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 ,,Cases,,,,,
 119 ,,Valid,,Missing,,Total,
 120 ,Manufacturer,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 ,,Cases,,,,,
 455 ,,Valid,,Missing,,Total,
 456 ,y,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_DATA([make-big-input.pl],
 682   [for ($i=0; $i<100000; $i++) { print "AB12\n" };
 683    for ($i=0; $i<100000; $i++) { print "AB04\n" };
 684 ])
 685 AT_CHECK([$PERL make-big-input.pl > large.txt])
 686 AT_DATA([examine.sps], [dnl
 687 DATA LIST FILE='large.txt' /S 1-2 (A) X 3 .
 688
 689
 690 AGGREGATE OUTFILE=* /BREAK=X /A=N.
 691
 692
 693 EXAMINE /A BY X.
 694 ])
 695 AT_CHECK([pspp -o pspp.csv examine.sps])
 696 dnl Ignore output -- this is just a no-crash check.
 697 AT_DATA([more-big-input.pl],
 698   [for ($i=0; $i<25000; $i++) { print "AB04\nAB12\n" };
 699 ])
 700 AT_CHECK([$PERL more-big-input.pl >> large.txt])
 701 AT_CHECK([pspp -o pspp.csv examine.sps])
 702 dnl Ignore output -- this is just a no-crash check.
 703 AT_CLEANUP
 704
 705
 706 dnl Test that the ID command works with non-numberic variables
 707 AT_SETUP([EXAMINE -- non-numeric ID])
 708 AT_KEYWORDS([categorical categoricals])
 709
 710 AT_DATA([examine-id.sps], [dnl
 711 data list notable list /x * y (a12).
 712 begin data.
 713 1  one
 714 2  two
 715 3  three
 716 4  four
 717 5  five
 718 6  six
 719 7  seven
 720 8  eight
 721 9  nine
 722 10 ten
 723 11 eleven
 724 12 twelve
 725 30 thirty
 726 300 threehundred
 727 end data.
 728
 729 examine x
 730         /statistics = extreme
 731         /id = y
 732         /plot = boxplot
 733         .
 734 ])
 735
 736 AT_CHECK([pspp -O format=csv examine-id.sps], [0], [dnl
 737 Table: Case Processing Summary
 738 ,Cases,,,,,
 739 ,Valid,,Missing,,Total,
 740 ,N,Percent,N,Percent,N,Percent
 741 x,14,100.0%,0,.0%,14,100.0%
 742
 743 Table: Extreme Values
 744 ,,,y,Value
 745 x,Highest,1,threehundred,300.00
 746 ,,2,thirty,30.00
 747 ,,3,twelve,12.00
 748 ,,4,eleven,11.00
 749 ,,5,ten,10.00
 750 ,Lowest,1,one,1.00
 751 ,,2,two,2.00
 752 ,,3,three,3.00
 753 ,,4,four,4.00
 754 ,,5,five,5.00
 755
 756 Table: Tests of Normality
 757 ,Shapiro-Wilk,,
 758 ,Statistic,df,Sig.
 759 x,.37,14,.00
 760 ])
 761
 762 AT_CLEANUP
 763
 764 dnl Test for a crash which happened on cleanup from a bad input syntax
 765 AT_SETUP([EXAMINE -- Bad Input])
 766 AT_KEYWORDS([categorical categoricals])
 767
 768 AT_DATA([examine-bad.sps], [dnl
 769 data list list /h * g *.
 770 begin data.
 771 1 1
 772 2 1
 773 3 1
 774 4 1
 775 5 2
 776 6 2
 777 7 2
 778 8 2
 779 9 2
 780 end data.
 781
 782 EXAMINE
 783         /VARIABLES= h
 784         BY  g
 785         /STATISTICS = DESCRIPTIVES EXTREME
 786         /PLOT = lkajsdas
 787         .
 788 ])
 789
 790 AT_CHECK([pspp -o pspp.csv examine-bad.sps], [1], [ignore])
 791
 792 AT_CLEANUP
 793
 794
 795 dnl Check the MISSING=REPORT option
 796 AT_SETUP([EXAMINE -- MISSING=REPORT])
 797 AT_KEYWORDS([categorical categoricals])
 798
 799 AT_DATA([examine-report.sps], [dnl
 800 set format = F22.0.
 801 data list list /x * g *.
 802 begin data.
