Assessment Report

Course: STAT216, Introduction to Statistics
Semester: Fall 2014 AY
Submitted by:  Jim Robison-Cox and Jade Schmidt

 

Overview

Number of students in course: Fall 820, Spring 620
Number of students assessed: 1140

Learning Outcomes

LO# Learning Outcomes
1. Interpret and draw inferences from mathematical or statistical models represented as formulas, graphs, or tables.
2. Represent mathematical or statistical information numerically and visually.
3. Employ quantitative methods such as arithmetic, algebra, geometry, or statistical inference to solve problems.

Assessment Questions

Aligned LO# Exam Questions
1. 250 people who frequently suffer from headaches agreed to participate in a study. 100 of these people were randomly assigned to recieve a new headache medication when they had a headache, and the other 150 people recieved the old headache medication. The time until the patient reported that they no longer had a headache was recorded. Choose the most valid conclusion for these data. 
2.

A research article reports the results of a new drug test. The drug is to be used to decrease vision loss in people with macular degeneration more effectively than the current treatment. The article gives a p-value of 0.04 in the analysis section. Is the following interpretation valid or invalid? 

"We conclude that the new drug is not effective because the difference in the proportion of macular degeneration patients with vision loss between the two treatments is only 0.04."
3. A graduate student is designing a research study. She is hoping to show that the results of an experiment are statistically significant. What type of p-value would she want to obtain?

Assessment

Criteria for Learning Outcomes LO 1 LO 2 LO 3
Total number assessed Fall 620, Spring 519 Fall 620, Spring 519 Fall 620, Spring 519
Number of students demonstrated acceptable level      
Proportion of students rated as acceptable Fall 77.6%, Spring 81.7% Fall 51.1%, Spring 49.6%

Fall 76.5%, Spring 75.8%

Fall 56.1%, Spring 56%

Fall 65.2%, Spring 64.8%
Does this meet minimum 2/3 threshold? yes no

yes

no

no
Comments/ideas for better alignment of course or assignment The Q core is designed to teach students quantitative reasoning, and introductory statistics should be a primary course in which students learn to “make sense of data”.  none We intend to focus more effort on confidence interval interpretation, and will build questions to probe this understanding into future final exams. We plan to reassess this effort in Spring 2016 with results from the final exam this semester.

 

 

 

Comments/ideas for improving the assessment process

 With the large numbers of students enrolled in Stat 216 each semester, we feel uncomfortable with assessing just a few assignments from a few students who were
taught by a subset of our instructors. We instead are using a nationally developed multiple choice web-based survey tool which had a high student participation rate. The NSF funded assessment will likely not be available in future years. As a substitute, we have adapted some similar questions (derived from the GOALS exam1) and use them as
part of our final exam, so these could be used to assess many of these learning outcomes. We need to make some changes or additions to get information on learning
outcomes 2 and 3 above.		 We need a better way to evaluate the second set of objectives. Our students are using graphs in class every day to separate out extreme observations in a null distribution, and they transfer such plots to their activity work books, so we think the assessment did not ask them the right
questions. We also need to think about what it means to “represent statistical information numerically”. Computing appropriate summary statistics could fit into this category, and certainly computing a p-value or a confidence interval could as well.		 none

 

PDF of STAT216 Q-Core Assessment Report