Dealing with a very unusual value in a variable

[Pratap Khattri]

Hello,

I have been working on cleaning a dataset to practice stata. In the dataset, there is a variable that contains data on land area. I was able to clean every entry in that variable except for two entries that are mixed fractions. I have never seen this before in a dataset and am not sure how I would convert them to decimal format (I would like to convert ½ to 0.5, and so forth). The output is shown below(the rows with the erroneous entries are highlighted in red):

tab 'varname'

Area Freq. Percent Cum.

1 628 25.41 25.41
2 521 21.08 46.50
3 257 10.40 56.90
4 203 8.22 65.12
5 146 5.91 71.02
10 85 3.44 74.46
0.5 81 3.28 77.74
6 72 2.91 80.66
7 44 1.78 82.44
1.5 41 1.66 84.10
20 41 1.66 85.75
50 37 1.50 87.25
8 37 1.50 88.75
15 33 1.34 90.08
100 30 1.21 91.30
12 21 0.85 92.15
30 16 0.65 92.80
9 15 0.61 93.40
2.5 13 0.53 93.93
25 11 0.45 94.37
.5 10 0.40 94.78
18 9 0.36 95.14
40 9 0.36 95.51
0.25 7 0.28 95.79
35 6 0.24 96.03
45 6 0.24 96.28
½ 6 0.24 96.52
.25 5 0.20 96.72
13 5 0.20 96.92
60 5 0.20 97.13
0.75 4 0.16 97.29
1.50 4 0.16 97.45
14 4 0.16 97.61
200 4 0.16 97.77
0.50 3 0.12 97.90
10.5 3 0.12 98.02
11 3 0.12 98.14
16 3 0.12 98.26
. 2 0.08 98.34
150 2 0.08 98.42
17 2 0.08 98.50
2.50 2 0.08 98.58
34 2 0.08 98.66
500 2 0.08 98.75
53 2 0.08 98.83
55 2 0.08 98.91
70 2 0.08 98.99
700 2 0.08 99.07
.8 1 0.04 99.11
0.7 1 0.04 99.15
10000 1 0.04 99.19
120 1 0.04 99.23
1600 1 0.04 99.27
19 1 0.04 99.31
2000 1 0.04 99.35
21 1 0.04 99.39
22 1 0.04 99.43
23 1 0.04 99.47
28 1 0.04 99.51
2½ 1 0.04 99.55
3.5 1 0.04 99.60
32 1 0.04 99.64
42 1 0.04 99.68
450 1 0.04 99.72
4900 1 0.04 99.76
63 1 0.04 99.80
65 1 0.04 99.84
67 1 0.04 99.88
75 1 0.04 99.92
80 1 0.04 99.96
8100 1 0.04 100.00

Total 2,471 100.00

Thanks for your help!

[Andrew Musau]

For your future posts, please familiarize yourself with the dataex command for presenting data examples (see FAQ Advice #12 for details). You will have to replace all fractional values, one at a time.

* Example generated by -dataex-. To install: ssc install dataex
clear
input str7 Area
"2¾"
"3.2"
"5½"
"4"
"7½"
"4¾" 
end

gen wanted=Area
replace wanted= subinstr(wanted, "½", ".5", .)
replace wanted= subinstr(wanted, "¾", ".75", .)
destring wanted, replace

Res.:

. l, sep(0)

     +---------------+
     | Area   wanted |
     |---------------|
  1. |   2¾     2.75 |
  2. |  3.2      3.2 |
  3. |   5½      5.5 |
  4. |    4        4 |
  5. |   7½      7.5 |
  6. |   4¾     4.75 |
     +---------------+