INTERMEDIATE TOPICS IN BIOSTATISTICS FOR HEALTH SCIENCESPHST 651

Course Data

Number:PHST 651

Title:Intermediate Topics in Biostatistics for Health Sciences

Credithours:3

Department:Bioinformatics and Biostatistics

School/College:School of Public Health and Information Sciences

Type:Lecture

Location:TBA

Meeting Time:TBA

Semester:TBA

Date Updated:12/1/2015

Course Description

This course provides an introduction to selected intermediate topics in biostatistics. Topics will include the following: linear models/regression, analysis of variance (ANOVA), categorical data analysis using the generalized linear model with logistic and Poisson regression, model selection techniques, nonparametric inference, quantile regression, and multivariate data analysis (factor models, clustering, and discriminant analysis). Heavy emphasis will be placed on implementing statistical methods in commercial statistical software packages.

Course Objectives

Upon successful completion of this course, students will be able to:

  1. Fit linear statistical models using a commercial statistical software package, including the analysis of variance (ANOVA) of single-factor, randomized block, factorial, and repeated measures data, simple, and multiple linear regression models, and analysis of covariance models (ANCOVA).
  2. Interpret output provided by linear models procedures from a commercial statistical software package, and conduct appropriate follow-up analyses including multiple comparisons procedures and diagnostic examinations of traditional linear model assumptions.
  3. Understand categorical data analyses, which include contingency tables, comparing two proportions, chi-squared test, logistic regression and multinomial logistic regression for categorical data, and log-linear model for count data.
  4. Interpret output provided by categorical analysis procedures from a commercial statistical software package, and successfully interpret the results.
  5. Understand and implement classical nonparametric statistical methods for one-sample, two samples inference, nonparametric regression, etc. Understand and implement quantile regression. Use statistical software to carry out various methods.
  6. Using statistical software, apply multivariate analysis techniques (discriminant, cluster, and factor analysis) to learn from data and successfully interpret the results

Prerequisites

PHST 500 and 501, equivalent introductory statistics or biostatistics coursework, or instructor permission.Enrollment in a Master’s or Ph.D. program in the School of Public Health and Information Sciences or instructor consent is required.

Course Instructors

This course will consist of four modules each taught by a different instructor.

Topic Area / Name / Office / Phone / Email
ANOVA and Regression / Doug Lorenz, Ph.D. / SPHIS 134 / 1-502-852-3635 /
Categorical Data Analysis / Maiying Kong, Ph.D. / SPHIS 138 / 1-502-852-3988 /
Nonparametrics and
Quantile Regression / Qi Zheng, Ph.D. / SPHIS 123 / 1-502-852-8780 /
Multivariate Statistical Analysis / Jeremy Gaskins, Ph.D. / SPHIS 130 / 1-502-852-3300 /

The course instructors welcome conversations with students outside of class. Students may correspond with instructors by email or set up appointments by contacting the instructor of a particular topic area.Students should also contact the relevant instructor for a given topic area with questions they might have regarding the mechanics or operation of the course. Questions that are not specific to a particular module and/or that regard the course in its entirety shall be directed to Dr. Doug Lorenz.

Course Topics and Schedule

The table below provides the topics to be covered for each subject area and each instructor. The content and form of each lecture is determined by the instructor of note. All instructional material – notes, readings, data sets, etc. – will be furnished by the instructor of note for the given lecture. Any questions about a lecture should be directed to the appropriate instructor for that lecture.

IMPORTANT NOTE: The schedule and topics may change as the course unfolds. Changes will be posted on Blackboard.

Week / Topic Area / Instructor / Lecture 1 / Lecture 2
Week 1 / ANOVA / Lorenz / One-Way ANOVA, Multiple Comparisons, Diagnostics / Multi-Way ANOVA – Factorial and RCB Designs
Week 2 / ANOVA / Lorenz / Repeated Measures ANOVA; Adjusting for continuous covariates (ANCOVA) / Nonparametric alternatives to ANOVA; Random and mixed effects ANOVA
Week 3 / Regression / Lorenz / Simple regression; Diagnostic and remedial measures / Multiple regression, categorical predictors (ANCOVA); Multicollinearity
Week 4 / Regression / Lorenz / Polynomial regression; Non-linear regression / Model selection and building; Stepwise methods and selection criteria
Week 5 / Contingency tables / Kong / Describe contingency tables; introduce concepts for relative risk and odds ratio. / Compare two proportions
Week 6 / Chi-squared test / Kong / Test independence in two-way contingency tables / Introduce logistic regression
Week 7 / Multiple/multinomial
logistic regression / Kong / Multiple logistic regression / Multinomial logistic regression
Week 8 / Log-linear model for count data / Kong / Log-linear models / Inference for log-linear models
Week 9 / Nonparametric
Tests / Zheng / Dichotomous data problem / One-sample location problem
Week 10 / Nonparametric
Tests / Zheng / Two-sample location problem / The independence problem
Week 11 / Nonparametric
Regression / Zheng / Nonparametric regression / Quantile regression
Week 12 / Multivariate / Gaskins / Intro to multivariate data; covariance matrices / 2-group discriminant analysis
Week 13 / Multivariate / Gaskins / Multi-group discriminant analysis / K-mean classification
Week 14 / Multivariate / Gaskins / Principle Components; Intro to Factor Analysis / Factor analysis

