Check for issues in the survey data — check_question

Functions to check the inital survey data for issues such as unexpected answers to questions or questions with multiple answers.

Usage

check_question_columns(
  data,
  question_columns,
  survey_values = OMESurvey::survey_values,
  format = "table",
  exclude_99 = TRUE
)

check_survey(data, ID_column = "Barcode_ID", ...)

check_IDs(
  data,
  ID_column = "Barcode_ID",
  first_name_column = "First_Name",
  surname_column = "Surname",
  true_df = NA
)

Arguments

data: survey data, formatted where each row is an individual and column names describe demographics or survey questions (of the format YY__ZZ, where XX is the theme and YY the question, where ' ' are replaced with '_' and '?' with 'XX'.)
question_columns: The index of the question columns.
survey_values: A list of possible survey question answers by default it takes the values from OMESurvey::survey_values.
format: Either 'list' where the result for each question is another element in a list or 'table' which concatenates all the results into a single table.
exclude_99: Logical (TRUE/FALSE) for whether to exclude '99' from the non-expected values.
ID_column: the name of the ID column. Default is "Barcode_ID".
...: parameters for downstream functions.
first_name_column: the name of the "First Name' column. Default is "First_Name".
surname_column: the name of the "Surname' column. Default is "Surname".
true_df: a data frame including the columns ID_column, first_name_column, surname_column which contain the correct ID, first name and surname combinations.

Details

check_survey() checks survey data for issues with barcodes, unexpected values and questions with multiple answers (i.e "99"). Uses check_question_columns() to genereate unexpected values and location "99"s.
check_question_columns() checks given columns for unexpected values where the expected values are auto-detected by comparison to survey_values.

See Preparation and initial survey analysis vignette for example.