Non-standard evaluation • ripeds

By default ripeds allows for non-standard evaluation (NSE) in the functions. This default setting will feel normal to those accustomed to working with packages in the Tidyverse. Users can also turn off NSE with ipeds_init(use_nse = FALSE).

library("ripeds")

Selecting variables

With NSE, variables can be separated by commas without quotation.

NSE

df <- ipeds_init() |>
  ipeds_select(instnm, stabbr) |>
  ipeds_year(2023) |>
  ipeds_get()

df |> head()
#>   unitid                              instnm stabbr year
#> 1 100654            Alabama A & M University     AL 2023
#> 2 100663 University of Alabama at Birmingham     AL 2023
#> 3 100690                  Amridge University     AL 2023
#> 4 100706 University of Alabama in Huntsville     AL 2023
#> 5 100724            Alabama State University     AL 2023
#> 6 100733 University of Alabama System Office     AL 2023

Using standard evaluation, quotation and concatenation c() is required.

Standard

df <- ipeds_init(use_nse = FALSE) |>
  ipeds_select(c("instnm", "stabbr")) |>
  ipeds_year(2023) |>
  ipeds_get()

df |> head()
#>   unitid                              instnm stabbr year
#> 1 100654            Alabama A & M University     AL 2023
#> 2 100663 University of Alabama at Birmingham     AL 2023
#> 3 100690                  Amridge University     AL 2023
#> 4 100706 University of Alabama in Huntsville     AL 2023
#> 5 100724            Alabama State University     AL 2023
#> 6 100733 University of Alabama System Office     AL 2023

Note that with standard evaluation, users can store variables names in an object and pass the object to ipeds_select(). This can be useful with large variable requests or dynamic requests.

## using object
vars_to_get <- c("instnm", "stabbr")
df <- ipeds_init(use_nse = FALSE) |>
  ipeds_select(vars_to_get) |>
  ipeds_year(2023) |>
  ipeds_get()

df |> head()
#>   unitid                              instnm stabbr year
#> 1 100654            Alabama A & M University     AL 2023
#> 2 100663 University of Alabama at Birmingham     AL 2023
#> 3 100690                  Amridge University     AL 2023
#> 4 100706 University of Alabama in Huntsville     AL 2023
#> 5 100724            Alabama State University     AL 2023
#> 6 100733 University of Alabama System Office     AL 2023

## dynamically
df_list <- purrr::map(c("instnm", "stabbr", "control"),
                      ~ ipeds_init(use_nse = FALSE) |>
                        ipeds_select(.x) |>
                        ipeds_year(2023) |>
                        ipeds_get())

lapply(df_list, head)
#> [[1]]
#>   unitid                              instnm year
#> 1 100654            Alabama A & M University 2023
#> 2 100663 University of Alabama at Birmingham 2023
#> 3 100690                  Amridge University 2023
#> 4 100706 University of Alabama in Huntsville 2023
#> 5 100724            Alabama State University 2023
#> 6 100733 University of Alabama System Office 2023
#> 
#> [[2]]
#>   unitid stabbr year
#> 1 100654     AL 2023
#> 2 100663     AL 2023
#> 3 100690     AL 2023
#> 4 100706     AL 2023
#> 5 100724     AL 2023
#> 6 100733     AL 2023
#> 
#> [[3]]
#>   unitid control year
#> 1 100654       1 2023
#> 2 100663       1 2023
#> 3 100690       2 2023
#> 4 100706       1 2023
#> 5 100724       1 2023
#> 6 100733       1 2023

Filters

Filters can also be applied using either NSE or standard evaluation. With NSE, multiple filters are separated with commas, which are treated as AND (&).

NSE

df <- ipeds_init() |>
  ipeds_select(instnm, stabbr) |>
  ipeds_year(2023) |>
  ipeds_filter(stabbr == "KY", control == 1) |> # stabbr == "KY" & control == 1
  ipeds_get()

df |> head()
#>   unitid year                                                instnm control
#> 1 156231 2023               Ashland Community and Technical College       1
#> 2 156338 2023 Southcentral Kentucky Community and Technical College       1
#> 3 156392 2023             Bluegrass Community and Technical College       1
#> 4 156620 2023                           Eastern Kentucky University       1
#> 5 156648 2023         Elizabethtown Community and Technical College       1
#> 6 156790 2023                Hazard Community and Technical College       1
#>   stabbr
#> 1     KY
#> 2     KY
#> 3     KY
#> 4     KY
#> 5     KY
#> 6     KY

df |> dplyr::distinct(stabbr, control)
#>   stabbr control
#> 1     KY       1

If you want a more complex filter, you will likely need to use standard evaluation.

Standard

df <- ipeds_init(use_nse = FALSE) |>
  ipeds_select(c("instnm", "stabbr")) |>
  ipeds_year(2023) |>
  ipeds_filter("stabbr == 'KY' & (control == 1 | control == 2)") |> 
  ipeds_get()

df |> head()
#>   unitid year                                  instnm control stabbr
#> 1 156189 2023                     Alice Lloyd College       2     KY
#> 2 156213 2023                       Asbury University       2     KY
#> 3 156222 2023             Asbury Theological Seminary       2     KY
#> 4 156231 2023 Ashland Community and Technical College       1     KY
#> 5 156286 2023                   Bellarmine University       2     KY
#> 6 156295 2023                           Berea College       2     KY

df |> dplyr::distinct(stabbr, control)
#>   stabbr control
#> 1     KY       2
#> 2     KY       1

Note that with ipeds_init(use_nse = FALSE) you will have to use standard evaluation throughout the call chain. For more information about how filters work, see the vignette on filter behavior.