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

   1 dnl PSPP - a program for statistical analysis.
   2 dnl Copyright (C) 2017 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%,.00,0%,24.00,100%
  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%,.00,0%,8.00,100%
 122 ,Bloggs,8.00,100%,.00,0%,8.00,100%
 123 ,Charlies,8.00,100%,.00,0%,8.00,100%
 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%,.00,0%,23.00,100%
 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
 253 AT_SETUP([EXAMINE -- extremes with fractional weights])
 254 AT_KEYWORDS([categorical categoricals])
 255 AT_DATA([extreme.sps], [dnl
 256 set format=F20.3.
 257 data list notable list /w * x *.
 258 begin data.
 259  0.88  300000
 260  0.86  320000
 261  0.98  480000
 262  0.93  960000
 263  1.35  960000
 264  1.31  960000
 265  0.88  960000
 266  0.88  1080000
 267  0.88  1080000
 268  0.95  1200000
 269  1.47  1200000
 270  0.93  1200000
 271  0.98  1320000
 272  1.31  1380000
 273  0.93  1440000
 274  0.88  1560000
 275  1.56  1560000
 276  1.47  1560000
 277 end data.
 278
 279 weight by w.
 280
 281
 282 EXAMINE
 283         x
 284         /STATISTICS = DESCRIPTIVES EXTREME (5)
 285         .
 286 ])
 287
 288 AT_CHECK([pspp -O format=csv  extreme.sps], [0], [dnl
 289 Table: Case Processing Summary
 290 ,Cases,,,,,
 291 ,Valid,,Missing,,Total,
 292 ,N,Percent,N,Percent,N,Percent
 293 x,19.430,100%,.000,0%,19.430,100%
 294
 295 Table: Extreme Values
 296 ,,,Case Number,Value
 297 x,Highest,1,18,1560000.000
 298 ,,2,17,1560000.000
 299 ,,3,16,1560000.000
 300 ,,4,15,1440000.000
 301 ,,5,14,1380000.000
 302 ,Lowest,1,1,300000.000
 303 ,,2,2,320000.000
 304 ,,3,3,480000.000
 305 ,,4,4,960000.000
 306 ,,5,5,960000.000
 307
 308 Table: Descriptives
 309 ,,,Statistic,Std. Error
 310 x,Mean,,1120010.293,86222.178
 311 ,95% Confidence Interval for Mean,Lower Bound,939166.693,
 312 ,,Upper Bound,1300853.894,
 313 ,5% Trimmed Mean,,1141017.899,
 314 ,Median,,1200000.000,
 315 ,Variance,,144447748124.869,
 316 ,Std. Deviation,,380062.821,
 317 ,Minimum,,300000.000,
 318 ,Maximum,,1560000.000,
 319 ,Range,,1260000.000,
 320 ,Interquartile Range,,467258.065,
 321 ,Skewness,,-.887,.519
 322 ,Kurtosis,,.340,1.005
 323 ])
 324
 325 AT_CLEANUP
 326
 327 dnl Test the PERCENTILES subcommand of the EXAMINE command.
 328 dnl In particular test that it behaves properly when there are only
 329 dnl a few cases.
 330 AT_SETUP([EXAMINE -- percentiles])
 331 AT_KEYWORDS([categorical categoricals])
 332 AT_DATA([examine.sps], [dnl
 333 DATA LIST LIST /X *.
 334 BEGIN DATA.
 335 2.00
 336 8.00
 337 5.00
 338 END DATA.
 339
 340 EXAMINE /x
 341         /PERCENTILES=HAVERAGE.
 342
 343 EXAMINE /x
 344         /PERCENTILES=WAVERAGE.
 345
 346 EXAMINE /x
 347         /PERCENTILES=ROUND.
 348
 349 EXAMINE /x
 350         /PERCENTILES=EMPIRICAL.
 351
 352 EXAMINE /x
 353         /PERCENTILES=AEMPIRICAL.
