sps])
 574 dnl Ignore output -- this is just a no-crash check.
 575 AT_CLEANUP
 576
 577 dnl Test that two consecutive EXAMINE commands don't crash PSPP.
 578 AT_SETUP([EXAMINE -- consecutive runs don't crash])
 579 AT_KEYWORDS([categorical categoricals])
 580 AT_DATA([examine.sps], [dnl
 581 data list list /y * z *.
 582 begin data.
 583 6 4
 584 5 3
 585 7 6
 586 end data.
 587
 588 EXAMINE /VARIABLES= z BY y.
 589
 590 EXAMINE /VARIABLES= z.
 591 ])
 592 AT_CHECK([pspp -o pspp.csv examine.sps])
 593 dnl Ignore output -- this is just a no-crash check.
 594 AT_CLEANUP
 595
 596 dnl Test that /DESCRIPTIVES does not crash in presence of missing values.
 597 AT_SETUP([EXAMINE -- missing values don't crash])
 598 AT_KEYWORDS([categorical categoricals])
 599 AT_DATA([examine.sps], [dnl
 600 data list list /x * y *.
 601 begin data.
 602 1 0
 603 2 0
 604 . 0
 605 3 1
 606 4 1
 607 end data.
 608 examine x by y /statistics=descriptives.
 609 ])
 610 AT_CHECK([pspp -o pspp.csv examine.sps])
 611 dnl Ignore output -- this is just a no-crash check.
 612 AT_CLEANUP
 613
 614 dnl Test that having only a single case doesn't crash.
 615 AT_SETUP([EXAMINE -- single case doesn't crash])
 616 AT_KEYWORDS([categorical categoricals])
 617 AT_DATA([examine.sps], [dnl
 618 DATA LIST LIST /quality * .
 619 BEGIN DATA
 620 3
 621 END DATA
 622
 623
 624 EXAMINE
 625         quality
 626         /STATISTICS descriptives
 627         /PLOT = histogram
 628         .
 629 ])
 630 AT_CHECK([pspp -o pspp.csv examine.sps], [0], [ignore])
 631 dnl Ignore output -- this is just a no-crash check.
 632 AT_CLEANUP
 633
 634 dnl Test that all-missing data doesn't crash.
 635 AT_SETUP([EXAMINE -- all-missing data doesn't crash])
 636 AT_KEYWORDS([categorical categoricals])
 637 AT_DATA([examine.sps], [dnl
 638 DATA LIST LIST /x *.
 639 BEGIN DATA.
 640 .
 641 .
 642 .
 643 .
 644 END DATA.
 645
 646 EXAMINE /x
 647         PLOT=HISTOGRAM BOXPLOT NPPLOT SPREADLEVEL(1) ALL
 648         /ID=x
 649         /STATISTICS = DESCRIPTIVES EXTREME (5) ALL
 650         /PERCENTILE=AEMPIRICAL
 651         .
 652 ])
 653 AT_CHECK([pspp -o pspp.csv examine.sps], [0], [ignore])
 654 dnl Ignore output -- this is just a no-crash check.
 655 AT_CLEANUP
 656
 657 dnl Test that big input doesn't crash (bug 11307).
 658 AT_SETUP([EXAMINE -- big input doesn't crash])
 659 AT_KEYWORDS([categorical categoricals slow])
 660 AT_DATA([examine.sps], [dnl
 661 INPUT PROGRAM.
 662         LOOP #I=1 TO 50000.
 663                 COMPUTE X=NORMAL(10).
 664                 END CASE.
 665         END LOOP.
 666         END FILE.
 667 END INPUT PROGRAM.
 668
 669
 670 EXAMINE /x
 671         /STATISTICS=DESCRIPTIVES.
 672 ])
 673 AT_CHECK([pspp -o pspp.csv examine.sps])
 674 dnl Ignore output -- this is just a no-crash check.
 675 AT_CLEANUP
 676
 677 dnl Another test that big input doesn't crash.
 678 dnl The actual bug that this checks for has been lost.
 679 AT_SETUP([EXAMINE -- big input doesn't crash 2])
 680 AT_KEYWORDS([categorical categoricals slow])
 681 AT_CHECK([$PYTHON3 -c '
 682 for i in range(100000): print("AB12")
 683 for i in range(100000): print("AB04")
 684 ' > large.txt])
 685 AT_DATA([examine.sps], [dnl
 686 DATA LIST FILE='large.txt' /S 1-2 (A) X 3 .
 687
 688
 689 AGGREGATE OUTFILE=* /BREAK=X /A=N.
 690
 691
 692 EXAMINE /A BY X.
 693 ])
 694 AT_CHECK([pspp -o pspp.csv examine.sps])
 695 dnl Ignore output -- this is just a no-crash check.
