Functions

enrollment_2022_2023 <- read_excel(path = "data-raw/fallmembershipreport_20222023.xlsx", sheet = "School 2022-23") |> clean_names() math_scores_2021_2022 <- read_excel ("data-raw/pagr schools math tot raceethnicity_2122.xlsx") |> clean_names()

math_scores_2021_2022 |> filter(student_group == "Total Population (All Students)") |> filter(grade_level == "Grade 3") glimpse select(academic_year, school_id, contains("number_level_")) |> view() pivot_longer(cols = contains("number_level_")), view() names_to = "proficiency_levels", values_to = "number_of_students")

third_grade_math_proficiency_2021_2022 |> mutate(proficiency_level = str_remove(proficiency_level, pattern = "number_level_")) third_grade_math_proficiency_2021_2022 |> mutate(proficiency_level = recode(proficiency_level, "number_level_4" = "4", "number_level_3" = "3", "number_level_2" = "2", "number_level_1" = "1"))

third_grade_math_proficiency_2021_2022 |> mutate(proficiency_level = case_when( proficiency_level == "number_level_4" ~"4" proficiency_level == "number_level_4" ~"3" proficiency_level == "number_level_4" ~"2" proficiency_level == "number_level_4" ~"1"

))

third_grade_math_proficiency_2021_2022 |> mutate(proficiency_level = parse_number(proficiency_level))

enrollment_by_race_ethnicity_2022_2023 <- enrollment_2022_2023 |> select(district_institution_id, x2022_23_american_indian_alaska_native:x2022_23_percent_multi_racial) |> select(-contains("percent")) |> pivot_longer(cols = -district_institution_id, names_to = "race_ethnicity", values_to = "number_of_students")

mutate(number_of_students = parse_number(number_of_students) bind_rows(third_grade_proficiency_2018_2019)

gss_cat |> view() third_grade_math_proficiency_2021_2022 |> group_by(school_id) mutate(pct = number_of_students / sum(number_of_students, na.rm =TRUE

                                    ))

third_grade_math_proficiency_2021_2022 |> enrollment_by_race_ethnicity_2022_2023 <- enrollment_2022_2023 |> select(district_institution_id, school_institution_id, x2022_23_american_indian_alaska_native:x2022_23_multi_racial) |> select(-contains("percent")) |> pivot_longer(cols = -c(district_institution_id, school_institution_id), names_to = "race_ethnicity", values_to = "number_of_students") |> mutate(race_ethnicity = str_remove(race_ethnicity, pattern = "x2022_23_")) |> mutate(race_ethnicity = case_when( race_ethnicity == "american_indian_alaska_native" ~ "American Indian Alaska Native", race_ethnicity == "asian" ~ "Asian", race_ethnicity == "black_african_american" ~ "Black/African American", race_ethnicity == "hispanic_latino" ~ "Hispanic/Latino", race_ethnicity == "multiracial" ~ "Multi-Racial", race_ethnicity == "native_hawaiian_pacific_islander" ~ "Native Hawaiian Pacific Islander", race_ethnicity == "white" ~ "White", race_ethnicity == "multi_racial" ~ "Multiracial" ))

Can you review my code and apply any corrective changes. There is something I am not incorporating correctly. The data export is not displaying as expected.

A question for code readability:

ATM we use set_names() to get a column name which we later on change with mutate(race_ethnicity = case_when(...)) Would it be cleaner to immediately change them to the final out come or is the code more readable/understandable that way?

Incorporating functions will be a gamechanger for me! I have a related workflow question. In my work, I have several use cases where I implement similar cleaning steps across many raw assessment files; however, there is slight variation between the raw files due to differences in variable names, order, response options, etc. over time (a separate issue to tackle!).

In this case, what would you recommend for script structure? I'm thinking one R script file for establishing the functions, and another R script file that cleans each assessment by year -- starting with glimpse() or something similar to understand the structure of each file, then carrying out cleaning steps specific to that file/not standard across files, and finally implementing the functions that are standard across files. Does this make sense, or is there a better/cleaner way to approach this combined approach of working with functions and also implementing "extra" cleaning outside the functions? Thanks!