|
DataMuseum.dkPresents historical artifacts from the history of: DKUUG/EUUG Conference tapes |
This is an automatic "excavation" of a thematic subset of
See our Wiki for more about DKUUG/EUUG Conference tapes Excavated with: AutoArchaeologist - Free & Open Source Software. |
top - downloadIndex: ┃ T ┃
Length: 17019 (0x427b)
Types: TextFile
Names: »TRY_ME.out«
└─⟦87ddcff64⟧ Bits:30001253 CPHDIST85 Tape, 1985 Autumn Conference Copenhagen
└─ ⟦this⟧ »cph85dist/stat/test/DATA/TRY_ME.out«
This is a test of the statistical programs from CSL
First, let's look at the anova for errors:
dm s1 s2 s3 s4 s5 < data | anova subject OS time difficulty errors
SOURCE: grand mean
OS time diffi N MEAN SD SE
66 -0.6515 2.5267 0.3110
SOURCE: OS
OS time diffi N MEAN SD SE
UNIX 36 -1.5556 1.8890 0.3148
UNIQ 30 0.4333 2.7877 0.5090
SOURCE: time
OS time diffi N MEAN SD SE
6hour 33 0.5455 2.2233 0.3870
month 33 -1.8485 2.2517 0.3920
SOURCE: OS time
OS time diffi N MEAN SD SE
UNIX 6hour 18 -0.4444 1.6881 0.3979
UNIX month 18 -2.6667 1.3720 0.3234
UNIQ 6hour 15 1.7333 2.2509 0.5812
UNIQ month 15 -0.8667 2.7220 0.7028
SOURCE: difficulty
OS time diffi N MEAN SD SE
easy 22 -2.3636 1.8910 0.4032
mediu 22 -0.9091 1.9978 0.4259
hard 22 1.3182 2.2336 0.4762
SOURCE: OS difficulty
OS time diffi N MEAN SD SE
UNIX easy 12 -2.8333 1.1146 0.3218
UNIX mediu 12 -2.0833 1.3114 0.3786
UNIX hard 12 0.2500 1.6583 0.4787
UNIQ easy 10 -1.8000 2.4855 0.7860
UNIQ mediu 10 0.5000 1.7795 0.5627
UNIQ hard 10 2.6000 2.2211 0.7024
SOURCE: time difficulty
OS time diffi N MEAN SD SE
6hour easy 11 -1.0909 1.6404 0.4946
6hour mediu 11 0.0000 1.5492 0.4671
6hour hard 11 2.7273 1.4894 0.4491
month easy 11 -3.6364 1.1201 0.3377
month mediu 11 -1.8182 2.0405 0.6152
month hard 11 -0.0909 1.9725 0.5947
SOURCE: OS time difficulty
OS time diffi N MEAN SD SE
UNIX 6hour easy 6 -2.0000 0.6325 0.2582
UNIX 6hour mediu 6 -1.0000 0.6325 0.2582
UNIX 6hour hard 6 1.6667 0.5164 0.2108
UNIX month easy 6 -3.6667 0.8165 0.3333
UNIX month mediu 6 -3.1667 0.7528 0.3073
UNIX month hard 6 -1.1667 0.9832 0.4014
UNIQ 6hour easy 5 0.0000 1.8708 0.8367
UNIQ 6hour mediu 5 1.2000 1.4832 0.6633
UNIQ 6hour hard 5 4.0000 1.2247 0.5477
UNIQ month easy 5 -3.6000 1.5166 0.6782
UNIQ month mediu 5 -0.2000 1.9235 0.8602
UNIQ month hard 5 1.2000 2.1679 0.9695
FACTOR: subject OS time difficulty errors
LEVELS: 11 2 2 3 66
TYPE : RANDOM BETWEEN WITHIN WITHIN DATA
SOURCE SS df MS F p
===============================================================
mean 28.0152 1 28.0152 27.741 0.001 ***
s/O 9.0889 9 1.0099
OS 64.7293 1 64.7293 64.096 0.000 ***
s/O 9.0889 9 1.0099
time 94.5606 1 94.5606 106.086 0.000 ***
ts/O 8.0222 9 0.8914
Ot 0.5838 1 0.5838 0.655 0.439