 696 AT_CHECK([$PYTHON3 -c 'for i in range(25000): print("AB04\nAB12")' >> large.txt])
 697 AT_CHECK([pspp -o pspp.csv examine.sps])
 698 dnl Ignore output -- this is just a no-crash check.
 699 AT_CLEANUP
 700
 701
 702 dnl Test that the ID command works with non-numberic variables
 703 AT_SETUP([EXAMINE -- non-numeric ID])
 704 AT_KEYWORDS([categorical categoricals])
 705
 706 AT_DATA([examine-id.sps], [dnl
 707 data list notable list /x * y (a12).
 708 begin data.
 709 1  one
 710 2  two
 711 3  three
 712 4  four
 713 5  five
 714 6  six
 715 7  seven
 716 8  eight
 717 9  nine
 718 10 ten
 719 11 eleven
 720 12 twelve
 721 30 thirty
 722 300 threehundred
 723 end data.
 724
 725 set small=0.
 726 examine x
 727         /statistics = extreme
 728         /id = y
 729         /plot = boxplot
 730         .
 731 ])
 732
 733 AT_CHECK([pspp -O format=csv examine-id.sps], [0], [dnl
 734 Table: Case Processing Summary
 735 ,Cases,,,,,
 736 ,Valid,,Missing,,Total,
 737 ,N,Percent,N,Percent,N,Percent
 738 x,14,100.0%,0,.0%,14,100.0%
 739
 740 Table: Extreme Values
 741 ,,,y,Value
 742 x,Highest,1,threehundred,300.00
 743 ,,2,thirty,30.00
 744 ,,3,twelve,12.00
 745 ,,4,eleven,11.00
 746 ,,5,ten,10.00
 747 ,Lowest,1,one,1.00
 748 ,,2,two,2.00
 749 ,,3,three,3.00
 750 ,,4,four,4.00
 751 ,,5,five,5.00
 752
 753 Table: Tests of Normality
 754 ,Shapiro-Wilk,,
 755 ,Statistic,df,Sig.
 756 x,.37,14,.00
 757 ])
 758
 759 AT_CLEANUP
 760
 761 dnl Test for a crash which happened on cleanup from a bad input syntax
 762 AT_SETUP([EXAMINE -- Bad Input])
 763 AT_KEYWORDS([categorical categoricals])
 764
 765 AT_DATA([examine-bad.sps], [dnl
 766 data list list /h * g *.
 767 begin data.
 768 1 1
 769 2 1
 770 3 1
 771 4 1
 772 5 2
 773 6 2
 774 7 2
 775 8 2
 776 9 2
 777 end data.
 778
 779 EXAMINE
 780         /VARIABLES= h
 781         BY  g
 782         /STATISTICS = DESCRIPTIVES EXTREME
 783         /PLOT = lkajsdas
 784         .
 785 ])
 786
 787 AT_CHECK([pspp -o pspp.csv examine-bad.sps], [1], [ignore])
 788
 789 AT_CLEANUP
 790
 791
 792 dnl Check the MISSING=REPORT option
 793 AT_SETUP([EXAMINE -- MISSING=REPORT])
 794 AT_KEYWORDS([categorical categoricals])
 795
 796 AT_DATA([examine-report.sps], [dnl
 797 set format = F22.0.
 798 data list list /x * g *.
 799 begin data.
 800 1   1
 801 2   1
 802 3   1
 803 4   1
 804 5   1
 805 6   1
 806 7   1
 807 8   1
 808 9   1
 809 10   2
 810 20   2
 811 30   2
 812 40   2
 813 50   2
 814 60   2
 815 70   2
 816 80   2
 817 90   2
 818 101   9
 819 201   9
 820 301   9
 821 401   9
 822 501   99
 823 601   99
 824 701   99
 825 801   99
 826 901   99
 827 1001  .
 828 2002  .
 829 3003  .
 830 4004  .
 831 end data.
 832
 833 MISSING VALUES g (9, 99, 999).
 834
 835 EXAMINE
 836         /VARIABLES = x
 837         BY  g
 838         /STATISTICS = EXTREME
 839         /NOTOTAL
 840         /MISSING = REPORT.