 354 ])
 355 AT_CHECK([pspp -o pspp.csv examine.sps])
 356 AT_CHECK([cat pspp.csv], [0], [dnl
 357 Table: Reading free-form data from INLINE.
 358 Variable,Format
 359 X,F8.0
 360
 361 Table: Case Processing Summary
 362 ,Cases,,,,,
 363 ,Valid,,Missing,,Total,
 364 ,N,Percent,N,Percent,N,Percent
 365 X,3,100%,0,0%,3,100%
 366
 367 Table: Percentiles
 368 ,,Percentiles,,,,,,
 369 ,,5,10,25,50,75,90,95
 370 X,HAverage,.40,.80,2.00,5.00,8.00,8.00,8.00
 371 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 372
 373 Table: Case Processing Summary
 374 ,Cases,,,,,
 375 ,Valid,,Missing,,Total,
 376 ,N,Percent,N,Percent,N,Percent
 377 X,3,100%,0,0%,3,100%
 378
 379 Table: Percentiles
 380 ,,Percentiles,,,,,,
 381 ,,5,10,25,50,75,90,95
 382 X,Weighted Average,.30,.60,1.50,3.50,5.75,7.10,7.55
 383 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 384
 385 Table: Case Processing Summary
 386 ,Cases,,,,,
 387 ,Valid,,Missing,,Total,
 388 ,N,Percent,N,Percent,N,Percent
 389 X,3,100%,0,0%,3,100%
 390
 391 Table: Percentiles
 392 ,,Percentiles,,,,,,
 393 ,,5,10,25,50,75,90,95
 394 X,Rounded,.00,.00,2.00,5.00,5.00,8.00,8.00
 395 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 396
 397 Table: Case Processing Summary
 398 ,Cases,,,,,
 399 ,Valid,,Missing,,Total,
 400 ,N,Percent,N,Percent,N,Percent
 401 X,3,100%,0,0%,3,100%
 402
 403 Table: Percentiles
 404 ,,Percentiles,,,,,,
 405 ,,5,10,25,50,75,90,95
 406 X,Empirical,2.00,2.00,2.00,5.00,8.00,8.00,8.00
 407 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 408
 409 Table: Case Processing Summary
 410 ,Cases,,,,,
 411 ,Valid,,Missing,,Total,
 412 ,N,Percent,N,Percent,N,Percent
 413 X,3,100%,0,0%,3,100%
 414
 415 Table: Percentiles
 416 ,,Percentiles,,,,,,
 417 ,,5,10,25,50,75,90,95
 418 X,Empirical with averaging,2.00,2.00,2.00,5.00,8.00,8.00,8.00
 419 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 420 ])
 421 AT_CLEANUP
 422
 423 AT_SETUP([EXAMINE -- missing values])
 424 AT_KEYWORDS([categorical categoricals])
 425 AT_DATA([examine.sps], [dnl
 426 DATA LIST LIST /x * y *.
 427 BEGIN DATA.
 428 1   1
 429 2   1
 430 3   1
 431 4   1
 432 5   2
 433 6   2
 434 .   2
 435 END DATA
 436
 437 EXAMINE /x by y
 438         /MISSING = PAIRWISE
 439         .
 440 ])
 441 AT_CHECK([pspp -o pspp.csv examine.sps])
 442 AT_CHECK([cat pspp.csv], [0], [dnl
 443 Table: Reading free-form data from INLINE.
 444 Variable,Format
 445 x,F8.0
 446 y,F8.0
 447
 448 Table: Case Processing Summary
 449 ,Cases,,,,,
 450 ,Valid,,Missing,,Total,
 451 ,N,Percent,N,Percent,N,Percent
 452 x,6,85.7143%,1,14.2857%,7,100%
 453
 454 Table: Case Processing Summary
 455 ,,Cases,,,,,
 456 ,,Valid,,Missing,,Total,
 457 ,y,N,Percent,N,Percent,N,Percent
 458 x,1.00,4,100%,0,0%,4,100%
 459 ,2.00,2,66.6667%,1,33.3333%,3,100%
 460 ])
 461 AT_CLEANUP
 462
 463
 464 AT_SETUP([EXAMINE -- user missing values])
 465 AT_KEYWORDS([categorical categoricals])
 466 AT_DATA([examine-m.sps], [dnl
 467 DATA LIST notable LIST /x * y *.
 468 BEGIN DATA.
 469 1                   2
 470 9999999999          2
 471 9999999999          99
 472 END DATA.
 473
 474 MISSING VALUES x (9999999999).
 475 MISSING VALUES y (99).
 476
 477 EXAMINE
 478         /VARIABLES= x y
 479         /MISSING=PAIRWISE.
