tests/language/stats/examine.at

   1 AT_BANNER([EXAMINE])
   2
   3 AT_SETUP([EXAMINE])
   4 AT_DATA([examine.sps], [
   5 DATA LIST LIST /QUALITY * W * BRAND * .
   6 BEGIN DATA
   7 3  1  1
   8 2  2  1
   9 1  2  1
  10 1  1  1
  11 4  1  1
  12 4  1  1
  13 5  1  2
  14 2  1  2
  15 4  4  2
  16 2  1  2
  17 3  1  2
  18 7  1  3
  19 4  2  3
  20 5  3  3
  21 3  1  3
  22 6  1  3
  23 END DATA
  24
  25 WEIGHT BY w.
  26
  27 VARIABLE LABELS brand   'Manufacturer'.
  28 VARIABLE LABELS quality 'Breaking Strain'.
  29
  30 VALUE LABELS /brand 1 'Aspeger' 2 'Bloggs' 3 'Charlies'.
  31
  32 LIST /FORMAT=NUMBERED.
  33
  34 EXAMINE
  35         quality BY brand
  36         /STATISTICS descriptives extreme(3)
  37         .
  38 ])
  39 AT_CHECK([pspp -o pspp.csv examine.sps])
  40 dnl In the following data, only the extreme values have been checked.
  41 dnl The descriptives have been blindly pasted.
  42 AT_CHECK([cat pspp.csv], [0], [dnl
  43 Table: Reading free-form data from INLINE.
  44 Variable,Format
  45 QUALITY,F8.0
  46 W,F8.0
  47 BRAND,F8.0
  48
  49 Table: Data List
  50 Case Number,QUALITY,W,BRAND
  51 1,3.00,1.00,1.00
  52 2,2.00,2.00,1.00
  53 3,1.00,2.00,1.00
  54 4,1.00,1.00,1.00
  55 5,4.00,1.00,1.00
  56 6,4.00,1.00,1.00
  57 7,5.00,1.00,2.00
  58 8,2.00,1.00,2.00
  59 9,4.00,4.00,2.00
  60 10,2.00,1.00,2.00
  61 11,3.00,1.00,2.00
  62 12,7.00,1.00,3.00
  63 13,4.00,2.00,3.00
  64 14,5.00,3.00,3.00
  65 15,3.00,1.00,3.00
  66 16,6.00,1.00,3.00
  67
  68 Table: Case Processing Summary
  69 ,Cases,,,,,
  70 ,Valid,,Missing,,Total,
  71 ,N,Percent,N,Percent,N,Percent
  72 Breaking Strain,24.00,100%,.00,0%,24.00,100%
  73
  74 Table: Extreme Values
  75 ,,,Case Number,Value
  76 Breaking Strain,Highest,1,12,7.00
  77 ,,2,16,6.00
  78 ,,3,7,5.00
  79 ,Lowest,1,3,1.00
  80 ,,2,3,1.00
  81 ,,3,4,1.00
  82
  83 Table: Descriptives
  84 ,,,Statistic,Std. Error
  85 Breaking Strain,Mean,,3.54,.32
  86 ,95% Confidence Interval for Mean,Lower Bound,2.87,
  87 ,,Upper Bound,4.21,
  88 ,5% Trimmed Mean,,3.50,
  89 ,Median,,4.00,
  90 ,Variance,,2.52,
  91 ,Std. Deviation,,1.59,
  92 ,Minimum,,1.00,
  93 ,Maximum,,7.00,
  94 ,Range,,6.00,
  95 ,Interquartile Range,,2.75,
  96 ,Skewness,,.06,.47
  97 ,Kurtosis,,-.36,.92
  98
  99 Table: Case Processing Summary
 100 ,,Cases,,,,,
 101 ,,Valid,,Missing,,Total,
 102 ,Manufacturer,N,Percent,N,Percent,N,Percent
 103 Breaking Strain,Aspeger,8.00,100%,.00,0%,8.00,100%
 104 ,Bloggs,8.00,100%,.00,0%,8.00,100%
 105 ,Charlies,8.00,100%,.00,0%,8.00,100%
 106
 107 Table: Extreme Values
 108 ,Manufacturer,,,Case Number,Value
 109 Breaking Strain,Aspeger,Highest,1,5,4.00
 110 ,,,2,6,4.00
 111 ,,,3,1,3.00
 112 ,,Lowest,1,3,1.00
 113 ,,,2,3,1.00
 114 ,,,3,4,1.00
 115 ,Bloggs,Highest,1,7,5.00
 116 ,,,2,9,4.00
 117 ,,,3,9,4.00
 118 ,,Lowest,1,8,2.00
 119 ,,,2,10,2.00
 120 ,,,3,11,3.00
