1DBCOLSTATSCORES(1) User Contributed Perl Documentation DBCOLSTATSCORES(1)
2
3
4
6 dbcolstatscores - compute z-scores or t-scores for each value in a
7 population
8
10 dbcolstatscores [-t] [--tmean=MEAN] [--tstddev=STDDEV] column
11
13 Compute statistics (z-score and optionally t-score) over a COLUMN of
14 numbers. Creates new columns called "zscore", "tscore". T-scores are
15 only computed if requested with the "-t" option, or if "--tmean" or
16 "--tstddev" are explicitly specified (defaults are mean of 50, standard
17 deviation of 10).
18
19 You may recall from your statistics class that a z-score is simply the
20 value normalized by mean and standard deviation, so that 0.0 is the
21 mean and positive or negative values are multiples of the standard
22 deviation. It assumes data follows a normal (Gaussian) distribution.
23
24 T-score scales the z-score to match a mean of 50 and a standard
25 deviation of 10. This program allows generalized t-scores that use any
26 mean and standard deviation.
27
28 Other scales are sometimes used as well. The Wechsler Adult
29 Intelligence Scale (one type of IQ test) is adjusted to a mean of 100
30 and a standard deviation of 15. Other tests scale to other standard
31 deviations.
32
33 This program requires two passes over the data, and consumes O(1)
34 memory and O(number of rows) disk space.
35
37 -a or --include-non-numeric
38 Compute stats over all records (treat non-numeric records as zero
39 rather than just ignoring them).
40
41 -t Compute t-scores in addition to z-scores.
42
43 --tmean MEAN
44 Use the given MEAN for t-scores.
45
46 --tstddev STDDEV or --tsd STDDEV
47 Use the given STDDEV for the standard deviation of the t-scores.
48
49 -f FORMAT or --format FORMAT
50 Specify a printf(3)-style format for output statistics. Defaults
51 to "%.5g".
52
53 -T TmpDir
54 where to put tmp files. Also uses environment variable TMPDIR, if
55 -T is not specified. Default is /tmp.
56
57 This module also supports the standard fsdb options:
58
59 -d Enable debugging output.
60
61 -i or --input InputSource
62 Read from InputSource, typically a file name, or "-" for standard
63 input, or (if in Perl) a IO::Handle, Fsdb::IO or Fsdb::BoundedQueue
64 objects.
65
66 -o or --output OutputDestination
67 Write to OutputDestination, typically a file name, or "-" for
68 standard output, or (if in Perl) a IO::Handle, Fsdb::IO or
69 Fsdb::BoundedQueue objects.
70
71 --autorun or --noautorun
72 By default, programs process automatically, but Fsdb::Filter
73 objects in Perl do not run until you invoke the run() method. The
74 "--(no)autorun" option controls that behavior within Perl.
75
76 --help
77 Show help.
78
79 --man
80 Show full manual.
81
83 Input:
84 #fsdb name id test1
85 a 1 80
86 b 2 70
87 c 3 65
88 d 4 90
89 e 5 70
90 f 6 90
91
92 Command:
93 cat DATA/grades.fsdb | dbcolstatscores --tmean 50 --tstddev 10 test1 | dbcolneaten
94
95 Output:
96 #fsdb name id test1 zscore tscore
97 a 1 80 0.23063 52.306
98 b 2 70 -0.69188 43.081
99 c 3 65 -1.1531 38.469
100 d 4 90 1.1531 61.531
101 e 5 70 -0.69188 43.081
102 f 6 90 1.1531 61.531
103 # | dbcolstatscores --tmean 50 --tstddev 10 test1
104 # | dbcolneaten
105
107 dbcolpercentile(1), dbcolstats(1), Fsdb, dbcolscorrelate
108
110 Copyright (C) 1991-2018 by John Heidemann <johnh@isi.edu>
111
112 This program is distributed under terms of the GNU general public
113 license, version 2. See the file COPYING with the distribution for
114 details.
115
116
117
118perl v5.32.0 2020-11-16 DBCOLSTATSCORES(1)