Skip to contents

Create Free Text Log

create_free_text_log.Rd

Collates free text responses from text fields in the survey data. Some language detection is performed and placed in the log notes section for possible translation.

Usage

create_free_text_log(
  response_data,
  form_schema,
  url,
  type_to_keep = "text",
  existing_log,
  columns_to_exclude = NULL
)

Arguments

response_data

data.frame of ODK questionnaire responses

form_schema

data.frame or flattened ODK form schema

url

The ODK submission URL excluding the uuid identifier

type_to_keep

String. Regex pattern passed to stringr::str_detect.

existing_log

data.frame Existing log used to create semi-clean data. Used to prevent double entry of items.

columns_to_exclude

Character. Character vector of columns to exclude from the log.

Value

data.frame validation log

Details

Since this a "type" based validation, we must provide the existing log to prevent the perptual addition of entries. Unlike other logs, this log looks for specific types of data in the schema. If they are found, a new record is created. Even if a record is marked validated, its type will not change and so we must ensure the record is only added to the log if it is unvalidated in the current log. Any records that are added to the log but are unvalidated (in their original state) and duplicates, will be removed when logs are combined in combine_logs.

Examples

if (FALSE) { # \dontrun{

odk_schema_data <- schema_from_odk_xlsx_template(file_path = "inst/example_odk_schema.xlsx")

create_free_text_log(response_data = semi_clean_data,
                      form_schema = odk_schema_data,
                      url = "https://odk.xyz.io/#/projects/project-name/submissions",
                      existing_log = existing_log)

# exclude certain columns from the free text log
create_free_text_log(response_data = semi_clean_data,
                      form_schema = odk_schema_data,
                      url = "https://odk.xyz.io/#/projects/project-name/submissions",
                      existing_log = existing_log,
                      columns_to_exclude = c("household_id", "human_id")
                      )
} # }