ph 142
problem set 8
partial key
30 points
two independent sample tests for comparing means
Baldi and Moore, exercise 18.39, pages 461
We need to very carefully check the assumptions for using the t distribution, as there are only
10 rats in each group.
P
ph 142
problem set 8
due Wednesday, Apr 8
two independent sample tests for comparing means
Baldi and Moore, exercise 18.39, pages 461
The Stata dataset is toxic_ink.dta.
There are 3 variables:
group
coded C or T
week0
rats weights at beginning of the stud
ph 142
problem set 10
partial key
20 points
1. Use the scatterplot to discuss the overall pattern of association and any unusual points.
0
percent body fat by siri method
10
20
30
40
50
siri values for percent body fat vs. bmi
20
25
30
35
Quetelet body ma
What is Health
Promotion?
and
How Can We Use it to
Make Cal a HealthPromoting University?
Cathy Kodama
University Health Services
the Tang Center
College
Health
Public
Health
Health
Promotion
The SocioEcological Model
Clicker Question
In your opinion, i
Your Action Project:
Making it Awesome
Sarah Gamble, Ph.D.
1.29.16
Lets practice some behavior
change!
Last 4 rows switch seats with first 4 rows
People in middle: move to opposite side (edges to
middle, middle to edges, or right to left. You pick.)
When
S. Alex Stalcup, M.D.
New Leaf Treatment Center
251 Lafayette Circle, Suite 150
Lafayette, CA 94549
Tel: 9252845200
Fax: 9252845204
alex@nltc.com
www.nltc.com
TEACHING ABOUT ADDICTION
THE CORE CONCEPTS OF THE MEDICAL MODEL OF
ADDICTION:
Students learn
SLEEEEEEEP
Why something everyone desperately needs is so hard to get
David Resnick & Alison Cohen, January 2016
Overview
Clicker poll
Trends in sleep for college students
Importance of sleep: consequences of sleep quantity and quality
Sleep quality
Ways
The table in Baldi and Moore, like most tables,
gives cumulative probabilities of the form P ( z <= 1.25 )
a re a u n d e r th e s ta n d a rd n o rm a l c u rv e b e lo w 1 .2 5
0
1. Probabilities of the form
1 .25
P ( z > 1.25 ) are found using the rela
checking the assumption that the population from which we've sampled
follows a normal curve
The procedures that we will use for confidence intervals and tests of significance for population
means and differences of population means based on the t distribu
Public Health Challenges
of the 21st Century
Stefano M Bertozzi
Public Health 14: Healthy People
February 1, 2016
2
LOREM IPSUM DOLOR
What do the colors on this map represent?
Header Georgia 22
A. Median age in census block
pt
B. Sex ratio in census bloc
Public Health 14
2/3/16
College Alcohol Use The Real Buzz
Karen Hughes, MPH
UHS Tang Center PartySafe@Cal
Every facet of higher education  creating a safe and stimulating learning environment, maintaining
world class academic standards, sustaining studen
introduction to probability and statistics in biology and public health
PUBLIC HEA 142

Spring 2016
ph 142
problem set 9
two independent sample tests for comparing means
acupuncture for treating migraines
Baldi and Moore, exercise 18.35, pages 459 and 460
Acupuncture is widely used to prevent migraine attacks, but does it work? A study investigated
the
ph 142
problem set 2
10 points
We have edited the problems from the text for clarity and to bring them closer to language
were using in lecture.
Baldi and Moore problems 10.10 and 10.11, page 246247
Use the following counts to set up a complete twoway t
ph 142
problem set 1
10 points
In his 1954 classic, How to Lie with Statistics, Darrell Huff and the illustrator Irving Geis use the
following humorous sketch to illustrate how averages can be misleading summaries with
skewed distributions:
These are the
ANNOUNCEMENT
You can only switch discussion sections if a section has
seats open.
If its not open, wait and see if it opens up. Section
begins Feb 1, so there is time.
ICLICKER: WHICH OF THE
FOLLOWING ARE THE CORE
FUNCTIONS OF PUBLIC HEALTH,
ACCORDING TO
introduction to probability and statistics in biology and public health
PUBLIC HEA 142

Spring 2016
ph 142
section workskeet Wednesday Apr 20 and Thursday Apr 21
Name _ SID _
pedometer steps and BMI in college women
1. Does a line give a good summary of the relationship between BMI and pedometer steps?
Explain your reasoning.
2. Is the variability of th
introduction to probability and statistics in biology and public health
PUBLIC HEA 142

Spring 2016
ph 142
problem set 8
confidence intervals and hypothesis tests about a proportion
19.34 matched pairs Onesample procedures for proportions, like those for means, are used
to analyze data from matched pairs designs. Here is an example.
Each of 50 subjects
introduction to probability and statistics in biology and public health
PUBLIC HEA 142

Spring 2016
ph 142
section workskeet Friday Apr 15
Name _ SID _
association between body mass index
and testosterone concentration in pubertal males
Note: We saw from our work with the graph of the testosterone values and bmi that there is a
concern with the assumpti
introduction to probability and statistics in biology and public health
PUBLIC HEA 142

Spring 2016
ph 142
section workskeet Wednesday Apr 13 and Thursday Apr 14
Name _ SID _
association between body mass index
and testosterone concentration in pubertal males
1. Does a line give a good summary of the relationship between testosterone level and bmi?
Expl
Gaussian or 'normal' distributions N( , 2 )
Baldi and Moore, chapter 11
family of distributions that are symmetric and "bell shaped"
mean is denoted by small Greek letter
(pronounced "mu" by mathematicians)
standard deviation is denoted by small Greek le
caffeine drinks (coffee, tea, cola)
facts we're given:
55% of adults drink coffee
25% of adults drink tea
45% of adults drink cola
15% drink both coffee and tea
5% drink all 3 beverages
25 % drink both coffee and cola
5 % drink only tea
coffee
tea
cola
assumption checking for regression models
1. are the data a random sample from a population?
Whenever we want to use a dataset to make inference about a population, we need to look at
how the data were collected.
We often need to define the population abo
more than two categorical variables
introduction to effect modification and confounding
overview
In the context of assessing association between an exposure and a disease, we often need to
look at this association over a number of subpopulations. The sub
outliers and influential points in regression
illustrated with the NHANES data
association between systolic blood pressure and blood lead levels
. regr bpxsy1 lbxbpb if lbxbpb
Source 
SS
df
MS
+Model  35908.7982
1 35908.7982
Residual  1634079.67 4809
is the slope the same for men as for women?
interaction
men = 1 and women = 2
regr bpxsy1 lbxbpb if lbxbpb < 40 & riagendr=1
Source 
SS
df
MS
+Model 
14380.264
1
14380.264
Residual  705102.357 2461 286.510507
+Total  719482.621 2462 292.235021
Num
precision of confidence intervals and power calculations for two sample problems
about means
introduction and overview
The basic ideas about sample size calculations and power of tests of significance that weve covered for
onesample problems can be exten
precision of confidence intervals and power calculations for two sample problems
about proportions
The details of the calculations differ because the standard error of the difference of two proportions
depends on the population proportions p 1 and p 2 .
I
ph 142
rough outline, part 1
Wednesday, Jan 21
course overview
introduction to probability
To the Student: Stat Thinking
chapter 9: 207217, 224230
Friday, Jan 23
basic probability: complements, unions, intersections
chapter 10:235236, 242244
Friday, Jan 2