 803 1   1
 804 2   1
 805 3   1
 806 4   1
 807 5   1
 808 6   1
 809 7   1
 810 8   1
 811 9   1
 812 10   2
 813 20   2
 814 30   2
 815 40   2
 816 50   2
 817 60   2
 818 70   2
 819 80   2
 820 90   2
 821 101   9
 822 201   9
 823 301   9
 824 401   9
 825 501   99
 826 601   99
 827 701   99
 828 801   99
 829 901   99
 830 1001  .
 831 2002  .
 832 3003  .
 833 4004  .
 834 end data.
 835
 836 MISSING VALUES g (9, 99, 999).
 837
 838 EXAMINE
 839         /VARIABLES = x
 840         BY  g
 841         /STATISTICS = EXTREME
 842         /NOTOTAL
 843         /MISSING = REPORT.
 844 ])
 845
 846
 847 AT_CHECK([pspp -o pspp.csv -o pspp.txt examine-report.sps])
 848 AT_CHECK([cat pspp.csv], [0],
 849   [[Table: Reading free-form data from INLINE.
 850 Variable,Format
 851 x,F8.0
 852 g,F8.0
 853
 854 Table: Case Processing Summary
 855 ,,Cases,,,,,
 856 ,,Valid,,Missing,,Total,
 857 ,g,N,Percent,N,Percent,N,Percent
 858 x,.,4,100.0%,0,.0%,4,100.0%
 859 ,1,9,100.0%,0,.0%,9,100.0%
 860 ,2,9,100.0%,0,.0%,9,100.0%
 861 ,9[a],4,100.0%,0,.0%,4,100.0%
 862 ,99[a],5,100.0%,0,.0%,5,100.0%
 863
 864 Footnotes:
 865 a,User-missing value.
 866
 867 Table: Extreme Values
 868 ,g,,,Case Number,Value
 869 x,.,Highest,1,31,4004
 870 ,,,2,30,3003
 871 ,,,3,29,2002
 872 ,,,4,28,1001
 873 ,,,5,0,0
 874 ,,Lowest,1,28,1001
 875 ,,,2,29,2002
 876 ,,,3,30,3003
 877 ,,,4,31,4004
 878 ,,,5,31,4004
 879 ,1,Highest,1,9,9
 880 ,,,2,8,8
 881 ,,,3,7,7
 882 ,,,4,6,6
 883 ,,,5,5,5
 884 ,,Lowest,1,1,1
 885 ,,,2,2,2
 886 ,,,3,3,3
 887 ,,,4,4,4
 888 ,,,5,5,5
 889 ,2,Highest,1,18,90
 890 ,,,2,17,80
 891 ,,,3,16,70
 892 ,,,4,15,60
 893 ,,,5,14,50
 894 ,,Lowest,1,10,10
 895 ,,,2,11,20
 896 ,,,3,12,30
 897 ,,,4,13,40
 898 ,,,5,14,50
 899 ,9[a],Highest,1,22,401
 900 ,,,2,21,301
 901 ,,,3,20,201
 902 ,,,4,19,101
 903 ,,,5,0,0
 904 ,,Lowest,1,19,101
 905 ,,,2,20,201
 906 ,,,3,21,301
 907 ,,,4,22,401
 908 ,,,5,22,401
 909 ,99[a],Highest,1,27,901
 910 ,,,2,26,801
 911 ,,,3,25,701
 912 ,,,4,24,601
 913 ,,,5,23,501
 914 ,,Lowest,1,23,501
 915 ,,,2,24,601
 916 ,,,3,25,701
 917 ,,,4,26,801
 918 ,,,5,27,901
 919
 920 Footnotes:
 921 a,User-missing value.
 922 ]])
 923
 924 AT_CLEANUP
 925
 926
 927 dnl Run a test of the basic STATISTICS using a "real"
 928 dnl dataset and comparing with "real" results kindly
 929 dnl provided by Olaf Nöhring
 930 AT_SETUP([EXAMINE -- sample unweighted])
 931 AT_KEYWORDS([categorical categoricals])
 932
 933 AT_DATA([sample.sps], [dnl
 934 set format = F22.4.
