Code for replicating some ONS Publications • tidylfs

As part of the QA checks for this package there is code to replicate official ONS publications.

Salary

EARN04

  lfs <- lfs_load()

  earn04 <- lfs %>%
    dplyr::filter(FTPTWK == "Full-time") %>%
    lfs_summarise_salary(QUARTER) %>%
    dplyr::select(QUARTER, mean_weekly_pay, median_weekly_pay)

EARN06

ONS Publication

  earn06 <- lfs %>%
    dplyr::filter(FTPTWK == "Full-time") %>%
    dplyr::filter(!is.na(OCCUPATION_MAJOR)) %>%
    lfs_summarise_salary(QUARTER, OCCUPATION_MAJOR) %>%
    tidyr::pivot_wider(
      id_cols = QUARTER,
      names_from = OCCUPATION_MAJOR,
      values_from = "mean_weekly_pay"
    )

EARN07

ONS Publication

Currently only accurate from 2009 onwards

  earn07 <- lfs %>%
    dplyr::filter(YEAR > 2009) %>%
    dplyr::filter(!is.na(INDUSTRY)) %>%
    dplyr::mutate(GROUPED_INDUSTRY = dplyr::case_when(
      substr(INDUSTRY, 1, 1) == "A" ~ "A",
      substr(INDUSTRY, 1, 1) %in% c("B", "D", "E") ~ "BDE",
      substr(INDUSTRY, 1, 1) %in% "C" ~ "C",
      substr(INDUSTRY, 1, 1) %in% "F" ~ "F",
      substr(INDUSTRY, 1, 1) %in% "G" ~ "G",
      substr(INDUSTRY, 1, 1) %in% "H" ~ "H",
      substr(INDUSTRY, 1, 1) %in% "I" ~ "I",
      substr(INDUSTRY, 1, 1) %in% "J" ~ "J",
      substr(INDUSTRY, 1, 1) %in% c("K", "L") ~ "KL",
      substr(INDUSTRY, 1, 1) %in% "M" ~ "M",
      substr(INDUSTRY, 1, 1) %in% "N" ~ "N",
      substr(INDUSTRY, 1, 1) %in% "O" ~ "O",
      substr(INDUSTRY, 1, 1) %in% "P" ~ "P",
      substr(INDUSTRY, 1, 1) %in% "Q" ~ "Q",
      substr(INDUSTRY, 1, 1) %in% c("R", "S", "T") ~ "RST"
      # substr(INDUSTRY, 1, 1) %in% "U" ~ "U"
    )) %>%
    dplyr::filter(!is.na(GROUPED_INDUSTRY)) %>%
    dplyr::filter(FTPTWK == "Full-time") %>%
    lfs_summarise_salary(QUARTER, GROUPED_INDUSTRY) %>%
    tidyr::pivot_wider(
      id_cols = QUARTER,
      names_from = GROUPED_INDUSTRY,
      values_from = "mean_weekly_pay"
    )

Unemployment

UNEM01 NSA

ONS Publication

Accurate from 1995 onwards, also replicates sex splits when adding SEX as a grouping variable.

  unem01 <- lfs %>%
    dplyr::filter(YEAR > 1995) %>%
    lfs_summarise_unemployment(QUARTER) %>%
    dplyr::select(QUARTER, unemployed_percentage) %>%
    dplyr::mutate(unemployed_percentage = unemployed_percentage * 100) %>%
    dplyr::filter(!is.na(unemployed_percentage))

Hours Worked

HOUR01 NSA

ONS Publication

  hour01 <- lfs %>%
    dplyr::mutate(FT = !FTPTWK == "Full-time") %>%
    lfs_summarise_hours(QUARTER, FT) %>%
    tidyr::pivot_wider(
      id_cols = QUARTER,
      names_from = FT,
      values_from = hours
    )