SYLLABUS PHE: 515 Introduction to Biostatistics SCHOOL OF COMMUNITY HEALTH Professor Alexis Dinno (503) 725-3076 alexis.dinno@pdx.edu WINTER QUARTER, 2016 Office: Office Hours: Classroom: Class Meetings: 450D Urban Center Tuesdays & Wednesdays 10:00–11:50 Neuberger Hall 59 Mondays & Wednesdays, 14:00–15:50 COURSE DESCRIPTION This course covers a broad range of basic statistical methods used in the health sciences. The course begins by covering methods of summarizing data through graphical displays and numerical measures. Basic probability concepts will be explored to establish the basis for statistical inference. Confidence intervals and hypothesis testing will be studied with emphasis on applying these methods to relevant situations. Both normal theory and nonparametric approaches will be studied including one- and twosample tests of population means and tests of independence for two-way tables. Students will be introduced to one-way analysis of variance (ANOVA), correlation, and simple linear regression. We focus on understanding when to use basic statistical methods, how to compute test statistics and how to interpret and communicate the results. We require software (Stata) as part of the course to introduce you to basic data management, reading output from statistics packages, interpreting and summarizing results. LEARNING COMPETENCIES (SEE APPENDED COURSE COMPETENCY MATRIX) 1) Select and generate graphical and numerical summaries of data. 2) Use principles of statistical inference to make conclusions about populations from samples. 3) Communicate statistical findings to others. 4) Use computer software to conduct simple statistical analysis. PHE 515 Introduction to Biostatistics is restricted to students in the Oregon Master of Public Health degree program. Conversely, the core Biostatistics requirement can be fulfilled only by this course or an equivalent Master of Public Health biostatistics course at OHSU. This course stresses biostatistical methods with example data and problems appropriate to public health (including the identification and access of public health data sets), explicit emphasis on analysis of data from epidemiologic study designs, inference about epidemiologic measures (e.g. relative risks, odds ratios), tests for equivalence alongside tests for difference, and stresses the links between statistical inference for public health concepts (e.g. the population perspective, the vulnerabilities perspective, etc.). TEXTS AND READING ASSIGNMENTS Pagano M, Gauvreau K. Principles of Biostatistics, 2nd Edition, Duxbury Press, Pacific Grove, CA, 2000. The book also includes a CD containing data needed for the exercises. The book is available at the PSU bookstore. All additional materials, including lecture slides (available after each lecture), are available only via my course website (http://web.pdx.edu/~adinno/index.html#PHE510) Used copies from many book vendors are listed here: http://www.fetchbook.info/compare.do?search=9780534229023. You should do the reading for a particular lecture before the lecture. You will have more pertinent questions, pick up the material more quickly, and while you can ask me a question about the last reading pretty successfully, you’ll get much less helpful answers if you ask the book a question about the last lecture. I intend optional readings to give you insight into alternatives to, histories of and applications of the methods that we cover in class: read or skim them, but don’t study them. (But do read them! ) PHE 510: INTRODUCTION TO BIOSTATISTICS WINTER QUARTER, 2016 STATISTICAL COMPUTING We will be using the statistical computing package Stata™ for this class. You may, at your option use another statistical package (R, SPSS™ or its open source equivalent, PSPP, SAS™, etc.), but will receive assistance for computing-related question for Stata only. The course materials include DINNO’S STATA CHEAT SHEET which offers both specific tips for using Stata, and links to online resources for using it. Stata (http://www.stata.com) R (http://cran.r-project.org/) PSPP (https://www.gnu.org/software/pspp/) SPSS (PSU maintains a site license through Self Service Software: http://www.pdx.edu/oit/self-service-software) SAS (http://www.sas.com/) METHODS OF EVALUATION Homework: (35% of grade) Note: Unless you contact me before the due date with a valid excuse, late homework will not be accepted. Except for the last assignment, homework will be assigned is due on paper—not electronically—one week from the date of assignment. In-class mid-term: (30% of grade) Monday, February 9. In-class final: (35% of grade) Wednesday, March 18 Note: my exams are open book/open note, but I do not permit networked digital devices such as laptops, tablets, e-readers, etc., so factor this into your textbook decisions. Extra credit: (Approximately 0–10% of course grade) Students will have the opportunity