From 2dc534d62fe4d7c03bc70c709b6a40de1e6e11a9 Mon Sep 17 00:00:00 2001
From: Reiko Okamoto <rokamoto@ohri.ca>
Date: Wed, 29 Jul 2026 16:25:01 -0400
Subject: [PATCH 1/4] Add main and helper functions to calculate SSI

---
 R/derive_sii.R                   | 374 +++++++++++++++++++++++++++++++
 tests/testthat/test-derive_sii.R |  32 +++
 2 files changed, 406 insertions(+)
 create mode 100644 R/derive_sii.R
 create mode 100644 tests/testthat/test-derive_sii.R

diff --git a/R/derive_sii.R b/R/derive_sii.R
new file mode 100644
index 0000000..e3706de
--- /dev/null
+++ b/R/derive_sii.R
@@ -0,0 +1,374 @@
+derive_sii <- function(df) {
+  SII <- df |>
+    derive_SO() |>
+    derive_FO() |>
+    derive_FS() |>
+    dplyr::mutate(
+      SII = (.data$SO + .data$FO + .data$FS) / 3
+    ) |>
+    dplyr::select(-dplyr::all_of(c("SO", "FO", "FS")))
+
+  SII
+}
+
+# STRUCTURAL OBJECTIVE
+derive_SO <- function(df) {
+  SO <- df |>
+    derive_A() |>
+    derive_B() |>
+    derive_C() |>
+    derive_D() |>
+    derive_E() |>
+    dplyr::mutate(
+      SO = (.data$A + .data$B + .data$C + .data$D + .data$E) / 5
+    ) |>
+    dplyr::select(-dplyr::all_of(c("A", "B", "C", "D", "E")))
+}
+
+# COMMUNITY PARTICIPATION
+derive_A <- function(df) {
+  expected_cols <- paste0(
+    "SPA_",
+    c("OUTS", "CHRCH", "CLUB", "EDUC", "NEIBR", "OTACT", "SPORT", "VOLUN")
+  )
+  validate_cols(df, expected_cols)
+
+  df |>
+    dplyr::mutate(
+      dplyr::across(
+        .cols = expected_cols,
+        .fns = \(x) {
+          dplyr::case_when(
+            x == 1 ~ 0, # at least once a day
+            x == 2 ~ 2.5, # at least once a week
+            x == 3 ~ 5, # at least once a month
+            x == 4 ~ 7.5, # at least once a year
+            x == 5 ~ 10, # never
+            .default = NA
+          )
+        },
+        .names = "{.col}_mpd"
+      ),
+      A = rowSums(dplyr::pick(dplyr::ends_with("_mpd"))) / 8
+    ) |>
+    dplyr::select(-dplyr::ends_with("_mpd"))
+}
+
+# SOCIAL NETWORK QUANTITY
+derive_B <- function(df) {
+  expected_cols <- paste(
+    "SN",
+    c(
+      "CHILD",
+      "FRND",
+      "NEIBR",
+      "SIBLIV",
+      "RELLIV",
+      "PERWSCH",
+      "PERCOM",
+      "PERACT"
+    ),
+    "NB",
+    sep = "_"
+  )
+  validate_cols(df, expected_cols)
+
+  df |>
+    dplyr::mutate(
+      # all of their min is 0
+      SN_CHILD_NB_mpd = rev_min_max(.data$SN_CHILD_NB, max = 20),
+      SN_FRND_NB_mpd = rev_min_max(.data$SN_FRND_NB, max = 90),
+      SN_NEIBR_NB_mpd = rev_min_max(.data$SN_NEIBR_NB, max = 90),
+      SN_SIBLIV_NB_mpd = rev_min_max(.data$SN_SIBLIV_NB, max = 50),
+      SN_RELLIV_NB_mpd = rev_min_max(.data$SN_RELLIV_NB, max = 100),
+      SN_PERWSCH_NB_mpd = rev_min_max(.data$SN_PERWSCH_NB, max = 100),
+      SN_PERCOM_NB_mpd = rev_min_max(.data$SN_PERCOM_NB, max = 100),
+      SN_PERACT_NB_mpd = rev_min_max(.data$SN_PERACT_NB, max = 100),
+      B = rowSums(dplyr::pick(dplyr::ends_with("mpd"))) / 8
+    ) |>
+    dplyr::select(-dplyr::ends_with("mpd"))
+}
+
+rev_min_max <- function(NB, max) {
+  10 - ((10 * NB) / max)
+}
+
+# LAST TIME VISITED
+derive_C <- function(df) {
+  expected_cols <- paste0(
+    "SN_SEE",
+    c("CHILD", "SIB", "REL", "FRND", "NEIBR")
+  )
+  validate_cols(df, expected_cols)
+
+  df |>
+    dplyr::mutate(
+      dplyr::across(
+        .cols = expected_cols,
+        .fns = \(x) {
+          dplyr::case_when(
+            x == 1 ~ 0, # within the last day or two
+            x == 2 ~ 2, # within the last week or two
+            x == 3 ~ 4, # within the past month
+            x == 4 ~ 6, # within the past 6 months
+            x == 5 ~ 8, # within the past year
+            x %in% c(6, 7) ~ 10, # more than 1 year ago; everyone lives in household
+            .default = NA
+          )
+        },
+        .names = "{.col}_mpd"
+      ),
+      C = rowSums(dplyr::pick(dplyr::ends_with("mpd"))) / 5
+    ) |>
+    dplyr::select(-dplyr::ends_with("mpd"))
+}
+
+#' Convert "Number of people living in household" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `SN_LIVH_NB`.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `D`, dichotomized to 0 or 10.
+#'
+#' @details
+#' Helper function for [derive_SO()].
+#'
+#' @keywords internal
+derive_D <- function(df) {
+  validate_cols(df, "SN_LIVH_NB")
+
+  df |>
+    dplyr::mutate(
+      # SN_LIVH_NB excludes the respondent
+      D = dplyr::if_else(.data$SN_LIVH_NB == 0, 10, 0, missing = NA)
+    )
+}
+
+#' Convert "Marital status" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `SDC_MRTL`, where 2 = Married or living with common-law partner.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `E`, dichotomized to 0 or 10.
+#'
+#' @details
+#' Helper function for [derive_SO()].
+#'
+#' @keywords internal
+derive_E <- function(df) {
+  validate_cols(df, "SDC_MRTL")
+
+  df |>
+    dplyr::mutate(
+      E = dplyr::if_else(.data$SDC_MRTL == 2, 0, 10, missing = NA)
+    )
+}
+
+#' Compute the functional objective dimension of the CLSA-SII
+#'
+#' @param df A data frame containing cleaned survey data. Must include the
+#'   following columns, each corresponding to a transformed subscale score the
+#'   Medical Outcomes Study (MOS) Social Support Survey:
+#'
+#'   * `SSA_DPAFF`: Affection
+#'   * `SSA_DPEMO`: Emotional and informational support
+#'   * `SSA_DPSOC`: Positive social interaction
+#'   * `SSA_DPTNG`: Tangible support
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `FO`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_sii()].
+#'
+#' @keywords internal
+derive_FO <- function(df) {
+  FO <- df |>
+    derive_x() |> # affection
+    derive_y() |> # emotional and informational support
+    derive_z() |> # positive social interaction
+    derive_infinity() |> # tangible support
+    dplyr::mutate(
+      FO = (.data$x + .data$y + .data$z + .data$infinity) / 4
+    ) |>
+    dplyr::select(-dplyr::all_of(c("x", "y", "z", "infinity")))
+}
+
+#' Convert "Affection" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `SSA_DPAFF` (Affection subscale of the MOS Social Support Survey). This
+#'   function expects `SSA_DPAFF` to be pre-transformed by CLSA.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `x`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_FO()].
+#'
+#' @keywords internal
+derive_x <- function(df) {
+  validate_cols(df, "SSA_DPAFF")
+
+  df |>
+    dplyr::mutate(
+      SUM_AFF = ((12 * .data$SSA_DPAFF) / 100) + 3,
+      x = 10 * ((15 - .data$SUM_AFF) / (15 - 3))
+    ) |>
+    dplyr::select(-.data$SUM_AFF)
+}
+
+#' Convert "Emotional and Informational Support" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `SSA_DPEMO` (Emotional and Informational Support subscale of the MOS Social
+#'   Support Survey). This function expects `SSA_DPEMO` to be pre-transformed by
+#'   CLSA.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `y`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_FO()].
+#'
+#' @keywords internal
+derive_y <- function(df) {
+  validate_cols(df, "SSA_DPEMO")
+
+  df |>
+    dplyr::mutate(
+      SUM_EMO = ((32 * .data$SSA_DPEMO) / 100) + 8,
+      y = 10 * ((40 - .data$SUM_EMO) / (40 - 8))
+    ) |>
+    dplyr::select(-.data$SUM_EMO)
+}
+
+#' Convert "Positive Social Interaction" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `SSA_DPSOC` (Positive Social Interaction subscale of the MOS Social Support
+#'   Survey). This function expects `SSA_DPSOC` to be pre-transformed by CLSA.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `z`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_FO()].
+#'
+#' @keywords internal
+derive_z <- function(df) {
+  validate_cols(df, "SSA_DPSOC")
+
+  df |>
+    dplyr::mutate(
+      SUM_SOC = ((12 * .data$SSA_DPSOC) / 100) + 3,
+      z = 10 * ((15 - .data$SUM_SOC) / (15 - 3))
+    ) |>
+    dplyr::select(-.data$SUM_SOC)
+}
+
+#' Convert "Tangible Support" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `SSA_DPTNG` (Tangible Support subscale of the MOS Social Support Survey).
+#'   This function expects `SSA_DPTNG` to be pre-transformed by CLSA.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `infinity`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_FO()].
+#'
+#' @keywords internal
+derive_infinity <- function(df) {
+  validate_cols(df, "SSA_DPTNG")
+
+  df |>
+    dplyr::mutate(
+      SUM_TNG = ((16 * .data$SSA_DPTNG) / 100) + 4,
+      infinity = 10 * ((20 - .data$SUM_TNG) / (20 - 4))
+    ) |>
+    dplyr::select(-.data$SUM_TNG)
+}
+
+#' Compute the functional subjective dimension of the CLSA-SII
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `DEP_LONLY` and `SPA_MORAC`.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `FS`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_sii()].
+#'
+#' @keywords internal
+derive_FS <- function(df) {
+  FS <- df |>
+    derive_alpha() |> # loneliness
+    derive_omega() |> # desire to participate more
+    dplyr::mutate(
+      FS = (.data$alpha + .data$omega) / 2
+    ) |>
+    dplyr::select(-dplyr::all_of(c("alpha", "omega")))
+}
+
+#' Convert "How often did you feel lonely" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `DEP_LONLY`.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `alpha`, dichotomized to 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_FS()].
+#'
+#' @keywords internal
+derive_alpha <- function(df) {
+  validate_cols(df, "DEP_LONLY")
+
+  df |>
+    dplyr::mutate(
+      alpha = dplyr::case_when(
+        DEP_LONLY == 1 ~ 10, # all of the time (5-7 days)
+        DEP_LONLY == 2 ~ 6.67, # occasionally (3-4 days)
+        DEP_LONLY == 3 ~ 3.33, # some of the time (1-2 days)
+        DEP_LONLY == 4 ~ 0, # rarely or never (less than 1 day),
+        .default = NA
+      )
+    )
+}
+
+#' Convert "Desire to participate in more activities" into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include
+#'   `SPA_MORAC`, where 1 = Yes and 2 = No.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `omega`, dichotomized to 0 or 10.
+#'
+#' @details
+#' Helper function for [derive_FS()].
+#'
+#' @keywords internal
+derive_omega <- function(df) {
+  validate_cols(df, "SPA_MORAC")
+
+  df |>
+    dplyr::mutate(
+      omega = dplyr::if_else(.data$SPA_MORAC == 1, 10, 0, missing = NA)
+    )
+}
diff --git a/tests/testthat/test-derive_sii.R b/tests/testthat/test-derive_sii.R
new file mode 100644
index 0000000..7c3b9fb
--- /dev/null
+++ b/tests/testthat/test-derive_sii.R
@@ -0,0 +1,32 @@
+test_df <- data.frame(
+  entity_id = 12345,
+  SPA_OUTS = 2,
+  SPA_CHRCH = 5,
+  SPA_CLUB = 5,
+  SPA_EDUC = 4,
+  SPA_NEIBR = 5,
+  SPA_OTACT = 3,
+  SPA_SPORT = 5,
+  SPA_VOLUN = 5,
+  SN_SEECHILD = 2,
+  SN_SEESIB = 6,
+  SN_SEEREL = 2,
+  SN_SEEFRND = 1,
+  SN_SEENEIBR = 2,
+  SN_LIVH_NB = 0,
+  SDC_MRTL = 2,
+  SN_CHILD_NB = 2,
+  SN_FRND_NB = 6,
+  SN_NEIBR_NB = 4,
+  SN_SIBLIV_NB = 0,
+  SN_RELLIV_NB = 75,
+  SN_PERWSCH_NB = 7,
+  SN_PERCOM_NB = 0,
+  SN_PERACT_NB = 0,
+  DEP_LONLY = 4,
+  SPA_MORAC = 2,
+  SSA_DPAFF = 75,
+  SSA_DPEMO = 56.25,
+  SSA_DPTNG = 87.50,
+  SSA_DPSOC = 66.67
+)