ts/O 8.0222 9 0.8914
difficu 151.3030 2 75.6515 47.374 0.000 ***
ds/O 28.7444 18 1.5969
Od 7.6192 2 3.8096 2.386 0.120
ds/O 28.7444 18 1.5969
td 2.9394 2 1.4697 0.629 0.545
tds/O 42.0778 18 2.3377
Otd 5.3162 2 2.6581 1.137 0.343
tds/O 42.0778 18 2.3377
Now let's look at the reaction time data:
dm s1 s2 s3 s4 s6 < data | anova subject OS time difficulty rt
SOURCE: grand mean
OS time diffi N MEAN SD SE
66 368.0758 236.5611 29.1187
SOURCE: OS
OS time diffi N MEAN SD SE
UNIX 36 351.0833 230.7209 38.4535
UNIQ 30 388.4667 245.7558 44.8687
SOURCE: time
OS time diffi N MEAN SD SE
6hour 33 492.0606 255.1066 44.4084
month 33 244.0909 129.9479 22.6210
SOURCE: OS time
OS time diffi N MEAN SD SE
UNIX 6hour 18 486.5000 251.9387 59.3825
UNIX month 18 215.6667 85.3801 20.1243
UNIQ 6hour 15 498.7333 267.5648 69.0849
UNIQ month 15 278.2000 165.7120 42.7866
SOURCE: difficulty
OS time diffi N MEAN SD SE
easy 22 260.0455 144.4535 30.7976
mediu 22 332.0455 162.2510 34.5920
hard 22 512.1364 301.1526 64.2059
SOURCE: OS difficulty
OS time diffi N MEAN SD SE
UNIX easy 12 230.1667 115.9646 33.4761
UNIX mediu 12 311.5000 132.7311 38.3162
UNIX hard 12 511.5833 304.2175 87.8200
UNIQ easy 10 295.9000 172.1462 54.4374
UNIQ mediu 10 356.7000 196.6091 62.1732
UNIQ hard 10 512.8000 313.8491 99.2478
SOURCE: time difficulty
OS time diffi N MEAN SD SE
6hour easy 11 344.6364 157.7988 47.5781
6hour mediu 11 410.1818 182.2453 54.9490
6hour hard 11 721.3636 247.0232 74.4803
month easy 11 175.4545 56.3691 16.9959
month mediu 11 253.9091 92.9424 28.0232
month hard 11 302.9091 182.0090 54.8778
SOURCE: OS time difficulty
OS time diffi N MEAN SD SE
UNIX 6hour easy 6 305.8333 119.8589 48.9322
UNIX 6hour mediu 6 394.5000 127.7885 52.1694
UNIX 6hour hard 6 759.1667 215.8021 88.1009
UNIX month easy 6 154.5000 38.4435 15.6945
UNIX month mediu 6 228.5000 76.7796 31.3451
UNIX month hard 6 264.0000 99.6072 40.6645
UNIQ 6hour easy 5 391.2000 198.3046 88.6845
UNIQ 6hour mediu 5 429.0000 248.6152 111.1841
UNIQ 6hour hard 5 676.0000 299.3693 133.8820
UNIQ month easy 5 200.6000 68.1711 30.4870
UNIQ month mediu 5 284.4000 109.9832 49.1860
UNIQ month hard 5 349.6000 255.7739 114.3856
FACTOR: subject OS time difficulty rt
LEVELS: 11 2 2 3 66
TYPE : RANDOM BETWEEN WITHIN WITHIN DATA
SOURCE SS df MS F p
===============================================================
mean 8941664.3788 1 8941664.3788 305.581 0.000 ***
s/O 263350.3833 9 29261.1537
OS 22868.4045 1 22868.4045 0.782 0.400
s/O 263350.3833 9 29261.1537
time 1014568.0152 1 1014568.0152 23.351 0.001 ***
ts/O 391045.4500 9 43449.4944
Ot 10350.3682 1 10350.3682 0.238 0.637
ts/O 391045.4500 9 43449.4944
difficu 741888.1212 2 370944.0606 21.074 0.000 ***