 121 ,Charlies,Highest,1,12,7.00
 122 ,,,2,16,6.00
 123 ,,,3,14,5.00
 124 ,,Lowest,1,15,3.00
 125 ,,,2,13,4.00
 126 ,,,3,13,4.00
 127
 128 Table: Descriptives
 129 ,Manufacturer,,,Statistic,Std. Error
 130 Breaking Strain,Aspeger,Mean,,2.25,.45
 131 ,,95% Confidence Interval for Mean,Lower Bound,1.18,
 132 ,,,Upper Bound,3.32,
 133 ,,5% Trimmed Mean,,2.22,
 134 ,,Median,,2.00,
 135 ,,Variance,,1.64,
 136 ,,Std. Deviation,,1.28,
 137 ,,Minimum,,1.00,
 138 ,,Maximum,,4.00,
 139 ,,Range,,3.00,
 140 ,,Interquartile Range,,2.75,
 141 ,,Skewness,,.47,.75
 142 ,,Kurtosis,,-1.55,1.48
 143 ,Bloggs,Mean,,3.50,.38
 144 ,,95% Confidence Interval for Mean,Lower Bound,2.61,
 145 ,,,Upper Bound,4.39,
 146 ,,5% Trimmed Mean,,3.50,
 147 ,,Median,,4.00,
 148 ,,Variance,,1.14,
 149 ,,Std. Deviation,,1.07,
 150 ,,Minimum,,2.00,
 151 ,,Maximum,,5.00,
 152 ,,Range,,3.00,
 153 ,,Interquartile Range,,1.75,
 154 ,,Skewness,,-.47,.75
 155 ,,Kurtosis,,-.83,1.48
 156 ,Charlies,Mean,,4.88,.44
 157 ,,95% Confidence Interval for Mean,Lower Bound,3.83,
 158 ,,,Upper Bound,5.92,
 159 ,,5% Trimmed Mean,,4.86,
 160 ,,Median,,5.00,
 161 ,,Variance,,1.55,
 162 ,,Std. Deviation,,1.25,
 163 ,,Minimum,,3.00,
 164 ,,Maximum,,7.00,
 165 ,,Range,,4.00,
 166 ,,Interquartile Range,,1.75,
 167 ,,Skewness,,.30,.75
 168 ,,Kurtosis,,.15,1.48
 169 ])
 170 AT_CLEANUP
 171
 172 AT_SETUP([EXAMINE -- extremes])
 173 AT_DATA([examine.sps], [dnl
 174 data list free /V1 W
 175 begin data.
 176 1  1
 177 2  1
 178 3  2
 179 3  1
 180 4  1
 181 5  1
 182 6  1
 183 7  1
 184 8  1
 185 9  1
 186 10 1
 187 11 1
 188 12 1
 189 13 1
 190 14 1
 191 15 1
 192 16 1
 193 17 1
 194 18 2
 195 19 1
 196 20 1
 197 end data.
 198
 199 weight by w.
 200
 201 examine v1
 202  /statistics=extreme(6)
 203  .
 204 ])
 205 AT_CHECK([pspp -o pspp.csv examine.sps])
 206 AT_CHECK([cat pspp.csv], [0], [dnl
 207 Table: Case Processing Summary
 208 ,Cases,,,,,
 209 ,Valid,,Missing,,Total,
 210 ,N,Percent,N,Percent,N,Percent
 211 V1,23.00,100%,.00,0%,23.00,100%
 212
 213 Table: Extreme Values
 214 ,,,Case Number,Value
 215 V1,Highest,1,21,20.00
 216 ,,2,20,19.00
 217 ,,3,19,18.00
 218 ,,4,19,18.00
 219 ,,5,18,17.00
 220 ,,6,17,16.00
 221 ,Lowest,1,1,1.00
 222 ,,2,2,2.00
 223 ,,3,3,3.00
 224 ,,4,3,3.00
 225 ,,5,4,3.00
 226 ,,6,5,4.00
 227 ])
 228 AT_CLEANUP
 229
 230 dnl Test the PERCENTILES subcommand of the EXAMINE command.
 231 dnl In particular test that it behaves properly when there are only
 232 dnl a few cases.
 233 AT_SETUP([EXAMINE -- percentiles])
 234 AT_DATA([examine.sps], [dnl
 235 DATA LIST LIST /X *.
 236 BEGIN DATA.
 237 2.00
 238 8.00
 239 5.00
 240 END DATA.
 241
 242 EXAMINE /x
 243         /PERCENTILES=HAVERAGE.
 244
 245 EXAMINE /x
 246         /PERCENTILES=WAVERAGE.
 247
 248 EXAMINE /x
 249         /PERCENTILES=ROUND.
 250
 251 EXAMINE /x
 252         /PERCENTILES=EMPIRICAL.