From 5fc06bf5503a8d491cc1e6e46c44aa50667ccc02 Mon Sep 17 00:00:00 2001
From: Reiko Okamoto <rokamoto@ohri.ca>
Date: Thu, 30 Jul 2026 11:38:29 -0400
Subject: [PATCH 2/4] Finish writing main and helper functions to calculate SSI

---
 DESCRIPTION                      |   2 +-
 NAMESPACE                        |   1 +
 R/derive_sii.R                   | 182 ++++++++++++++++++++++++++++---
 man/clsatools-package.Rd         |   1 +
 man/derive_A.Rd                  |  35 ++++++
 man/derive_B.Rd                  |  34 ++++++
 man/derive_C.Rd                  |  31 ++++++
 man/derive_D.Rd                  |  23 ++++
 man/derive_E.Rd                  |  23 ++++
 man/derive_FO.Rd                 |  30 +++++
 man/derive_FS.Rd                 |  23 ++++
 man/derive_SO.Rd                 |  35 ++++++
 man/derive_alpha.Rd              |  23 ++++
 man/derive_infinity.Rd           |  24 ++++
 man/derive_omega.Rd              |  23 ++++
 man/derive_sii.Rd                |  65 +++++++++++
 man/derive_x.Rd                  |  24 ++++
 man/derive_y.Rd                  |  25 +++++
 man/derive_z.Rd                  |  24 ++++
 tests/testthat/test-derive_sii.R |  32 ------
 20 files changed, 613 insertions(+), 47 deletions(-)
 create mode 100644 man/derive_A.Rd
 create mode 100644 man/derive_B.Rd
 create mode 100644 man/derive_C.Rd
 create mode 100644 man/derive_D.Rd
 create mode 100644 man/derive_E.Rd
 create mode 100644 man/derive_FO.Rd
 create mode 100644 man/derive_FS.Rd
 create mode 100644 man/derive_SO.Rd
 create mode 100644 man/derive_alpha.Rd
 create mode 100644 man/derive_infinity.Rd
 create mode 100644 man/derive_omega.Rd
 create mode 100644 man/derive_sii.Rd
 create mode 100644 man/derive_x.Rd
 create mode 100644 man/derive_y.Rd
 create mode 100644 man/derive_z.Rd

