Data-Based Decisions

Implemented Action:

In response to high failure and withdrawal rates in pre-calculus and first semester calculus courses, and in introductory statistics course, computerized placement tests have been implemented for all mathematics courses up to the first semester of calculus and introductory statistics. The same ACCUPLACER tests have been adopted by all state funded institutions of higher education in Utah. The ACCUPLACER College Level Mathematics test was first administered on February 6, 2007 and fully implemented for the Logan, UT campus for Fall semester 2007.

• In Fall semester 2008 the use of the ACCUPLACER mathematics placement tests was extended to cover USU’s regional campuses and distance education sites and concurrent enrollment sections of mathematics taught in high schools statewide.
• There is research in the mathematics education literature that indicates that mathematics skills atrophy very quickly when not use. So, a one-year time limit on the acceptability of mathematics prerequisites for all mathematics courses through Calculus I and introductory statistics courses has been adopted. Students wishing to register for one of these classes that have not passed an appropriate college mathematics or statistics course, or scored high enough on a nationally administered test, such as ACT, SAT, or AP, within the last year must take the ACCUPLACER mathematics test.
• For students that have been away from mathematics for more than one year, the Department has instituted a set of refresher courses that are offered the week before classes begin in Fall and Spring semesters. The refresher courses cover material from Math 0900 (Elements of Algebra), Math 1010 (Intermediate Algebra), Math 1050 (College Algebra) and Math 1060 (Trigonometry). The fee for the refresher course includes the cost of taking the placement test at the beginning and the end of the refresher course. Almost all students do better on the second testing; many qualify for admission to one class higher than that which they qualified for before taking the refresher course.
• In response to concerns about uneven quality of instruction and assessment in Math 1100 (Calculus Techniques), a common final examination was instituted.
• In response to low enrollment numbers, two statistics electives, Stat 5110 and 5300, have been dropped from the Department’s course listings. The most relevant material from Stat 5110 (Theory of Linear Models) has been moved to a 6000-level course in which it is more appropriate. Stat 5300 (Statistical Process Control) primarily served students in other departments outside of the College of Science, and when the enrollments dropped below 10 consistently the course was removed from the regular rotation of topics classes in statistics. These two electives have been replaced— beginning Fall semester 2006—by Stat 5410/6410 (Spatial Statistics) and Stat 5570/6670 (Bioinformatics), which are more relevant for both undergraduate and graduate students in statistics.
• In response to low enrollments and reduced budgets, the quantitative literacy course Math 1030 (Quantitative Reasoning) is no longer offered. Students that might have taken this course in the past will now be advised to take Stat 1040 (Introduction to Statistics).
• In response to changes in skills needed by students with graduate degrees in statistics, a number of changes have been made in the collection of courses taken by M.S. and Ph.D. students in statistics.

Actions Under Discussion:

• To address concerns about differences in exams used at USU’s Logan campus and those used at regional sites, under consideration is a proposal to use a group of common problems in all final exams for Math 1010, irrespective of where the course is being delivered from.
• Students that continue to take the placement examination until USU’s add deadline at the end of the third week of classes sometimes find that when they finally attend the section of the course for which they are qualified for admission, they are so far behind that they cannot successfully complete the course. The Department is considering a proposal to stop giving the placement exam after the end of the first week of classes.
• In order to address the question of how students that qualified for admission to courses in different ways perform in the class, a pilot study is being carried out in Math 1100 (Calculus Techniques). The method by which students qualified for admission to Math 1100 has been recorded and student performance will be evaluated at the end of the semester.
• Material in Stat 5110 (Theory of Linear Models) and Stat 6120 (Generalized Linear Models) has been augmented with new material to create a new course, Stat 6100 (Advanced Regression Analysis), that replaced Stat 5110 and 6120.
• Stat 6510 (Resampling Methods) and 6520 (Smoothing and Density Estimation) have been reduced and combined into a single new course, Stat 6530 (Modern Nonparametric Statistics).
• A new course, Stat 6190 (Wavelet Methods for Time Series), has been created to alternate with Stat 6180 (Time Series) so as to provide with both traditional and modern time series methods.
• Stat 6740 (Bayesian Statistics) has been added to the curriculum, reflecting the important part of theoretical and applied statistics that is Bayesian.
• Stat 6600 (Multivariate Statistics) has been replaced by Stat 6650 (Statistical Learning: Multivariate Statistical Methods for Bioinformatics, Data Mining, and Machine Learning), which is the modern approach to multivariate statistics.

Further Topics:

• Monitoring of all courses affected by the placement test is ongoing. Of particular interest are the D-F-W rates and performance of persons that qualify for admission via the placement test as opposed to other means.
• Issues of consistency in sections of Calculus I and Calculus II have been raised. Mechanisms for coordinating instructors and a common final have been discussed.
• For many years, Stat 2300 (Business Statistics) has been taught in a large lecture format with recitations supplementing the lectures. Starting Fall semester 2009 five additional classes are being taught in a large lecture, three lectures and two recitation sections per week format: Stat 1040 (Introduction to Statistics), Math 1050 (College Algebra), Math 1100 (Calculus Techniques), Stat 3000 (Statistics for Scientists), and Math 2250 (Linear Algebra and Differential Equations), which is taught with four lectures and no recitation sections per week. We will be monitoring student performance in these courses very closely, and comparing outcomes data with the same data from previous years.