 480 ])
 481 AT_CHECK([pspp -O format=csv examine-m.sps], [0], [dnl
 482 Table: Case Processing Summary
 483 ,Cases,,,,,
 484 ,Valid,,Missing,,Total,
 485 ,N,Percent,N,Percent,N,Percent
 486 x,1,33.3333%,2,66.6667%,3,100%
 487 y,2,66.6667%,1,33.3333%,3,100%
 488 ])
 489 AT_CLEANUP
 490
 491 AT_SETUP([EXAMINE -- missing values and percentiles])
 492 AT_KEYWORDS([categorical categoricals])
 493 AT_DATA([examine.sps], [dnl
 494 DATA LIST LIST /X *.
 495 BEGIN DATA.
 496 99
 497 99
 498 5.00
 499 END DATA.
 500
 501 MISSING VALUE X (99).
 502
 503 EXAMINE /x
 504         /PERCENTILES=HAVERAGE.
 505 ])
 506 AT_CHECK([pspp -o pspp.csv examine.sps])
 507 dnl Ignore output -- this is just a no-crash check.
 508 AT_CLEANUP
 509
 510 dnl Tests the trimmed mean calculation in the case
 511 dnl where the data is weighted towards the centre.
 512 AT_SETUP([EXAMINE -- trimmed mean])
 513 AT_KEYWORDS([categorical categoricals])
 514 AT_DATA([examine.sps], [dnl
 515 DATA LIST LIST /X * C *.
 516 BEGIN DATA.
 517 1 1
 518 2 49
 519 3 2
 520 END DATA.
 521
 522 WEIGHT BY c.
 523
 524 EXAMINE
 525         x
 526         /STATISTICS=DESCRIPTIVES
 527         .
 528 ])
 529 AT_CHECK([pspp -o pspp.csv examine.sps])
 530 AT_CHECK([cat pspp.csv], [0], [dnl
 531 Table: Reading free-form data from INLINE.
 532 Variable,Format
 533 X,F8.0
 534 C,F8.0
 535
 536 Table: Case Processing Summary
 537 ,Cases,,,,,
 538 ,Valid,,Missing,,Total,
 539 ,N,Percent,N,Percent,N,Percent
 540 X,52.00,100%,.00,0%,52.00,100%
 541
 542 Table: Descriptives
 543 ,,,Statistic,Std. Error
 544 X,Mean,,2.02,.03
 545 ,95% Confidence Interval for Mean,Lower Bound,1.95,
 546 ,,Upper Bound,2.09,
 547 ,5% Trimmed Mean,,2.00,
 548 ,Median,,2.00,
 549 ,Variance,,.06,
 550 ,Std. Deviation,,.24,
 551 ,Minimum,,1.00,
 552 ,Maximum,,3.00,
 553 ,Range,,2.00,
 554 ,Interquartile Range,,.00,
 555 ,Skewness,,1.19,.33
 556 ,Kurtosis,,15.73,.65
 557 ])
 558 AT_CLEANUP
 559
 560 AT_SETUP([EXAMINE -- crash bug])
 561 AT_KEYWORDS([categorical categoricals])
 562 AT_DATA([examine.sps], [dnl
 563 data list list /a * x * y *.
 564 begin data.
 565 3 1 3
 566 5 1 4
 567 7 2 3
 568 end data.
 569
 570 examine a by x by y
 571         /statistics=DESCRIPTIVES
 572         .
 573 ])
 574 AT_CHECK([pspp -o pspp.csv examine.sps])
 575 dnl Ignore output -- this is just a no-crash check.
 576 AT_CLEANUP
 577
 578 dnl Test that two consecutive EXAMINE commands don't crash PSPP.
 579 AT_SETUP([EXAMINE -- consecutive runs don't crash])
 580 AT_KEYWORDS([categorical categoricals])
 581 AT_DATA([examine.sps], [dnl
 582 data list list /y * z *.
 583 begin data.
 584 6 4
 585 5 3
 586 7 6
 587 end data.
 588
 589 EXAMINE /VARIABLES= z BY y.
 590
 591 EXAMINE /VARIABLES= z.
 592 ])
 593 AT_CHECK([pspp -o pspp.csv examine.sps])
 594 dnl Ignore output -- this is just a no-crash check.
 595 AT_CLEANUP
 596
 597 dnl Test that /DESCRIPTIVES does not crash in presence of missing values.