 841 ])
 842
 843
 844 AT_CHECK([pspp -o pspp.csv -o pspp.txt examine-report.sps])
 845 AT_CHECK([cat pspp.csv], [0],
 846   [[Table: Reading free-form data from INLINE.
 847 Variable,Format
 848 x,F8.0
 849 g,F8.0
 850
 851 Table: Case Processing Summary
 852 ,g,Cases,,,,,
 853 ,,Valid,,Missing,,Total,
 854 ,,N,Percent,N,Percent,N,Percent
 855 x,.,4,100.0%,0,.0%,4,100.0%
 856 ,1,9,100.0%,0,.0%,9,100.0%
 857 ,2,9,100.0%,0,.0%,9,100.0%
 858 ,9[a],4,100.0%,0,.0%,4,100.0%
 859 ,99[a],5,100.0%,0,.0%,5,100.0%
 860 Footnote: a. User-missing value.
 861
 862 Table: Extreme Values
 863 ,g,,,Case Number,Value
 864 x,.,Highest,1,31,4004
 865 ,,,2,30,3003
 866 ,,,3,29,2002
 867 ,,,4,28,1001
 868 ,,,5,0,0
 869 ,,Lowest,1,28,1001
 870 ,,,2,29,2002
 871 ,,,3,30,3003
 872 ,,,4,31,4004
 873 ,,,5,31,4004
 874 ,1,Highest,1,9,9
 875 ,,,2,8,8
 876 ,,,3,7,7
 877 ,,,4,6,6
 878 ,,,5,5,5
 879 ,,Lowest,1,1,1
 880 ,,,2,2,2
 881 ,,,3,3,3
 882 ,,,4,4,4
 883 ,,,5,5,5
 884 ,2,Highest,1,18,90
 885 ,,,2,17,80
 886 ,,,3,16,70
 887 ,,,4,15,60
 888 ,,,5,14,50
 889 ,,Lowest,1,10,10
 890 ,,,2,11,20
 891 ,,,3,12,30
 892 ,,,4,13,40
 893 ,,,5,14,50
 894 ,9[a],Highest,1,22,401
 895 ,,,2,21,301
 896 ,,,3,20,201
 897 ,,,4,19,101
 898 ,,,5,0,0
 899 ,,Lowest,1,19,101
 900 ,,,2,20,201
 901 ,,,3,21,301
 902 ,,,4,22,401
 903 ,,,5,22,401
 904 ,99[a],Highest,1,27,901
 905 ,,,2,26,801
 906 ,,,3,25,701
 907 ,,,4,24,601
 908 ,,,5,23,501
 909 ,,Lowest,1,23,501
 910 ,,,2,24,601
 911 ,,,3,25,701
 912 ,,,4,26,801
 913 ,,,5,27,901
 914 Footnote: a. User-missing value.
 915 ]])
 916
 917 AT_CLEANUP
 918
 919
 920 dnl Run a test of the basic STATISTICS using a "real"
 921 dnl dataset and comparing with "real" results kindly
 922 dnl provided by Olaf Nöhring
 923 AT_SETUP([EXAMINE -- sample unweighted])
 924 AT_KEYWORDS([categorical categoricals])
 925
 926 AT_DATA([sample.sps], [dnl
 927 set format = F22.4.
