Use of Sample Items to Measure Student Ability to Apply Computing Knowledge and Mathematics

Student Learning Outcomes:

Students will have the ability to apply knowledge of computing and mathematics appropriate to their discipline. Performance indicators for this SLO are based on 2 factors: demonstrating knowledge of the material and the ability to apply it.

1. Students are considered to have satisfactory knowledge if they can show:

a) an understanding of computing principles, methods, and techniques related to the discipline

b) an understanding of the related mathematics to the problem

2. Students are considered to be able to have met the application by:

a) apply principles, methods, and techniques related to the field to solve a problem

b) apply or use the required mathematical approaches to solving the problem              

This learning outcome is considered met if the weighted mean ratings of sample items indicate at least 80% of students receive a rating of “C.”

Methods:

Using sample items designed to reflect SLO criteria, data was collected by the instructor in Spring 2012 from 67 students enrolled in the course CPSC 335 Problem Solving Strategies, and in Fall 2011 from 31 students enrolled in the course CPSC 481 Artificial Intelligence.

Sample questions by criterion:

Knowledge

1a: Understand Big O notation: Indicate True or False: 3nlog n + 7 ∈ θ (n2)

1b: Find Big O of sums: Rewrite (Σ2n−1i=0) + 3 in closed form and prove efficiency class.

Application

2a: Solve recurrence relation: (n) = 4(n/2) + 3n2, T(1) = 1

2b: Find Efficiency class of following code: What is the efficiency of this algorithm in the worst case?

     Alg (A[0…n-1]):

     do:

                flag = False

                for i from 1 to n-1:

                                  if A[i-1] > A[i]:

                                               swap(A[i-1], A[i])

                                               flag = True

     while flag;

     return A

2a & b: Compute the information gain for each attribute and determine the attribute that must be used for the root node of the decision tree for the following weather data set that consists of two attributes, “outlook” and “temperature”, and a class “picnic” that makes a binary decision.

Picnic	Outlook	Temperature
Yes	Sunny	Cool
Yes	Overcast	Hot
No	Rainy	Mild
Yes	Overcast	Cool
Yes	Sunny	Hot
Yes	Sunny	Mild

Results:

Responses were rated on a 4-point scale (A. Excellent, B. Satisfactory, C. Developing, D. Unsatisfactory,). With more than one set of data collected for this outcome, the weighted average was calculated and results are shown in the graph below.

 

Using the weighted mean of all samples collected for this learning outcome, it was revealed that 89% of Computer Science students enrolled in CPSC 335 and CPSC 481 recieve a rating of "C" or higher on sample items.  These results indicate the criteria for satisfying this SLO were met.

Improvement Actions:

Since the assessment process is tied to specific courses in the Department of Computer Science, it was recommended that an exit survey or examination be developed to provide a more objective assessment of student learning.