Course Materials

Blackboard

The primary mechanism for communication in this course, other than class meetings, is UofL’s Blackboard system at Instructors may use Blackboard to make assignments, provide materials, communicate changes or additions to the course materials or course schedule, and to communicate with students other aspects of the course. It is imperative that students familiarize themselves with Blackboard, check Blackboard frequently for possible announcements, and make sure that their e-mail account in Blackboard is correct, active, and checked frequently.

Required Textbook

None.

Required Software

IBM SPSS. Student versions are available at the University’s software resales store, iTechXpress, found at A University account is required to log in to the store and purchase software online.Students may also purchase software at the iTech Xpress store, located in the lower level of the Miller Information Technology Center on the Belknap campus

Recommended Textbooks

The primary instructional resource will be the lecture notes provided by each individual instructor.The following textbooks are recommended for further reading and as reference texts on each topic area, but are not required for completion of the course.

Linear Models – ANOVA and Regression

  • Weisberg, S. (2005). Applied Linear Regression. 4th edition. Wiley, New York. Specific to regression, fairly easy to follow with a heavy bent toward applications. A companion website is available:
  • Montgomery, D.C. (2001). Design and Analysis of Experiments, Wiley, New York. Comprehensive coverage of linear models, at a more applied and less theoretical level than Kutner, et al.
  • Kutner M.H., Nachsteim C.J., Neter, J. (2004).Applied Linear Statistical Models.5th edition.McGraw-Hill/Irwin.This text is somewhat advanced, but fairly comprehensive in scope, covering ANOVA, ANCOVA, regression, and even dabbling in random and mixed effects models.An excellent (albeit expensive) reference text.

Categorical Data Analysis

  • Agresti, A. (2007).An Introduction to Categorical Data Analysis, 2nd edition. Wiley-Interscience.
  • Agresti, A. (2002). Categorical Data Analysis, 2nd edition. Wiley & Sons, Inc., Hoboken, New Jersey.

Nonparametrics and Quantile Regression

  • Sprent, P., Smeeton, N.C.(2007) Applied Nonparametric Statistics. 4th edition. Chapman and Hall.
  • Daniel, W.W.(2000) Applied Nonparametric Statistics. 2nd edition. Duxbury.
  • Koenker, R. (2005) Quantile Regression. Cambridge University Press.

Multivariate Statistical Analysis

  • Johnson, R.A., Wichern, D.W. (2007) Applied Multivariate Statistical Analysis. 6th edition. Pearson Prentice Hall
  • Stephens, J. (2009) Applied Multivariate Statistics for the Social Sciences. 5th edition. Routledge

Prepared Materials Used by Instructors

Materials used by instructors in class are available to students via Blackboard no later than 24 hours following the class. These may include outlines, citations, slide presentations, and other materials. There is no assurance that the materials include everything discussed in the class.

Course Policies

Attendance and Class Participation

Students are expected to either attend class regularly or keep up with online versions or recordings of the lectures.Class participation is encouraged but not considered as part of a student’s grade.

Student Evaluation and Grading

Course grades will be assigned based on the student’s performance on homework assignments, in-class quizzes, and two reports. Each report will be worth 5% of the course grade (10% total), quiz scores will be worth 30% of the course grade, and homework assignments will be worth 60% of the course grade.

Students will be assigned homework assignments to complete for grade. Each of the four modules will have two or three assignments. Each assignment will contain 3-5 problems that will require the student to implement and interpret the data analysis techniques learned in class, and assignments will require the use of statistical software. Homework will be announced a week in advance of the due date, and late submissions will not be accepted.

