MEDS 300: Statistical Methods for Biomedical Research
Professors: Dr. Elahe Nezami
E-mail:
Health Promotion & Disease Prevention Program:
Angela Almer Turk
3375 S. Hoover St.
University Village, Ste. E210
Los Angeles, CA 90089
; (213) 821-1601
Course Description and Goals
Throughout this class, students pursuing pre-medical education will learn basic concepts in statistics and research methods. Students will be familiarized with various statistical tests (parametric and non-parametric), the underlying assumptions, and statistical analyses used in health fields. Upon completion of this course, students will appreciate the role and quality of statistical methods in published medical research, and will know how to appraise the evidence relevant to clinical decisions.
Students will receive detailed instruction in each of the following areas:
· Basic concepts, uses and misuses of statistics
· Scientific method
· Research hypothesis development and testing
· Theories, constructs and variables (theoretical vs. operations definitions)
· Self-report measures (open vs. closed-ended, measurement scales, survey development)
· Research ethics
· Questionnaire development— reliability, validity (external and internal) and threats to validity
· Experimental studies (study designs, threats, causation)
· Sampling (random sampling, stratified sampling, convenience sampling)
· Study design (observational and experimental)
· Diagnostic tests
· Probability distribution
· Concepts of population and sample
· Descriptive statistics
· Importance of normal distribution, “normal ranges” for laboratory tests
· Confidence intervals and p-values
· The “art” of significance testing
· Parametric and non-parametric tests
· Interpreting statistical results
· Appraising the evidence
· Meta analysis and regression analysis
· Reading medical literature
Textbooks
Rosnow, R.L., & Rosenthal, R. (2008). Beginning Behavioral Research: A Conceptual Primer. 6th Edition. New Jersey: Prentice Hall.
NOTE: A copy will be available on reserve at Leavey Library. The earlier 2005 5th Edition can also be used for this course.
Kiess, H.O. & Green, B.A. (2008) Statistical Concepts for the Behavioral Sciences (4th ed.). Boston: Allyn and Bacon.
Bock, David E. et al. Stats: Modeling the World. 3rd edition; Boston. Yates, Moore, & Starnes. The Practice of Statistics, 2008.
Grade ComponentsHomework/Lab Assignments / 10%
Project :
due the final exam day / 20%
Mid-term Exam / 30%
Final Exam / 30%
Participation
(class activities, in-class assignments, discussions) / 10%
Grade Scale
A 95-100%
A 90-94%
B+ 87-89%
B 83-86%
B 80-82% / C+ 77-79%
C 73-76%
C 70-72%
D+ 67-69%
D 63-66%
D 60-62%
F 61%-0%
Course Outline:
Class Week / General TopicWeek 1 / Part 1 – Sampling Distribution Models
· Simulating a Sampling Distribution Model
· Sampling Variability
· Describing the Sampling Distribution Models for Sample Proportions in terms of Center, Spread, and Shape
· Assumptions and Conditions for the Sampling Distribution Model of Sample Proportions
· Calculating Probabilities Based on the Sampling Distribution Model of Sample Proportions
· Describing the Sampling Distribution Models for Sample Means in terms of Center, Spread, and Shape
· Central Limit Theorem
· Assumptions and Conditions for the Sampling Distribution Model of Sample Means
· Calculating Probabilities Based on the Sampling Distribution Model of Sample Means
· Law of Diminishing Returns
· Standard Error of the Sampling Distribution Model
Part 2 – Confidence Intervals for Proportions and sample parameters
· Sampling Variability
· Estimating Population Parameters
· Point Estimates
· Margin of Error
· Interpreting Confidence Levels
· Critical Values of z*
· Creating a One-Proportion Z-Interval
· Interpreting Confidence Intervals
· Assumptions and Conditions for a One-Proportion Z-Interval
· Calculating Minimum Sample Size for a given Margin of Error
Week 2 / Part 3 – Testing Hypotheses About Proportions
· Logic of a Hypothesis Test
· Null vs. Alternate Hypotheses
· Idea of Rejecting vs. Retaining the Null Hypothesis
· Conducting a One-Proportion Z-Test
· Calculating a Probability Value (P-Value)
· Assumptions and Conditions for a One-Proportion Z-Test
· One-sided vs. Two-sided Hypothesis Tests
· Drawing Conclusions from our Data
· How Hypothesis Tests and Confidence Intervals are Related
Part-4 – More About Tests
· P-values as a Conditional Probability
· Making a Decision based on an Alpha Level
· Critical Values for a Hypothesis Test
· Comparing a Hypothesis Test to a Confidence Interval
· Type I and Type II Errors
· Power of the Test
· The Relationship between Alpha, Beta, and Power Effect Size
Week 3 / Part-5 – Comparing Two Proportions
· Sampling Distribution Model for the Difference Between Two Independent Proportions
· Assumptions and Conditions for Two-Proportion Inference
· Creating a Two-Proportion Z-Interval
· Idea of Pooling
· Conducting a Two-Proportion Z-Test
· Relationship between an Interval and a Test
Week 4 / Part-5 – Comparing Means
· Sampling Distribution Model for the Difference Between Two Independent Means