ds/O 316841.3000 18 17602.2944
Od 11851.9121 2 5925.9561 0.337 0.719
ds/O 316841.3000 18 17602.2944
td 240245.2121 2 120122.6061 3.645 0.047 *
tds/O 593249.4333 18 32958.3019
Otd 31216.0212 2 15608.0106 0.474 0.630
tds/O 593249.4333 18 32958.3019
Now let's combine these and look at the efficiency (= #correct/rt):
dm s1 s2 s3 s4 '(10-x5)/x6' < data | anova subject OS time difficulty efficiency
SOURCE: grand mean
OS time diffi N MEAN SD SE
66 0.0434 0.0298 0.0037
SOURCE: OS
OS time diffi N MEAN SD SE
UNIX 36 0.0490 0.0313 0.0052
UNIQ 30 0.0366 0.0268 0.0049
SOURCE: time
OS time diffi N MEAN SD SE
6hour 33 0.0263 0.0172 0.0030
month 33 0.0604 0.0301 0.0052
SOURCE: OS time
OS time diffi N MEAN SD SE
UNIX 6hour 18 0.0295 0.0191 0.0045
UNIX month 18 0.0685 0.0292 0.0069
UNIQ 6hour 15 0.0225 0.0144 0.0037
UNIQ month 15 0.0508 0.0292 0.0075
SOURCE: difficulty
OS time diffi N MEAN SD SE
easy 22 0.0618 0.0331 0.0071
mediu 22 0.0415 0.0240 0.0051
hard 22 0.0268 0.0207 0.0044
SOURCE: OS difficulty
OS time diffi N MEAN SD SE
UNIX easy 12 0.0688 0.0325 0.0094
UNIX mediu 12 0.0487 0.0275 0.0079
UNIX hard 12 0.0295 0.0214 0.0062
UNIQ easy 10 0.0533 0.0335 0.0106
UNIQ mediu 10 0.0330 0.0165 0.0052
UNIQ hard 10 0.0236 0.0204 0.0064
SOURCE: time difficulty
OS time diffi N MEAN SD SE
6hour easy 11 0.0380 0.0166 0.0050
6hour mediu 11 0.0295 0.0158 0.0048
6hour hard 11 0.0115 0.0051 0.0015
month easy 11 0.0856 0.0279 0.0084
month mediu 11 0.0535 0.0254 0.0077
month hard 11 0.0421 0.0188 0.0057
SOURCE: OS time difficulty
OS time diffi N MEAN SD SE
UNIX 6hour easy 6 0.0442 0.0166 0.0068
UNIX 6hour mediu 6 0.0325 0.0175 0.0072
UNIX 6hour hard 6 0.0120 0.0043 0.0017
UNIX month easy 6 0.0934 0.0245 0.0100
UNIX month mediu 6 0.0649 0.0270 0.0110
UNIX month hard 6 0.0470 0.0158 0.0064
UNIQ 6hour easy 5 0.0305 0.0146 0.0065
UNIQ 6hour mediu 5 0.0260 0.0146 0.0065
UNIQ 6hour hard 5 0.0110 0.0064 0.0029
UNIQ month easy 5 0.0762 0.0316 0.0141
UNIQ month mediu 5 0.0399 0.0167 0.0074
UNIQ month hard 5 0.0363 0.0222 0.0099
FACTOR: subject OS time difficulty efficiency
LEVELS: 11 2 2 3 66
TYPE : RANDOM BETWEEN WITHIN WITHIN DATA
SOURCE SS df MS F p
===============================================================
mean 0.1242 1 0.1242 268.814 0.000 ***
s/O 0.0042 9 0.0005
OS 0.0025 1 0.0025 5.410 0.045 *
s/O 0.0042 9 0.0005
time 0.0192 1 0.0192 93.221 0.000 ***
ts/O 0.0019 9 0.0002
Ot 0.0005 1 0.0005 2.249 0.168
ts/O 0.0019 9 0.0002
difficu 0.0135 2 0.0068 16.519 0.000 ***
ds/O 0.0074 18 0.0004
Od 0.0003 2 0.0002 0.420 0.663
ds/O 0.0074 18 0.0004
td 0.0016 2 0.0008 2.280 0.131
tds/O 0.0064 18 0.0004
Otd 0.0002 2 0.0001 0.221 0.804
tds/O 0.0064 18 0.0004
Let's look at the linear relation between reaction time and errors:
dm s6 s5 < data | regress -p rt errors
Analysis for 66 points of 2 variables:
Variable rt errors
Min 103.0000 -5.0000
Max 999.0000 5.0000
Sum 24293.0000 -43.0000
Mean 368.0758 -0.6515
SD 236.5611 2.5267
Correlation Matrix:
rt 1.0000
errors 0.6440 1.0000
Variable rt errors
Regression Equation for rt:
rt = 60.3 errors + 407.36
Significance test for prediction of rt
Mult-R R-Squared F(1,64) prob (F)
0.6440 0.4148 45.3611 0.0000
Significance test(s) for predictor(s) of rt
Predictor beta b Rsq t(64) F(1,64) p
errors 0.6440 60.2968 0.0000 6.7351 45.3611 0.0000
We can get similar information from pair as there are only two variables:
dm s6 s5 < data | pair -ps -x rt -y errors
rt errors Difference
Minimums 103.0000 -5.0000 106.0000
Maximums 999.0000 5.0000 994.0000
Sums 24293.0000 -43.0000 24336.0000
SumSquares 12579139.0000 443.0000 12561192.0000
Means 368.0758 -0.6515 368.7273
SDs 236.5611 2.5267 234.9417
t(65) 12.6405 -2.0948 12.7502
p 0.0000 0.0401 0.0000
Correlation r-squared t(64) p
0.6440 0.4148 6.7351 0.0000
Intercept Slope
-3.1835 0.0069
|--------------------------------------------------|5
| 2|
| |
| 1 1 1 |
| |
| 1 2 |
| |
| 1 1 1 1 1 2 |
| |
| 11 1 1 1 |
| |
| 11 1 11 1 1 1 |
| |
| 1 111 1 1 1 1 11 |
| |
| 11 112 1 21 2 |
| |
|1 1112 11 |
| |
| 11 3 |
| 21 |
|--------------------------------------------------|-5
103.000 999.000
We can look at the skew of the distribution or RT's after taking the log:
dm 'Log(x6)' < data | desc -so -h -i 0.1 -m 2 -M 3 -cfp
------------------------------------------------------------
Under Range In Range Over Range Sum
0 66 0 164.330
------------------------------------------------------------
Mean Median Midpoint Geometric Harmonic
2.490 2.462 2.506 2.477 2.465
------------------------------------------------------------
SD Quart Dev Range SE mean
0.254 0.196 0.987 0.031
------------------------------------------------------------
Minimum Quartile 1 Quartile 2 Quartile 3 Maximum
2.013 2.290 2.462 2.682 3.000
------------------------------------------------------------
Skew Kurtosis
0.346 2.225
------------------------------------------------------------
Midpt Freq Cum Prop Cum
2.050 1 1 0.015 0.015 *
2.150 8 9 0.121 0.136 ********
2.250 9 18 0.136 0.273 *********
2.350 7 25 0.106 0.379 *******
2.450 12 37 0.182 0.561 ************
2.550 8 45 0.121 0.682 ********
2.650 5 50 0.076 0.758 *****
2.750 7 57 0.106 0.864 *******
2.850 3 60 0.045 0.909 ***
2.950 6 66 0.091 1.000 ******
Here's a strange way to get a mean:
dm x6 < data | transpose | dm "'MEAN ='" SUM/N
MEAN = 368.076
Hope these examples are useful.
You might want to try out the textbook examples
to make sure things are working right.
Gary Perlman