diff --git a/DESCRIPTION b/DESCRIPTION
index 883df55..1da0494 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -11,7 +11,6 @@ Description: What the package does (one paragraph).
 License: MIT + file LICENSE
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.3.2
 Imports: 
     dplyr,
     glue,
@@ -27,3 +26,4 @@ Suggests:
 Config/testthat/edition: 3
 URL: https://big-life-lab.github.io/clsatools/, https://github.com/Big-Life-Lab/clsatools
 VignetteBuilder: knitr
+Config/roxygen2/version: 8.0.0
diff --git a/NAMESPACE b/NAMESPACE
index 9dc2dfb..b5d7fcc 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -5,6 +5,7 @@ export(derive_adl_dcls)
 export(derive_cag_fpas)
 export(derive_cr1_frhc)
 export(derive_cr2_frhc)
+export(derive_sii)
 export(derive_sls_dcls)
 export(find_matching_pairs)
 export(recode_missing)
diff --git a/R/derive_sii.R b/R/derive_sii.R
index e3706de..8836495 100644
--- a/R/derive_sii.R
+++ b/R/derive_sii.R
@@ -1,31 +1,135 @@
+#' Derive the CLSA Social Isolation Index (SII)
+#'
+#' @param df A data frame containing cleaned data for a survey cycle. Must
+#'   include all columns required to compute the structural (SO),
+#'   functional overall (FO), and functional support (FS) dimensions. See
+#'   `@examples` for a minimal working input.
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, plus a new column
+#' `SII` indicating the respondent's social isolation index score.
+#'
+#' @references
+#' Wister A, Cosco T, Mitchell B, Menec V, Fyffe I. Development and
+#' concurrent validity of a composite social isolation index for older
+#' adults using the CLSA. Can J Aging. 2019;38(2):180-192.
+#' doi:10.1017/S0714980818000612
+#'
+#' Canadian Longitudinal Study on Aging. Derived Variables — Social Support
+#' Availability (SSA). 2018.
+#' \url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_ssa_11apr2018_v1.1.pdf}
+#'
+#' @export
+#'
+#' @examples
+#' df <- data.frame(
+#'   SPA_OUTS = c(3, 3, 2),
+#'   SPA_CHRCH = c(3, 5, 5),
+#'   SPA_CLUB = c(5, 5, 5),
+#'   SPA_EDUC = c(4, 4, 3),
+#'   SPA_NEIBR = c(5, 4, 4),
+#'   SPA_OTACT = c(4, 3, 3),
+#'   SPA_SPORT = c(1, 3, 2),
+#'   SPA_VOLUN = c(5, 4, 5),
+#'   SN_CHILD_NB = c(2, 4, 0),
+#'   SN_FRND_NB = c(0, 0, 7),
+#'   SN_NEIBR_NB = c(3, 31, 2),
+#'   SN_SIBLIV_NB = c(5, 6, 3),
+#'   SN_RELLIV_NB = c(12, 60, 23),
+#'   SN_PERWSCH_NB = c(50, 95, 100),
+#'   SN_PERCOM_NB = c(0, 0, 0),
+#'   SN_PERACT_NB = c(4, 10, 0),
+#'   SN_SEECHILD = c(2, 1, NA),
+#'   SN_SEESIB = c(2, 3, 4),
+#'   SN_SEEREL = c(2, 4, 3),
+#'   SN_SEEFRND = c(1, 1, 1),
+#'   SN_SEENEIBR = c(2, 2, 6),
+#'   SN_LIVH_NB = c(2, 5, 0),
+#'   SDC_MRTL = c(2, 2, 1),
+#'   SSA_DPAFF = c(100, 100, 50),
+#'   SSA_DPEMO = c(62.5, 34.38, 75),
+#'   SSA_DPSOC = c(58.33, 83.33, 75),
+#'   SSA_DPTNG = c(50, 43.75, 62.5),
+#'   DEP_LONLY = c(2, 4, 4),
+#'   SPA_MORAC = c(2, 2, 2)
+#' )
+#' derive_sii(df)
 derive_sii <- function(df) {
-  SII <- df |>
+  sii <- df |>
     derive_SO() |>
     derive_FO() |>
     derive_FS() |>
     dplyr::mutate(
-      SII = (.data$SO + .data$FO + .data$FS) / 3
+      sii = (.data$SO + .data$FO + .data$FS) / 3
     ) |>
     dplyr::select(-dplyr::all_of(c("SO", "FO", "FS")))
 
-  SII
+  sii
 }
 
-# STRUCTURAL OBJECTIVE
+#' Compute the structural objective dimension of the CLSA-SII
+#'
+#' @param df A data frame containing cleaned survey data. Must include the
+#'   following columns:
+#'
+#'   * `SPA_OUTS`, `SPA_CHRCH`, `SPA_CLUB`, `SPA_EDUC`, `SPA_NEIBR`,
+#'     `SPA_OTACT`, `SPA_SPORT`, `SPA_VOLUN` (frequency of participation in
+#'     community activities)
+#'   * `SN_CHILD_NB`, `SN_FRND_NB`, `SN_NEIBR_NB`, `SN_SIBLIV_NB`,
+#'     `SN_RELLIV_NB`, `SN_PERWSCH_NB`, `SN_PERCOM_NB`, `SN_PERACT_NB`
+#'     (social network quantity)
+#'   * `SN_SEECHILD`, `SN_SEESIB`, `SN_SEEREL`, `SN_SEEFRND`, `SN_SEENEIBR`
+#'     (recency of visits outside household)
+#'   * `SN_LIVH_NB` (number of people living in household)
+#'   * `SDC_MRTL` (marital status)
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `SO`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_sii()].
+#'
+#' @keywords internal
 derive_SO <- function(df) {
   SO <- df |>
-    derive_A() |>
-    derive_B() |>
-    derive_C() |>
-    derive_D() |>
-    derive_E() |>
+    derive_A() |> # frequency of participation in community activities
+    derive_B() |> # social network quantity
+    derive_C() |> # recency of visits outside household
+    derive_D() |> # number of people living in household
+    derive_E() |> # marital status
     dplyr::mutate(
       SO = (.data$A + .data$B + .data$C + .data$D + .data$E) / 5
     ) |>
     dplyr::select(-dplyr::all_of(c("A", "B", "C", "D", "E")))
+
+  SO
 }
 
-# COMMUNITY PARTICIPATION
+#' Summarize "Community participation" survey items into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include the
+#'   following columns, each indicating frequency of participation in an
+#'   activity:
+#'
+#'   * `SPA_OUTS`: Activities with family/friends outside of household
+#'   * `SPA_CHRCH`: Church or religious activities
+#'   * `SPA_CLUB`: Service club or fraternal organization activities
+#'   * `SPA_EDUC`: Educational or cultural activities
+#'   * `SPA_NEIBR`: Neighbourhood, community, or professional association
+#'     activities
+#'   * `SPA_OTACT`: Any other recreational activities involving others
+#'   * `SPA_SPORT`: Sports or physical activities involving others
+#'   * `SPA_VOLUN`: Volunteer or charity work
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `A`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_SO()].
+#'
+#' @keywords internal
 derive_A <- function(df) {
   expected_cols <- paste0(
     "SPA_",
@@ -54,7 +158,29 @@ derive_A <- function(df) {
     dplyr::select(-dplyr::ends_with("_mpd"))
 }
 
-# SOCIAL NETWORK QUANTITY
+#' Summarize "Social network quantity" survey items into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include the
+#'   following columns, each indicating a count of people in the respondent's
+#'   network:
+#'
+#'   * `SN_CHILD_NB`: Living children
+#'   * `SN_FRND_NB`: Close friends
+#'   * `SN_NEIBR_NB`: Neighbours
+#'   * `SN_SIBLIV_NB`: Living siblings
+#'   * `SN_RELLIV_NB`: Living relatives
+#'   * `SN_PERWSCH_NB`: People known through work or school
+#'   * `SN_PERCOM_NB`: People known through community involvement
+#'   * `SN_PERACT_NB`: People known through other activities
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `B`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_SO()].
+#'
+#' @keywords internal
 derive_B <- function(df) {
   expected_cols <- paste(
     "SN",
@@ -75,7 +201,7 @@ derive_B <- function(df) {
 
   df |>
     dplyr::mutate(
-      # all of their min is 0
+      # apply reverse min-max scaling; all min is 0
       SN_CHILD_NB_mpd = rev_min_max(.data$SN_CHILD_NB, max = 20),
       SN_FRND_NB_mpd = rev_min_max(.data$SN_FRND_NB, max = 90),
       SN_NEIBR_NB_mpd = rev_min_max(.data$SN_NEIBR_NB, max = 90),
@@ -89,11 +215,31 @@ derive_B <- function(df) {
     dplyr::select(-dplyr::ends_with("mpd"))
 }
 
+#' @noRd
 rev_min_max <- function(NB, max) {
   10 - ((10 * NB) / max)
 }
 
-# LAST TIME VISITED
+#' Summarize "Last time visited" survey items into a 0-10 score
+#'
+#' @param df A data frame containing cleaned survey data. Must include the
+#'   following columns, each indicating how recently the respondent last visited
+#'   someone outside their household:
+#'
+#'   * `SN_SEECHILD`: Children
+#'   * `SN_SEESIB`: Siblings
+#'   * `SN_SEEREL`: Relatives
+#'   * `SN_SEEFRND`: Close friends
+#'   * `SN_SEENEIBR`: Neighbours
+#'
+#' @returns
+#' A data frame with the same number of rows as `df`, with a new column
+#' `C`, ranging from 0 to 10.
+#'
+#' @details
+#' Helper function for [derive_SO()].
+#'
+#' @keywords internal
 derive_C <- function(df) {
   expected_cols <- paste0(
     "SN_SEE",
@@ -112,7 +258,7 @@ derive_C <- function(df) {
             x == 3 ~ 4, # within the past month
             x == 4 ~ 6, # within the past 6 months
             x == 5 ~ 8, # within the past year
-            x %in% c(6, 7) ~ 10, # more than 1 year ago; everyone lives in household
+            x %in% c(6, 7) ~ 10, # more than 1 year ago; NA everyone lives in household
             .default = NA
           )
         },
@@ -197,6 +343,8 @@ derive_FO <- function(df) {
       FO = (.data$x + .data$y + .data$z + .data$infinity) / 4
     ) |>
     dplyr::select(-dplyr::all_of(c("x", "y", "z", "infinity")))
+
+  FO
 }
 
