Course Description:
This course builds on the ideas and tools presented in Biostat I.
We will present the essential ideas behind statistical inference, and learn how to explore and analyze data.
Through this course, you will develop skills necessary to formulate scientific questions, and translate those
questions into statistical models. After going over estimation, hypothesis testing, and confidence intervals,
we will study regression using linear, logistic and log-linear models.
|
Instructor:
Sandrah (Sandy) Eckel, seckel@jhsph.edu website
Prerequisites:
Chuck Rohde's Biostat I course or permission of the instructor.
Software:
We will work mainly with the statistical software R.
John Verzani's
SimpleR
may be a useful reference. Install R on your own computer
following these directions.
Time:
Lectures: Generally Mon, Tue, Thu, Fri, 8.30-10.15am
Labs: Generally Mon, Tue, Thu, Fri, 10.30-12.30pm
Important! There are many exceptions to this general schedule.
See the course
schedule for more details.
Place:
Aleksandria Learning Centre, Classroom K130, City centre campus
Address: Fabianinkatu 28 (entrance through the Fabianinkatu 26 gateway)
|
|
Office hours:
Mondays (each week) and Tuesdays (week 18 and 21) or
Thursdays (week 17, 19 and 20), 2-3pm in my office at Ruskeasuo campus, Room 6241
Grading:
Lab participation = 10%
Three homework assignments = 60% (20% each)
Final exam = 30%
Attendance Policy:
You may miss up to 3 days of class (either lecture and lab on a given day) with no penalty. If you do miss a class
it is your responsibility to go over the material that we covered that day on your own before you attend the next class. If
you have questions on material that you missed, come to my office hour or ask me after class.
For every day beyond three days that you miss, you will have 5 percentage points deducted from the lab participation
portion of your grade. For example, if you missed a total of 4 days of class, you would be able to receive at most 5/10
of the 10% of your grade that comes from lab participation. If you missed 5 days, you would get at most 0/10 (none) of the 10%
of your grade coming from lab participation. If you miss more than 5 days, we'll talk. It is very important to attend class!
Homework Policy:
Homework assignments are a chance for you to get hands on practice with the material covered in lectures and labs.
You are encouraged to work in groups of two to prepare your homework solutions; partners turning in a single assignment
will share the same grade.
Homeworks will be completed by collaborating WITHIN groups but NOT BETWEEN groups. If two or more groups turn in assignments
that are excessively similar, all involved groups will recieve a grade of 0 on the assignment. To receive full credit on your
assignment, show YOUR work! If you have questions about the homework assignment, please see me before class, during break,
after class, or during my office hours.
Assignments are due at the beginning of class on the day they are due. Homeworks will not be accepted after 9am on
the day that they are due. If you cannot arrive at class on time, make sure your partner can turn it in before the 9am
deadline. Otherwise, you are free to turn it in the day before it is due. I will make no exceptions to this rule.
Final Exam:
Wednesday, May 21 from 10.00-12.00pm
LUENTOSALI 2 (A wing, 2nd floor), Mannerheimintie 172/Kytosuontie 9
Pre-Final Exam Office Hours:
Wednesday, May 21 from 8.30-10.00am
Multimedia classroom (C wing, 2nd floor), Mannerheimintie 172/Kytosuontie 9
S: Slides
N: Notes (more convenient form of slides for printing)
L: Lab material / Exercises
H: Homework
| Date |
S |
N |
L |
H |
Topic |
|
| April | 21 |
 |
|
|
|
Review from fall, exploratory data analysis |
| 22 |
|
|
|
|
Probability and distributions |
| 24 |
|
|
|
|
Normal distribution, sampling distribution and CLT |
| 25 |
|
|
|
|
Statistical inference: confidence intervals |
|
| 28 |
|
|
|
|
Statistical inference: hypothesis testing |
| 29 |
|
|
|
|
Dichotomous variables and Chi-square tests |
| 30 |
|
|
|
|
Analysis of variance |
|
| May | 5 |
|
|
|
|
Correlation and simple linear regression |
| 6 |
|
|
|
|
Linear regression: multiple covariates and confounding |
| 8 |
|
|
|
|
Linear regression: model assumptions, diagnostic plots |
| 9 |
|
|
|
|
Linear regression: model assessment, nested models, hypothesis testing |
|
| 12 |
|
|
|
|
Linear regression: Interactions and splines |
| 13 |
|
|
|
|
Introduction to logistic regression |
| 15 |
|
|
|
|
Interpreting logistic regression models |
| 16 |
|
|
|
|
Effect modification and confounding in logistic regression |
|
| 19 |
|
|
|
|
Logistic regression diagnostics, splines and interactions |
| 20 |
|
|
|
|
Review, Question and answer |
| 21 |
|
|
|
|
Exam |