data-raw/making_mcod_fwf_dicts.R
In mkiang/narcan: Helper functions for working with multiple cause of death data
## Create the MCOD fixed width format dictionary
library(dplyr)
library(tibble)
library(usethis)
library(here)
library(purrr)
library(fs)
## For 1979 to 2004, just use NBER dictionaries
if (!file_exists(here("inst", "extdata", "fwf_dicts_1979_2004.RDS"))) {
fwf_dicts_1979_2004 <- purrr::map_dfr(.x = 1979:2004,
.f = ~ narcan:::.dct_to_fwf_df(.x) %>%
mutate(year = .x))
saveRDS(fwf_dicts_1979_2004,
here("inst", "extdata", "fwf_dicts_1979_2004.RDS"),
compress = "xz")
} else {
fwf_dicts_1979_2004 <- readRDS(here("inst", "extdata", "fwf_dicts_1979_2004.RDS"))
}
## Starting in 2005, they added race recode 40 and the public files
## dropped geographic information so we will just repeat. They also added
## a few columns we will add here.
##
## NOTE: The race recode 40 variable exists in the 2005-2011 data BUT IT
## DOES NOT DO ANYTHING. The variable isn't actually available until 2012 so
## this core removes it from the FWF dictionary (at the end of the file)
## just to avoid the confusion.
mcod_fwf_dicts <- bind_rows(
fwf_dicts_1979_2004,
fwf_dicts_1979_2004 %>%
filter(year == 2004) %>%
mutate(year = 2005) %>%
add_case(
name = "racer40",
type = "n",
start = 489,
end = 490,
year = 2005
) %>%
add_case(
name = "tobacco_use",
type = "c",
start = 142,
end = 142,
year = 2005
) %>%
add_case(
name = "pregnancy_status",
type = "n",
start = 143,
end = 143,
year = 2005
) %>%
add_case(
name = "certifier",
type = "c",
start = 110,
end = 110,
year = 2005
)
)
## No changes for 2006 to 2019 so we just repeat
for (y in 2006:2019) {
mcod_fwf_dicts <- mcod_fwf_dicts %>%
bind_rows(
mcod_fwf_dicts %>%
filter(year == 2005) %>%
mutate(year = y)
)
}
## 2020 added occupation
mcod_fwf_dicts <- mcod_fwf_dicts %>%
bind_rows(
mcod_fwf_dicts %>%
filter(year == 2019) %>%
add_case(
name = "occupation",
type = "c",
start = 806,
end = 809
) %>%
add_case(
name = "occupationr",
type = "c",
start = 810,
end = 811
) %>%
add_case(
name = "industry",
type = "c",
start = 812,
end = 815
) %>%
add_case(
name = "industryr",
type = "c",
start = 816,
end = 817
) %>%
mutate(year = 2020)
)