 598 AT_SETUP([EXAMINE -- missing values don't crash])
 599 AT_KEYWORDS([categorical categoricals])
 600 AT_DATA([examine.sps], [dnl
 601 data list list /x * y *.
 602 begin data.
 603 1 0
 604 2 0
 605 . 0
 606 3 1
 607 4 1
 608 end data.
 609 examine x by y /statistics=descriptives.
 610 ])
 611 AT_CHECK([pspp -o pspp.csv examine.sps])
 612 dnl Ignore output -- this is just a no-crash check.
 613 AT_CLEANUP
 614
 615 dnl Test that having only a single case doesn't crash.
 616 AT_SETUP([EXAMINE -- single case doesn't crash])
 617 AT_KEYWORDS([categorical categoricals])
 618 AT_DATA([examine.sps], [dnl
 619 DATA LIST LIST /quality * .
 620 BEGIN DATA
 621 3
 622 END DATA
 623
 624
 625 EXAMINE
 626         quality
 627         /STATISTICS descriptives
 628         /PLOT = histogram
 629         .
 630 ])
 631 AT_CHECK([pspp -o pspp.csv examine.sps], [0], [ignore])
 632 dnl Ignore output -- this is just a no-crash check.
 633 AT_CLEANUP
 634
 635 dnl Test that all-missing data doesn't crash.
 636 AT_SETUP([EXAMINE -- all-missing data doesn't crash])
 637 AT_KEYWORDS([categorical categoricals])
 638 AT_DATA([examine.sps], [dnl
 639 DATA LIST LIST /x *.
 640 BEGIN DATA.
 641 .
 642 .
 643 .
 644 .
 645 END DATA.
 646
 647 EXAMINE /x
 648         PLOT=HISTOGRAM BOXPLOT NPPLOT SPREADLEVEL(1) ALL
 649         /ID=x
 650         /STATISTICS = DESCRIPTIVES EXTREME (5) ALL
 651         /PERCENTILE=AEMPIRICAL
 652         .
 653 ])
 654 AT_CHECK([pspp -o pspp.csv examine.sps], [0], [ignore])
 655 dnl Ignore output -- this is just a no-crash check.
 656 AT_CLEANUP
 657
 658 dnl Test that big input doesn't crash (bug 11307).
 659 AT_SETUP([EXAMINE -- big input doesn't crash])
 660 AT_KEYWORDS([categorical categoricals])
 661 AT_DATA([examine.sps], [dnl
 662 INPUT PROGRAM.
 663         LOOP #I=1 TO 50000.
 664                 COMPUTE X=NORMAL(10).
 665                 END CASE.
 666         END LOOP.
 667         END FILE.
 668 END INPUT PROGRAM.
 669
 670
 671 EXAMINE /x
 672         /STATISTICS=DESCRIPTIVES.
 673 ])
 674 AT_CHECK([pspp -o pspp.csv examine.sps])
 675 dnl Ignore output -- this is just a no-crash check.
 676 AT_CLEANUP
 677
 678 dnl Another test that big input doesn't crash.
 679 dnl The actual bug that this checks for has been lost.
 680 AT_SETUP([EXAMINE -- big input doesn't crash 2])
 681 AT_KEYWORDS([categorical categoricals])
 682 AT_DATA([make-big-input.pl],
 683   [for ($i=0; $i<100000; $i++) { print "AB12\n" };
 684    for ($i=0; $i<100000; $i++) { print "AB04\n" };
 685 ])
 686 AT_CHECK([$PERL make-big-input.pl > large.txt])
 687 AT_DATA([examine.sps], [dnl
 688 DATA LIST FILE='large.txt' /S 1-2 (A) X 3 .
 689
 690
 691 AGGREGATE OUTFILE=* /BREAK=X /A=N.
 692
 693
 694 EXAMINE /A BY X.
 695 ])
 696 AT_CHECK([pspp -o pspp.csv examine.sps])
 697 dnl Ignore output -- this is just a no-crash check.
 698 AT_DATA([more-big-input.pl],
 699   [for ($i=0; $i<25000; $i++) { print "AB04\nAB12\n" };
 700 ])
 701 AT_CHECK([$PERL more-big-input.pl >> large.txt])
 702 AT_CHECK([pspp -o pspp.csv examine.sps])
 703 dnl Ignore output -- this is just a no-crash check.
