Dynamic answer-dependent multiple-choice questions and holistic assessment analysis in high-enrollment courses

Area of Science:

• Educational Technology
• Higher Education Pedagogy
• Computer Science Applications in Education

Background:

• High student-to-instructor ratios in 4-year public institutions create significant pedagogical challenges.
• Traditional grading methods for multiple-choice questions are time-consuming, limiting faculty interaction with students.

Purpose of the Study:

• To develop and demonstrate an automated workflow for grading multiple-choice questions using the R programming language.
• To enhance the efficiency of the teaching team by automating grading and enabling more time for student engagement.
• To provide a versatile tool for detailed assessment analysis, including question performance and demographic integration.

Main Methods:

• Implementation of a custom workflow in the R programming language for grading multiple-choice assessments.
• Development of functionality to handle errors and grade answer-dependent style multiple-choice questions.
• Integration of student demographic data for comprehensive performance analysis.

Main Results:

• The R workflow enables rapid grading and accurate adjustment for errors in multiple-choice questions.
• The system supports the analysis of answer-dependent questions, categorized by factors like Fundamental Statements or Bloom's Taxonomy Levels.
• Demonstrated ability to integrate student demographics for a holistic view of course performance.

Conclusions:

• The developed R workflow significantly enhances the efficiency of grading multiple-choice assessments in higher education.
• This approach allows instructors to identify specific factors influencing student performance and adapt teaching strategies accordingly.
• The workflow provides a dynamic and versatile tool for assessment analysis, ultimately fostering a more effective learning environment.