## 2021 had dramatic changes. Namely, they no longer use bridged race,
## they also removed population information and education coding, there
## is no more hispanic recoding, they added race checkboxes, they moved the
## location of race recode 40, etc. To keep this transparent, I copied the
## raw tribble information here, which just shows the changes from 2020 to
## 2021.
##
## NOTE: While the data dictionary shows the following columns the
## ACTUAL DATA FILE DOES NOT SEEM TO MATCH THE DICTIONARY. Need to use the
## 2020 data file so just going to comment this out for now.
# mcod_fwf_dicts <- mcod_fwf_dicts %>%
# bind_rows(
# tibble::tribble(
# ~name, ~type, ~start, ~end,
# "rectype", "n", 19, 19,
# "restatus", "n", 20, 20,
# "stateoc", "c", 21, 22,
# "countyoc", "c", 21, 25,
# "exstatoc", "c", 26, 27,
# "popsizoc", "n", 28, 28,
# "staters", "c", 29, 30,
# "statersr", "c", 33, 34,
# "countyrs", "c", 33, 37,
# "cityrs", "c", 38, 42,
# "popsize", "c", 43, 43,
# "metro", "c", 44, 44,
# "exstares", "c", 45, 46,
# # "pmsares", "c", 47, 50,
# "popsizrs", "c", 51, 51,
# # "popmsa", "c", 52, 52,
# # "cmsares", "n", 53, 54,
# "statbth", "c", 55, 56,
# "statbthr", "c", 59, 60,
# # "educ89", "n", 61, 62,
# "educ", "n", 63, 63,
# "educflag", "n", 64, 64,
# "monthdth", "n", 65, 66,
# "sex", "c", 69, 69,
# "age", "n", 70, 73,
# "ageflag", "n", 74, 74,
# "ager52", "n", 75, 76,
# "ager27", "n", 77, 78,
# "ager12", "n", 79, 80,
# "ager22", "n", 81, 82,
# "placdth", "n", 83, 83,
# "marstat", "c", 84, 84,
# "weekday", "n", 85, 85,
# "year", "n", 102, 105,
# "injwork", "c", 106, 106,
# "mandeath", "n", 107, 107,
# "methdisp", "c", 108, 108,
# "autopsy", "c", 109, 109,
# "certifier", "c", 110, 110,
# "tobacco_use", "c", 142, 142,
# "pregnancy_status", "n", 143, 143,
# "activity", "n", 144, 144,
# "injury", "n", 145, 145,
# "ucod", "c", 146, 149,
# "ucr358", "n", 150, 152,
# "ucr113", "n", 154, 156,
# "ucr130", "n", 157, 159,
# "ucr39", "n", 160, 161,
# "eanum", "n", 163, 164,
# "econdp_1", "n", 165, 165,
# "econds_1", "n", 166, 166,
# "enicon_1", "c", 167, 170,
# "econdp_2", "n", 172, 172,
# "econds_2", "n", 173, 173,
# "enicon_2", "c", 174, 177,
# "econdp_3", "n", 179, 179,
# "econds_3", "n", 180, 180,
# "enicon_3", "c", 181, 184,
# "econdp_4", "n", 186, 186,
# "econds_4", "n", 187, 187,
# "enicon_4", "c", 188, 191,
# "econdp_5", "n", 193, 193,
# "econds_5", "n", 194, 194,
# "enicon_5", "c", 195, 198,
# "econdp_6", "n", 200, 200,
# "econds_6", "n", 201, 201,
# "enicon_6", "c", 202, 205,
# "econdp_7", "n", 207, 207,
# "econds_7", "n", 208, 208,
# "enicon_7", "c", 209, 212,
# "econdp_8", "n", 214, 214,
# "econds_8", "n", 215, 215,
# "enicon_8", "c", 216, 219,
# "econdp_9", "n", 221, 221,
# "econds_9", "n", 222, 222,
# "enicon_9", "c", 223, 226,
# "econdp_10", "n", 228, 228,
# "econds_10", "n", 229, 229,
# "enicon_10", "c", 230, 233,
# "econdp_11", "n", 235, 235,
# "econds_11", "n", 236, 236,
# "enicon_11", "c", 237, 240,
# "econdp_12", "n", 242, 242,
# "econds_12", "n", 243, 243,
# "enicon_12", "c", 244, 247,
# "econdp_13", "n", 249, 249,
# "econds_13", "n", 250, 250,
# "enicon_13", "c", 251, 254,
# "econdp_14", "n", 256, 256,
# "econds_14", "n", 257, 257,
# "enicon_14", "c", 258, 261,
# "econdp_15", "n", 263, 263,
# "econds_15", "n", 264, 264,
# "enicon_15", "c", 265, 268,
# "econdp_16", "n", 270, 270,
# "econds_16", "n", 271, 271,
# "enicon_16", "c", 272, 275,
# "econdp_17", "n", 277, 277,
# "econds_17", "n", 278, 278,
# "enicon_17", "c", 279, 282,
# "econdp_18", "n", 284, 284,
# "econds_18", "n", 285, 285,
# "enicon_18", "c", 286, 289,
# "econdp_19", "n", 291, 291,
# "econds_19", "n", 292, 292,
# "enicon_19", "c", 293, 296,
# "econdp_20", "n", 298, 298,
# "econds_20", "n", 299, 299,
# "enicon_20", "c", 300, 303,
# "ranum", "n", 341, 342,
# "record_1", "c", 344, 347,
# "record_2", "c", 349, 352,
# "record_3", "c", 354, 357,
# "record_4", "c", 359, 362,
# "record_5", "c", 364, 367,
# "record_6", "c", 369, 372,
# "record_7", "c", 374, 377,
# "record_8", "c", 379, 382,
# "record_9", "c", 384, 387,