· When to Use the Z-distribution vs. the T-distribution
· Assumptions and Conditions for Two-Sample Inference for Unpaired Means
· Creating a Two-Sample T-Interval for Unpaired Means
· Idea of Pooling
· Conducting a Two-Sample T-Test for Unpaired Means
· Relationship between an Interval and a Test
· TI: Calculating a Two-Sample T-Interval for Unpaired Means, Calculating a Two-Sample T-Test for Unpaired Means
Part- 6 – Paired Samples and Blocks
· Paired Data vs. Independent Samples
· Assumptions and Conditions for Inference for Paired Means
· Creating a Matched-Pairs T-Interval for Means
· Conducting a Matched-Pairs T-Test for Means
Week 5 / Part-7 – Comparing Counts
· Intro to ANOVA
· One-Factor ANOVA
· Multiple Comparison Tests
· Intro to Linear Models
· Two-Factor ANOVA
Week 6 / Part-8 – Comparing Counts
· Chi-Square Distribution
· Chi-Square Test of Goodness of Fit
· Assumptions and Conditions for Chi-Square Tests
· Expected Counts vs. Observed Counts
· Chi-Square Test of Homogeneity
· Chi-Square Test of Independence
Week 7 / Part-9 –Non parametric tests and Inferences for Regression
· Idealized Regression Model
· Assumptions and Conditions for Inference for Regression
· Sampling Distribution Model for the Slope of the Regression Line
· Single & multiple linear regression
· Polynomial regression
· Pearson, Spearman correlation
Week 8 / Midterm
Weeks 8 & 9 / Part-10-Clinical Trials
· Basic concepts and designs:Different types of designs (Experimental, qualitative, and quantitative). Controlled and uncontrolled clinical trials; historical controls; protocol; placebo; randomization; blind and double blind trials; ethical issues; protocol deviations.
· Size of trials.
· Multiplicity and meta-analysis:interim analyses; multi-centre trials; combining trials.
· Cross-over trials.
· Binary response data:logistic regression modelling; McNemar's test, relative risks, odds ratios.
Week 10 / Part-11-Survival Data Analysis
· Basic concepts:survivor function; hazard function; censoring.
· Single sample methods:lifetables; Kaplan-Meier survival curve; parametric models.
· Two sample methods: log-rank test; parametric comparisons.
· Regression models: inclusion of covariates; Cox's proportional hazards model; parametric and accelerated failure time models.
Weeks 11 and 12 / Part-11-Design of Experiments
· Notation:Design region. Variation and blocking. Stages in experimental research. Randomisation.
· Criteria for a good experiment:
Optimality criteria.
· General theory of block designs
· Factorial designs:
Estimability. Blocking. Confounding. Screening.
· Optimum Design theory:
General Equivalence Theorem. Experiments with constraints. Design construction.
Weeks 12, 13 and 14
(project is due the final exam day) / Part-12-Final Project will demonstrate students understanding of the statistical and design concepts
· Choose a question of interest.
· Design an appropriate study or experiment.
· Collect data and describe the data using graphical displays, and summery statistics.
· Make inference about the population and test the results.
· Justify methods and state conclusions
· Make a presentation for the class, and submit a report.
Week 15 / Final exam
Academic Dishonesty
Academic dishonesty on exams, assignments, and other aspects of the course is grounds for failure on the assignment, failure in the course, or expulsion from the university. This is another strict policy. Students are expected to understand what constitutes plagiarism and other forms of cheating, as well as the consequences. For information, see http://www.usc.edu/dept/publications/SCAMPUS/gov/. Click on University Governance and then Behavior Violating University Standards and Appropriate Sanctions to find definitions of cheating. Sanctions are listed under 11.80. The minimum official consequence for cheating is course failure.
Students with Disabilities
Any student requesting academic accommodations based on a disability is required to register with Disability Services and Programs (DSP) each semester. A letter of verification for approved accommodations can be obtained from DSP. Please be certain the letter is delivered to the Instructor as early in the semester as possible. DSP is located on the University Park campus in STU 301 and is open 8:30 a.m. – 5:00 p.m., Monday through Friday. The phone number is (213) 740-0776.
Student Athletes
Student athletes are expected to keep the Professor or Teaching Assistant informed of any necessary situations which might lead to a missed class.
Emergency Preparedness/Course Continuity:
In case of emergency, and travel to campus is difficult, USC executive leadership will announce an electronic way for instructors to teach students in their residence halls or homes using a combination of Blackboard, teleconferencing, and other technologies. Instructors should be prepared to assign students a "Plan B" project that can be completed at a distance. For additional information about maintaining your classes in an emergency please access: http://cst.usc.edu/services/emergencyprep.html