to complete extra credit assignments and participate in competitions for extra credit throughout the course. These opportunities are entirely optional, although pursuing them will learning opportunities. Grading Policy: Homework Exam I Exam II 35% 30% 35% Grading Scale Thresholds: ≥90%: A ≥84%: B+ ≥80%: B ≥74%: C+ ≥70%: C ≥64%: D+ ≥60%: D <57%: F ≥87%: ≥77%: ≥67%: ≥57%: A– B– C– D– PSU DISABILITY RESOURCE CENTER (DRC) Accommodations are collaborative efforts between students, faculty, and the Disability Resource Center (http://drc.pdx.edu/). Students with accommodations approved through the DRC are responsible for contacting the family member in charge of the course, prior to or during the first week of the term, to discuss accommodations. Students who believe they are eligible for accommodations but who have not yet obtained approval through the DRC should immediately contact the DRC, at 503-725-4150. SAFE CAMPUS MODULE If you have not done so already, please complete the Safe Campus Module in d21. The module should take approximately 30 to 40 minutes to complete and contains important information and resources. If you or someone you know was been harassed or assaulted, you can find the appropriate resources on PSU’s Enrollment Management and Student Affairs: Sexual Prevention and Response website at http://www.pdx.edu/sexual-assault/. PSU’s Student Code of Conduct makes It clear that violence and harassment based on sex and gender are strictly prohibited and offenses are subject to the full realm of sanctions, up to and including suspension and expulsion. PHE 510: INTRODUCTION TO BIOSTATISTICS CLASS SCHEDULE LECTURE # 1 2 DATE M Jan 5 Required reading: W Jan 7 Required reading: Optional reading: 3 M Jan 12 Required reading: Optional reading: 4 W Jan 14 Required reading: M Jan 19 TOPIC WINTER QUARTER, 2016 HOMEWORK DUE (PAGE COUNT) Overview of statistics; data presentation; obtaining public health data • Pagano & Gauvreau Sections 2.1–2.3 (17) Numerical summaries; rates, population health measures; population pyramids; direct and indirect age standardization • Pagano & Gauvreau Sections 3.1–3.3, 4.1–4.2 • Gould, S. J. (1985). The median isn’t the message. Discover, 6(6):42–44. (33) (3) Visual distribution summaries; probability; risk; relative risk; risk difference; odds ratio • Pagano & Gauvreau Sections 6.1–6.5 • Burnham et al. (2006). The Lancet, 368(9545):1421–1428. • Correspondence over Burnham’s article included for optional reading (25) (8) (4) Theoretical probability distributions; discrete and continuous distributions population versus sample quantities; PMFs and PDFs • Pagano & Gauvreau Sections 7.1–7.5 HW 1 (30) NO CLASS: Martin Luther King Day 5 W Jan 21 Required reading: The z-score; sample distribution of the mean; standard error of the mean • Pagano & Gauvreau Chapters 8.1–8.3 HW 2 (8) 6 M Jan 26 Required reading: Statistical inference; confidence intervals; the t-distribution • Pagano & Gauvreau Chapters 9.1–9.3 HW 3 (11) 7 W Jan 28 Required reading: 8 M Feb 2 Required reading: Optional reading: Hypothesis testing; tests for difference (positivist hypotheses); prelude to tests for equivalence (negativist hypotheses) • Pagano & Gauvreau Chapters 10.1–10.6 Comparison of two means; confidence intervals versus hypothesis testing; statistical power; equivalence hypothesis testing; relevance tests • Pagano & Gauvreau Chapters 11.1–11.2 • Dinno “Applying t tests of equivalence using two one-sided tests” • Dinno, A. (2014). Comment on “The Effect of Same-Sex Marriage Laws on Different-Sex Marriage: Evidence From the Netherlands”. Demography, 51(6):2343–2347. • Cumming, G. (2009). Inference by eye: reading the overlap of independent confidence intervals. Statistics In Medicine, 28(2):205–220. Schuirmann, D. A. (1987). A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability. Pharmacometrics, 15(6):657–680. Westlake, W. J. (1976). Symmetric confidence intervals for bioequivalence trials. Biometrics, 32:741–744. • • W Feb 4 Review session for the mid-term M Feb 9 Mid-term exam (18) HW 4 (14) (5) (5) (16) (24) (4) PHE 510: INTRODUCTION TO BIOSTATISTICS CLASS SCHEDULE (CONTINUED) LECTURE # 9 10 DATE W Feb 11 Required reading: Homework will require: M Feb 16 Required reading: Optional reading: 11 W Feb 18 Required reading: Optional reading 12 M Feb 23 TOPIC WINTER QUARTER, 2016 HOMEWORK DUE (PAGE COUNT) Analysis of Variance (ANOVA) & F-test; repeated measures ANOVA • Pagano & Gauvreau Chapters 12.1–12.2 • Glantz (2005) primer of biostatistics. 6th edition. 347–350 ANOVA & F-test; multiple comparisons; family-wise error rate; false discovery rate; introducing nonparametric methods • Pagano & Gauvreau Chapters 12.1–12.2 • Shaffer J. P. (1995). Multiple Hypothesis Testing. Annual Review of Psychology. 46:561–584. • Benjamini Y. & Hochberg Y. (2000) On the Adaptive Control of the False Discovery Rate in Multiple Testing with Independent Statistics. Journal of Educational and Behavioral Statistics. 