Tidies a Statistics Canada full-table download (the “CSV for database
loading” form, as the data service mirrors it under
statcan_indigenous/tbl_<pid>): VALUE as a number, and the
quality flags kept beside it in quality.
Value
The data frame with VALUE numeric and a factor column
quality with levels "ok", "E", "F", "x",
"..", "...", of class c("rmbl_statcan", "data.frame").
Details
Statistics Canada flags every estimate: E use with caution,
F too unreliable to be published, x suppressed to meet the
confidentiality requirements of the Statistics Act, .. not available,
... not applicable. An F, x, .. or ...
cell has no value here (NA), whatever the file holds, and nothing in
the package imputes it or works it back from totals and margins: a suppressed
cell is suppressed to protect people, and small First Nations, Metis and
Inuit communities most of all. E values are kept and marked.
The census data tables (98-10-xxxx) come in a wide layout: one column per
member of a dimension ("Indigenous identity (9):Inuk (Inuit)[6]"),
each followed by its Symbol column, and no VALUE. They are made
long first: the dimension becomes a column of member labels
("Indigenous identity (9)"), the values VALUE and the symbols
STATUS, so every Statistics Canada table reads the same way.
A label that names two members of one dimension (the census tables repeat
labels under different parents, and the file drops the hierarchy) gets its
member number from the coordinate: "Without children [3]" and
"Without children [7]". Other labels are left as they are.
Examples
# the layout of a downloaded table, with one cell suppressed
d <- data.frame(REF_DATE = "2019/2022", GEO = "Canada",
`Indigenous identity` = c("First Nations", "Inuk (Inuit)",
"Non-Indigenous"),
VALUE = c("45.8", "", "61.5"), STATUS = c("", "x", ""),
check.names = FALSE)
sctidy(d)
#> ╭── Statistics Canada table, tidied ──────────────────────────────────────╮
#> │ Rows 3 │
#> │ Columns 6 │
#> │ Reference periods 2019/2022 │
#> │ Geographies Canada │
#> │ Quality ok 2, x 1 │
#> │ │
#> │ REF_DATE GEO Indigenous identity VALUE quality │
#> │ --------- ------ ------------------- ----- ------- │
#> │ 2019/2022 Canada First Nations 45.80 ok │
#> │ 2019/2022 Canada Inuk (Inuit) x │
#> │ 2019/2022 Canada Non-Indigenous 61.50 ok │
#> │ │
#> │ VALUE is numeric; a suppressed cell (F, x, .., ...) is NA, with its │
#> │ flag kept in quality. │
#> │ │
#> │ Symbols: E use with caution; F too unreliable to be published; │
#> │ x suppressed (confidentiality); .. not available; ... not applicable. │
#> ╰─────────────────────────────────────────────────────────────────────────╯
if (FALSE) { # interactive()
# needs a data service key: sign in once with dlogin()
t924 <- sctidy("13100924")
}