 #' Convert "Affection" into a 0-10 score
@@ -219,6 +367,7 @@ derive_x <- function(df) {
   df |>
     dplyr::mutate(
       SUM_AFF = ((12 * .data$SSA_DPAFF) / 100) + 3,
+      # apply reverse min-max scaling
       x = 10 * ((15 - .data$SUM_AFF) / (15 - 3))
     ) |>
     dplyr::select(-.data$SUM_AFF)
@@ -245,6 +394,7 @@ derive_y <- function(df) {
   df |>
     dplyr::mutate(
       SUM_EMO = ((32 * .data$SSA_DPEMO) / 100) + 8,
+      # apply reverse min-max scaling
       y = 10 * ((40 - .data$SUM_EMO) / (40 - 8))
     ) |>
     dplyr::select(-.data$SUM_EMO)
@@ -270,6 +420,7 @@ derive_z <- function(df) {
   df |>
     dplyr::mutate(
       SUM_SOC = ((12 * .data$SSA_DPSOC) / 100) + 3,
+      # apply reverse min-max scaling
       z = 10 * ((15 - .data$SUM_SOC) / (15 - 3))
     ) |>
     dplyr::select(-.data$SUM_SOC)
@@ -295,6 +446,7 @@ derive_infinity <- function(df) {
   df |>
     dplyr::mutate(
       SUM_TNG = ((16 * .data$SSA_DPTNG) / 100) + 4,
+      # apply reverse min-max scaling
       infinity = 10 * ((20 - .data$SUM_TNG) / (20 - 4))
     ) |>
     dplyr::select(-.data$SUM_TNG)
@@ -321,6 +473,8 @@ derive_FS <- function(df) {
       FS = (.data$alpha + .data$omega) / 2
     ) |>
     dplyr::select(-dplyr::all_of(c("alpha", "omega")))
+
+  FS
 }
 
 #' Convert "How often did you feel lonely" into a 0-10 score
diff --git a/man/clsatools-package.Rd b/man/clsatools-package.Rd
index 4d89ebf..2c6d1a2 100644
--- a/man/clsatools-package.Rd
+++ b/man/clsatools-package.Rd
@@ -23,6 +23,7 @@ Useful links:
 
 Authors:
 \itemize{
+  \item Reiko Okamoto \email{rokamoto@ohri.ca} (\href{https://orcid.org/0009-0006-3293-7940}{ORCID})
   \item Wenshan Li \email{wensli@toh.ca} (\href{https://orcid.org/0000-0002-6225-5549}{ORCID})
 }
 
