Survey of statistical methods that introduces experimental design and analysis of variance; multiple linear regression; analysis of residuals to assess model fit; categorical data analysis including chi-squared tests and logistic regression; non-parametric tests; and statistical power. Emphasizes precise statistical communication and implementation using software with programming capabilities. Major project required.
Athena Title
Statistical Methods (Honors)
Equivalent Courses
Not open to students with credit in STAT 4210
Prerequisite
(STAT 2000 or STAT 2000E or STAT 2100H or BUSN 3000 or BUSN 3000E or BUSN 3000H or BIOS 2010 or BIOS 2010E) and permission of Honors
Semester Course Offered
Offered spring
Grading System
A - F (Traditional)
Student learning Outcomes
Students will evaluate potential study designs to select one that provides adequate statistical power and supports the intended scope of the study’s conclusions.
Students will use statistical software to calculate summary statistics, construct graphs, and fit models that can be used to make predictions and quantify explained and unexplained variability.
Students will use multivariable models to describe the relationship between variables of interest while controlling for potential confounding variables and nuisance variables that increase variability.
Students will assess model fit and select a final model that accounts for important data features (e.g., interaction between variables, non-linear relationships).
Students will use the statistical problem-solving process as a framework to evaluate alignment between the research question, data collection, analysis methods, and conclusions.
Topical Outline
Introduction to the statistical problem-solving process
Introduction to multivariable thinking
Parametric and nonparametric inference methods to compare means
Statistical power
Sums of squares (analysis of variance) to quantify sources of variation
One-way and two-way ANOVA F-tests
Multivariable models that account for interactions between predictors
Full-factorial designs and randomized block designs
Multiple regression models with quantitative and/or categorical predictors
Analysis of residuals and model building
Transformations for modeling nonlinear associations
Chi-squared tests of independence and goodness-of-fit
Logistic regression
Institutional Competencies Learning Outcomes
Analytical Thinking
The ability to reason, interpret, analyze, and solve problems from a wide array of authentic contexts.