At the end of each module, students will take an in-class quiz to assess their understanding of the material covered in that module. The quizzes will test for understanding of fundamental concepts and the ability to correctly interpret analyses. Each quiz will be given during the first 20-30 minutes of the first class period following the completion of a module.

For each report, the student will need to find an article from a journal in his/her field of study (health management, environmental health, etc.) that uses one of the methods discussed in class. The student will write a one-to-two page (single spaced) report discussing the article and critiquing the application of the statistical methods. In particular, the student should focus on whether appropriate methods were chosen and if they were interpreted correctly. The first report will be due [** week 8 **] and will critique an article that uses a statistical method discussed during the first 7 weeks of the course. The second report will be due [** finals day **] and will discuss an article using a method learned after week 7.

The course letter grade will be assigned as follows: A: 90-100, B: 80-89, C: 70-79, D: 60-90, F: 0-59. Students taking the course as Pass/Fail will need to earn a C grade or better (70 and above) to earn a grade of “Pass”.

Other Policies

UofL Policy on Sexual Misconduct (Sexual Harassment, Sexual Assault, and Sexual/Dating/Domestic Violence) and Sex Discrimination

Sexual misconduct (including sexual harassment, sexual assault, and any other nonconsensual behavior of a sexual nature) and sex discrimination violate University policies.Students experiencing such behavior may obtain confidential support from the PEACC Program (852-2663), Counseling Center (852-6585), and Campus Health Services (852-6479). To report sexual misconduct or sex discrimination, contact the Dean of Students (852-5787) or University of Louisville Police (852-6111).

Disclosure to University faculty or instructors of sexual misconduct, domestic violence, dating violence, or sex discrimination occurring on campus, in a University-sponsored program, or involving a campus visitor or University student or employee (whether current or former) is not confidential under Title IX.Faculty and instructors must forward such reports, including names and circumstances, to the University’s Title IX officer.

For more information, see the Sexual Misconduct Resource Guide.

Syllabus Revision

The course director reserves the right to modify any portion of this syllabus. A best effort is made to provide an opportunity for students to comment on a proposed change before the change takes place.

Inclement Weather

This course adheres to the University’s policy and decisions regarding cancellation or delayed class schedules. Adjustments are made to the class schedule as necessary to take into account any delays or cancellations of this class. Local television and radio stations broadcast University delays or closings. The UofL web site and telephone information line (1-502-852-5555) also broadcast delays or closings.

Grievances

A student who has grievances regarding the course shouldseek to have the matter resolved through informal discussion and through administrative channels, such as the course director, chair of the course’s department, associate dean for student affairs, and university grievance officer. If the issue remains unresolved, the student may file a formal grievance. More information is located at Summary of SPHIS Student Academic Grievance Procedure in Student Academic Grievance Committee.

Disabilities

In accordance with the Americans with Disabilities Act, students with bona fide disabilities are afforded reasonable accommodation. The Disability Resource Center certifies a disability and advises faculty members of reasonable accommodations. More information is located at the University’s disability resource center page.

Academic Honesty

Students are required to comply with the academic honesty policies of the university and School of Public Health and Information Sciences. These policies prohibit plagiarism, cheating, and other violations of academic honesty. More information is located in the SPHIS policy on student academic honesty.

Course instructors use a range of strategies (including plagiarism-prevention software provided by the university) to compare student works with private and public information resources in order to identify possible plagiarism and academic dishonesty. Comparisons of student works require students to submit electronic copies of their final works to the plagiarism-prevention service. The service delivers the works to instructors along with originality reports detailing the presence or lack of possible problems. The service retains copies of final works and may request students’ permission to share copies with other universities for the sole and limited purpose of plagiarism prevention and detection.

In addition instructors provide the opportunity for students to submit preliminary drafts of their works to the service to receive reports of possible problems. Such reports are available only to the submitting student. Copies of preliminary drafts are not retained by the service.

Continuity of Instruction Plan

A plan for continuity of instruction for this course has been developed and published. All plans are available online. Continuity of instruction plans provide guidance for how instruction may be modified to lessen disruptionby events that affect transportation, communication, or personal interaction. Such events may be weather-related (e.g., floods, blizzards, tornados), health-related (e.g., epidemics), or other widespreadoccurrences or threats.

Additional Policy Information

Additional policy information is available in the following:

SPHIS Catalog

SPHIS Policies and Procedures

UofL Graduate Catalog

Page 1 of 8