NSDUH Data Example (#15)

* Data creation program

* Add docs

* Add to readme

* Update labels, zap formats

Author: szimmer

R/data.R

@@ -287,6 +287,35 @@
 #' @source \url{https://electionstudies.org/data-center/2020-time-series-study/}
 "anes_2020"
 
+#' @title NATIONAL SURVEY ON DRUG USE AND HEALTH (NSDUH) (2023) data
+#' @description A subset of variables from the NSDUH 2023 Public Use File
+#' @format A data frame with 56705 rows and 22 variables:
+#' \describe{
+#'    \item{\code{QUESTID2}}{double RESPONDENT IDENTIFICATION}
+#'    \item{\code{ANALWT2_C}}{double FIN PRSN-LEVEL SMPLE WGHT 2}
+#'    \item{\code{VESTR_C}}{double VARIANCE STRATUM}
+#'    \item{\code{VEREP}}{double VARIANCE PRIMARY SAMPLING UNIT}
+#'    \item{\code{NICVAPMON}}{integer RC-NICOTINE VAPING - PAST MONTH USE}
+#'    \item{\code{TOBMON}}{integer RC-ANY TOBACCO - PAST MONTH USE}
+#'    \item{\code{ALCMON}}{integer RC-ALCOHOL - PAST MONTH USE}
+#'    \item{\code{ILLMON}}{integer RC-ANY ILLICIT DRUG - PAST MONTH USE}
+#'    \item{\code{ILTOBVAPALC}}{integer RC-TOBACCO, NICOTINE VAPING, ALCOHOL, ILLICIT DRUG - PST MON}
+#'    \item{\code{BNGDRKMON}}{integer RC-BINGE ALCOHOL USE PAST 30 DAYS}
+#'    \item{\code{IRPYUD5ALC}}{integer ALCOHOL USE DISORDER IN THE PAST YEAR - IMP REV}
+#'    \item{\code{UD5ILLANY}}{integer RC-DRUG USE DISORDER - PAST YEAR USERS}
+#'    \item{\code{UD5ILALANY}}{integer RC-DRUG OR ALCOHOL USE DISORDER - PAST YEAR USERS}
+#'    \item{\code{YMDELT}}{factor RC-YOUTH: LIFETIME MAJOR DEPRESSIVE EPISODE (MDE)}
+#'    \item{\code{YMDEYR}}{factor RC-YOUTH: PAST YEAR MAJOR DEPRESSIVE EPISODE (MDE)}
+#'    \item{\code{MDEIMPY}}{factor RC-YOUTH: MDE WITH SEVERE ROLE IMPAIRMENT}
+#'    \item{\code{AMIPY}}{integer RC-IMP AMI IND (1/0) BASED ON PREDICTED SMI PROB PY}
+#'    \item{\code{SMIPY}}{integer RC-IMP SMI IND (1/0) BASED ON PREDICTED SMI PROB PY}
+#'    \item{\code{AGE3}}{factor RECODE - FINAL EDITED AGE}
+#'    \item{\code{NEWRACE2}}{factor RC-RACE/HISPANICITY RECODE (7 LEVELS)}
+#'    \item{\code{IRSEX}}{factor SEX AT BIRTH - IMPUTATION REVISED}
+#'    \item{\code{POVERTY3}}{factor RC-POVERTY LEVEL-NEW INC (% OF US CENSUS POVERTY THRESHOLD)}
+#' }
+#' @source \url{https://www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health/datafiles}
+"nsduh_2023"
 
 #' @title California Health Interview Survey (CHIS) (2023) data
 #' @description A subset of variables from the CHIS 2023 Public Use File
@@ -313,4 +342,4 @@
 #'    \item{\code{RAKEDW1 - RAKEDW80}}{CHIS2023 RAKED WEIGHT - REPLICATE 1 through REPLICATE 80}
 #' }
 #' @source \url{https://healthpolicy.ucla.edu/our-work/public-use-files/one-year-public-use-files-pufs}
-"chis_2023"
+"chis_2023"

README.Rmd

@@ -20,7 +20,6 @@ knitr::opts_chunk$set(
   fig.align = "center",
   tidy = "styler"
 )
-library(tidyverse)
 ```
 
 The **srvyexploR** package provides datasets used in the book [Exploring Complex Survey Data Analysis Using R: A Tidy Introduction with {srvyr} and {survey}](https://tidy-survey-r.github.io/tidy-survey-book/). This will help readers follow along with the examples and work through the exercises.
@@ -77,6 +76,16 @@ head(ncvs_2021_person)
 head(ncvs_2021_incident)
 ```
 