 928 DATA LIST notable LIST /X *
 929 BEGIN DATA.
 930 461.19000000
 931 466.38000000
 932 479.46000000
 933 480.10000000
 934 483.43000000
 935 488.30000000
 936 489.00000000
 937 491.62000000
 938 505.62000000
 939 511.30000000
 940 521.53000000
 941 526.70000000
 942 528.25000000
 943 538.70000000
 944 540.22000000
 945 540.58000000
 946 546.10000000
 947 548.17000000
 948 553.99000000
 949 566.21000000
 950 575.90000000
 951 584.38000000
 952 593.40000000
 953 357.05000000
 954 359.73000000
 955 360.48000000
 956 373.98000000
 957 374.13000000
 958 381.45000000
 959 383.72000000
 960 390.00000000
 961 400.34000000
 962 415.32000000
 963 415.91000000
 964 418.30000000
 965 421.03000000
 966 422.43000000
 967 426.93000000
 968 433.25000000
 969 436.89000000
 970 445.33000000
 971 446.33000000
 972 446.55000000
 973 456.44000000
 974 689.49000000
 975 691.92000000
 976 695.00000000
 977 695.36000000
 978 698.21000000
 979 699.46000000
 980 706.61000000
 981 710.69000000
 982 715.82000000
 983 715.82000000
 984 741.39000000
 985 752.27000000
 986 756.73000000
 987 757.74000000
 988 759.57000000
 989 796.07000000
 990 813.78000000
 991 817.25000000
 992 825.48000000
 993 831.28000000
 994 849.24000000
 995 890.00000000
 996 894.78000000
 997 935.65000000
 998 935.90000000
 999 945.90000000
1000 1012.8600000
1001 1022.6000000
1002 1061.8100000
1003 1063.5000000
1004 1077.2300000
1005 1151.6300000
1006 1355.2800000
1007 598.88000000
1008 606.91000000
1009 621.60000000
1010 624.80000000
1011 636.13000000
1012 637.38000000
1013 640.32000000
1014 649.35000000
1015 656.51000000
1016 662.55000000
1017 664.69000000
1018 106.22000000
1019 132.24000000
1020 174.76000000
1021 204.85000000
1022 264.93000000
1023 264.99000000
1024 269.84000000
1025 325.12000000
1026 331.67000000
1027 337.26000000
1028 347.68000000
1029 354.91000000
1030 END DATA.
1031
1032 EXAMINE
1033         x
1034         /STATISTICS=DESCRIPTIVES
1035         .
1036 ])
1037
1038 AT_CHECK([pspp -O format=csv sample.sps], [0], [dnl
1039 Table: Case Processing Summary
1040 ,Cases,,,,,
1041 ,Valid,,Missing,,Total,
1042 ,N,Percent,N,Percent,N,Percent
1043 X,100,100.0%,0,.0%,100,100.0%
1044
1045 Table: Descriptives
1046 ,,,Statistic,Std. Error
1047 X,Mean,,587.6603,23.2665
1048 ,95% Confidence Interval for Mean,Lower Bound,541.4946,
1049 ,,Upper Bound,633.8260,
1050 ,5% Trimmed Mean,,579.7064,
1051 ,Median,,547.1350,
1052 ,Variance,,54132.8466,
1053 ,Std. Deviation,,232.6647,
1054 ,Minimum,,106.2200,
1055 ,Maximum,,1355.2800,
1056 ,Range,,1249.0600,
1057 ,Interquartile Range,,293.1575,
1058 ,Skewness,,.6331,.2414
1059 ,Kurtosis,,.5300,.4783
1060 ])
1061
1062 AT_CLEANUP
1063
1064
1065
1066 dnl Test for a crash which happened on bad input syntax
1067 AT_SETUP([EXAMINE -- Empty Parentheses])
1068 AT_KEYWORDS([categorical categoricals])
1069
1070 AT_DATA([examine-empty-parens.sps], [dnl
1071 DATA LIST notable LIST /X *
1072 BEGIN DATA.
1073 2
1074 3
1075 END DATA.
1076
1077
1078 EXAMINE
1079         x
1080         /PLOT = SPREADLEVEL()
1081         .
1082 ])
1083
1084 AT_CHECK([pspp -o pspp.csv examine-empty-parens.sps], [1], [ignore])
1085
1086 AT_CLEANUP
1087
1088
1089
1090
1091 dnl Test for another crash which happened on bad input syntax
1092 AT_SETUP([EXAMINE -- Bad variable])
1093 AT_KEYWORDS([categorical categoricals])
1094
1095 AT_DATA([examine-bad-variable.sps], [dnl
1096 data list list /h * g *.
1097 begin data.
1098 3 1
1099 4 1
1100 5 2
1101 end data.
1102
1103 EXAMINE
1104         /VARIABLES/ h
1105         BY  g
1106         .
1107 ])
1108
1109 AT_CHECK([pspp -o pspp.csv examine-bad-variable.sps], [1], [ignore])
1110
1111 AT_CLEANUP
1112
1113
1114
1115 dnl Test for yet another crash. This time for extremes vs. missing weight values.\0
1116 AT_SETUP([EXAMINE -- Extremes vs. Missing Weights])
1117 AT_KEYWORDS([categorical categoricals])
1118
1119 AT_DATA([examine-missing-weights.sps], [dnl
1120 data list notable list /h * g *.
1121 begin data.
1122 3 1
1123 4 .
1124 5 1
1125 2 1
1126 end data.
1127
1128 WEIGHT BY g.
1129
1130 EXAMINE h
1131         /STATISTICS extreme(3)
1132         .