diff --git a/man/derive_A.Rd b/man/derive_A.Rd
new file mode 100644
index 0000000..482d41a
--- /dev/null
+++ b/man/derive_A.Rd
@@ -0,0 +1,35 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_A}
+\alias{derive_A}
+\title{Summarize "Community participation" survey items into a 0-10 score}
+\usage{
+derive_A(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include the
+following columns, each indicating frequency of participation in an
+activity:
+\itemize{
+\item \code{SPA_OUTS}: Activities with family/friends outside of household
+\item \code{SPA_CHRCH}: Church or religious activities
+\item \code{SPA_CLUB}: Service club or fraternal organization activities
+\item \code{SPA_EDUC}: Educational or cultural activities
+\item \code{SPA_NEIBR}: Neighbourhood, community, or professional association
+activities
+\item \code{SPA_OTACT}: Any other recreational activities involving others
+\item \code{SPA_SPORT}: Sports or physical activities involving others
+\item \code{SPA_VOLUN}: Volunteer or charity work
+}}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{A}, ranging from 0 to 10.
+}
+\description{
+Summarize "Community participation" survey items into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_SO]{derive_SO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_B.Rd b/man/derive_B.Rd
new file mode 100644
index 0000000..c18294c
--- /dev/null
+++ b/man/derive_B.Rd
@@ -0,0 +1,34 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_B}
+\alias{derive_B}
+\title{Summarize "Social network quantity" survey items into a 0-10 score}
+\usage{
+derive_B(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include the
+following columns, each indicating a count of people in the respondent's
+network:
+\itemize{
+\item \code{SN_CHILD_NB}: Living children
+\item \code{SN_FRND_NB}: Close friends
+\item \code{SN_NEIBR_NB}: Neighbours
+\item \code{SN_SIBLIV_NB}: Living siblings
+\item \code{SN_RELLIV_NB}: Living relatives
+\item \code{SN_PERWSCH_NB}: People known through work or school
+\item \code{SN_PERCOM_NB}: People known through community involvement
+\item \code{SN_PERACT_NB}: People known through other activities
+}}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{B}, ranging from 0 to 10.
+}
+\description{
+Summarize "Social network quantity" survey items into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_SO]{derive_SO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_C.Rd b/man/derive_C.Rd
new file mode 100644
index 0000000..8c34437
--- /dev/null
+++ b/man/derive_C.Rd
@@ -0,0 +1,31 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_C}
+\alias{derive_C}
+\title{Summarize "Last time visited" survey items into a 0-10 score}
+\usage{
+derive_C(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include the
+following columns, each indicating how recently the respondent last visited
+someone outside their household:
+\itemize{
+\item \code{SN_SEECHILD}: Children
+\item \code{SN_SEESIB}: Siblings
+\item \code{SN_SEEREL}: Relatives
+\item \code{SN_SEEFRND}: Close friends
+\item \code{SN_SEENEIBR}: Neighbours
+}}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{C}, ranging from 0 to 10.
+}
+\description{
+Summarize "Last time visited" survey items into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_SO]{derive_SO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_D.Rd b/man/derive_D.Rd
new file mode 100644
index 0000000..0858dfc
--- /dev/null
+++ b/man/derive_D.Rd
@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_D}
+\alias{derive_D}
+\title{Convert "Number of people living in household" into a 0-10 score}
+\usage{
+derive_D(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{SN_LIVH_NB}.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{D}, dichotomized to 0 or 10.
+}
+\description{
+Convert "Number of people living in household" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_SO]{derive_SO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_E.Rd b/man/derive_E.Rd
new file mode 100644
index 0000000..1f80228
--- /dev/null
+++ b/man/derive_E.Rd
@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_E}
+\alias{derive_E}
+\title{Convert "Marital status" into a 0-10 score}
+\usage{
+derive_E(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{SDC_MRTL}, where 2 = Married or living with common-law partner.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{E}, dichotomized to 0 or 10.
+}
+\description{
+Convert "Marital status" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_SO]{derive_SO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_FO.Rd b/man/derive_FO.Rd
new file mode 100644
index 0000000..e5ed1f7
--- /dev/null
+++ b/man/derive_FO.Rd
@@ -0,0 +1,30 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_FO}
+\alias{derive_FO}
+\title{Compute the functional objective dimension of the CLSA-SII}
+\usage{
+derive_FO(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include the
+following columns, each corresponding to a transformed subscale score the
+Medical Outcomes Study (MOS) Social Support Survey:
+\itemize{
+\item \code{SSA_DPAFF}: Affection
+\item \code{SSA_DPEMO}: Emotional and informational support
+\item \code{SSA_DPSOC}: Positive social interaction
+\item \code{SSA_DPTNG}: Tangible support
+}}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{FO}, ranging from 0 to 10.
+}
+\description{
+Compute the functional objective dimension of the CLSA-SII
+}
+\details{
+Helper function for \code{\link[=derive_sii]{derive_sii()}}.
+}
+\keyword{internal}
diff --git a/man/derive_FS.Rd b/man/derive_FS.Rd
new file mode 100644
index 0000000..8ac253d
--- /dev/null
+++ b/man/derive_FS.Rd
@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_FS}
+\alias{derive_FS}
+\title{Compute the functional subjective dimension of the CLSA-SII}
+\usage{
+derive_FS(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{DEP_LONLY} and \code{SPA_MORAC}.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{FS}, ranging from 0 to 10.
+}
+\description{
+Compute the functional subjective dimension of the CLSA-SII
+}
+\details{
+Helper function for \code{\link[=derive_sii]{derive_sii()}}.
+}
+\keyword{internal}
diff --git a/man/derive_SO.Rd b/man/derive_SO.Rd
new file mode 100644
index 0000000..1ba0e61
--- /dev/null
+++ b/man/derive_SO.Rd
@@ -0,0 +1,35 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_SO}
+\alias{derive_SO}
+\title{Compute the structural objective dimension of the CLSA-SII}
+\usage{
+derive_SO(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include the
+following columns:
+\itemize{
+\item \code{SPA_OUTS}, \code{SPA_CHRCH}, \code{SPA_CLUB}, \code{SPA_EDUC}, \code{SPA_NEIBR},
+\code{SPA_OTACT}, \code{SPA_SPORT}, \code{SPA_VOLUN} (frequency of participation in
+community activities)
+\item \code{SN_CHILD_NB}, \code{SN_FRND_NB}, \code{SN_NEIBR_NB}, \code{SN_SIBLIV_NB},
+\code{SN_RELLIV_NB}, \code{SN_PERWSCH_NB}, \code{SN_PERCOM_NB}, \code{SN_PERACT_NB}
+(social network quantity)
+\item \code{SN_SEECHILD}, \code{SN_SEESIB}, \code{SN_SEEREL}, \code{SN_SEEFRND}, \code{SN_SEENEIBR}
+(recency of visits outside household)
+\item \code{SN_LIVH_NB} (number of people living in household)
+\item \code{SDC_MRTL} (marital status)
+}}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{SO}, ranging from 0 to 10.
+}
+\description{
+Compute the structural objective dimension of the CLSA-SII
+}
+\details{
+Helper function for \code{\link[=derive_sii]{derive_sii()}}.
+}
+\keyword{internal}
diff --git a/man/derive_alpha.Rd b/man/derive_alpha.Rd
new file mode 100644
index 0000000..6175f4b
--- /dev/null
+++ b/man/derive_alpha.Rd
@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_alpha}
+\alias{derive_alpha}
+\title{Convert "How often did you feel lonely" into a 0-10 score}
+\usage{
+derive_alpha(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{DEP_LONLY}.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{alpha}, dichotomized to 0 to 10.
+}
+\description{
+Convert "How often did you feel lonely" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_FS]{derive_FS()}}.
+}
+\keyword{internal}
diff --git a/man/derive_infinity.Rd b/man/derive_infinity.Rd
new file mode 100644
index 0000000..a2b7087
--- /dev/null
+++ b/man/derive_infinity.Rd
@@ -0,0 +1,24 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_infinity}
+\alias{derive_infinity}
+\title{Convert "Tangible Support" into a 0-10 score}
+\usage{
+derive_infinity(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{SSA_DPTNG} (Tangible Support subscale of the MOS Social Support Survey).
+This function expects \code{SSA_DPTNG} to be pre-transformed by CLSA.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{infinity}, ranging from 0 to 10.
+}
+\description{
+Convert "Tangible Support" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_FO]{derive_FO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_omega.Rd b/man/derive_omega.Rd
new file mode 100644
index 0000000..b11053c
--- /dev/null
+++ b/man/derive_omega.Rd
@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_omega}
+\alias{derive_omega}
+\title{Convert "Desire to participate in more activities" into a 0-10 score}
+\usage{
+derive_omega(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{SPA_MORAC}, where 1 = Yes and 2 = No.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{omega}, dichotomized to 0 or 10.
+}
+\description{
+Convert "Desire to participate in more activities" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_FS]{derive_FS()}}.
+}
+\keyword{internal}
diff --git a/man/derive_sii.Rd b/man/derive_sii.Rd
new file mode 100644
index 0000000..93ee5ab
--- /dev/null
+++ b/man/derive_sii.Rd
@@ -0,0 +1,65 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_sii}
+\alias{derive_sii}
+\title{Derive the CLSA Social Isolation Index (SII)}
+\usage{
+derive_sii(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned data for a survey cycle. Must
+include all columns required to compute the structural (SO),
+functional overall (FO), and functional support (FS) dimensions. See
+\verb{@examples} for a minimal working input.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, plus a new column
+\code{SII} indicating the respondent's social isolation index score.
+}
+\description{
+Derive the CLSA Social Isolation Index (SII)
+}
+\examples{
+df <- data.frame(
+  SPA_OUTS = c(3, 3, 2),
+  SPA_CHRCH = c(3, 5, 5),
+  SPA_CLUB = c(5, 5, 5),
+  SPA_EDUC = c(4, 4, 3),
+  SPA_NEIBR = c(5, 4, 4),
+  SPA_OTACT = c(4, 3, 3),
+  SPA_SPORT = c(1, 3, 2),
+  SPA_VOLUN = c(5, 4, 5),
+  SN_CHILD_NB = c(2, 4, 0),
+  SN_FRND_NB = c(0, 0, 7),
+  SN_NEIBR_NB = c(3, 31, 2),
+  SN_SIBLIV_NB = c(5, 6, 3),
+  SN_RELLIV_NB = c(12, 60, 23),
+  SN_PERWSCH_NB = c(50, 95, 100),
+  SN_PERCOM_NB = c(0, 0, 0),
+  SN_PERACT_NB = c(4, 10, 0),
+  SN_SEECHILD = c(2, 1, NA),
+  SN_SEESIB = c(2, 3, 4),
+  SN_SEEREL = c(2, 4, 3),
+  SN_SEEFRND = c(1, 1, 1),
+  SN_SEENEIBR = c(2, 2, 6),
+  SN_LIVH_NB = c(2, 5, 0),
+  SDC_MRTL = c(2, 2, 1),
+  SSA_DPAFF = c(100, 100, 50),
+  SSA_DPEMO = c(62.5, 34.38, 75),
+  SSA_DPSOC = c(58.33, 83.33, 75),
+  SSA_DPTNG = c(50, 43.75, 62.5),
+  DEP_LONLY = c(2, 4, 4),
+  SPA_MORAC = c(2, 2, 2)
+)
+derive_sii(df)
+}
+\references{
+Wister A, Cosco T, Mitchell B, Menec V, Fyffe I. Development and
+concurrent validity of a composite social isolation index for older
+adults using the CLSA. Can J Aging. 2019;38(2):180-192.
+doi:10.1017/S0714980818000612
+
+Canadian Longitudinal Study on Aging. Derived Variables — Social Support
+Availability (SSA). 2018.
+\url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_ssa_11apr2018_v1.1.pdf}
+}
diff --git a/man/derive_x.Rd b/man/derive_x.Rd
new file mode 100644
index 0000000..98cb169
--- /dev/null
+++ b/man/derive_x.Rd
@@ -0,0 +1,24 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_x}
+\alias{derive_x}
+\title{Convert "Affection" into a 0-10 score}
+\usage{
+derive_x(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{SSA_DPAFF} (Affection subscale of the MOS Social Support Survey). This
+function expects \code{SSA_DPAFF} to be pre-transformed by CLSA.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{x}, ranging from 0 to 10.
+}
+\description{
+Convert "Affection" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_FO]{derive_FO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_y.Rd b/man/derive_y.Rd
new file mode 100644
index 0000000..a3c9daf
--- /dev/null
+++ b/man/derive_y.Rd
@@ -0,0 +1,25 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_y}
+\alias{derive_y}
+\title{Convert "Emotional and Informational Support" into a 0-10 score}
+\usage{
+derive_y(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{SSA_DPEMO} (Emotional and Informational Support subscale of the MOS Social
+Support Survey). This function expects \code{SSA_DPEMO} to be pre-transformed by
+CLSA.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{y}, ranging from 0 to 10.
+}
+\description{
+Convert "Emotional and Informational Support" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_FO]{derive_FO()}}.
+}
+\keyword{internal}
diff --git a/man/derive_z.Rd b/man/derive_z.Rd
new file mode 100644
index 0000000..0372fda
--- /dev/null
+++ b/man/derive_z.Rd
@@ -0,0 +1,24 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/derive_sii.R
+\name{derive_z}
+\alias{derive_z}
+\title{Convert "Positive Social Interaction" into a 0-10 score}
+\usage{
+derive_z(df)
+}
+\arguments{
+\item{df}{A data frame containing cleaned survey data. Must include
+\code{SSA_DPSOC} (Positive Social Interaction subscale of the MOS Social Support
+Survey). This function expects \code{SSA_DPSOC} to be pre-transformed by CLSA.}
+}
+\value{
+A data frame with the same number of rows as \code{df}, with a new column
+\code{z}, ranging from 0 to 10.
+}
+\description{
+Convert "Positive Social Interaction" into a 0-10 score
+}
+\details{
+Helper function for \code{\link[=derive_FO]{derive_FO()}}.
+}
+\keyword{internal}
diff --git a/tests/testthat/test-derive_sii.R b/tests/testthat/test-derive_sii.R
index 7c3b9fb..e69de29 100644
--- a/tests/testthat/test-derive_sii.R
+++ b/tests/testthat/test-derive_sii.R
@@ -1,32 +0,0 @@
-test_df <- data.frame(
-  entity_id = 12345,
-  SPA_OUTS = 2,
-  SPA_CHRCH = 5,
-  SPA_CLUB = 5,
-  SPA_EDUC = 4,
-  SPA_NEIBR = 5,
-  SPA_OTACT = 3,
-  SPA_SPORT = 5,
-  SPA_VOLUN = 5,
-  SN_SEECHILD = 2,
-  SN_SEESIB = 6,
-  SN_SEEREL = 2,
-  SN_SEEFRND = 1,
-  SN_SEENEIBR = 2,
-  SN_LIVH_NB = 0,
-  SDC_MRTL = 2,
-  SN_CHILD_NB = 2,
-  SN_FRND_NB = 6,
-  SN_NEIBR_NB = 4,
-  SN_SIBLIV_NB = 0,
-  SN_RELLIV_NB = 75,
-  SN_PERWSCH_NB = 7,
-  SN_PERCOM_NB = 0,
-  SN_PERACT_NB = 0,
-  DEP_LONLY = 4,
-  SPA_MORAC = 2,
-  SSA_DPAFF = 75,
-  SSA_DPEMO = 56.25,
-  SSA_DPTNG = 87.50,
-  SSA_DPSOC = 66.67
-)