25:60–83. Nonparametric tests: sign, sign rank, rank sum, Kruskal-Wallis & Dunn’s post hoc test for difference; equivalence and relevance • Pagano & Gauvreau Chapters 13.1–13.4 • Kruskal, W. H. and Wallis, A. (1952). Use of ranks in one-criterion variance analysis. Journal of the American Statistical Association, 47(260):583–621. • Dunn, O. J. (1964). Multiple comparisons using rank sums. Technometrics, 6(3):241–252. HW 5 (10) (4) HW 6 (10) (24) (24) (11) (39) (12) Confidence intervals for proportions; tests for proportion difference; equivalence and relevance; introducing contingency tables; testing relative risks • Pagano & Gauvreau Chapters 14.1–14.6 • Agresti, A. and Caffo, B. (2000). Simple and effective confidence intervals for proportions and differences of proportions result from adding two successes and two failures. The American Statistician, 54(4):280–288. • Cochran, W. G. (1950). The comparison of percentages. Biometrika, 37(3/4):256–266. HW 7 HW 8 Required reading: Categorical data analysis (contingency tables); McNemar’s tests for difference, equivalence and relevance; Cochran’s test • Pagano & Gauvreau Chapters 15.1–15.3 14 M Mar 2 Required reading: Correlation; prelude to linear regression • Pagano & Gauvreau Chapters 17.1–17.3 HW 9 (10) 15 W Mar 4 Required reading: Optional reading: Linear regression • Pagano & Gauvreau Chapters 18.1–18.3 • Reshef, D., et al. (2011). Detecting novel associations in large data sets. Science, 334(6062):1518–1524. HW 10 (24) (7) 16 M Mar 9 Linear regression and inference: tests of parameter difference equivalence, and relevance; introduction to survival analysis • Pagano & Gauvreau Chapters 18.1–18.3, 21.1–21.2 HW 11 W Mar 11 Review session for the final exam HW 12 W Mar 18 Final exam: NOTE DIFFERENT TIME! (12:30–14:20) Required reading: Optional reading: 13 W Feb 25 Required reading: (13) (9) (11) (16) (36) PHE 510: INTRODUCTION TO BIOSTATISTICS HOMEWORK ASSIGNMENTS HW # MATERIAL COVERED 1 Chapter 2, Chapter 3 2 Chapter 4 3 Chapter 6, Chapter 7 4 Chapter 8, Chapter 9 5 Chapter 10 6 Chapter 11, Dinno handout and article 7 Chapter 12, Shaffer, Benjamini, Glantz, lecture 8 Chapter 13 plus Kruskal-Wallis from lecture 9 Chapter 14, Agresti, 10 Chapter 15 11 Chapter 17 12 Chapter 18, Chapter 21 WINTER QUARTER, 2016 DUE IN CLASS Lecture 4 Lecture 5 Lecture 6 Lecture 8 Lecture 9 Lecture 10 Lecture 12 Lecture 13 Lecture 14 Lecture 15 Lecture 16 Lecture 17 TOTAL POINTS (%) 51 (11) 46 (10) 48 (11) 35 (8) 34 (8) 35 (8) 43 (10) 30 (7) 41 (9) 29 (7) 27 (6) 36 (8) Students are strongly urged to study in groups every week and to co-teach, to collaborate on working through the pencil and paper homework, as well as the computer assignments. However, each student must turn in her or his own homework. PHE 510: INTRODUCTION TO BIOSTATISTICS WINTER QUARTER, 2015 Core Course: Introduction to Biostatistics PHE 510 (PSU), PHPM 524 (SOM, OHSU), CPHN 530 (SON, OHSU) Credits: 4 credits COURSE COMPETENCY MATRIX Competencies 1. Select and generate graphical and numerical summaries of data 2. Use principles of statistical inference to make conclusions about populations from samples. 3. Communicate statistical findings to others Related Components Use graphical methods to display features of data. 2. Compute numerical summaries to summarize features of data. 3. Interpret graphical and numerical summaries to describe data. Competency Demonstrations Menu of Options Quizzes/Exam(s) Homework Learning Activities Menu of Options Utilize web sources Example and case study Statistical software examples End-of-unit exercise Class session Computer lab session Written class note or/and present Class session Reading Case study Using statistical software(s) Public questions – practice exercises End-of-unit exercise Application self-tests Computer lab session Guided practice and feedback Case study Individual or team project End-of-unit exercises Computer lab session Utilize internet resources Case study Guided practice and feedback 1. Apply principles of probability laws/distributions, interval estimation, and hypothesis testing. 2. Select and perform statistical procedures based on type of data and assumptions for approaches used. 3. Construct and interpret point and interval estimates for population parameters using sample data. 1. Provide a written state or verbally present: 1. Statistical methods used 2. Results obtained 3. Conclusions drawn 4. Limitations of conclusions related to study design and analysis 4. Use computer software to 1. Enter and read data into a statistical software package conduct simple statistical 2. Manipulate and transform data elements analysis 3. Use program to perform statistical analysis. 4. Interpret the output from statistical software. Quizzes/Exam(s) Homework Quizzes/Exam(s) Homework Homework