 704 AT_CLEANUP
 705
 706
 707 dnl Test that the ID command works with non-numberic variables
 708 AT_SETUP([EXAMINE -- non-numeric ID])
 709 AT_KEYWORDS([categorical categoricals])
 710
 711 AT_DATA([examine-id.sps], [dnl
 712 data list notable list /x * y (a12).
 713 begin data.
 714 1  one
 715 2  two
 716 3  three
 717 4  four
 718 5  five
 719 6  six
 720 7  seven
 721 8  eight
 722 9  nine
 723 10 ten
 724 11 eleven
 725 12 twelve
 726 30 thirty
 727 300 threehundred
 728 end data.
 729
 730 examine x
 731         /statistics = extreme
 732         /id = y
 733         /plot = boxplot
 734         .
 735 ])
 736
 737 AT_CHECK([pspp -O format=csv examine-id.sps], [0],
 738 [Table: Case Processing Summary
 739 ,Cases,,,,,
 740 ,Valid,,Missing,,Total,
 741 ,N,Percent,N,Percent,N,Percent
 742 x,14,100%,0,0%,14,100%
 743
 744 Table: Extreme Values
 745 ,,,y,Value
 746 x,Highest,1,threehundred,300.00
 747 ,,2,thirty      ,30.00
 748 ,,3,twelve      ,12.00
 749 ,,4,eleven      ,11.00
 750 ,,5,ten         ,10.00
 751 ,Lowest,1,one         ,1.00
 752 ,,2,two         ,2.00
 753 ,,3,three       ,3.00
 754 ,,4,four        ,4.00
 755 ,,5,five        ,5.00
 756 ])
 757
 758 AT_CLEANUP
 759
 760 dnl Test for a crash which happened on cleanup from a bad input syntax
 761 AT_SETUP([EXAMINE -- Bad Input])
 762 AT_KEYWORDS([categorical categoricals])
 763
 764 AT_DATA([examine-bad.sps], [dnl
 765 data list list /h * g *.
 766 begin data.
 767 1 1
 768 2 1
 769 3 1
 770 4 1
 771 5 2
 772 6 2
 773 7 2
 774 8 2
 775 9 2
 776 end data.
 777
 778 EXAMINE
 779         /VARIABLES= h
 780         BY  g
 781         /STATISTICS = DESCRIPTIVES EXTREME
 782         /PLOT = lkajsdas
 783         .
 784 ])
 785
 786 AT_CHECK([pspp -o pspp.csv examine-bad.sps], [1], [ignore])
 787
 788 AT_CLEANUP
 789
 790
 791 dnl Check the MISSING=REPORT option
 792 AT_SETUP([EXAMINE -- MISSING=REPORT])
 793 AT_KEYWORDS([categorical categoricals])
 794
 795 AT_DATA([examine-report.sps], [dnl
 796 set format = F22.0.
 797 data list list /x * g *.
 798 begin data.
 799 1   1
 800 2   1
 801 3   1
 802 4   1
 803 5   1
 804 6   1
 805 7   1
 806 8   1
 807 9   1
 808 10   2
 809 20   2
 810 30   2
 811 40   2
 812 50   2
 813 60   2
 814 70   2
 815 80   2
 816 90   2
 817 101   9
 818 201   9
 819 301   9
 820 401   9
 821 501   99
 822 601   99
 823 701   99
 824 801   99
 825 901   99
 826 1001  .
 827 2002  .
 828 3003  .
 829 4004  .
 830 end data.
 831
 832 MISSING VALUES g (9, 99, 999).
 833
 834 EXAMINE
 835         /VARIABLES = x
 836         BY  g
 837         /STATISTICS = EXTREME
 838         /NOTOTAL
 839         /MISSING = REPORT.
 840 ])