 253
 254 EXAMINE /x
 255         /PERCENTILES=AEMPIRICAL.
 256 ])
 257 AT_CHECK([pspp -o pspp.csv examine.sps])
 258 AT_CHECK([cat pspp.csv], [0], [dnl
 259 Table: Reading free-form data from INLINE.
 260 Variable,Format
 261 X,F8.0
 262
 263 Table: Case Processing Summary
 264 ,Cases,,,,,
 265 ,Valid,,Missing,,Total,
 266 ,N,Percent,N,Percent,N,Percent
 267 X,3,100%,0,0%,3,100%
 268
 269 Table: Percentiles
 270 ,,Percentiles,,,,,,
 271 ,,5,10,25,50,75,90,95
 272 X,HAverage,.40,.80,2.00,5.00,8.00,8.00,8.00
 273 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 274
 275 Table: Case Processing Summary
 276 ,Cases,,,,,
 277 ,Valid,,Missing,,Total,
 278 ,N,Percent,N,Percent,N,Percent
 279 X,3,100%,0,0%,3,100%
 280
 281 Table: Percentiles
 282 ,,Percentiles,,,,,,
 283 ,,5,10,25,50,75,90,95
 284 X,Weighted Average,.30,.60,1.50,3.50,5.75,7.10,7.55
 285 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 286
 287 Table: Case Processing Summary
 288 ,Cases,,,,,
 289 ,Valid,,Missing,,Total,
 290 ,N,Percent,N,Percent,N,Percent
 291 X,3,100%,0,0%,3,100%
 292
 293 Table: Percentiles
 294 ,,Percentiles,,,,,,
 295 ,,5,10,25,50,75,90,95
 296 X,Rounded,.00,.00,2.00,5.00,5.00,8.00,8.00
 297 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 298
 299 Table: Case Processing Summary
 300 ,Cases,,,,,
 301 ,Valid,,Missing,,Total,
 302 ,N,Percent,N,Percent,N,Percent
 303 X,3,100%,0,0%,3,100%
 304
 305 Table: Percentiles
 306 ,,Percentiles,,,,,,
 307 ,,5,10,25,50,75,90,95
 308 X,Empirical,2.00,2.00,2.00,5.00,8.00,8.00,8.00
 309 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 310
 311 Table: Case Processing Summary
 312 ,Cases,,,,,
 313 ,Valid,,Missing,,Total,
 314 ,N,Percent,N,Percent,N,Percent
 315 X,3,100%,0,0%,3,100%
 316
 317 Table: Percentiles
 318 ,,Percentiles,,,,,,
 319 ,,5,10,25,50,75,90,95
 320 X,Empirical with averaging,2.00,2.00,2.00,5.00,8.00,8.00,8.00
 321 ,Tukey's Hinges,,,3.50,5.00,6.50,,
 322 ])
 323 AT_CLEANUP
 324
 325 AT_SETUP([EXAMINE -- missing values])
 326 AT_DATA([examine.sps], [dnl
 327 DATA LIST LIST /x * y *.
 328 BEGIN DATA.
 329 1   1
 330 2   1
 331 3   1
 332 4   1
 333 5   2
 334 6   2
 335 .   2
 336 END DATA
 337
 338 EXAMINE /x by y.
 339 ])
 340 AT_CHECK([pspp -o pspp.csv examine.sps])
 341 AT_CHECK([cat pspp.csv], [0], [dnl
 342 Table: Reading free-form data from INLINE.
 343 Variable,Format
 344 x,F8.0
 345 y,F8.0
 346
 347 Table: Case Processing Summary
 348 ,Cases,,,,,
 349 ,Valid,,Missing,,Total,
 350 ,N,Percent,N,Percent,N,Percent
 351 x,6,85.7143%,1,14.2857%,7,100%
 352
 353 Table: Case Processing Summary
 354 ,,Cases,,,,,
 355 ,,Valid,,Missing,,Total,
 356 ,y,N,Percent,N,Percent,N,Percent
 357 x,1.00,4,100%,0,0%,4,100%
 358 ,2.00,2,66.6667%,1,33.3333%,3,100%
 359 ])
 360 AT_CLEANUP
 361
 362 AT_SETUP([EXAMINE -- missing values and percentiles])
 363 AT_DATA([examine.sps], [dnl
 364 DATA LIST LIST /X *.
 365 BEGIN DATA.
 366 99
 367 99
 368 5.00
 369 END DATA.
 370
 371 MISSING VALUE X (99).
 372
 373 EXAMINE /x
 374         /PERCENTILES=HAVERAGE.
 375 ])
 376 AT_CHECK([pspp -o pspp.csv examine.sps])
 377 dnl Ignore output -- this is just a no-crash check.
 378 AT_CLEANUP
 379
 380 dnl Tests the trimmed mean calculation in the case
 381 dnl where the data is weighted towards the centre.
