Assessing Student Learning in an Introductory Computer Science Course: Predictors of Success in Subsequent Computer Science Courses

San Diego State University Foundation, San Diego CA

Investigators

Abstract

This project seeks to address a challenging issue in Computer Science (CS) education: determining the readiness of students who complete the introductory CS course to be successful in subsequent CS courses. Accurately assessing student learning in introductory programming courses is complex and challenging. Research studies suggest that many students who receive passing grades in their first programming class (CS1) are unprepared to succeed in subsequent classes in the CS curriculum. Several studies have reported failure rates of 30% or more in students' second programing course (CS2). This project seeks to improve methods for assessing student learning in CS1 and for identifying students who are likely to be able to succeed in CS2 and later courses. The results of the project could help advisors and instructors guide CS1 students in choosing an appropriate educational path. What is learned through this project's research agenda could apply not just to CS majors, but also to all students who need to learn about computer programming. Finding effective ways to assess student preparation may also help address the persistent underrepresentation of women and other groups of students in the CS major. Objective evidence that a student is likely to succeed in later courses might encourage that student to persist in the major, even if the student is not completely confident about being prepared. Some CS education researchers have recommended the use of laboratory exams in CS1 as a direct measure of students' programming ability. In this type of exam, students are asked to write a program in a closed-lab setting with a fixed time limit and access to a compiler. Other researchers assert that the primary objective of CS1 should not be to produce accomplished programmers, but rather to give students mastery of all the basic programming concepts. Still others suggest that CS1 should foster students' abstract reasoning abilities and should provide appropriate learning resources to promote development of more advanced skills. This project will investigate and provide evidence concerning these different points of view. Students in CS1 will complete a series of programming tasks in a controlled laboratory setting to assess their programming skills. They will also take written exams that assess their mastery of fundamental programming concepts and their level of abstract reasoning ability. The results from these assessments will be compared with students' subsequent performance in the following CS2 course. Evaluation of project outcomes will use standard quantitative statistical methods and qualitative data derived from interviews with selected students. In addition to performance in CS2, the evaluation criteria will include factors such as students' perceived levels of test anxiety and their overall pattern of performance on the various types of assessment. These evaluations will also consider subgroups of students according to gender, ethnicity, prior programming experience, and other factors. The results may provide important information about the applicability and equity of using different assessment methods as a primary means of measuring student learning of fundamental CS skills. Project research findings may benefit CS departments across the nation regarding successful introductory CS courses which foster the talent of a broad spectrum of students. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.