 841
 842
 843 AT_CHECK([pspp -O format=csv examine-report.sps], [0], [dnl
 844 Table: Reading free-form data from INLINE.
 845 Variable,Format
 846 x,F8.0
 847 g,F8.0
 848
 849 Table: Case Processing Summary
 850 ,,Cases,,,,,
 851 ,,Valid,,Missing,,Total,
 852 ,g,N,Percent,N,Percent,N,Percent
 853 x,. (missing),4,100%,0,0%,4,100%
 854 ,1,9,100%,0,0%,9,100%
 855 ,2,9,100%,0,0%,9,100%
 856 ,9 (missing),4,100%,0,0%,4,100%
 857 ,99 (missing),5,100%,0,0%,5,100%
 858
 859 Table: Extreme Values
 860 ,g,,,Case Number,Value
 861 x,. (missing),Highest,1,31,4004
 862 ,,,2,30,3003
 863 ,,,3,29,2002
 864 ,,,4,28,1001
 865 ,,,5,0,0
 866 ,,Lowest,1,28,1001
 867 ,,,2,29,2002
 868 ,,,3,30,3003
 869 ,,,4,31,4004
 870 ,,,5,31,4004
 871 ,1,Highest,1,9,9
 872 ,,,2,8,8
 873 ,,,3,7,7
 874 ,,,4,6,6
 875 ,,,5,5,5
 876 ,,Lowest,1,1,1
 877 ,,,2,2,2
 878 ,,,3,3,3
 879 ,,,4,4,4
 880 ,,,5,5,5
 881 ,2,Highest,1,18,90
 882 ,,,2,17,80
 883 ,,,3,16,70
 884 ,,,4,15,60
 885 ,,,5,14,50
 886 ,,Lowest,1,10,10
 887 ,,,2,11,20
 888 ,,,3,12,30
 889 ,,,4,13,40
 890 ,,,5,14,50
 891 ,9 (missing),Highest,1,22,401
 892 ,,,2,21,301
 893 ,,,3,20,201
 894 ,,,4,19,101
 895 ,,,5,0,0
 896 ,,Lowest,1,19,101
 897 ,,,2,20,201
 898 ,,,3,21,301
 899 ,,,4,22,401
 900 ,,,5,22,401
 901 ,99 (missing),Highest,1,27,901
 902 ,,,2,26,801
 903 ,,,3,25,701
 904 ,,,4,24,601
 905 ,,,5,23,501
 906 ,,Lowest,1,23,501
 907 ,,,2,24,601
 908 ,,,3,25,701
 909 ,,,4,26,801
 910 ,,,5,27,901
 911 ])
 912
 913
 914 AT_CLEANUP
 915
 916
 917 dnl Run a test of the basic STATISTICS using a "real"
 918 dnl dataset and comparing with "real" results kindly
 919 dnl provided by Olaf Nöhring
 920 AT_SETUP([EXAMINE -- sample unweighted])
 921 AT_KEYWORDS([categorical categoricals])
 922
 923 AT_DATA([sample.sps], [dnl
 924 set format = F22.4.
 925 DATA LIST notable LIST /X *
 926 BEGIN DATA.
 927 461.19000000
 928 466.38000000
 929 479.46000000
 930 480.10000000
 931 483.43000000
 932 488.30000000
 933 489.00000000
 934 491.62000000
 935 505.62000000
 936 511.30000000
 937 521.53000000
 938 526.70000000
 939 528.25000000
 940 538.70000000
 941 540.22000000
 942 540.58000000
 943 546.10000000
 944 548.17000000
 945 553.99000000
 946 566.21000000
 947 575.90000000
 948 584.38000000
 949 593.40000000
 950 357.05000000
 951 359.73000000
 952 360.48000000
 953 373.98000000
 954 374.13000000
 955 381.45000000
 956 383.72000000
 957 390.00000000
 958 400.34000000
 959 415.32000000
 960 415.91000000
 961 418.30000000
 962 421.03000000
 963 422.43000000
 964 426.93000000
 965 433.25000000
 966 436.89000000
 967 445.33000000
 968 446.33000000
 969 446.55000000
 970 456.44000000
 971 689.49000000
 972 691.92000000
 973 695.00000000
 974 695.36000000
 975 698.21000000
 976 699.46000000
 977 706.61000000
 978 710.69000000
 979 715.82000000
 980 715.82000000
 981 741.39000000
 982 752.27000000
 983 756.73000000
 984 757.74000000
 985 759.57000000
 986 796.07000000
 987 813.78000000
 988 817.25000000
 989 825.48000000
 990 831.28000000
 991 849.24000000
 992 890.00000000
 993 894.78000000
 994 935.65000000
 995 935.90000000
 996 945.90000000
 997 1012.8600000
 998 1022.6000000
 999 1061.8100000
1000 1063.5000000
1001 1077.2300000
1002 1151.6300000
1003 1355.2800000
1004 598.88000000
1005 606.91000000
1006 621.60000000
1007 624.80000000
1008 636.13000000
1009 637.38000000
1010 640.32000000
1011 649.35000000
1012 656.51000000
1013 662.55000000
1014 664.69000000
1015 106.22000000
1016 132.24000000
1017 174.76000000
1018 204.85000000
1019 264.93000000
1020 264.99000000
1021 269.84000000
1022 325.12000000
1023 331.67000000
1024 337.26000000
1025 347.68000000
1026 354.91000000
1027 END DATA.
1028
1029 EXAMINE
1030         x
1031         /STATISTICS=DESCRIPTIVES
1032         .
