A Method Of Predicting The Chances Of Successor Failure For Individual Students In Large Introductory Engineering Physics Classes

NOTE: The first page of text has been automatically extracted and included below in lieu of an abstract - .-. Session 2530 A Method of Predicting the Chances of Successor Failure for Individual Students in Large introductory Engineering Physics Classes Highly Correlative Discriminant Indicators Are Used to Determine a Student’s Chances for Success in an Australian First- Year Engineering Physics Course Scott Grenquist Department of Physics University of Newcastle, Australia ABSTRACT: It would be except ionally beneficial to know which students were going to pass and which studculs were going to Ihil a subject prior to beginning the class. I Iowever, due to the dynamic nature of the teaching method, where instructo]-s @ to help poorer students, and students rise to the challenge of the new material being taught, it is nearly impossible to pi-edict the final distribution of marks in a subject. This is especially true for small classes whine the instmctol- can assess the students continually and individually, and respond to ihcir various problems. [t is also true tbr subjects taught in the third and tburth years of an engineering student’s university education, due to smaller student class sl~es and a more cohesive student grouping. 1 Iowever, in introductory subjects, such as }~ngine.x-x-ing Physics, where the student class sizes range from 170 to 350 students, there is Iitlle chance tbr the Iect urer to assess the students individually or continually. Also, instead of the students being a cohesive group of students, as they are in the later years of their education, they come from extremely ciiversc educational, economic and cultural backgrounds. ‘1’bc large population sizes and hetemgenous mixture of these introductory subjects, in concert with the exhaustive testing ofthe students’ capabilities both from their Tertiary Ilntrance Scores and their introductory university achievement tests, allow highly correlative predictive indicators to be applied that give students a probabilistic assessment of their chances to successfully complete their introductory Engirwcring Physics subject, In this way, a student knows kom the very beginning where they stand in the class, and the efh-t they must put forward to compctc successfully in the class, In addition, their progressive assessment throughout the course of the subject can be used by the student as a positive feedback mechanism concerning their progress through the subject. The results of the analysis show that students can be ad~rised as to their chances of failing or passing to a high probability, but obviously not with absolute certainty. 1 Iowever, some students that have exceedingly high scores or exceedingly low scores in their high school and university testing can be given almost total assurance that they will pass or fail the subject respectively. INTRODUCTION At the Universi~ of Newcastle, there are two separate sequences of two semester-based engineering physics subjects that al-e ol~ercd by the Departrnemt of Physics to students that are enrolled in courses leading to degrees in 1 {engineering or the Physical Sciences. The less rigorous of these two sequences includes 1>1 lYS 111 and PI IYS 112, which are required for all engineering students that are not enrolled in either E1eetrical Engineering or Computer Engineering. The engineering degrt.xx serviced by this sequence arc Mechanical Engineering, Chemical Engineering, industrial Engineering, Civil Engineering, Materials Engineering, Environmental Engineering and Surveying. PHYS 111 is also required separately by several other degwc programs located in the h’acuity of Sciences and Mathematics. This subject is by fw the largest of the iutroductmy engineering physics courses, incorporating between 300-400 students each year. In eompari.son, the PHYS 113 and PI IYS 114 sequence, the more rigorous of the two enginecling physics sequences, is required only iix the Electrical Engineering, Computer Engineering and Physics students. Due to the greater selectiviv and rigour of the PI IYS 113 and PI IYS 114 sequence, it has traditionally contained a much smaller cohort of st udents, ranging tlorn 1 ~0-200 studen[s cacb year. Due to the large numbers of students that are required to be enrolled in these two sequences (greater than 500 students each year), and to the substantial amount that arc unsuccessful in tbcir first attempt to pass the classes (~s’%o-q~~.), it has bccorne important to objectively and quantitatively discern the probability that an individual student has of passing the course during their initial attempt. For instance, if it were found that a student had a high probability of passing the UXMSCS in their respective sequence during their tlrst attempt, they would instructed in that regard, and may direct (heir effo~ls into other courses in which they may be lacking. On the other hand, if a student was found to have a low probability of passing the course, then tbcy would be advised to that et~ect, and be directed toww-d available help sessions and tutors, or toward remedial instruction that would better prepare tkn fbr these introductory engineering physics sequences. Traditionally, the Department has adopted the policy that two key indicators wwdd be consulted to determine the probability of a prospective student’s success in either of the two sequences. These two indicators WXX-C the student’s scores on their I Iighw School Cellificate (1 MC) ‘Fertiary Entrance Examination, and their beginning-of-the-year Physics Attaimncnt Test. If their scores on both of the two tests were sufficiently high, then they were infbrnwd that they should have iew problems in passing the subjects. On the other *, . .- . .- @Xi~ 1996 ASEE Annual Conference Proceedings ‘.,,,pyllj .