qualtRics 3.2.0

• Changed how multiple choice questions are mapped to an R factor with the convert argument to fetch_survey(), to now excluding NA as a factor level (#315)

• Deprecated the save_dir and force_request arguments for fetch_survey(), so that survey response downloads are no longer cached; calls to fetch_survey() now always query the Qualtrics API (#317)

• Added new tmp_dir argument to fetch_survey() so users can specify where survey results are briefly stored on disk (#327)

qualtRics 3.1.7

• Refactored code for checking arguments and errors, thanks to @jmobrien (#263)

• Fixed bug in fetch_survey() for include_* = NA, thanks to @jmobrien (#277)

• Updated test fixture for new version of vcr

qualtRics 3.1.6

• Add fetch_distribution_history() and list_distribution_links() for more handling of distribution data, thanks to @chrisumphlett and @dsen6644 (#221, #239)

• Changed handling of literal "NA" text input from users so it is no longer converted to an R NA value thanks to @jmobrien (#244)

• Use httr::RETRY() instead of httr::VERB() in qualtrics_api_request() to implement consistent API error-handling across all of the functions in the package. They will be retried up to 3 times if there is any non-4xx error. Thanks to @chrisumphlett (#217)

• Update fetch_distributions() for changes to the Qualtrics API, thanks to @chrisumphlett (#250)

qualtRics 3.1.5

• Add fetch_description() to download complete survey description metadata from v3 API endpoint (more up-to-date than older metadata()) thanks to @jmobrien (#207)
• Warn users about possible incorrect results from API when breakout_sets and label are both FALSE
• Refactor internal URL creation for API calls thanks to @jmobrien (#225)
• Add fetch_id() to return a surveyID based on a unique survey name as it appears in the Qualtrics UI thanks to @markjrieke (#230).

qualtRics 3.1.4

• Add fetch_distributions() to access distribution data for a specific survey thanks to @dsen6644 (#169)
• Handle mailing list embedded data better thanks to @dsen6644 (#175)
• Updated links to API documentation
• Create unique column names for questions using choiceId thanks to @lyh970817 (#182)
• Fix bug when include_questions only contains one QID thanks to @lyh970817 (#197)
• Generate correct/updated column mapping for survey responses thanks to @jmobrien (#199). These column mappings are available as an attribute on survey results or via the new extract_colmap() function.

qualtRics 3.1.3

• Update include_questions argument to use correct name in API request.
• Build API payloads with jsonlite (#155) thanks to @jmobrien
• Convert tests to webmockr and vcr (#140 and #161) thanks to @shaun-jacks and @dsen6644
• Allow user to specify column types for both fetch_survey() and read_survey() (#162) thanks to @jntrcs

qualtRics v3.1.2

• Fix bug in infer_data_types() to avoid errors with factors/numeric values
• Improvements to documentation, error checking
• Allow user to access column mapping for questions and IDs (#115)
• Deprecate registerOptions() to avoid unexpected behavior with options
• Make data import more robust with more condition and error checking, as well as better defaults
• For empty surveys, return zero row dataframe (#127)
• Remove unnecessary dependency on yaml and deprecate qualtRicsConfigFile(), to avoid unexpected behavior
• Deprecate old versions of functions: getSurveys(), getSurveyQuestions(), getSurvey(), readSurvey()
• Move to updated version of Qualtrics API (#130) thanks to @jmobrien
• Correctly handle time zone conversions (#137) thanks to @jmobrien
• Add breakout_sets parameter thanks to @shaun-jacks
• Fix another bug in infer_data_types() for answers choices that include HTML

qualtRics v3.1.0

• New maintainer: Julia Silge
• Add all previous contributors to DESCRIPTION as ctb
• Declare testthat dependency in DESCRIPTION (reason for previous archiving from CRAN)
• Simpler approach for storing API credentials as environment variables with qualtrics_api_credentials() (registerOptions() is now soft deprecated with a warning)
• Simplify README (keep all existing detailed workflow documentation in vignette)
• Relicense from GPL-3 to MIT. See consent from authors here.
• Improvements to documentation throughout
• Renaming (with warnings on old versions) of key functions for clarity and reduction in confusion, plus improvements:
  • all_surveys() (from old version of getSurveys())
  • survey_questions() (from old version of getSurveyQuestions())
  • fetch_survey() (from old version of getSurvey())
  • read_survey() (from old version of readSurvey())

qualtRics v3.01

NEW FEATURES

• Added 'metadata' function that allows the user to retrieve detailed metadata about survey.
• User can now convert specific question types automatically. See this page for more information.

MINOR IMPROVEMENTS

• Using package httptest for mock API requests so that API calls can be tested.
• getSurveys() and getSurveyQuestions() now return a tibble

BUG FIXES

• Added .onDetach conditions so that environment variables (root url and API key) are removed when package is unloaded. This prevents issues if user decides to load the package again.
• We found that surveys that use new lines in the questions break the readSurvey function.
  The problem is, that read.csv (and read.table as well as the readr library implementation) ignore the quote = """ option when a skip = 2 or skip = 3 parameter is set. As a result the read function slices off the questions row somewhere in the middle when first importing just the table body using skip.

DEPRECATED AND DEFUNCT

• convertstandardcolumns deprecated since readr::read_csv does this automagically. It has been changed in config file to 'convertvariables'.

qualtRics v2.3

We found that surveys that use new lines in the questions break the readSurvey function.
The problem is, that read.csv (and read.table as well as the readr library implementation) ignore the quote = """ option when a skip = 2 or skip = 3 parameter is set.
As a result the read function slices off the questions row somewhere in the middle when first importing just the table body using skip.