Answers for Unit 12 Assignment
3 e 1 P o m
For the population of low birth weight infants, a signiﬁcant linear relationship was found to
exist between systolic blood pressure and gestational age (data set lowbwt), etc.
a. Construct a two-way scatter plot

Answers to Assignment for Unit 9
h r14rlm
As part of a study conducted in France investigating the effectiveness of the drug
mifepristone (RU 486) for terminating early pregnancy, 488 women were administered
mifepristone followed 48 hours later by the sin

Unit 14 Assignment
Chapter 21, Problem 7
In the table in the textbook on page 511 you will find survival times in months since diagnosis for
10 AIDS patients suffering from concomitant esophageal candidiasis, an infection due to
Candida yeast, and cytomeg

Answers to Unit 11 Assignment
oft e ro ms
thpte: 17, Problem 6
Thirty-five patients with ischemic heart disease, a suppression of blood flow to the heart,
took part in a series of tests designed to evaluate the perception of pain. In one part of the
study

ACTIVE CLASS 3.1
JHU Cocoa
Content in
Chocolate Tasting
Trial (C3T2)
Designing and
implementing a research
study
QuODDID
Question of scientific interest
Outcome measures
Design
Data analysis
Interpretation
Dissemination
Public Heath Biosta0s0cs

ACTIVE CLASS 2.3
Comparing two
distributions
Confidence intervals for
difference in means
Log $ vs $ scale
Brief review of
visualization
Confidence interval recipe
CLT&
&
Truth&
&
&
&
Data&
&
&
estimate (t.025 or z.025 ) se estimate
We estimate that th

ACTIVE CLASS 3.3
Tests of
Hypotheses
Error rates
t-statistics and p-values
CIs for proportions
Finishing up from last class
Public Heath Biosta0s0cs
2
Hypothesis testing made easy
1. Specify a precise null hypothesis about the population.
The pop

Unit 4 Assignment
Chapter 9, Problem 5
The distribution of systolic and diastolic blood pressure for female diabetics between the ages of
30 and 34 have unknown means. However the standard deviations are s = 11.8 mm and d =
9.1 mm Hg, respectively.
a. A r

ACTIVE CLASS 4.1
Introduction to
College Alcohol
Survey and Linear
Regression
College alcohol survey
Intro to linear regression
QuODDID
Question: How does drinking in college affect academic
performance?
Outcome: GPA
Design: Nationally representative su

ACTIVE CLASS 2.4
Stratification and
pooling
Calculating a pooled
estimate from stratified
analysis
Log $ vs $ scale
Illustration of confounding and stratification
0 1 2 3 4 5 6
$ (log scale)
Boxplots of medical expenditures, log10($+1)
Age <= 65
Not Poo

ACTIVE CLASS 2.2
Confidence
intervals
Confidence intervals
Warm up question
Q 2.8: In 10,000 turns of the lucky number game, we expect to
see _ of each possible outcome. The reason we dont see
exactly this number is due to _.
400 800
1400
Distribution of

ACTIVE CLASS 1.2
Probability
Probability
2x2 tables
Screening tests
Review: Meanings of probability
Frequentist: long term relative frequency
Examples: Tossing a coin; disease rates
Bayesian (subjectivist): measure of personal belief
Examples: Prob

ACTIVE CLASS 1.3
Probability
Discrete probability
distributions
Binomial distribution
Binomial likelihood
Warm-up (Question 1.7)
Consider whether or not someone smokes (T) as a
screening test for whether that person has lung cancer (D).
Suppose 90% of

ACTIVE CLASS 1.1
Course
Introduction
Course information
What is probability?
The scientific method
Biostatistics 345
Objective: to enable each student to enhance his/her
quantitative, scientific reasoning
General Method: To achieve mastery, a student

ACTIVE CLASS 4.2
Linear Regression
Details of linear regression
Intro to multiple regression
Heights of fathers and sons
Pearson & Lee, Biometrika 2:357-462, 1906
10/19/15
Public Heath Biosta5s5cs
2
Heights of fathers and sons
10/19/15

ACTIVE CLASS 5.2
Fitting logistic
regression
models, by hand
and by computer
Logistic regression by
hand
Method of maximum
likelihood
Objective of analysis
Estimate risk of infant mortality as a function of gestational
age, parity and other factors
Pub