+### NSDUH
+
+The National Survey on Drug Use and Health (NSDUH) is an annual survey of the civilian, non-institutionalized population in the United States who are at least 12 years old. Topics include substance use (tobacco, alcohol, and illicit drugs including marijuana), mental health, and general health. This package provides a subset of the variables from the 2023 Public Use File. For more details about the study and the data, refer to the [Methodological Summary and Definitions](https://www.samhsa.gov/data/sites/default/files/reports/rpt47098/Methodological%20Summary%20and%20Definitions/2023-nsduh-method-summary-defs.pdf), [Data User's Guide](https://www.samhsa.gov/data/sites/default/files/reports/rpt56198/2023-nsduh-puf-data-users-guide.pdf), and [Codebook](https://www.samhsa.gov/data/system/files/media-puf-file/NSDUH-2023-DS0001-info-codebook_v1.pdf).
+
+```{r}
+#| label: nsduh-show
+
+head(nsduh_2023)
+```
+
 ### RECS
 
 Three files are included associated with RECS - a dataset with the 2015 data with some derived variables created for the book (`recs_2015`), the 2020 data with some derived variables created for the book (`recs_2020`), and the 2020 data with the original variables (`recs_2020_raw`). RECS is a survey about energy consumption and expenditure among residential households in the United States and has been conducted since 1979 by the Energy Information Administration. More information about the original data is available at the [RECS website](https://www.eia.gov/consumption/residential/data/2020/).
@@ -152,7 +161,7 @@ Anyone interested in redistributing the ANES data should refer to the [ANES FAQ
 
 ANES: 
 
-+ American National Election Studies. 2021. ANES 2020 Time Series Study Full Release [dataset and documentation]. July 19, 2021 version. https://www.electionstudies.org
++ American National Election Studies, 2021. ANES 2020 Time Series Study Full Release [dataset and documentation]. July 19, 2021 version. https://www.electionstudies.org
 
 CHIS:
 
@@ -162,7 +171,12 @@ NCVS:
 
 + United States. Bureau of Justice Statistics. National Crime Victimization Survey, [United States], 2021. Inter-university Consortium for Political and Social Research [distributor], 2022-09-19. https://doi.org/10.3886/ICPSR38429.v1
 
+NSDUH:
+
++ Center for Behavioral Health Statistics and Quality, 2025. 2023 National Survey on Drug Use
+and Health: Public use file data users’ guide. https://www.samhsa.gov/data/data-wecollect/nsduh/datafiles
+
 RECS:
 
-+ U.S. Energy Information Administration. 2024. Residential Energy Consumption 2020 Survey Data. [dataset and documentation]. January 2024 version. https://www.eia.gov/consumption/residential/data/2020/index.php?view=microdata
-+ U.S. Energy Information Administration. 2018 Residential Energy Consumption 2015 Survey Data. [dataset and documentation]. December 2018 version. https://www.eia.gov/consumption/residential/data/2015/index.php?view=microdata 
++ U.S. Energy Information Administration, 2024. Residential Energy Consumption 2020 Survey Data. [dataset and documentation]. January 2024 version. https://www.eia.gov/consumption/residential/data/2020/index.php?view=microdata
++ U.S. Energy Information Administration, 2018 Residential Energy Consumption 2015 Survey Data. [dataset and documentation]. December 2018 version. https://www.eia.gov/consumption/residential/data/2015/index.php?view=microdata 

