Discussant Webster West.

Slides:



Advertisements
Similar presentations
Implementation and Order of Topics at Hope College.
Advertisements

Panel at 2013 Joint Mathematics Meetings
An Active Approach to Statistical Inference using Randomization Methods Todd Swanson & Jill VanderStoep Hope College Holland, Michigan.
Chapter 4: Designing Studies
Understanding the Variability of Your Data: Dependent Variable.
A new approach to introductory statistics Nathan Tintle Hope College.
Using Insights from SCHEMATYC Workshop in my Statistics Course Naomi Schmidt Department of Economics, Applied Statistics, and International Business New.
Introduction to Econometrics The Statistical Analysis of Economic (and related) Data.
Copyright (c) Bani K. Mallick1 STAT 651 Lecture #16.
Click on image for full.pdf article Links in article to access datasets.
Overview of STAT 270 Ch 1-9 of Devore + Various Applications.
Robert delMas (Univ. of Minnesota, USA) Ann Ooms (Kingston College, UK) Joan Garfield (Univ. of Minnesota, USA) Beth Chance (Cal Poly State Univ., USA)
Who is Really Responsible for On-line Students’ Technical Support? James R. Lackey, Ph.D. Oklahoma State University Stillwater, Oklahoma.
Statistics: Unlocking the Power of Data Lock 5 Hypothesis Testing: Hypotheses STAT 101 Dr. Kari Lock Morgan SECTION 4.1 Statistical test Null and alternative.
Understanding the Variability of Your Data: Dependent Variable Two "Sources" of Variability in DV (Response Variable) –Independent (Predictor/Explanatory)
Introducing Inference with Simulation Methods; Implementation at Duke University Kari Lock Morgan Department of Statistical Science, Duke University
Advanced Higher Statistics Data Analysis and Modelling Hypothesis Testing Statistical Inference AH.
Ch 10 – Intro To Inference 10.1: Estimating with Confidence 10.2 Tests of Significance 10.3 Making Sense of Statistical Significance 10.4 Inference as.
Copyright © 2011 Pearson Education, Inc. Putting Statistics to Work.
Maths Study Centre CB Open 11am – 5pm Semester Weekdays Check out This presentation.
Stat 112 Notes 5 Today: –Chapter 3.7 (Cautions in interpreting regression results) –Normal Quantile Plots –Chapter 3.6 (Fitting a linear time trend to.
+ Using StatCrunch to Teach Statistics Using Resampling Techniques Webster West Texas A&M University.
Reflections on Using Simulation Based Methods to Teach Statistical Methods Amanda Ellis and Melissa Pittard University of Kentucky, Department of Statistics.
Yandell - Econ 216 Chap 1-1 Chapter 1 Introduction and Data Collection.
Joan Donohue University of South Carolina
Scott Elliot, SEG Measurement Gerry Bogatz, MarketingWorks
Section 9.4 Day 3.
More on Inference.
Advanced Higher Statistics
Stat 100 March 20 Chapter 19, Problems 1-7 Reread Chapter 4.
Teaching Statistics with Simulation
Teaching Introductory Statistics
8-1 of 23.
Introduction Osborn.
Observational Study vs. Experimental Design
Statistical inference: distribution, hypothesis testing
Simulation: Sensitivity, Bootstrap, and Power
More on Inference.
Stat 217 – Day 28 Review Stat 217.
(Random) Advice For Instructors
Lesson Comparing Two Means.
EQ: How well does the line fit the data?
Statistical Inference
Unit 3 – Linear regression
Chapter 4: Designing Studies
(or why should we learn this stuff?)
Chapter 4: Designing Studies
Introduction to Econometrics
Using Simulation Methods to Introduce Inference
Sampling and Sample Size Calculations
Using Simulation Methods to Introduce Inference
Lesson Using Studies Wisely.
Chapter 4: Designing Studies
CHAPTER 12 Inference for Proportions
CHAPTER 12 Inference for Proportions
Chapter 4: Designing Studies
Chapter 4: Designing Studies
Inference for Who? Students at I.S.U. What? Time (minutes).
Chapter 4: Designing Studies
Chapter 4: Designing Studies
Chapter 4: Designing Studies
What is different? Student Reactions Student demographics
Chapter 4: Designing Studies
Chapter 4: Designing Studies
Analysis Miss Johnson.
Chapter 4: Designing Studies
STT215 Course Overview Collecting Data Exploring Data
Chapter 4: Designing Studies
Statistical Inference for the Mean: t-test
Chapter 4: Designing Studies
Presentation transcript:

Discussant Webster West

My experience Taught a simulation based intro stat course at NCSU over several semesters. Collected large amounts of data on how students learn via simulation No strong evidence that students learn more when I teach with simulation Tactile simulations do seem to improve the learning process a bit.

Comments/Questions for Karsten In many ways, this is the most well designed study with the most randomization. Odd that the only improvements with simulation came for confidence intervals. Very disappointing that the P-value is still elusive even with simulation. How does the use of the ARTIST instrument make this study compare to the others that are CAOS based?

Comments/Questions for Beth This study appears to do the most with covariate information. Pre/post design is very nice. There still appears to be some nasty lurking variables primarily related to instructor/curriculum. CAOS type questions are probably much less likely to be used in traditional courses. Why aren’t we seeing more improvements at the end of the semester?

Comments/Questions for Nathan The focus on background ability is interesting. Pre/post is a nice design. There is a lack of randomization and accounting for confounding factors – instructor, etc. CAOS type questions are probably much less likely to be used in traditional courses. Why aren’t we seeing more improvements at the end of the semester?

Comments/Questions for Bob Lack of pre/post data and other covariates (like instructor info) makes interpreting the results very difficult. The instructor effect is very clear in that the Minnesota student performed much better than others. CAOS type questions are probably much less likely to be used in traditional courses. Why aren’t we seeing more improvements at the end of the semester (outside of Minnesota)?

General Comments What are we missing with simulation? Standardization is very important. Inference with summary statistics One sample hypothesis testing problems are not natural with simulation. Why are we so concerned with inference? Random samples are almost impossible to come by Basic inference is often not very interesting with even moderately sized data sets. Can’t we do something better?