 935 DATA LIST notable LIST /X *
 936 BEGIN DATA.
 937 461.19000000
 938 466.38000000
 939 479.46000000
 940 480.10000000
 941 483.43000000
 942 488.30000000
 943 489.00000000
 944 491.62000000
 945 505.62000000
 946 511.30000000
 947 521.53000000
 948 526.70000000
 949 528.25000000
 950 538.70000000
 951 540.22000000
 952 540.58000000
 953 546.10000000
 954 548.17000000
 955 553.99000000
 956 566.21000000
 957 575.90000000
 958 584.38000000
 959 593.40000000
 960 357.05000000
 961 359.73000000
 962 360.48000000
 963 373.98000000
 964 374.13000000
 965 381.45000000
 966 383.72000000
 967 390.00000000
 968 400.34000000
 969 415.32000000
 970 415.91000000
 971 418.30000000
 972 421.03000000
 973 422.43000000
 974 426.93000000
 975 433.25000000
 976 436.89000000
 977 445.33000000
 978 446.33000000
 979 446.55000000
 980 456.44000000
 981 689.49000000
 982 691.92000000
 983 695.00000000
 984 695.36000000
 985 698.21000000
 986 699.46000000
 987 706.61000000
 988 710.69000000
 989 715.82000000
 990 715.82000000
 991 741.39000000
 992 752.27000000
 993 756.73000000
 994 757.74000000
 995 759.57000000
 996 796.07000000
 997 813.78000000
 998 817.25000000
 999 825.48000000
1000 831.28000000
1001 849.24000000
1002 890.00000000
1003 894.78000000
1004 935.65000000
1005 935.90000000
1006 945.90000000
1007 1012.8600000
1008 1022.6000000
1009 1061.8100000
1010 1063.5000000
1011 1077.2300000
1012 1151.6300000
1013 1355.2800000
1014 598.88000000
1015 606.91000000
1016 621.60000000
1017 624.80000000
1018 636.13000000
1019 637.38000000
1020 640.32000000
1021 649.35000000
1022 656.51000000
1023 662.55000000
1024 664.69000000
1025 106.22000000
1026 132.24000000
1027 174.76000000
1028 204.85000000
1029 264.93000000
1030 264.99000000
1031 269.84000000
1032 325.12000000
1033 331.67000000
1034 337.26000000
1035 347.68000000
1036 354.91000000
1037 END DATA.
1038
1039 EXAMINE
1040         x
1041         /STATISTICS=DESCRIPTIVES
1042         .
1043 ])
1044
1045 AT_CHECK([pspp -O format=csv sample.sps], [0], [dnl
1046 Table: Case Processing Summary
1047 ,Cases,,,,,
1048 ,Valid,,Missing,,Total,
1049 ,N,Percent,N,Percent,N,Percent
1050 X,100,100.0%,0,.0%,100,100.0%
1051
1052 Table: Descriptives
1053 ,,,Statistic,Std. Error
1054 X,Mean,,587.6603,23.2665
1055 ,95% Confidence Interval for Mean,Lower Bound,541.4946,
1056 ,,Upper Bound,633.8260,
1057 ,5% Trimmed Mean,,579.7064,
1058 ,Median,,547.1350,
1059 ,Variance,,54132.8466,
1060 ,Std. Deviation,,232.6647,
1061 ,Minimum,,106.2200,
1062 ,Maximum,,1355.2800,
1063 ,Range,,1249.0600,
1064 ,Interquartile Range,,293.1575,
1065 ,Skewness,,.6331,.2414
1066 ,Kurtosis,,.5300,.4783
1067 ])
1068
1069 AT_CLEANUP
1070
1071
1072
1073 dnl Test for a crash which happened on bad input syntax
1074 AT_SETUP([EXAMINE -- Empty Parentheses])
1075 AT_KEYWORDS([categorical categoricals])
1076
1077 AT_DATA([examine-empty-parens.sps], [dnl
1078 DATA LIST notable LIST /X *
1079 BEGIN DATA.
1080 2
1081 3
1082 END DATA.
1083
1084
1085 EXAMINE
1086         x
1087         /PLOT = SPREADLEVEL()
1088         .
1089 ])
1090
1091 AT_CHECK([pspp -o pspp.csv examine-empty-parens.sps], [1], [ignore])
1092
1093 AT_CLEANUP
1094
1095
1096
1097
1098 dnl Test for another crash which happened on bad input syntax
1099 AT_SETUP([EXAMINE -- Bad variable])
1100 AT_KEYWORDS([categorical categoricals])
1101
1102 AT_DATA([examine-bad-variable.sps], [dnl
1103 data list list /h * g *.
1104 begin data.
1105 3 1
1106 4 1
1107 5 2
1108 end data.
1109
1110 EXAMINE
1111         /VARIABLES/ h
1112         BY  g
1113         .
1114 ])
1115
1116 AT_CHECK([pspp -o pspp.csv examine-bad-variable.sps], [1], [ignore])
1117
1118 AT_CLEANUP
1119
1120
1121
1122 dnl Test for yet another crash. This time for extremes vs. missing weight values.\0
1123 AT_SETUP([EXAMINE -- Extremes vs. Missing Weights])
1124 AT_KEYWORDS([categorical categoricals])
1125
1126 AT_DATA([examine-missing-weights.sps], [dnl
1127 data list notable list /h * g *.