README.md

@@ -47,22 +47,76 @@ Once the package is loaded, you can use the data immediately as follows:
 
 ``` r
 head(anes_2020)
-#> # A tibble: 6 × 65
-#>   V200001 CaseID V200002 InterviewMode V200010b Weight V200010c VarUnit V200010d
-#>     <dbl>  <dbl> <hvn_l> <fct>            <dbl>  <dbl>    <dbl> <fct>      <dbl>
-#> 1  200015 200015       3 Web              1.01   1.01         2 2              9
-#> 2  200022 200022       3 Web              1.16   1.16         2 2             26
-#> 3  200039 200039       3 Web              0.769  0.769        1 1             41
-#> 4  200046 200046       3 Web              0.521  0.521        2 2             29
-#> 5  200053 200053       3 Web              0.966  0.966        1 1             23
-#> 6  200060 200060       3 Web              0.235  0.235        2 2             37
-#> # ℹ 56 more variables: Stratum <fct>, V201006 <hvn_lbll>,
-#> #   CampaignInterest <fct>, V201023 <hvn_lbll>, EarlyVote2020 <fct>,
-#> #   V201024 <hvn_lbll>, V201025x <hvn_lbll>, V201028 <hvn_lbll>,
-#> #   V201029 <hvn_lbll>, V201101 <hvn_lbll>, V201102 <hvn_lbll>,
-#> #   VotedPres2016 <fct>, V201103 <hvn_lbll>, VotedPres2016_selection <fct>,
-#> #   V201228 <hvn_lbll>, V201229 <hvn_lbll>, V201230 <hvn_lbll>,
-#> #   V201231x <hvn_lbll>, PartyID <fct>, V201233 <hvn_lbll>, …
+#>   V200001 CaseID V200002 InterviewMode  V200010b    Weight V200010c VarUnit
+#> 1  200015 200015       3           Web 1.0057375 1.0057375        2       2
+#> 2  200022 200022       3           Web 1.1634731 1.1634731        2       2
+#> 3  200039 200039       3           Web 0.7686811 0.7686811        1       1
+#> 4  200046 200046       3           Web 0.5210195 0.5210195        2       2
+#> 5  200053 200053       3           Web 0.9657892 0.9657892        1       1
+#> 6  200060 200060       3           Web 0.2347078 0.2347078        2       2
+#>   V200010d Stratum V201006     CampaignInterest V201023 EarlyVote2020 V201024
+#> 1        9       9       2  Somewhat interested      -1          <NA>      -1
+#> 2       26      26       3  Not much interested      -1          <NA>      -1
+#> 3       41      41       2  Somewhat interested      -1          <NA>      -1
+#> 4       29      29       3  Not much interested      -1          <NA>      -1
+#> 5       23      23       2  Somewhat interested      -1          <NA>      -1
+#> 6       37      37       1 Very much interested      -1          <NA>      -1
+#>   V201025x V201028 V201029 V201101 V201102 VotedPres2016 V201103
+#> 1        3      -1      -1      -1       1           Yes       2
+#> 2        3      -1      -1      -1       1           Yes       5
+#> 3        3      -1      -1      -1       1           Yes       1
+#> 4        3      -1      -1      -1       1           Yes       1
+#> 5        3      -1      -1      -1       1           Yes       2
+#> 6        3      -1      -1      -1       2            No      -1
+#>   VotedPres2016_selection V201228 V201229 V201230 V201231x
+#> 1                   Trump       2       1      -1        7
+#> 2                   Other       5      -1       2        4
+#> 3                 Clinton       3      -1       3        3
+#> 4                 Clinton       2       2      -1        6
+#> 5                   Trump       3      -1       2        4
+#> 6                    <NA>       3      -1       3        3
+#>                      PartyID V201233     TrustGovernment V201237
+#> 1          Strong republican       5               Never       3
+#> 2                Independent       5               Never       4
+#> 3       Independent-democrat       4    Some of the time       4
+#> 4 Not very strong republican       3 About half the time       2
+#> 5                Independent       5               Never       4
+#> 6       Independent-democrat       4    Some of the time       2
+#>           TrustPeople V201507x Age    AgeGroup V201510   Education V201546
+#> 1 About half the time       46  46       40-49       6  Bachelor's       1
+#> 2    Some of the time       37  37       30-39       3     Post HS       2
+#> 3    Some of the time       40  40       40-49       2 High school       2
+#> 4    Most of the time       41  41       40-49       4     Post HS       2
+#> 5    Some of the time       72  72 70 or older       8    Graduate       2
+#> 6    Most of the time       71  71 70 or older       3     Post HS       2
+#>   V201547a V201547b V201547c V201547d V201547e V201547z V201549x      RaceEth
+#> 1       -3       -3       -3       -3       -3       -3        3     Hispanic
+#> 2       -3       -3       -3       -3       -3       -3        4 Asian, NH/PI
+#> 3       -3       -3       -3       -3       -3       -3        1        White
+#> 4       -3       -3       -3       -3       -3       -3        4 Asian, NH/PI
+#> 5       -3       -3       -3       -3       -3       -3        5        AI/AN
+#> 6       -3       -3       -3       -3       -3       -3        1        White
+#>   V201600 Gender V201607 V201610 V201611 V201613 V201615 V201616 V201617x
+#> 1       1   Male      -3      -3      -3      -3      -3      -3       21
+#> 2       2 Female      -3      -3      -3      -3      -3      -3       13
+#> 3       2 Female      -3      -3      -3      -3      -3      -3       17
+#> 4       1   Male      -3      -3      -3      -3      -3      -3        7
+#> 5       1   Male      -3      -3      -3      -3      -3      -3       22
+#> 6       2 Female      -3      -3      -3      -3      -3      -3        3
+#>             Income         Income7 V202051 V202066 V202072 VotedPres2020
+#> 1 $175,000-249,999   $125k or more      -1       1      -1          <NA>
+#> 2   $70,000-74,999   $60k to < 80k      -1       4       1           Yes
+#> 3 $100,000-109,999 $100k to < 125k      -1       4       1           Yes
+#> 4   $35,000-39,999   $20k to < 40k      -1       4       1           Yes
+#> 5 $250,000 or more   $125k or more      -1       4       1           Yes
+#> 6   $15,000-19,999      Under $20k      -1       4       1           Yes
+#>   V202073 V202109x V202110x VotedPres2020_selection
+#> 1      -1        0       -1                    <NA>
+#> 2       3        1        3                   Other
+#> 3       1        1        1                   Biden
+#> 4       1        1        1                   Biden
+#> 5       2        1        2                   Trump
+#> 6       1        1        1                   Biden
 ```
 