From bc437f27e7bde809b654e2f1f3aab8d41a3422f1 Mon Sep 17 00:00:00 2001
From: Reiko Okamoto <rokamoto@ohri.ca>
Date: Fri, 31 Jul 2026 10:31:25 -0400
Subject: [PATCH 3/4] Add tests for derive_sii, update docs

---
 R/derive_sii.R                   |   6 +-
 man/derive_alpha.Rd              |   2 +-
 tests/testthat/test-derive_sii.R | 499 +++++++++++++++++++++++++++++++
 3 files changed, 503 insertions(+), 4 deletions(-)

diff --git a/R/derive_sii.R b/R/derive_sii.R
index 8836495..85577bc 100644
--- a/R/derive_sii.R
+++ b/R/derive_sii.R
@@ -140,7 +140,7 @@ derive_A <- function(df) {
   df |>
     dplyr::mutate(
       dplyr::across(
-        .cols = expected_cols,
+        .cols = dplyr::all_of(expected_cols),
         .fns = \(x) {
           dplyr::case_when(
             x == 1 ~ 0, # at least once a day
@@ -250,7 +250,7 @@ derive_C <- function(df) {
   df |>
     dplyr::mutate(
       dplyr::across(
-        .cols = expected_cols,
+        .cols = dplyr::all_of(expected_cols),
         .fns = \(x) {
           dplyr::case_when(
             x == 1 ~ 0, # within the last day or two
@@ -484,7 +484,7 @@ derive_FS <- function(df) {
 #'
 #' @returns
 #' A data frame with the same number of rows as `df`, with a new column
-#' `alpha`, dichotomized to 0 to 10.
+#' `alpha`, ranging from 0 to 10.
 #'
 #' @details
 #' Helper function for [derive_FS()].
diff --git a/man/derive_alpha.Rd b/man/derive_alpha.Rd
index 6175f4b..e0d63af 100644
--- a/man/derive_alpha.Rd
+++ b/man/derive_alpha.Rd
@@ -12,7 +12,7 @@ derive_alpha(df)
 }
 \value{
 A data frame with the same number of rows as \code{df}, with a new column
-\code{alpha}, dichotomized to 0 to 10.
+\code{alpha}, ranging from 0 to 10.
 }
 \description{
 Convert "How often did you feel lonely" into a 0-10 score
diff --git a/tests/testthat/test-derive_sii.R b/tests/testthat/test-derive_sii.R
index e69de29..de00041 100644
--- a/tests/testthat/test-derive_sii.R
+++ b/tests/testthat/test-derive_sii.R
@@ -0,0 +1,499 @@
+df_A <- data.frame(
+  SPA_OUTS = 3,
+  SPA_CHRCH = 3,
+  SPA_CLUB = 5,
+  SPA_EDUC = 4,
+  SPA_NEIBR = 5,
+  SPA_OTACT = 4,
+  SPA_SPORT = 1,
+  SPA_VOLUN = 2
+)
+
+test_that("derive_A maps codes and averages correctly", {
+  result <- derive_A(df_A)
+
+  # (5 + 5 + 10 + 7.5 + 10 + 7.5 + 0 + 2.5) / 8 = 5.9375
+  expect_equal(result$A, 5.9375)
+})
+
+test_that("derive_A returns NA for invalid codes", {
+  result <- df_A |>
+    dplyr::mutate(SPA_OUTS = 6) |>
+    derive_A()
+
+  expect_true(is.na(result$A))
+})
+
+test_that("derive_A drops intermediate columns", {
+  result <- derive_A(df_A)
+
+  expect_false(any(stringr::str_detect(colnames(result), "_mpd$")))
+})
+
+test_that("derive_A errors if a required column is missing", {
+  df_A |>
+    dplyr::select(-SPA_OUTS) |>
+    derive_A() |>
+    expect_error()
+})
+
+df_B <- data.frame(
+  SN_CHILD_NB = 2,
+  SN_FRND_NB = 0,
+  SN_NEIBR_NB = 3,
+  SN_SIBLIV_NB = 5,
+  SN_RELLIV_NB = 12,
+  SN_PERWSCH_NB = 50,
+  SN_PERCOM_NB = 0,
+  SN_PERACT_NB = 4
+)
+
+test_that("derive_B transforms inputs and averages correctly", {
+  result <- derive_B(df_B)
+
+  # (9 + 10 + 9.6667 + 9 + 8.8 + 5 + 10 + 9.6) / 8 = 8.883337
+  expect_equal(result$B, 8.883337, tolerance = 1e-3)
+})
+
+test_that("derive_B returns NA when an input count is NA", {
+  result <- df_B |>
+    dplyr::mutate(SN_CHILD_NB = NA) |>
+    derive_B()
+
+  expect_true(is.na(result$B))
+})
+
+test_that("derive_B drops intermediate columns", {
+  result <- derive_B(df_B)
+
+  expect_false(any(stringr::str_detect(colnames(result), "_mpd$")))
+})
+
+test_that("derive_B errors if a required column is missing", {
+  df_B |>
+    dplyr::select(-SN_CHILD_NB) |>
+    derive_B() |>
+    expect_error()
+})
+
+df_C <- data.frame(
+  SN_SEECHILD = 5,
+  SN_SEESIB = 4,
+  SN_SEEREL = 2,
+  SN_SEEFRND = 1,
+  SN_SEENEIBR = 3
+)
+
+test_that("derive_C maps codes and averages correctly", {
+  result <- derive_C(df_C)
+
+  # (8 + 6 + 2 + 0 + 4) / 5
+  expect_equal(result$C, 4)
+})
+
+test_that("derive_C returns NA when input is NA", {
+  result <- df_C |>
+    dplyr::mutate(SN_SEECHILD = NA) |>
+    derive_C()
+
+  expect_true(is.na(result$C))
+})
+
+test_that("derive_C drops intermediate columns", {
+  result <- derive_C(df_C)
+
+  expect_false(any(stringr::str_detect(colnames(result), "_mpd$")))
+})
+
+test_that("derive_C errors if a required column is missing", {
+  df_C |>
+    dplyr::select(-SN_SEECHILD) |>
+    derive_C() |>
+    expect_error()
+})
+
+df_D <- data.frame(
+  SN_LIVH_NB = 0
+)
+
+test_that("derive_D maps 0 to 10 and other values to 0", {
+  result <- derive_D(df_D)
+
+  expect_equal(result$D, 10)
+
+  result_not_alone <- df_D |>
+    dplyr::mutate(SN_LIVH_NB = 3) |>
+    derive_D()
+
+  expect_equal(result_not_alone$D, 0)
+})
+
+test_that("derive_D returns NA when input is NA", {
+  result <- df_D |>
+    dplyr::mutate(SN_LIVH_NB = NA) |>
+    derive_D()
+
+  expect_true(is.na(result$D))
+})
+
+test_that("derive_D errors if the required column is missing", {
+  df_D |>
+    dplyr::select(-SN_LIVH_NB) |>
+    derive_D() |>
+    expect_error()
+})
+
+df_E <- data.frame(
+  SDC_MRTL = 2
+)
+
+test_that("derive_E maps 2 to 0 and other values to 10", {
+  result <- derive_E(df_E)
+
+  expect_equal(result$E, 0)
+
+  result_not_married <- df_E |>
+    dplyr::mutate(SDC_MRTL = 1) |>
+    derive_E()
+
+  expect_equal(result_not_married$E, 10)
+})
+
+test_that("derive_E returns NA when input is NA", {
+  result <- df_E |>
+    dplyr::mutate(SDC_MRTL = NA) |>
+    derive_E()
+
+  expect_true(is.na(result$E))
+})
+
+test_that("derive_E errors if the required column is missing", {
+  df_E |>
+    dplyr::select(-SDC_MRTL) |>
+    derive_E() |>
+    expect_error()
+})
+
+df_SO <- dplyr::bind_cols(
+  df_A,
+  df_B,
+  df_C,
+  df_D,
+  df_E
+)
+
+test_that("derive_SO averages A-E correctly", {
+  result <- derive_SO(df_SO)
+
+  # A = 5.9375, B = 8.883337, C = 4, D = 10, E = 0
+  # (5.9375 + 8.883337 + 4 + 10 + 0) / 5
+  expect_equal(result$SO, 5.764167, tolerance = 1e-3)
+})
+
+test_that("derive_SO drops intermediate columns", {
+  result <- derive_SO(df_SO)
+
+  expect_false(any(c("A", "B", "C", "D", "E") %in% colnames(result)))
+})
+
+test_that("derive_SO errors if a required column is missing", {
+  df_SO |>
+    dplyr::select(-SDC_MRTL) |>
+    derive_SO() |>
+    expect_error()
+})
+
+df_x <- data.frame(
+  SSA_DPAFF = 50
+)
+
+test_that("derive_x transforms input correctly", {
+  result <- derive_x(df_x)
+
+  # SUM_AFF = (12 * 50 / 100) + 3 = 9
+  # x = 10 * (15 - 9) / (15 - 3) = 5
+  expect_equal(result$x, 5)
+})
+
+test_that("derive_x is bounded 0-10 at extremes", {
+  low <- df_x |> dplyr::mutate(SSA_DPAFF = 0) |> derive_x()
+  high <- df_x |> dplyr::mutate(SSA_DPAFF = 100) |> derive_x()