1128 begin data.
1129 3 1
1130 4 .
1131 5 1
1132 2 1
1133 end data.
1134
1135 WEIGHT BY g.
1136
1137 EXAMINE h
1138         /STATISTICS extreme(3)
1139         .
1140 ])
1141
1142 AT_CHECK([pspp -O format=csv  examine-missing-weights.sps], [0], [dnl
1143 "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."
1144
1145 Table: Case Processing Summary
1146 ,Cases,,,,,
1147 ,Valid,,Missing,,Total,
1148 ,N,Percent,N,Percent,N,Percent
1149 h,3.00,100.0%,.00,.0%,3.00,100.0%
1150
1151 Table: Extreme Values
1152 ,,,Case Number,Value
1153 h,Highest,1,3,5.00
1154 ,,2,2,4.00
1155 ,,3,1,3.00
1156 ,Lowest,1,4,2.00
1157 ,,2,1,3.00
1158 ,,3,2,4.00
1159 ])
1160
1161 AT_CLEANUP
1162
1163 dnl This is an example from doc/tutorial.texi
1164 dnl So if the results of this have to be changed in any way,
1165 dnl make sure to update that file.
1166 AT_SETUP([EXAMINE tutorial example 1])
1167 cp $top_srcdir/examples/repairs.sav .
1168 AT_DATA([repairs.sps], [dnl
1169 GET FILE='repairs.sav'.
1170 EXAMINE mtbf /STATISTICS=DESCRIPTIVES.
1171 COMPUTE mtbf_ln = LN (mtbf).
1172 EXAMINE mtbf_ln /STATISTICS=DESCRIPTIVES.
1173 ])
1174 AT_CHECK([pspp -o pspp.csv -o pspp.txt repairs.sps])
1175 AT_CHECK([cat pspp.csv], [0], [dnl
1176 Table: Case Processing Summary
1177 ,Cases,,,,,
1178 ,Valid,,Missing,,Total,
1179 ,N,Percent,N,Percent,N,Percent
1180 Mean time between failures (months) ,15,100.0%,0,.0%,15,100.0%
1181
1182 Table: Descriptives
1183 ,,,Statistic,Std. Error
1184 Mean time between failures (months) ,Mean,,8.32,1.62
1185 ,95% Confidence Interval for Mean,Lower Bound,4.85,
1186 ,,Upper Bound,11.79,
1187 ,5% Trimmed Mean,,7.69,
1188 ,Median,,8.12,
1189 ,Variance,,39.21,
1190 ,Std. Deviation,,6.26,
1191 ,Minimum,,1.63,
1192 ,Maximum,,26.47,
1193 ,Range,,24.84,
1194 ,Interquartile Range,,5.83,
1195 ,Skewness,,1.85,.58
1196 ,Kurtosis,,4.49,1.12
1197
1198 Table: Case Processing Summary
1199 ,Cases,,,,,
1200 ,Valid,,Missing,,Total,
1201 ,N,Percent,N,Percent,N,Percent
1202 mtbf_ln,15,100.0%,0,.0%,15,100.0%
1203
1204 Table: Descriptives
1205 ,,,Statistic,Std. Error
1206 mtbf_ln,Mean,,1.88,.19
1207 ,95% Confidence Interval for Mean,Lower Bound,1.47,
1208 ,,Upper Bound,2.29,
1209 ,5% Trimmed Mean,,1.88,
1210 ,Median,,2.09,
1211 ,Variance,,.54,
1212 ,Std. Deviation,,.74,
1213 ,Minimum,,.49,
1214 ,Maximum,,3.28,
1215 ,Range,,2.79,
1216 ,Interquartile Range,,.92,
1217 ,Skewness,,-.16,.58
1218 ,Kurtosis,,-.09,1.12
1219 ])
1220 AT_CLEANUP
1221
1222 dnl This is an example from doc/tutorial.texi
1223 dnl So if the results of this have to be changed in any way,
1224 dnl make sure to update that file.
1225 AT_SETUP([EXAMINE tutorial example 2])
1226 cp $top_srcdir/examples/physiology.sav .