# "record_10", "c", 389, 392,
# "record_11", "c", 394, 397,
# "record_12", "c", 399, 402,
# "record_13", "c", 404, 407,
# "record_14", "c", 409, 412,
# "record_15", "c", 414, 417,
# "record_16", "c", 419, 422,
# "record_17", "c", 424, 427,
# "record_18", "c", 429, 432,
# "record_19", "c", 434, 437,
# "record_20", "c", 439, 442,
# # "race", "n", 445, 446,
# # "brace", "n", 447, 447,
# "raceimp", "n", 448, 448,
# # "racer3", "n", 449, 449,
# # "racer5", "n", 450, 450,
# "hispanic", "n", 484, 486,
# # "hspanicr", "n", 488, 488,
# # "racer40", "n", 489, 490,
# "check_white", "n", 489, 489,
# "check_black", "n", 490, 490,
# "check_aian", "n", 491, 491,
# "check_asianindian", "n", 492, 492,
# "check_chinese", "n", 493, 493,
# "check_filipino", "n", 494, 494,
# "check_japanese", "n", 495, 495,
# "check_korean", "n", 496, 496,
# "check_vietnamese", "n", 497, 497,
# "check_otherasian", "n", 498, 498,
# "check_hawaiian", "n", 499, 499,
# "check_guamanian", "n", 500, 500,
# "check_samoan", "n", 501, 501,
# "check_otherpi", "n", 502, 502,
# "check_other", "n", 503, 503,
# "racer40", "n", 804, 805,
# "occupation", "c", 806, 809,
# "occupationr", "c", 810, 811,
# "industry", "c", 812, 815,
# "industryr", "c", 816, 817
# ) %>%
# mutate(year = 2021)
# )
mcod_fwf_dicts <- bind_rows(
mcod_fwf_dicts,
mcod_fwf_dicts %>%
filter(year == 2020) %>%
mutate(year = 2021),
mcod_fwf_dicts %>%
filter(year == 2020) %>%
mutate(year = 2022)
)
## Remove racer40 from 2005 to 2011 because the variable exists (i.e., it is
## in the data dictionary) but does not actually contain the values until 2012.
mcod_fwf_dicts <- mcod_fwf_dicts[-which(mcod_fwf_dicts$year < 2012 &
mcod_fwf_dicts$name == "racer40"), ]
mcod_fwf_dicts <- mcod_fwf_dicts %>%
arrange(year, start)
usethis::use_data(mcod_fwf_dicts, overwrite = TRUE)
usethis::use_data(mcod_fwf_dicts, internal = TRUE, overwrite = TRUE)
mkiang/narcan documentation built on Sept. 26, 2024, 7:18 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
## Create the MCOD fixed width format dictionary
library(dplyr)
library(tibble)
library(usethis)
library(here)
library(purrr)
library(fs)
## For 1979 to 2004, just use NBER dictionaries
if (!file_exists(here("inst", "extdata", "fwf_dicts_1979_2004.RDS"))) {
fwf_dicts_1979_2004 <- purrr::map_dfr(.x = 1979:2004,
.f = ~ narcan:::.dct_to_fwf_df(.x) %>%
mutate(year = .x))
saveRDS(fwf_dicts_1979_2004,
here("inst", "extdata", "fwf_dicts_1979_2004.RDS"),
compress = "xz")
} else {
fwf_dicts_1979_2004 <- readRDS(here("inst", "extdata", "fwf_dicts_1979_2004.RDS"))
}
## Starting in 2005, they added race recode 40 and the public files
## dropped geographic information so we will just repeat. They also added
## a few columns we will add here.
##
## NOTE: The race recode 40 variable exists in the 2005-2011 data BUT IT
## DOES NOT DO ANYTHING. The variable isn't actually available until 2012 so
## this core removes it from the FWF dictionary (at the end of the file)
## just to avoid the confusion.
mcod_fwf_dicts <- bind_rows(
fwf_dicts_1979_2004,
fwf_dicts_1979_2004 %>%
filter(year == 2004) %>%
mutate(year = 2005) %>%
add_case(
name = "racer40",
type = "n",
start = 489,
end = 490,
year = 2005
) %>%
add_case(
name = "tobacco_use",
type = "c",
start = 142,
end = 142,
year = 2005
) %>%
add_case(
name = "pregnancy_status",
type = "n",
start = 143,
end = 143,
year = 2005
) %>%
add_case(
name = "certifier",
type = "c",
start = 110,
end = 110,
year = 2005
)
)
## No changes for 2006 to 2019 so we just repeat
for (y in 2006:2019) {
mcod_fwf_dicts <- mcod_fwf_dicts %>%
bind_rows(
mcod_fwf_dicts %>%
filter(year == 2005) %>%
mutate(year = y)
)
}
## 2020 added occupation
mcod_fwf_dicts <- mcod_fwf_dicts %>%
bind_rows(
mcod_fwf_dicts %>%
filter(year == 2019) %>%
add_case(
name = "occupation",
type = "c",
start = 806,
end = 809
) %>%
add_case(
name = "occupationr",
type = "c",
start = 810,
end = 811
) %>%
add_case(
name = "industry",
type = "c",
start = 812,
end = 815
) %>%
add_case(
name = "industryr",
type = "c",
start = 816,
end = 817
) %>%
mutate(year = 2020)
)