+
+  expect_equal(low$x, 10)
+  expect_equal(high$x, 0)
+})
+
+test_that("derive_x returns NA when input is NA", {
+  result <- df_x |>
+    dplyr::mutate(SSA_DPAFF = NA) |>
+    derive_x()
+
+  expect_true(is.na(result$x))
+})
+
+test_that("derive_x drops intermediate column", {
+  result <- derive_x(df_x)
+
+  expect_false("SUM_AFF" %in% colnames(result))
+})
+
+test_that("derive_x errors if the required column is missing", {
+  df_x |>
+    dplyr::select(-SSA_DPAFF) |>
+    derive_x() |>
+    expect_error()
+})
+
+df_y <- data.frame(
+  SSA_DPEMO = 50
+)
+
+test_that("derive_y transforms input correctly", {
+  result <- derive_y(df_y)
+
+  # SUM_EMO = (32 * 50 / 100) + 8 = 24
+  # y = 10 * (40 - 24) / (40 - 8) = 5
+  expect_equal(result$y, 5)
+})
+
+test_that("derive_y is bounded 0-10 at extremes", {
+  low <- df_y |> dplyr::mutate(SSA_DPEMO = 0) |> derive_y()
+  high <- df_y |> dplyr::mutate(SSA_DPEMO = 100) |> derive_y()
+
+  expect_equal(low$y, 10)
+  expect_equal(high$y, 0)
+})
+
+test_that("derive_y returns NA when input is NA", {
+  result <- df_y |>
+    dplyr::mutate(SSA_DPEMO = NA) |>
+    derive_y()
+
+  expect_true(is.na(result$y))
+})
+
+test_that("derive_y drops intermediate column", {
+  result <- derive_y(df_y)
+
+  expect_false("SUM_EMO" %in% colnames(result))
+})
+
+test_that("derive_y errors if the required column is missing", {
+  df_y |>
+    dplyr::select(-SSA_DPEMO) |>
+    derive_y() |>
+    expect_error()
+})
+
+df_z <- data.frame(
+  SSA_DPSOC = 50
+)
+
+test_that("derive_z transforms input correctly", {
+  result <- derive_z(df_z)
+
+  # SUM_SOC = (12 * 50 / 100) + 3 = 9
+  # z = 10 * (15 - 9) / (15 - 3) = 5
+  expect_equal(result$z, 5)
+})
+
+test_that("derive_z is bounded 0-10 at extremes", {
+  low <- df_z |> dplyr::mutate(SSA_DPSOC = 0) |> derive_z()
+  high <- df_z |> dplyr::mutate(SSA_DPSOC = 100) |> derive_z()
+
+  expect_equal(low$z, 10)
+  expect_equal(high$z, 0)
+})
+
+test_that("derive_z returns NA when input is NA", {
+  result <- df_z |>
+    dplyr::mutate(SSA_DPSOC = NA) |>
+    derive_z()
+
+  expect_true(is.na(result$z))
+})
+
+test_that("derive_z drops intermediate column", {
+  result <- derive_z(df_z)
+
+  expect_false("SUM_SOC" %in% colnames(result))
+})
+
+test_that("derive_z errors if the required column is missing", {
+  df_z |>
+    dplyr::select(-SSA_DPSOC) |>
+    derive_z() |>
+    expect_error()
+})
+
+df_infinity <- data.frame(
+  SSA_DPTNG = 50
+)
+
+test_that("derive_infinity transforms input correctly", {
+  result <- derive_infinity(df_infinity)
+
+  # SUM_TNG = (16 * 50 / 100) + 4 = 12
+  # infinity = 10 * (20 - 12) / (20 - 4) = 5
+  expect_equal(result$infinity, 5)
+})
+
+test_that("derive_infinity is bounded 0-10 at extremes", {
+  low <- df_infinity |> dplyr::mutate(SSA_DPTNG = 0) |> derive_infinity()
+  high <- df_infinity |> dplyr::mutate(SSA_DPTNG = 100) |> derive_infinity()
+
+  expect_equal(low$infinity, 10)
+  expect_equal(high$infinity, 0)
+})
+
+test_that("derive_infinity returns NA when input is NA", {
+  result <- df_infinity |>
+    dplyr::mutate(SSA_DPTNG = NA) |>
+    derive_infinity()
+
+  expect_true(is.na(result$infinity))
+})
+
+test_that("derive_infinity drops intermediate column", {
+  result <- derive_infinity(df_infinity)
+
+  expect_false("SUM_TNG" %in% colnames(result))
+})
+
+test_that("derive_infinity errors if the required column is missing", {
+  df_infinity |>
+    dplyr::select(-SSA_DPTNG) |>
+    derive_infinity() |>
+    expect_error()
+})
+
+df_FO <- dplyr::bind_cols(df_x, df_y, df_z, df_infinity)
+
+test_that("derive_FO averages x/y/z/infinity correctly", {
+  result <- derive_FO(df_FO)
+
+  # x = 5, y = 5, z = 5, infinity = 5
+  # (5 + 5 + 5 + 5) / 4 = 5
+  expect_equal(result$FO, 5)
+})
+
+test_that("derive_FO drops intermediate columns", {
+  result <- derive_FO(df_FO)
+
+  expect_false(any(c("x", "y", "z", "infinity") %in% colnames(result)))
+})
+
+test_that("derive_FO errors if a required column is missing", {
+  df_FO |>
+    dplyr::select(-SSA_DPTNG) |>
+    derive_FO() |>
+    expect_error()
+})
+
+df_alpha <- data.frame(
+  DEP_LONLY = 1
+)
+
+test_that("derive_alpha maps codes correctly", {
+  result <- derive_alpha(df_alpha)
+  expect_equal(result$alpha, 10)
+
+  result2 <- df_alpha |> dplyr::mutate(DEP_LONLY = 2) |> derive_alpha()
+  expect_equal(result2$alpha, 6.67)
+
+  result3 <- df_alpha |> dplyr::mutate(DEP_LONLY = 3) |> derive_alpha()
+  expect_equal(result3$alpha, 3.33)
+
+  result4 <- df_alpha |> dplyr::mutate(DEP_LONLY = 4) |> derive_alpha()
+  expect_equal(result4$alpha, 0)
+})
+
+test_that("derive_alpha returns NA for invalid inputs", {
+  result <- df_alpha |>
+    dplyr::mutate(DEP_LONLY = 5) |>
+    derive_alpha()
+
+  expect_true(is.na(result$alpha))
+})
+
+test_that("derive_alpha errors if the required column is missing", {
+  df_alpha |>
+    dplyr::select(-DEP_LONLY) |>
+    derive_alpha() |>
+    expect_error()
+})
+
+df_omega <- data.frame(
+  SPA_MORAC = 1
+)
+
+test_that("derive_omega maps codes correctly", {
+  result <- derive_omega(df_omega)
+  expect_equal(result$omega, 10)
+
+  result_no <- df_omega |>
+    dplyr::mutate(SPA_MORAC = 2) |>
+    derive_omega()
+  expect_equal(result_no$omega, 0)
+})
+
+test_that("derive_omega returns NA when input is NA", {
+  result <- df_omega |>
+    dplyr::mutate(SPA_MORAC = NA) |>
+    derive_omega()
+
+  expect_true(is.na(result$omega))
+})
+
+test_that("derive_omega errors if the required column is missing", {
+  df_omega |>
+    dplyr::select(-SPA_MORAC) |>
+    derive_omega() |>
+    expect_error()
+})
+
+df_FS <- dplyr::bind_cols(df_alpha, df_omega)
+
+test_that("derive_FS averages alpha/omega correctly", {
+  result <- derive_FS(df_FS)
+
+  # alpha = 10, omega = 10
+  # (10 + 10) / 2 = 10
+  expect_equal(result$FS, 10)
+})
+
+test_that("derive_FS drops intermediate columns", {
+  result <- derive_FS(df_FS)
+
+  expect_false(any(c("alpha", "omega") %in% colnames(result)))
+})
+
+test_that("derive_FS errors if a required column is missing", {
+  df_FS |>
+    dplyr::select(-SPA_MORAC) |>
+    derive_FS() |>
+    expect_error()
+})
+
+df_sii <- dplyr::bind_cols(df_SO, df_FO, df_FS)
+
+test_that("derive_sii averages SO/FO/FS correctly", {
+  result <- derive_sii(df_sii)
+
+  # SO = 5.764167, FO = 5, FS = 10
+  # (5.764167 + 5 + 10) / 3 = 6.921389
+  expect_equal(result$sii, 6.921389, tolerance = 1e-3)
+})
+
+test_that("derive_sii drops intermediate columns", {
+  result <- derive_sii(df_sii)
+
+  expect_false(any(c("SO", "FO", "FS") %in% colnames(result)))
+})
+
+test_that("derive_sii errors if a required column is missing", {
+  df_sii |>
+    dplyr::select(-SDC_MRTL) |>
+    derive_sii() |>
+    expect_error()
+})

