1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/variableKey.R
\name{keyDiagnostic}
\alias{keyDiagnostic}
\title{Diagnose accuracy of result from applying variable key to data}
\usage{
keyDiagnostic(
dfold,
dfnew,
keylist,
max.values = 20,
nametrunc = 18,
wide = 200,
confidential = FALSE
)
}
\arguments{
\item{dfold}{Original data frame}
\item{dfnew}{The new recoded data frame}
\item{keylist}{The imported variable key that was used to
transform dfold into dfnew.}
\item{max.values}{Show up to this number of values for the old
variable}
\item{nametrunc}{Truncate column and row names. Needed if there
are long factor labels and we want to fit more information on
table. Default = 18 for new name, old name is 10 more
characters (18 + 10 = 28).}
\item{wide}{Number of characters per row in printed
output. Suggest very wide screen, default = 200.}
\item{confidential}{Should numbers in table be rounded to nearest
"10" to comply with security standard enforced by some
American research departments.}
}
\description{
Compare the old and new data frames, checking for accuracy of
calculations in various ways.
}
\details{
CAUTION: This can print WIDE matrices. Because the on-screen
output will be WIDE, make the display window WIDE!
Crosstabulate new variable versus old variable to see the coding
mismatches. For tables of up to 10 values or so, that will be
satisfactory.
For numeric variables, it appears there is no good thing to do
except possibly to re-apply any transformations.
}
\author{
Paul Johnson <pauljohn@ku.edu>
}
|