 See `?anes_2020` for more information about the data.
@@ -129,6 +183,37 @@ head(ncvs_2021_incident)
 #> #   V4267 <fct>, V4268 <fct>, V4269 <fct>, V4270 <fct>, V4271 <fct>, …
 ```
 
+### NSDUH
+
+The National Survey on Drug Use and Health (NSDUH) is an annual survey
+of the civilian, non-institutionalized population in the United States
+who are at least 12 years old. Topics include substance use (tobacco,
+alcohol, and illicit drugs including marijuana), mental health, and
+general health. This package provides a subset of the variables from the
+2023 Public Use File. For more details about the study and the data,
+refer to the [Methodological Summary and
+Definitions](https://www.samhsa.gov/data/sites/default/files/reports/rpt47098/Methodological%20Summary%20and%20Definitions/2023-nsduh-method-summary-defs.pdf),
+[Data User’s
+Guide](https://www.samhsa.gov/data/sites/default/files/reports/rpt56198/2023-nsduh-puf-data-users-guide.pdf),
+and
+[Codebook](https://www.samhsa.gov/data/system/files/media-puf-file/NSDUH-2023-DS0001-info-codebook_v1.pdf).
+
+``` r
+head(nsduh_2023)
+#> # A tibble: 6 × 22
+#>   QUESTID2 ANALWT2_C VESTR_C VEREP NICVAPMON TOBMON ALCMON ILLMON ILTOBVAPALC
+#>      <dbl>     <dbl>   <dbl> <dbl>     <int>  <int>  <int>  <int>       <int>
+#> 1 10000053     3276.   40031     2         0      0      1      0           1
+#> 2 10000679    15630.   40021     2         0      1      1      0           1
+#> 3 10001208     4018.   40043     1         0      1      0      1           1
+#> 4 10001260    10712.   40030     2         0      0      0      0           0
+#> 5 10001588     8195.   40023     2         0      0      1      0           1
+#> 6 10004996     3771.   40048     1         1      1      1      0           1
+#> # ℹ 13 more variables: BNGDRKMON <int>, IRPYUD5ALC <int>, UD5ILLANY <int>,
+#> #   UD5ILALANY <int>, YMDELT <fct>, YMDEYR <fct>, MDEIMPY <fct>, AMIPY <int>,
+#> #   SMIPY <int>, AGE3 <fct>, NEWRACE2 <fct>, IRSEX <fct>, POVERTY3 <fct>
+```
+
 ### RECS
 
 Three files are included associated with RECS - a dataset with the 2015
@@ -309,7 +394,7 @@ Anyone interested in redistributing the ANES data should refer to the
 
 ANES:
 
-- American National Election Studies. 2021. ANES 2020 Time Series Study
+- American National Election Studies, 2021. ANES 2020 Time Series Study
   Full Release \[dataset and documentation\]. July 19, 2021 version.
   <https://www.electionstudies.org>
 
@@ -326,13 +411,19 @@ NCVS:
   Consortium for Political and Social Research \[distributor\],
   2022-09-19. <https://doi.org/10.3886/ICPSR38429.v1>
 
+NSDUH:
+
+- Center for Behavioral Health Statistics and Quality, 2025. 2023
+  National Survey on Drug Use and Health: Public use file data users’
+  guide. <https://www.samhsa.gov/data/data-wecollect/nsduh/datafiles>
+
 RECS:
 
-- U.S. Energy Information Administration. 2024. Residential Energy
+- U.S. Energy Information Administration, 2024. Residential Energy
   Consumption 2020 Survey Data. \[dataset and documentation\]. January
   2024 version.
   <https://www.eia.gov/consumption/residential/data/2020/index.php?view=microdata>
-- U.S. Energy Information Administration. 2018 Residential Energy