Unit 5 Assignment
Chapter 10, Problem 9
The distribution of diastolic blood pressures for the population of female diabetics between the
ages of 30 and 34 has an unknown mean d and a standard deviation d = 9.1 mm Hg. It may
be useful to physicians to know

Paired Samples versus Independent
Samples
Paired Design 1
With paired data, we are interested in
comparing the responses within each
pair. We will analyze the differences of
the responses that form each pair.
Paired Data: Response = Annual Salary (in $100

Unit 2 Assignment
Chapter 3, Problem 7
In Massachusetts, eight individuals experienced an unexplained episode of vitamin D
intoxication that required hospitalization; it was thought that these unusual occurrences might be
the result of excessive supplemen

Unit 6 Assignment
Chapter 11, Problem 5
A crossover study was conducted to investigate whether oat bran cereal helps to lower
cholesterol levels in hypercholesterolemic males. Fourteen such individuals were randomly
placed on a diet that included either o

Unit 3 Assignment
Chapter 7, Problem 12
According to the Behavioral Risk Factor Surveillance System, 58% of all Americans adhere to a
sedentary lifestyle (sedentary means does not exercise).
a. If you selected repeated samples of 12 from the U.S. populati

ACTIVE CLASS 3.4
Multiplicity
Practice!
Multiplicity
Question 3.13: Practice
A study was done to determine whether AZT helps to reduce
the transmission of HIV from mother to baby (Connor et al.,
1994). Of the 180 babies whose mothers received AZT, 13
ba

ACTIVE CLASS 3.2
Tests of
Hypotheses
QuODDID
Question of scientific interest
Outcome measures
Design
Data analysis
Interpretation
Dissemination
3/27/14
Public Heath Biosta6s6cs
2
Question of scientific interest
Primary: Does the population

ACTIVE CLASS 2.1
Introduction to question of
interest
The medical
costs of smoking
Review of cause and intro
to confounding
Investigating the unknown
truth with the CLT
U.S. Department of Justice Sues Big
Tobacco for $280 Billion
The U.S. Department of

ACTIVE CLASS 5.4
Comparing
models and
assessing model
fit
Comparing different
logistic regression
models
Assessing model fit
Warm-up questions
>summary.glm( glm( death~gestage,family=binomial(link="logit") )
Coefficients:
(Intercept)
gestage
Estimate St

ACTIVE CLASS 5.1
Introduction to
NNIPS-II study
and logistic
regression
NNIPS-II study
Intro to logistic
regression
11/3/14%
Public%Heath%Biosta4s4cs%
2%
11/3/14%
Public%Heath%Biosta4s4cs%
3%
11/3/14%
Public%Heath%Biosta4s4cs%
4%
11/3/14%
Public%Heath%B

ACTIVE CLASS 5.3
Broken-line model
for Nepal infant
mortality data
Fitting a broken-line
model to the data
Warm-up questions
>summary.glm( glm( death~gestage,family=binomial(link="logit") )!
!
Coefficients:!
Estimate Std. Error
z value
Pr(>|z|)
!
(Interc

ACTIVE CLASS 1.1
Course Introduction
Course information
What is probability?
The scientific method
Biostatistics 345
Objective: to enable each student to enhance his/her
quantitative, scientific reasoning
General Method: To achieve mastery, a student

PRE-CLASS VIDEO 1.4
Characteristics of
Distributions
Measures of center
Measures of spread
Types of shapes
Discrete probability distributions
A random variable is a numerical function of the outcomes
of an experiment.
A random variable is discrete if

PRE-CLASS VIDEO 1.3
Screening Tests
Sensitivity/specificity
Positive/negative
predictive value
Screening test for Down Syndrome
A hypothetical screening test for Down Syndrome
(Sullivan, 2012)
Test
result
3/27/14
Affected
fetus
Unaffected
fetus
Total

PRE-CLASS VIDEO 1.5
Binomial
Distribution
Discrete probability distributions
A random variable is a numerical function of the outcomes
of an experiment.
Example: Flip a coin 5 times. Let X = the number of heads.
A random variable is discrete if it can

PRE-CLASS VIDEO 1.2
Probability
What is probability?
Discrete distributions
Joint, marginal,
conditional probabilities
What is probability?
Frequentist: long term relative frequency
Examples: Tossing a coin; disease rates (w/ large populations)
Baye