## 2021 had dramatic changes. Namely, they no longer use bridged race,
## they also removed population information and education coding, there
## is no more hispanic recoding, they added race checkboxes, they moved the
## location of race recode 40, etc. To keep this transparent, I copied the
## raw tribble information here, which just shows the changes from 2020 to
## 2021.
##
## NOTE: While the data dictionary shows the following columns the
## ACTUAL DATA FILE DOES NOT SEEM TO MATCH THE DICTIONARY. Need to use the
## 2020 data file so just going to comment this out for now.
# mcod_fwf_dicts <- mcod_fwf_dicts %>%
# bind_rows(
# tibble::tribble(
# ~name, ~type, ~start, ~end,
# "rectype", "n", 19, 19,
# "restatus", "n", 20, 20,
# "stateoc", "c", 21, 22,
# "countyoc", "c", 21, 25,
# "exstatoc", "c", 26, 27,
# "popsizoc", "n", 28, 28,
# "staters", "c", 29, 30,
# "statersr", "c", 33, 34,
# "countyrs", "c", 33, 37,
# "cityrs", "c", 38, 42,
# "popsize", "c", 43, 43,
# "metro", "c", 44, 44,
# "exstares", "c", 45, 46,
# # "pmsares", "c", 47, 50,
# "popsizrs", "c", 51, 51,
# # "popmsa", "c", 52, 52,
# # "cmsares", "n", 53, 54,
# "statbth", "c", 55, 56,
# "statbthr", "c", 59, 60,
# # "educ89", "n", 61, 62,
# "educ", "n", 63, 63,
# "educflag", "n", 64, 64,
# "monthdth", "n", 65, 66,
# "sex", "c", 69, 69,
# "age", "n", 70, 73,
# "ageflag", "n", 74, 74,
# "ager52", "n", 75, 76,
# "ager27", "n", 77, 78,
# "ager12", "n", 79, 80,
# "ager22", "n", 81, 82,
# "placdth", "n", 83, 83,
# "marstat", "c", 84, 84,
# "weekday", "n", 85, 85,
# "year", "n", 102, 105,
# "injwork", "c", 106, 106,
# "mandeath", "n", 107, 107,
# "methdisp", "c", 108, 108,
# "autopsy", "c", 109, 109,
# "certifier", "c", 110, 110,
# "tobacco_use", "c", 142, 142,
# "pregnancy_status", "n", 143, 143,
# "activity", "n", 144, 144,
# "injury", "n", 145, 145,
# "ucod", "c", 146, 149,
# "ucr358", "n", 150, 152,
# "ucr113", "n", 154, 156,
# "ucr130", "n", 157, 159,
# "ucr39", "n", 160, 161,
# "eanum", "n", 163, 164,
# "econdp_1", "n", 165, 165,
# "econds_1", "n", 166, 166,
# "enicon_1", "c", 167, 170,
# "econdp_2", "n", 172, 172,
# "econds_2", "n", 173, 173,
# "enicon_2", "c", 174, 177,
# "econdp_3", "n", 179, 179,
# "econds_3", "n", 180, 180,
# "enicon_3", "c", 181, 184,
# "econdp_4", "n", 186, 186,
# "econds_4", "n", 187, 187,
# "enicon_4", "c", 188, 191,
# "econdp_5", "n", 193, 193,
# "econds_5", "n", 194, 194,
# "enicon_5", "c", 195, 198,
# "econdp_6", "n", 200, 200,
# "econds_6", "n", 201, 201,
# "enicon_6", "c", 202, 205,
# "econdp_7", "n", 207, 207,
# "econds_7", "n", 208, 208,
# "enicon_7", "c", 209, 212,
# "econdp_8", "n", 214, 214,
# "econds_8", "n", 215, 215,
# "enicon_8", "c", 216, 219,
# "econdp_9", "n", 221, 221,
# "econds_9", "n", 222, 222,
# "enicon_9", "c", 223, 226,
# "econdp_10", "n", 228, 228,
# "econds_10", "n", 229, 229,
# "enicon_10", "c", 230, 233,
# "econdp_11", "n", 235, 235,
# "econds_11", "n", 236, 236,
# "enicon_11", "c", 237, 240,
# "econdp_12", "n", 242, 242,
