Exporting Data

To analyze learner submissions for a given problem in a workbook or practice session, Codespec provides a CSV export, which you can then import into DataShop. Each row represents a line in the learner’s submitted code; for block-based submissions, each row is a block; for fix- and write-code submissions, each row is a line. Although fix- and write-code submissions are more open-ended, for the purposes of comparing the rows with their block-based counterparts, the learner’s code is still evaluated against the closest-matching Parsons solution (for columns like expected_order, expected_indentation, line_outcome, etc).

Table with column names and definitions coming soon.

Survey responses maybe viewed by navigating to the "Results" page in the console for a given survey. There you will find a list of visualizations by question. Single- and multi-answer multiple choice questions, as well as numeric and Likert scale questions, display results in the form of a bar chart. Input question responses are displayed in tabular form.

For any question, you may download a CSV of the results. Columns consist of:

• row_id - an auto-incrementing field for each row

• survey - the code/id for the survey to which the question belongs

• page - the code/id for the survey page to which the question belongs

• question - the code/id for the question to which the response belongs

• survey_user_id - the respondent's anonymous id (e.g., learner_001_01). "learner_001" is the "base" anonymous id for the respondent; "_01" represents an instance of the respondent taking this particular survey. If this number is greater than 1, for example "_03", it means this response is from the third time the respondent took the survey.

• user_question_id - the id for the specific instance the respondent answered the corresponding question. This is useful for grouping responses to a multi-answer multiple choice question for a given survey user.

• user_question_input_value - if the question was an input field type, this column will contain the survey user's response

• user_question_selected_choice - if the question was a numeric or Likert scale question, this column will contain the id/code of the selected choice

• user_choice_choice_id - if the question was a single- or multi-answer multiple choice question, this column will contain the id/code of the selected choice. There will be 1 row per selected choice.

• user_choice_input_value - if the selected choice displayed an input field (e.g., for an "other" elaboration), this column will contain any text the respondent entered

• timestamp - the date and time the survey user's response was created

• timezone - the timezone of the timestamp (always UTC)