Comparing Student Performance Using Calculators With Performance Using Excel (Must You Know How To Ride A Horse If You Want To Drive A Car?)

Our school offers a senior year course in engineering economics. For several years students have been allowed to take routine tests in this course by using either computers with (and) spreadsheet programs(s) or conventional interest tables and calculators. The tests are the same for both sets of students and have been constructed and graded in such a way as to render differences in performance between using spreadsheets or calculators statistically insignificant. There are however significant differences between the approaches used by students using spreadsheets and those used by students using calculators on the same questions. There are also differences in the sources of difficulty and error. This paper discusses these differences and the methodology used in creating and grading these tests. A final observation is that over a period of several years the use of spreadsheets has increased. Introduction At our school, Engineering Economics is taught in a conventional classroom with several one hour examinations at regular class times during the semester. There is also a final examination that is also conducted in a regular classroom at the end of the semester. Starting in the Fall of 2001, the author allowed students to use laptops during in-class tests. Students who wanted to use laptops during a test could. Those who wanted to use their calculators could continue to use them. The course content was modified to include examples of both spreadsheet and calculator methods so that both computational approaches would be taught. Also, students were told that the FE requires calculators. This paper discusses how that change has impacted teaching engineering economics at our school. Tests were graded by the instructor as normal after allowing laptops in class. After each test, the class was told the average grades of students using laptops and of students using calculators. From the test averages students knew there was no incentive from the standpoint of grades for the use of either tool. Laptop availability was not a problem because there were a number of laptops available from the university and, many students borrowed laptops from friends. All students who used laptops used the Excel spreadsheet so in this paper I use “computer solution” and “spreadsheet solution” interchangeably. Allowing both the use of laptops AND calculators in the same test required some revision of the test procedures. Some important considerations regarding the nature of the test questions are: 1. All tests (hour quizzes and the Final Exam) were “open book, open notes”. This policy was already in place so that students could use the interest tables provided in the text. Because students already had access to notes and tables, the fact that the students could P ge 11338.2 use laptops to store notes or crib sheets presented no perceived advantage. In 2001 the school did not have a wireless network so students were not able to use laptops to communicate. 2. Test questions were written to reduce the number of repetitive calculations. 2.1 Students were told to perform no more than two iterations on any problem requiring an iterative solution whether or not they had obtained convergence. Two iterations allows the instructor to determine if the student has the correct solution procedure but does not burden a student who is using a calculator with time consuming repetitive calculations more easily performed on a computer. 2.2 Problems which had complicated cash flows over multiple years (such as calculations of After Tax Cash Flow with depreciation and loan payments) were performed for only 1 or 2 years. 2.3 Calculations involving long time periods (more than 4 years) with many different amounts were not used. 3. Until 2004, students using laptops provided their test answers on physical media; floppy disks, CD’s or flash drives provided by the instructor. After 2004, Widener a wireless network and many of the students were able to e-mail the results. 4. There was no statistical difference between the grades of students who used laptops and those who did not and the pooled results were available to the classes. Success of this policy is indicated by the following written response to a survey of students. “In this class, the tests were structured so that doing the test with a laptop or by hand did not favor either party.” Student Acceptance of the Use of Laptops The initial response to allowing the use of laptops during tests was positive based on one on one conversations. The percent of students using laptops increased as shown in Figure 1. The increased use of laptops indicates a growing acceptance of spreadsheet use in tests.