Download presentation
Presentation is loading. Please wait.
1
Joan Donohue University of South Carolina
Updating a Business Statistics Course: Incorporating Ideas from the ASA Curriculum Guidelines for Under-graduate Programs in Data Science Joan Donohue University of South Carolina
2
Topics Currently Covered in our Introductory Business Statistics Course
Descriptive Statistics Probability Discrete Probability Distributions Continuous Probability Distributions Sampling Distributions Confidence Intervals for Means and Proportions Hypothesis Tests for Means and Proportions Simple Linear Regression The above course is now being taught by the Statistics Department to all NEW undergraduate business students. Business faculty will instead teach a new Business Statistics course covering more applied and advanced topics. This semester a trial version of the new course is being taught to a small section of 18 Honors students.
3
Content in the New Business Statistics Course
Data Visualization Review Descriptive Statistics and Probability Distributions Data Exploration (obtaining and cleaning data) Working with “Big Data” (using pivot tables) Review of Hypothesis Tests on Means and Proportions ANOVA Multiple Regression Time Series Chi-Square Tests Simulation, Quality Control, or Nonparametrics 3 case studies 2 individual projects, 3 group projects 6 on-line homework assignments 3 tests
4
Examples of ASA Guidelines Being Incorporated into the New Course
Section 2.3, Data at the Core Data Science is highly experiential – a practiced art and developed skill. Students must encounter project-based, real-world applications with real data. There should be a heavy emphasis on data analysis, with more weight on “data” than “analysis.” Section 2.6, Flexibility We must prepare students to be able to learn techniques and methods that do not even exist today. Students should work with increasingly varied forms of data while incorporating practical data science skills.
5
Examples of ASA Guidelines Being Incorporated into the New Course
Section 3.3, Model Building and Assessment Students need informal modeling skills such as data visualization to identify potential sources of variation and help determine appropriate models. Students should be aware of how data issues such as sampling methods and sources of bias impact formal models and generalization of statistical findings. Section 3.6, Knowledge Transference Students should gain experience using oral, written, and visual modes to communicate effectively to a variety of audiences. Students need ethical training regarding citation, data ownership, security and sensitivity of data.
6
Project/Case Studies Done So Far in the Pilot Run of the New Course
1. Data Visualization Project – Tufte’s Rules Use darwinator.com to set up a tournament for students to evaluate each other’s projects. 2. Cardiac Care Case Study Use provided data to evaluate the care provided at local hospitals (guideline questions provided). 3. Data Exploration Project (done in pairs) Summarize any chosen data set from a suggested or other website (practice gathering their own data). 4. Big Data Manipulation Assignment Answer questions related to a real-world “Big Data” set (use pivot tables to manipulate and summarize big data sets).
7
1. Data Visualization – Tufte’s Rules
8
1. Data Visualization – Tufte’s Rules
9
1. Data Visualization – Tufte’s Rules
10
1. Data Visualization Provide students with 5 examples of poorly displayed data and ask them use the data from one of those graphs to prepare their own graph that improves upon the current presentation. Students submit a jpg image of their graph on darwinator.com in the contest set up by the instructor. After submissions are received, students then randomly pick 10 submissions to view, score, and leave comments.
11
1. Data Visualization
12
1. Data Visualization
13
1. Data Visualization
14
1. Data Visualization
15
1. Data Visualization
16
1. Data Visualization
17
2. Cardiac Care Case Study Students use provided data to evaluate care provided to cardiac patients at local hospitals. 3 guideline questions are provided. Students turned in a 2 page report. This would be difficult to scale up to the masses without changing the data from one student to the next.
18
2. Cardiac Care Case Study Questions a) How good is the quality of care at each location? Is one of the hospitals clearly outperforming the other? b) How many of the patients admitted in December 2016 would we expect to be readmitted within 30 days? c) Should Columbia invest in upgrades to its ambulance fleet? Justify.
19
3. Data Exploration Project (done in pairs)
Students are asked to summarize any chosen data set of about rows. Suggested websites: stats.nba.com data.gov quandl.com Turn in a 2 page report that “tells a story” using visual and numerical summaries of the data.
20
3. Data Exploration Project Example
21
3. Data Exploration Project Example
22
3. Data Exploration Project Example
23
4. Big Data Manipulation Assignment
Students answer questions related to a “Big Data” set with 14,000 ATM transactions in a provided Excel file. Question pools were set up so each student answered slightly different questions. The use of pivot tables was suggested for manipulation of the data.
24
4. Big Data Manipulation Assignment
25
Future Projects/Cases
2-sample hypothesis test – Group Project Students collect their own survey data. ANOVA – Case study Multiple Regression – Group project Students find their own data set. Time Series – Case study Kaggle.com – enter prediction contest
26
“Kaggle.com” Project
27
“Kaggle.com” Project
28
“Kaggle.com” Project
29
“Kaggle.com” Project
30
“Kaggle.com” Project
31
“Kaggle.com” Project (79 explanatory variables)
32
ASA Guidelines Incorporated into the New Course
Project-based. Real-world applications with real data. Work with varied forms of data. Develop data visualization and informal modeling skills. Sampling methods and sources of bias. Communication and ethics.
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.