1--- 2title: "Conversion semantics" 3output: rmarkdown::html_vignette 4vignette: > 5 %\VignetteIndexEntry{Conversion semantics} 6 %\VignetteEngine{knitr::rmarkdown} 7 %\VignetteEncoding{UTF-8} 8--- 9 10```{r, include = FALSE} 11library(haven) 12knitr::opts_chunk$set( 13 collapse = TRUE, 14 comment = "#>" 15) 16``` 17 18There are some differences between the way that R, SAS, SPSS, and Stata represented labelled data and missing values. While SAS, SPSS, and Stata share some obvious similarities, R is little different. This vignette explores the differences, and shows you how haven bridges the gap. 19 20## Value labels 21 22Base R has one data type that effectively maintains a mapping between integers and character labels: the factor. This however, is not the primary use of factors: they are instead designed to automatically generate useful contrasts for linear models. Factors differ from the labelled values provided by the other tools in important ways: 23 24* SPSS and SAS can label numeric and character values, not just 25 integer values. 26 27* The value do not need to be exhaustive. It is common to label the 28 special missing values (e.g. `.D` = did not respond, `.N` = 29 not applicable), while leaving other values as is. 30 31Value labels in SAS are a little different again. In SAS, labels are just special case of general formats. Formats include currencies and dates, but user-defined just assigns labels to individual values (including special missings value). Formats have names and existing independently of the variables they are associated with. You create a named format with `PROC FORMAT` and then associated with variables in a `DATA` step (the names of character formats thealways start with `$`). 32 33### `labelled()` 34 35To allow you to import labelled vectors into R, haven provides the S3 labelled class, created with `labelled()`. This class allows you to associated arbitrary labels with numeric or character vectors: 36 37```{r} 38x1 <- labelled( 39 sample(1:5), 40 c(Good = 1, Bad = 5) 41) 42x1 43 44x2 <- labelled( 45 c("M", "F", "F", "F", "M"), 46 c(Male = "M", Female = "F") 47) 48x2 49``` 50 51The goal of haven is not to provide a labelled vector that you can use everywhere in your analysis. The goal is to provide an intermediate datastructure that you can convert into a regular R data frame. You can do this by either converting to a factor or stripping the labels: 52 53```{r} 54as_factor(x1) 55zap_labels(x1) 56 57as_factor(x2) 58zap_labels(x2) 59``` 60 61See the documentation for `as_factor()` for more options to control exactly what the factor uses for levels. 62 63Both `as_factor()` and `zap_labels()` have data frame methods if you want to apply the same strategy to every column in a data frame: 64 65```{r} 66df <- tibble::data_frame(x1, x2, z = 1:5) 67df 68 69zap_labels(df) 70as_factor(df) 71``` 72 73## Missing values 74 75All three tools provide a global "system missing value" which is displayed as `.`. This is roughly equivalent to R's `NA`, although neither Stata nor SAS propagate missingness in numeric comparisons: SAS treats the missing value as the smallest possible number (i.e. `-inf`), and Stata treats it as the largest possible number (i.e. `inf`). 76 77Each tool also provides a mechanism for recording multiple types of missingness: 78 79* Stata has "extended" missing values, `.A` through `.Z`. 80 81* SAS has "special" missing values, `.A` through `.Z` plus `._`. 82 83* SPSS has per-column "user" missing values. Each column can declare 84 up to three distinct values or a range of values (plus one distinct 85 value) that should be treated as missing. 86 87Stata and SAS only support tagged missing values for numeric columns. SPSS supports up to three distinct values for character columns. Generally, operations involving a user-missing type return a system missing value. 88 89Haven models these missing values in two different ways: 90 91* For SAS and Stata, haven provides "tagged" missing values which extend R's 92 regular `NA` to add a single character label. 93 94* For SPSS, haven provides a subclass of `labelled` that also provides 95 user defined values and ranges. 96 97### Tagged missing values 98 99To support Stata's extended and SAS's special missing value, haven implements a tagged NA. It does this by taking advantage of the internal structure of a floating point NA. That allows these values to behave identical to NA in regular R operations, while still preserving the value of the tag. 100 101The R interface for creating with tagged NAs is a little clunky because generally they'll be created by haven for you. But you can create your own with `tagged_na()`: 102 103```{r} 104x <- c(1:3, tagged_na("a", "z"), 3:1) 105x 106``` 107 108Note these tagged NAs behave identically to regular NAs, even when printing. To see their tags, use `print_tagged_na()`: 109 110```{r} 111print_tagged_na(x) 112``` 113 114To test if a value is a tagged NA, use `is_tagged_na()`, and to extract the value of the tag, use `na_tag()`: 115 116```{r} 117is_tagged_na(x) 118is_tagged_na(x, "a") 119 120na_tag(x) 121``` 122 123My expectation is that tagged missings are most often used in conjuction with labels (described below), so labelled vectors print the tags for you, and `as_factor()` knows how to relabel: 124 125```{r} 126y <- labelled(x, c("Not home" = tagged_na("a"), "Refused" = tagged_na("z"))) 127y 128 129as_factor(y) 130``` 131 132### User defined missing values 133 134SPSS's user-defined values work differently to SAS and Stata. Each column can have either up to three distinct values that are considered as missing, or a range. Haven provides `labelled_spss()` as a subclass of `labelled()` to model these additional user-defined missings. 135 136```{r} 137x1 <- labelled_spss(c(1:10, 99), c(Missing = 99), na_value = 99) 138x2 <- labelled_spss(c(1:10, 99), c(Missing = 99), na_range = c(90, Inf)) 139 140x1 141x2 142``` 143 144These objects are somewhat dangerous to work with in R because most R functions don't know those values are missing: 145 146```{r} 147mean(x1) 148``` 149 150Because of that danger, the default behaviour of `read_spss()` is to return regular labelled objects where user-defined missing values have been converted to `NA`s. To get `read_spss()` to return `labelled_spss()` objects, you'll need to set `user_na = TRUE`. 151 152I've defined an `is.na()` method so you can find them yourself: 153 154```{r} 155is.na(x1) 156``` 157 158And the presence of that method does mean many functions with an `na.rm` argument will work correctly: 159 160```{r} 161mean(x1, na.rm = TRUE) 162``` 163 164But generally you should either convert to a factor, convert to regular missing vaues, or strip the all the labels: 165 166```{r} 167as_factor(x1) 168zap_missing(x1) 169zap_labels(x1) 170``` 171 172 173