preschool_sel

[ben-domingue]

decompose to multiple tables?

Children aged 2 - 5 answer item that assess their social and emotional skills, cognitive skills, early math, early reading, and other indicators of kindergarten readiness

[saviranadela]

hi @ben-domingue ! how can i find the link to the raw data?

[ben-domingue]

see here: https://github.com/ben-domingue/irw/blob/main/data/preschool_sel.R
and here:
Children aged 2 - 5 answer item that assess their social and emotional skills, cognitive skills, early math, early reading, and other indicators of kindergarten readiness (CC BY 4.0) [link] Bailey, C. S., Korucu, İrem, Eveleigh, A., Schnur, G., Costello, L., Tuttle, M., Knox-Lane, T., Cassidy, C., Ondrusek, A., McNaboe, T., Mazhar, A., & Xie, F. (2023). Preschool Social and Emotional Development Study—Connecticut Dataset. LDbase. https://doi.org/10.33009/ldbase.1680213217.8ed0 Get BibTex

[saviranadela]

codebook: https://ldbase.org/system/files/documents/2023-10/R305A180293_child-level_CT_v5_0.pdf

[saviranadela]

data:
preschool_sel.zip

code:

library(tidyverse)
library(haven)
library(labelled)

df <- read_sav('R305A180293_child-level_CT_v35.sav')

names(df) <- tolower(names(df))

df <- df |>
  select(contains('ss1'),
         contains('ss2'),
         contains('ss3'),
         contains('ss4'),
         contains('ss5'),
         contains('ss6'),
         contains('ss7'),
         contains('ss8'),
         contains('ss9'),
         contains('ss10'),
         contains('as1'),
         contains('as2'),
         contains('as3'),
         contains('as4'),
         contains('as5'),
         contains('as6'),
         contains('as7'),
         contains('as8'),
         contains('as9'),
         contains('as10'),
         starts_with('wj_lw'),
         starts_with('wj_ap'),
         starts_with('dn_1'),
         starts_with('dn_2'),
         starts_with('dn_3'),
         starts_with('dn_4'),
         starts_with('dn_5'),
         starts_with('dn_6'),
         starts_with('dn_7'),
         starts_with('dn_8'),
         starts_with('dn_9'),
         contains('_1s_'),
         contains('_2s_'),
         contains('_3s_'),
         contains('_4s_'),
         contains('_5s_'),
         contains('_6s_'),
         starts_with('box1'),
         starts_with('box2'),
         starts_with('htks1_1'),
         starts_with('htks1_2'),
         starts_with('htks1_3'),
         starts_with('htks1_4'),
         starts_with('htks1_5'),
         starts_with('htks1_6'),
         starts_with('htks1_7'),
         starts_with('htks1_8'),
         starts_with('htks1_9')) |> 
  select(-starts_with('emt1'),
         -starts_with('emt2'),
         -starts_with('emt3'),
         -starts_with('emt4'),
         -ends_with('hapb'),
         -ends_with('sadb'),
         -ends_with('angb'),
         -ends_with('afrb'),
         -ends_with('hapa'),
         -ends_with('sada'),
         -ends_with('anga'),
         -ends_with('afra'),
         -wj_lww_t1,
         -wj_lwss_t1,
         -wj_apw_t1,
         -wj_apss_t1,
         -contains('notes')) |>
  # replace invalid values with NA, change others to 0/1 wrong/right binary
  mutate(across(starts_with('dn'), ~if_else(. == 1, NA, .)),
         across(starts_with('dn'), ~if_else(. == 2, 1, .)),
         across(contains('_1s_') | contains('_2s_') | contains('_3s_') | contains('_4s_') | contains('_5s_') | contains('_6s_'), 
                ~if_else(. == 1, 0, .)),
         across(contains('_1s_') | contains('_2s_') | contains('_3s_') | contains('_4s_') | contains('_5s_') | contains('_6s_'), 
                ~if_else(. == 2, 1, .)),
         across(starts_with('htks'), ~if_else(. == 1, 0, .)),
         across(starts_with('htks'), ~if_else(. == 2, 1, .)),
         across(starts_with('box'), ~if_else(. == 0.5, NA, .)),
         # create participant ID
         id = row_number())

# find variables with no response or single responses to drop
# put them in a list to drop
drop_vars <- c()

for (i in 1:ncol(df)) {
  unique_vals <- unique(df[[i]])
  unique_len <- length(unique_vals)
  
  if (unique_len == 1 & is.na(unique(unique_vals[1]))) {
    drop_vars <- append(drop_vars, names(df)[i])
  }
  
  if (unique_len == 2) {
    if (is.na(unique_vals[1]) | is.na(unique_vals[2])) {
      drop_vars <- append(drop_vars, names(df)[i])
    }
  }
}

# drop variables with no responses or singular resposes
df <- df |>
  select(-all_of(drop_vars)) |>
  # pivot df to be long by item
  pivot_longer(cols = -id,
               names_to = 'item',
               values_to = 'resp',
               values_drop_na = T)

df_pl <- df %>%
  filter(grepl("^pl", item))

df_emt <- df %>%
  filter(grepl("^emt", item))

df_box <- df %>%
  filter(grepl("^box", item))

df_htks <- df %>%
  filter(grepl("^htks", item))

df_dn <- df %>%
  filter(grepl("^dn", item))

df_akt <- df %>%
  filter(grepl("^akt", item))

df_wj <- df %>%
  filter(grepl("^wj", item))

write.csv(df_pl, "preschool_sel_pl.csv", row.names=FALSE)
write.csv(df_emt, "preschool_sel_emt.csv", row.names=FALSE)
write.csv(df_box, "preschool_sel_box.csv", row.names=FALSE)
write.csv(df_htks, "preschool_sel_htks.csv", row.names=FALSE)
write.csv(df_dn, "preschool_sel_dn.csv", row.names=FALSE)
write.csv(df_akt, "preschool_sel_akt.csv", row.names=FALSE)
write.csv(df_wj, "preschool_sel_wj.csv", row.names=FALSE)

[saviranadela]

PR: #751

[ben-domingue]

@saviranadela this is great. one question. what is PL? see row 741 in green
https://docs.google.com/spreadsheets/d/1nhPyvuAm3JO8c9oa1swPvQZghAvmnf4xlYgbvsFH99s/edit?gid=0#gid=0

[saviranadela]

PL = preLAS. it's a language profienciency assessment for early learners! two subsets were used here: Simon Says (ss) and Art Show (as). see more: https://laslinks.com/prelas/

[ben-domingue]

fantastic, thank you!