Analyze rater comments in BTA

Text Analytics can be applied to both published and unpublished projects, and questions can be assigned to and removed from dictionaries at any time during the life of the project, so it is possible that feedback has not been analyzed by the Text Analytics engine or that the analysis was performed using a different dictionary.

Blue has been designed to handle these situations.

The following assumes that you have a published project with Text Analytics already setup.

Follow this simple procedure to verify and perform the proper analysis for all feedback.

1. Manage the published project and navigate to Text Analytics.
2. Verify when feedback was last analyzed and compare the number of analyzed comments to the total number of comments.
3. Decide if all the comments should be analyzed or if only new comments should be analyzed.
4. Click Start to analyze the comments.

Calculating the number of Comments

The total number of comments calculated in Text Analytics - Analyze Comment Responses can grow quickly because of several contributing factors: the number of dictionaries in the project, the number of comment questions, the total number of completed evaluations, and whether questions are course specific or instructor specific (course specific questions are counted once, whereas instructor specific are counted once for each instructor).

In other words, the following formula can be used to calculate the total number of comments to be analyzed in any scenario:

(# of completed questionnaires) x (# of Comments questions to analyze) x (# of dictionaries analyzing the questions) x (# of question occurrences)

For example, if there are 30,000 completed questionnaires that answer two Comments questions, analyzed by two dictionaries for a course question, the total number of comments equal 30,000 x 2 x 2 x 1 = 180,000.

If there is a team-taught course, then the completed tasks will be multiplied by the number of teachers. For example, if there are 30 completed questionnaires for 1 course with 3 teachers, being analyzed by 2 dictionaries then there are 30 x 1 x 3 x 2 = 180 comments to be analyzed regardless of how many comments actually included a response.