From 79d72847b6268b1bb6fa004fe6b1cf0f81cfeded Mon Sep 17 00:00:00 2001
From: Reiko Okamoto <rokamoto@ohri.ca>
Date: Fri, 31 Jul 2026 10:32:09 -0400
Subject: [PATCH 4/4] Update CLSA reference in docs

---
 R/derive_adl_dcls.R    | 8 +++++---
 R/derive_cag_fpas.R    | 6 ++++--
 R/derive_cr1_frhc.R    | 7 +++++--
 R/derive_cr2_frhc.R    | 7 +++++--
 R/derive_sls_dcls.R    | 6 ++++--
 man/derive_adl_dcls.Rd | 5 ++++-
 man/derive_cag_fpas.Rd | 4 +++-
 man/derive_cr1_frhc.Rd | 5 ++++-
 man/derive_cr2_frhc.Rd | 5 ++++-
 man/derive_sls_dcls.Rd | 4 +++-
 10 files changed, 41 insertions(+), 16 deletions(-)

diff --git a/R/derive_adl_dcls.R b/R/derive_adl_dcls.R
index 02c3c53..1473846 100644
--- a/R/derive_adl_dcls.R
+++ b/R/derive_adl_dcls.R
@@ -12,9 +12,11 @@
 #' Intended for use with Follow-up 2 data, for which this variable was not
 #' derived by the data provider.
 #'
-#' @references [Derived Variables - Basic Activities of Daily Living (ADL) &
-#' Instrumental Activities of Daily Living
-#' (IAL)](https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_adl_10aug2018.pdf)
+#' @references
+#' Canadian Longitudinal Study on Aging. Derived Variables — Basic
+#' Activities of Daily Living (ADL) & Instrumental Activities of Daily
+#' Living (IAL). 2018.
+#' \url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_adl_10aug2018.pdf}
 #'
 #' @export
 #' @examples
diff --git a/R/derive_cag_fpas.R b/R/derive_cag_fpas.R
index d5b55bc..6d11239 100644
--- a/R/derive_cag_fpas.R
+++ b/R/derive_cag_fpas.R
@@ -29,8 +29,10 @@
 #' `CAG_HLT_HC`) are excluded from the function because they are unavailable at
 #' Follow-up 2.
 #'
-#' @references [Derived Variables – Care Giving
-#' (CAG)](https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_caregiv_v1.4_2019feb01.pdf)
+#' @references
+#' Canadian Longitudinal Study on Aging. Derived Variables — Care Giving
+#' (CAG). 2019.
+#' \url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_caregiv_v1.4_2019feb01.pdf}
 #'
 #' @export
 #' @examples
diff --git a/R/derive_cr1_frhc.R b/R/derive_cr1_frhc.R
index a20aa79..132da13 100644
--- a/R/derive_cr1_frhc.R
+++ b/R/derive_cr1_frhc.R
@@ -27,8 +27,11 @@
 #' algorithm (`CR1_PRO_MC` and `CR1_PRO_MISSING`) are excluded from the function
 #' because they are unavailable at Follow-up 2.
 #'
-#' @references [Derived Variables – Care Receiving 1/Formal Care
-#' (CR1)](https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf)
+#' @references
+#' Canadian Longitudinal Study on Aging. Derived Variables — Care
+#' Receiving 1/Formal Care (CR1) & Care Receiving 2/Informal Care (CR2).
+#' 2017.
+#' \url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf}
 #'
 #' @export
 #' @examples
diff --git a/R/derive_cr2_frhc.R b/R/derive_cr2_frhc.R
index e47925a..aa724bb 100644
--- a/R/derive_cr2_frhc.R
+++ b/R/derive_cr2_frhc.R
@@ -25,8 +25,11 @@
 #' algorithm (`CR2_FAM_CS`, `CR2_FAM_MB`, `CR2_FAM_MISSING`, and `CR2_FAM_MN`)
 #' are excluded from the function because they are unavailable at Follow-up 2.
 #'
-#' @references [Derived Variables – Care Receiving 2/Informal Care (CR2)
-#' ](https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf)
+#' @references
+#' Canadian Longitudinal Study on Aging. Derived Variables — Care
+#' Receiving 1/Formal Care (CR1) & Care Receiving 2/Informal Care (CR2).
+#' 2017.
+#' \url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf}
 #'
 #' @export
 #' @examples
diff --git a/R/derive_sls_dcls.R b/R/derive_sls_dcls.R
index 653d303..8851cb6 100644
--- a/R/derive_sls_dcls.R
+++ b/R/derive_sls_dcls.R
@@ -6,8 +6,10 @@
 #' A data frame with the same number of rows as `df` plus a new column
 #' `SLS_DCLS` indicating the respondent's satisfaction with life.
 #'
-#' @references [Derived Variables – Satisfaction With Life (SLS)
-#' ](https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_sls_4dec2017.pdf)
+#' @references
+#' Canadian Longitudinal Study on Aging. Derived Variables — Satisfaction
+#' With Life (SLS). 2017.
+#' \url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_sls_4dec2017.pdf}
 #'
 #' @export
 #' @examples
diff --git a/man/derive_adl_dcls.Rd b/man/derive_adl_dcls.Rd
index 03957e7..23d6e03 100644
--- a/man/derive_adl_dcls.Rd
+++ b/man/derive_adl_dcls.Rd
@@ -49,5 +49,8 @@ df <- setNames(
 derive_adl_dcls(df)
 }
 \references{
-\href{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_adl_10aug2018.pdf}{Derived Variables - Basic Activities of Daily Living (ADL) & Instrumental Activities of Daily Living (IAL)}
+Canadian Longitudinal Study on Aging. Derived Variables — Basic
+Activities of Daily Living (ADL) & Instrumental Activities of Daily
+Living (IAL). 2018.
+\url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_adl_10aug2018.pdf}
 }
diff --git a/man/derive_cag_fpas.Rd b/man/derive_cag_fpas.Rd
index 67a7f4e..da806d6 100644
--- a/man/derive_cag_fpas.Rd
+++ b/man/derive_cag_fpas.Rd
@@ -59,5 +59,7 @@ df <- data.frame(
 derive_cag_fpas(df)
 }
 \references{
-\href{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_caregiv_v1.4_2019feb01.pdf}{Derived Variables – Care Giving (CAG)}
+Canadian Longitudinal Study on Aging. Derived Variables — Care Giving
+(CAG). 2019.
+\url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_caregiv_v1.4_2019feb01.pdf}
 }
diff --git a/man/derive_cr1_frhc.Rd b/man/derive_cr1_frhc.Rd
index 6b16d12..8c45340 100644
--- a/man/derive_cr1_frhc.Rd
+++ b/man/derive_cr1_frhc.Rd
@@ -56,5 +56,8 @@ df <- data.frame(
 derive_cr1_frhc(df)
 }
 \references{
-\href{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf}{Derived Variables – Care Receiving 1/Formal Care (CR1)}
+Canadian Longitudinal Study on Aging. Derived Variables — Care
+Receiving 1/Formal Care (CR1) & Care Receiving 2/Informal Care (CR2).
+2017.
+\url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf}
 }
diff --git a/man/derive_cr2_frhc.Rd b/man/derive_cr2_frhc.Rd
index 51139fa..7baf882 100644
--- a/man/derive_cr2_frhc.Rd
+++ b/man/derive_cr2_frhc.Rd
@@ -52,5 +52,8 @@ df <- data.frame(
 derive_cr2_frhc(df)
 }
 \references{
-\href{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf}{Derived Variables – Care Receiving 2/Informal Care (CR2) }
+Canadian Longitudinal Study on Aging. Derived Variables — Care
+Receiving 1/Formal Care (CR1) & Care Receiving 2/Informal Care (CR2).
+2017.
+\url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_cr_4dec2017.pdf}
 }
diff --git a/man/derive_sls_dcls.Rd b/man/derive_sls_dcls.Rd
index b5bd81e..13aaca9 100644
--- a/man/derive_sls_dcls.Rd
+++ b/man/derive_sls_dcls.Rd
@@ -38,5 +38,7 @@ df <- data.frame(
 derive_sls_dcls(df)
 }
 \references{
-\href{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_sls_4dec2017.pdf}{Derived Variables – Satisfaction With Life (SLS) }
+Canadian Longitudinal Study on Aging. Derived Variables — Satisfaction
+With Life (SLS). 2017.
+\url{https://www.clsa-elcv.ca/wp-content/uploads/2023/06/dv_sls_4dec2017.pdf}
 }