1033 ])
1034
1035 AT_CHECK([pspp -O format=csv sample.sps], [0], [dnl
1036 Table: Case Processing Summary
1037 ,Cases,,,,,
1038 ,Valid,,Missing,,Total,
1039 ,N,Percent,N,Percent,N,Percent
1040 X,100,100%,0,0%,100,100%
1041
1042 Table: Descriptives
1043 ,,,Statistic,Std. Error
1044 X,Mean,,587.6603,23.2665
1045 ,95% Confidence Interval for Mean,Lower Bound,541.4946,
1046 ,,Upper Bound,633.8260,
1047 ,5% Trimmed Mean,,579.7064,
1048 ,Median,,547.1350,
1049 ,Variance,,54132.8466,
1050 ,Std. Deviation,,232.6647,
1051 ,Minimum,,106.2200,
1052 ,Maximum,,1355.2800,
1053 ,Range,,1249.0600,
1054 ,Interquartile Range,,293.1575,
1055 ,Skewness,,.6331,.2414
1056 ,Kurtosis,,.5300,.4783
1057 ])
1058
1059 AT_CLEANUP
1060
1061
1062
1063 dnl Test for a crash which happened on bad input syntax
1064 AT_SETUP([EXAMINE -- Empty Parentheses])
1065 AT_KEYWORDS([categorical categoricals])
1066
1067 AT_DATA([examine-empty-parens.sps], [dnl
1068 DATA LIST notable LIST /X *
1069 BEGIN DATA.
1070 2
1071 3
1072 END DATA.
1073
1074
1075 EXAMINE
1076         x
1077         /PLOT = SPREADLEVEL()
1078         .
1079 ])
1080
1081 AT_CHECK([pspp -o pspp.csv examine-empty-parens.sps], [1], [ignore])
1082
1083 AT_CLEANUP
1084
1085
1086
1087
1088 dnl Test for another crash which happened on bad input syntax
1089 AT_SETUP([EXAMINE -- Bad variable])
1090 AT_KEYWORDS([categorical categoricals])
1091
1092 AT_DATA([examine-bad-variable.sps], [dnl
1093 data list list /h * g *.
1094 begin data.
1095 3 1
1096 4 1
1097 5 2
1098 end data.
1099
1100 EXAMINE
1101         /VARIABLES/ h
1102         BY  g
1103         .
1104 ])
1105
1106 AT_CHECK([pspp -o pspp.csv examine-bad-variable.sps], [1], [ignore])
1107
1108 AT_CLEANUP
1109
1110
1111
1112 dnl Test for yet another crash. This time for extremes vs. missing weight values.\0
1113 AT_SETUP([EXAMINE -- Extremes vs. Missing Weights])
1114 AT_KEYWORDS([categorical categoricals])
1115
1116 AT_DATA([examine-missing-weights.sps], [dnl
1117 data list notable list /h * g *.
1118 begin data.
1119 3 1
1120 4 .
1121 5 1
1122 2 1
1123 end data.
1124
1125 WEIGHT BY g.
1126
1127 EXAMINE h
1128         /STATISTICS extreme(3)
1129         .
1130 ])
1131
1132 AT_CHECK([pspp -O format=csv  examine-missing-weights.sps], [0], [dnl
1133 "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."
1134
1135 Table: Case Processing Summary
1136 ,Cases,,,,,
1137 ,Valid,,Missing,,Total,
1138 ,N,Percent,N,Percent,N,Percent
1139 h,3.00,100%,.00,0%,3.00,100%
1140
1141 Table: Extreme Values
1142 ,,,Case Number,Value
1143 h,Highest,1,3,5.00
1144 ,,2,2,4.00
1145 ,,3,1,3.00
1146 ,Lowest,1,4,2.00
1147 ,,2,1,3.00
1148 ,,3,2,4.00
1149 ])
1150
1151 AT_CLEANUP
1152
1153
1154