+- U.S. Energy Information Administration, 2018 Residential Energy
   Consumption 2015 Survey Data. \[dataset and documentation\]. December
   2018 version.
   <https://www.eia.gov/consumption/residential/data/2015/index.php?view=microdata>

data-raw/nsduh_2023.R

@@ -0,0 +1,100 @@
+## code to prepare `nsduh_2023` dataset goes here
+
+library(tidyverse)
+
+if (!dir.exists("nsduh-temp")) dir.create(here::here("nsduh-temp"))
+download.file(
+  "https://www.samhsa.gov/data/system/files/media-puf-file/NSDUH-2023-DS0001-bndl-data-r_v1.zip",
+  here::here("nsduh-temp", "nsduh-2023.zip")
+)
+
+unzip(here::here("nsduh-temp", "nsduh-2023.zip"),
+  exdir = here::here("nsduh-temp")
+)
+
+load(here::here("nsduh-temp", "NSDUH_2023.Rdata"))
+
+varinfo <- tibble(
+  Variables = names(puf2023_102124),
+  Label = map_chr(names(puf2023_102124), ~ attr(puf2023_102124[[.x]], "label"))
+)
+
+openxlsx2::write_xlsx(varinfo, here::here("nsduh-temp", "variable-list.xlsx"))
+
+fct_yesno_12 <- function(x) {
+  factor(
+    if_else(x %in% c(1:2), x, NA),
+    labels = c("Yes", "No")
+  )
+}
+
+nsduh_slim <- puf2023_102124 %>%
+  select(
+    QUESTID2, ANALWT2_C, VESTR_C, VEREP,
+    NICVAPMON, TOBMON, ALCMON, ILLMON, ILTOBVAPALC, BNGDRKMON,
+    IRPYUD5ALC, UD5ILLANY, UD5ILALANY,
+    YMDELT, YMDEYR, MDEIMPY,
+    AMIPY, SMIPY,
+    AGE3, NEWRACE2, IRSEX, POVERTY3
+  ) %>%
+  mutate(
+    across(c(NICVAPMON, TOBMON, ALCMON, ILLMON, ILTOBVAPALC, BNGDRKMON, IRPYUD5ALC, UD5ILLANY, UD5ILALANY, AMIPY, SMIPY), as.integer),
+    across(c(YMDELT, YMDEYR, MDEIMPY), fct_yesno_12),
+    AGE3 = factor(AGE3, labels = c("12-13", "14-15", "16-17", "18-20", "21-23", "24-25", "26-29", "30-34", "35-49", "50-64", "65+")),
+    NEWRACE2 = factor(NEWRACE2, labels = c("White, NH", "Black, NH", "Native Am/AK Native, NH", "Native HI/PI, NH", "Asian, NH", "More than one race, NH", "Other")),
+    IRSEX = factor(IRSEX, labels = c("Male", "Female")),
+    POVERTY3 = factor(POVERTY3, labels = c("0-100% FPL", "101-200% FPL", "201%+ FPL"))
+  )
+
+check_vars <- function(var) {
+  message(var)
+  table(nsduh_slim[[var]], puf2023_102124[[var]], useNA = "ifany") %>%
+    print()
+}
+
+nsduh_slim %>%
+  select(where(is.factor)) %>%
+  names() %>%
+  walk(check_vars)
+
+nsduh_slim %>%
+  select(-where(is.factor))
+
+# Update labels and zap formats
+
+update_label <- function(var) {
+  attr(nsduh_slim[[var]], "label") <<- attr(puf2023_102124[[var]], "label")
+}
+
+walk(names(nsduh_slim), update_label)
+
+haven::zap_formats(nsduh_slim)
+
+str(nsduh_slim)
+
+nsduh_slim_md <- tibble(
+  Variable = names(nsduh_slim),
+  Class = sapply(nsduh_slim, class),
+  Label = map_chr(names(nsduh_slim), \(x) attr(nsduh_slim[[x]], "label"))
+) %>%
+  mutate(
+    Class2 = map_chr(Class, ~ str_flatten(.x, collapse = ";")),
+    Class2 = if_else(Class2 == "numeric", "double", Class2)
+  )
+
+nsduh_slim_md %>%
+  mutate(
+    roxy = str_c("#'    \\item{\\code{", Variable, "}}{", Class2, " ", Label, "}")
+  ) %>%
+  pull(roxy) %>%
+  cat(sep = "\n")
+
+nsduh_2023 <- nsduh_slim
+
+summary(nsduh_2023)
+nrow(nsduh_2023)
+ncol(nsduh_2023)
+
+usethis::use_data(nsduh_2023, overwrite = TRUE)
+
+unlink(here::here("nsduh-temp"), recursive = TRUE)