1227 AT_DATA([examine.sps], [dnl
1228 GET FILE='physiology.sav'.
1229 EXAMINE height, weight /STATISTICS=EXTREME(3).
1230 ])
1231 AT_CHECK([pspp -o pspp.csv -o pspp.txt examine.sps])
1232 AT_CHECK([cat pspp.csv], [0], [dnl
1233 Table: Case Processing Summary
1234 ,Cases,,,,,
1235 ,Valid,,Missing,,Total,
1236 ,N,Percent,N,Percent,N,Percent
1237 Height in millimeters   ,40,100.0%,0,.0%,40,100.0%
1238 Weight in kilograms ,40,100.0%,0,.0%,40,100.0%
1239
1240 Table: Extreme Values
1241 ,,,Case Number,Value
1242 Height in millimeters   ,Highest,1,14,1903
1243 ,,2,15,1884
1244 ,,3,12,1802
1245 ,Lowest,1,30,179
1246 ,,2,31,1598
1247 ,,3,28,1601
1248 Weight in kilograms ,Highest,1,13,92.1
1249 ,,2,5,92.1
1250 ,,3,17,91.7
1251 ,Lowest,1,38,-55.6
1252 ,,2,39,54.5
1253 ,,3,33,55.4
1254 ])
1255 AT_CLEANUP
1256
1257
1258
1259 AT_SETUP([EXAMINE -- Crash on unrepresentable graphs])
1260 AT_DATA([examine.sps], [dnl
1261 data list notable list /x * g *.
1262 begin data.
1263 96 1
1264 end data.
1265
1266 examine x  by g
1267         /nototal
1268         /plot = all.
1269 ])
1270 dnl This bug only manifested itself on cairo based drivers.
1271 AT_CHECK([pspp -O format=pdf examine.sps], [1], [ignore])
1272 AT_CLEANUP
1273
1274
1275 dnl This example comes from the web site:
1276 dnl  https://www.spsstests.com/2018/11/shapiro-wilk-normality-test-spss.html
1277 AT_SETUP([EXAMINE -- shapiro-wilk 1])
1278 AT_KEYWORDS([shapiro wilk])
1279 AT_DATA([shapiro-wilk.sps], [dnl
1280 data list notable list /x * g *.
1281 begin data.
1282 96 1
1283 98 1
1284 95 1
1285 89 1
1286 90 1
1287 92 1
1288 94 1
1289 93 1
1290 97 1
1291 100 1
1292 99 2
1293 96 2
1294 80 2
1295 89 2
1296 91 2
1297 92 2
1298 93 2
1299 94 2
1300 99 2
1301 80 2
1302 end data.
1303
1304 set format F22.3.
1305
1306 examine x  by g
1307         /nototal
1308         /plot = all.
1309 ])
1310
1311 AT_CHECK([pspp -O format=csv shapiro-wilk.sps], [0],[dnl
1312 Table: Case Processing Summary
1313 ,,Cases,,,,,
1314 ,,Valid,,Missing,,Total,
1315 ,g,N,Percent,N,Percent,N,Percent
1316 x,1.00,10,100.0%,0,.0%,10,100.0%
1317 ,2.00,10,100.0%,0,.0%,10,100.0%
1318
1319 Table: Tests of Normality
1320 ,,Shapiro-Wilk,,
1321 ,g,Statistic,df,Sig.
1322 x,1.00,.984,10,.983
1323 ,2.00,.882,10,.136
1324 ])
1325
1326 AT_CLEANUP
1327
1328
1329 dnl This example comes from the web site:
1330 dnl  http://www.real-statistics.com/tests-normality-and-symmetry/statistical-tests-normality-symmetry/shapiro-wilk-expanded-test/
1331 dnl It uses a dataset larger than 11 samples. Hence the alternative method for
1332 dnl signficance is used.
1333 AT_SETUP([EXAMINE -- shapiro-wilk 2])
1334 AT_KEYWORDS([shapiro wilk])
1335 AT_DATA([shapiro-wilk2.sps], [dnl
1336 data list notable list /x *.
1337 begin data.
1338 65
1339 61
1340 63
1341 86
1342 70
1343 55
1344 74
1345 35
1346 72
1347 68
1348 45
1349 58
1350 end data.
1351
1352 set format F22.3.
1353
1354 examine x
1355         /plot = boxplot.
1356 ])
1357
1358 AT_CHECK([pspp -O format=csv shapiro-wilk2.sps], [0],[dnl
1359 Table: Case Processing Summary
1360 ,Cases,,,,,
1361 ,Valid,,Missing,,Total,
1362 ,N,Percent,N,Percent,N,Percent
1363 x,12,100.0%,0,.0%,12,100.0%
1364
1365 Table: Tests of Normality
1366 ,Shapiro-Wilk,,
1367 ,Statistic,df,Sig.
1368 x,.971,12,.922
1369 ])
1370
1371 AT_CLEANUP