# "econds_12", "n", 243, 243,
# "enicon_12", "c", 244, 247,
# "econdp_13", "n", 249, 249,
# "econds_13", "n", 250, 250,
# "enicon_13", "c", 251, 254,
# "econdp_14", "n", 256, 256,
# "econds_14", "n", 257, 257,
# "enicon_14", "c", 258, 261,
# "econdp_15", "n", 263, 263,
# "econds_15", "n", 264, 264,
# "enicon_15", "c", 265, 268,
# "econdp_16", "n", 270, 270,
# "econds_16", "n", 271, 271,
# "enicon_16", "c", 272, 275,
# "econdp_17", "n", 277, 277,
# "econds_17", "n", 278, 278,
# "enicon_17", "c", 279, 282,
# "econdp_18", "n", 284, 284,
# "econds_18", "n", 285, 285,
# "enicon_18", "c", 286, 289,
# "econdp_19", "n", 291, 291,
# "econds_19", "n", 292, 292,
# "enicon_19", "c", 293, 296,
# "econdp_20", "n", 298, 298,
# "econds_20", "n", 299, 299,
# "enicon_20", "c", 300, 303,
# "ranum", "n", 341, 342,
# "record_1", "c", 344, 347,
# "record_2", "c", 349, 352,
# "record_3", "c", 354, 357,
# "record_4", "c", 359, 362,
# "record_5", "c", 364, 367,
# "record_6", "c", 369, 372,
# "record_7", "c", 374, 377,
# "record_8", "c", 379, 382,
# "record_9", "c", 384, 387,
# "record_10", "c", 389, 392,
# "record_11", "c", 394, 397,
# "record_12", "c", 399, 402,
# "record_13", "c", 404, 407,
# "record_14", "c", 409, 412,
# "record_15", "c", 414, 417,
# "record_16", "c", 419, 422,
# "record_17", "c", 424, 427,
# "record_18", "c", 429, 432,
# "record_19", "c", 434, 437,
# "record_20", "c", 439, 442,
# # "race", "n", 445, 446,
# # "brace", "n", 447, 447,
# "raceimp", "n", 448, 448,
# # "racer3", "n", 449, 449,
# # "racer5", "n", 450, 450,
# "hispanic", "n", 484, 486,
# # "hspanicr", "n", 488, 488,
# # "racer40", "n", 489, 490,
# "check_white", "n", 489, 489,
# "check_black", "n", 490, 490,
# "check_aian", "n", 491, 491,
# "check_asianindian", "n", 492, 492,
# "check_chinese", "n", 493, 493,
# "check_filipino", "n", 494, 494,
# "check_japanese", "n", 495, 495,
# "check_korean", "n", 496, 496,
# "check_vietnamese", "n", 497, 497,
# "check_otherasian", "n", 498, 498,
# "check_hawaiian", "n", 499, 499,
# "check_guamanian", "n", 500, 500,
# "check_samoan", "n", 501, 501,
# "check_otherpi", "n", 502, 502,
# "check_other", "n", 503, 503,
# "racer40", "n", 804, 805,
# "occupation", "c", 806, 809,
# "occupationr", "c", 810, 811,
# "industry", "c", 812, 815,
# "industryr", "c", 816, 817
# ) %>%
# mutate(year = 2021)
# )
mcod_fwf_dicts <- bind_rows(
mcod_fwf_dicts,
mcod_fwf_dicts %>%
filter(year == 2020) %>%
mutate(year = 2021),
mcod_fwf_dicts %>%
filter(year == 2020) %>%
mutate(year = 2022)
)
## Remove racer40 from 2005 to 2011 because the variable exists (i.e., it is
## in the data dictionary) but does not actually contain the values until 2012.
mcod_fwf_dicts <- mcod_fwf_dicts[-which(mcod_fwf_dicts$year < 2012 &
mcod_fwf_dicts$name == "racer40"), ]
mcod_fwf_dicts <- mcod_fwf_dicts %>%
arrange(year, start)
usethis::use_data(mcod_fwf_dicts, overwrite = TRUE)
usethis::use_data(mcod_fwf_dicts, internal = TRUE, overwrite = TRUE)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.