1133 ])
1134
1135 AT_CHECK([pspp -O format=csv  examine-missing-weights.sps], [0], [dnl
1136 "examine-missing-weights.sps:13: warning: EXAMINE: At least one case in the data file had a weight value that was user-missing, system-missing, zero, or negative.  These case(s) were ignored."
1137
1138 Table: Case Processing Summary
1139 ,Cases,,,,,
1140 ,Valid,,Missing,,Total,
1141 ,N,Percent,N,Percent,N,Percent
1142 h,3.00,100.0%,.00,.0%,3.00,100.0%
1143
1144 Table: Extreme Values
1145 ,,,Case Number,Value
1146 h,Highest,1,3,5.00
1147 ,,2,2,4.00
1148 ,,3,1,3.00
1149 ,Lowest,1,4,2.00
1150 ,,2,1,3.00
1151 ,,3,2,4.00
1152 ])
1153
1154 AT_CLEANUP
1155
1156 dnl This is an example from doc/tutorial.texi
1157 dnl So if the results of this have to be changed in any way,
1158 dnl make sure to update that file.
1159 AT_SETUP([EXAMINE tutorial example 1])
1160 cp $top_srcdir/examples/repairs.sav .
1161 AT_DATA([repairs.sps], [dnl
1162 GET FILE='repairs.sav'.
1163 EXAMINE mtbf /STATISTICS=DESCRIPTIVES.
1164 COMPUTE mtbf_ln = LN (mtbf).
1165 EXAMINE mtbf_ln /STATISTICS=DESCRIPTIVES.
1166 ])
1167
1168 AT_CHECK([pspp -O format=csv repairs.sps], [0], [dnl
1169 Table: Case Processing Summary
1170 ,Cases,,,,,
1171 ,Valid,,Missing,,Total,
1172 ,N,Percent,N,Percent,N,Percent
1173 Mean time between failures (months) ,30,100.0%,0,.0%,30,100.0%
1174
1175 Table: Descriptives
1176 ,,,Statistic,Std. Error
1177 Mean time between failures (months) ,Mean,,8.78,1.10
1178 ,95% Confidence Interval for Mean,Lower Bound,6.53,
1179 ,,Upper Bound,11.04,
1180 ,5% Trimmed Mean,,8.20,
1181 ,Median,,8.29,
1182 ,Variance,,36.34,
1183 ,Std. Deviation,,6.03,
1184 ,Minimum,,1.63,
1185 ,Maximum,,26.47,
1186 ,Range,,24.84,
1187 ,Interquartile Range,,6.03,
1188 ,Skewness,,1.65,.43
1189 ,Kurtosis,,3.41,.83
1190
1191 Table: Case Processing Summary
1192 ,Cases,,,,,
1193 ,Valid,,Missing,,Total,
1194 ,N,Percent,N,Percent,N,Percent
1195 mtbf_ln,30,100.0%,0,.0%,30,100.0%
1196
1197 Table: Descriptives
1198 ,,,Statistic,Std. Error
1199 mtbf_ln,Mean,,1.95,.13
1200 ,95% Confidence Interval for Mean,Lower Bound,1.69,
1201 ,,Upper Bound,2.22,
1202 ,5% Trimmed Mean,,1.96,
1203 ,Median,,2.11,
1204 ,Variance,,.49,
1205 ,Std. Deviation,,.70,
1206 ,Minimum,,.49,
1207 ,Maximum,,3.28,
1208 ,Range,,2.79,
1209 ,Interquartile Range,,.88,
1210 ,Skewness,,-.37,.43
1211 ,Kurtosis,,.01,.83
1212 ])
1213
1214 AT_CLEANUP
1215
1216 dnl This is an example from doc/tutorial.texi
1217 dnl So if the results of this have to be changed in any way,
1218 dnl make sure to update that file.
1219 AT_SETUP([EXAMINE tutorial example 2])
1220 cp $top_srcdir/examples/physiology.sav .