 382 AT_SETUP([EXAMINE -- trimmed mean])
 383 AT_DATA([examine.sps], [dnl
 384 DATA LIST LIST /X * C *.
 385 BEGIN DATA.
 386 1 1
 387 2 49
 388 3 2
 389 END DATA.
 390
 391 WEIGHT BY c.
 392
 393 EXAMINE
 394         x
 395         /STATISTICS=DESCRIPTIVES
 396         .
 397 ])
 398 AT_CHECK([pspp -o pspp.csv examine.sps])
 399 AT_CHECK([cat pspp.csv], [0], [dnl
 400 Table: Reading free-form data from INLINE.
 401 Variable,Format
 402 X,F8.0
 403 C,F8.0
 404
 405 Table: Case Processing Summary
 406 ,Cases,,,,,
 407 ,Valid,,Missing,,Total,
 408 ,N,Percent,N,Percent,N,Percent
 409 X,52.00,100%,.00,0%,52.00,100%
 410
 411 Table: Descriptives
 412 ,,,Statistic,Std. Error
 413 X,Mean,,2.02,.03
 414 ,95% Confidence Interval for Mean,Lower Bound,1.95,
 415 ,,Upper Bound,2.09,
 416 ,5% Trimmed Mean,,2.00,
 417 ,Median,,2.00,
 418 ,Variance,,.06,
 419 ,Std. Deviation,,.24,
 420 ,Minimum,,1.00,
 421 ,Maximum,,3.00,
 422 ,Range,,2.00,
 423 ,Interquartile Range,,.00,
 424 ,Skewness,,1.19,.33
 425 ,Kurtosis,,15.73,.65
 426 ])
 427 AT_CLEANUP
 428
 429 AT_SETUP([EXAMINE -- crash bug])
 430 AT_DATA([examine.sps], [dnl
 431 data list list /a * x * y *.
 432 begin data.
 433 3 1 3
 434 5 1 4
 435 7 2 3
 436 end data.
 437
 438 examine a by x by y
 439         /statistics=DESCRIPTIVES
 440         .
 441 ])
 442 AT_CHECK([pspp -o pspp.csv examine.sps])
 443 dnl Ignore output -- this is just a no-crash check.
 444 AT_CLEANUP
 445
 446 dnl Test that two consecutive EXAMINE commands don't crash PSPP.
 447 AT_SETUP([EXAMINE -- consecutive runs don't crash])
 448 AT_DATA([examine.sps], [dnl
 449 data list list /y * z *.
 450 begin data.
 451 6 4
 452 5 3
 453 7 6
 454 end data.
 455
 456 EXAMINE /VARIABLES= z BY y.
 457
 458 EXAMINE /VARIABLES= z.
 459 ])
 460 AT_CHECK([pspp -o pspp.csv examine.sps])
 461 dnl Ignore output -- this is just a no-crash check.
 462 AT_CLEANUP
 463
 464 dnl Test that /DESCRIPTIVES does not crash in presence of missing values.
 465 AT_SETUP([EXAMINE -- missing values don't crash])
 466 AT_DATA([examine.sps], [dnl
 467 data list list /x * y *.
 468 begin data.
 469 1 0
 470 2 0
 471 . 0
 472 3 1
 473 4 1
 474 end data.
 475 examine x by y /statistics=descriptives.
 476 ])
 477 AT_CHECK([pspp -o pspp.csv examine.sps])
 478 dnl Ignore output -- this is just a no-crash check.
 479 AT_CLEANUP
 480
 481 dnl Test that having only a single case doesn't crash.
 482 AT_SETUP([EXAMINE -- single case doesn't crash])
 483 AT_DATA([examine.sps], [dnl
 484 DATA LIST LIST /quality * .
 485 BEGIN DATA
 486 3
 487 END DATA
 488
 489
 490 EXAMINE
 491         quality
 492         /STATISTICS descriptives
 493         .
 494 ])
 495 AT_CHECK([pspp -o pspp.csv examine.sps])
 496 dnl Ignore output -- this is just a no-crash check.
 497 AT_CLEANUP
 498
 499 dnl Test that all-missing data doesn't crash.
 500 AT_SETUP([EXAMINE -- all-missing data doesn't crash])
 501 AT_DATA([examine.sps], [dnl
 502 DATA LIST LIST /x *.
 503 BEGIN DATA.
 504 .
 505 .
 506 .
 507 .
 508 END DATA.
 509
 510 EXAMINE /x PLOT=HISTOGRAM.
 511 ])
 512 AT_CHECK([pspp -o pspp.csv examine.sps], [0], [dnl
 513 warning: Not creating plot because data set is empty.
 514 ])
 515 dnl Ignore output -- this is just a no-crash check.
 516 AT_CLEANUP