1221 AT_DATA([examine.sps], [dnl
1222 GET FILE='physiology.sav'.
1223 EXAMINE height, weight /STATISTICS=EXTREME(3).
1224 ])
1225 AT_CHECK([pspp -o pspp.csv -o pspp.txt examine.sps])
1226 AT_CHECK([cat pspp.csv], [0], [dnl
1227 Table: Case Processing Summary
1228 ,Cases,,,,,
1229 ,Valid,,Missing,,Total,
1230 ,N,Percent,N,Percent,N,Percent
1231 Height in millimeters   ,40,100.0%,0,.0%,40,100.0%
1232 Weight in kilograms ,40,100.0%,0,.0%,40,100.0%
1233
1234 Table: Extreme Values
1235 ,,,Case Number,Value
1236 Height in millimeters   ,Highest,1,14,1903
1237 ,,2,15,1884
1238 ,,3,12,1802
1239 ,Lowest,1,30,179
1240 ,,2,31,1598
1241 ,,3,28,1601
1242 Weight in kilograms ,Highest,1,13,92.1
1243 ,,2,5,92.1
1244 ,,3,17,91.7
1245 ,Lowest,1,38,-55.6
1246 ,,2,39,54.5
1247 ,,3,33,55.4
1248 ])
1249 AT_CLEANUP
1250
1251
1252
1253 AT_SETUP([EXAMINE -- Crash on unrepresentable graphs])
1254 AT_DATA([examine.sps], [dnl
1255 data list notable list /x * g *.
1256 begin data.
1257 96 1
1258 end data.
1259
1260 examine x  by g
1261         /nototal
1262         /plot = all.
1263 ])
1264 dnl This bug only manifested itself on cairo based drivers.
1265 AT_CHECK([pspp -O format=pdf examine.sps], [0], [ignore], [ignore])
1266 AT_CLEANUP
1267
1268
1269 dnl This example comes from the web site:
1270 dnl  https://www.spsstests.com/2018/11/shapiro-wilk-normality-test-spss.html
1271 AT_SETUP([EXAMINE -- shapiro-wilk 1])
1272 AT_KEYWORDS([shapiro wilk])
1273 AT_DATA([shapiro-wilk.sps], [dnl
1274 data list notable list /x * g *.
1275 begin data.
1276 96 1
1277 98 1
1278 95 1
1279 89 1
1280 90 1
1281 92 1
1282 94 1
1283 93 1
1284 97 1
1285 100 1
1286 99 2
1287 96 2
1288 80 2
1289 89 2
1290 91 2
1291 92 2
1292 93 2
1293 94 2
1294 99 2
1295 80 2
1296 end data.
1297
1298 set format F22.3.
1299
1300 examine x  by g
1301         /nototal
1302         /plot = all.
1303 ])
1304
1305 AT_CHECK([pspp -O format=csv shapiro-wilk.sps], [0],[dnl
1306 Table: Case Processing Summary
1307 ,g,Cases,,,,,
1308 ,,Valid,,Missing,,Total,
1309 ,,N,Percent,N,Percent,N,Percent
1310 x,1.00,10,100.0%,0,.0%,10,100.0%
1311 ,2.00,10,100.0%,0,.0%,10,100.0%
1312
1313 Table: Tests of Normality
1314 ,g,Shapiro-Wilk,,
1315 ,,Statistic,df,Sig.
1316 x,1.00,.984,10,.983
1317 ,2.00,.882,10,.136
1318 ])
1319
1320 AT_CLEANUP
1321
1322
1323 dnl This example comes from the web site:
1324 dnl  http://www.real-statistics.com/tests-normality-and-symmetry/statistical-tests-normality-symmetry/shapiro-wilk-expanded-test/
1325 dnl It uses a dataset larger than 11 samples. Hence the alternative method for
1326 dnl signficance is used.
1327 AT_SETUP([EXAMINE -- shapiro-wilk 2])
1328 AT_KEYWORDS([shapiro wilk])
1329 AT_DATA([shapiro-wilk2.sps], [dnl
1330 data list notable list /x *.
1331 begin data.
1332 65
1333 61
1334 63
1335 86
1336 70
1337 55
1338 74
1339 35
1340 72
1341 68
1342 45
1343 58
1344 end data.
1345
1346 set format F22.3.
1347
1348 examine x
1349         /plot = boxplot.
1350 ])
1351
1352 AT_CHECK([pspp -O format=csv shapiro-wilk2.sps], [0],[dnl
1353 Table: Case Processing Summary
1354 ,Cases,,,,,
1355 ,Valid,,Missing,,Total,
1356 ,N,Percent,N,Percent,N,Percent
1357 x,12,100.0%,0,.0%,12,100.0%
1358
1359 Table: Tests of Normality
1360 ,Shapiro-Wilk,,
1361 ,Statistic,df,Sig.
1362 x,.971,12,.922
1363 ])
1364
1365 AT_CLEANUP