Intro Lecture
Download
Report
Transcript Intro Lecture
STAT 1301
Introduction to Statistics
Wayne Woodward
Office: 143 Heroy
Phone: 768-2457
Hours: 2:00-3:00 W
3:00-4:00 Th
2:00-3:00 F
others by appointment
Text: Statistics, by Freedman, Pisani,
and Purves (3rd edition)
Lecture Sessions
a short quiz will be given at the beginning
of each lecture session over the reading
assignment for that day
– these will start on Thursday, January 20
counts 10% of total grade
(lowest 3 quiz grades dropped)
Homework
(for the previous week) will be taken up at
beginning of lab on Monday
counts 10% of grade
(2 lowest scores will be dropped)
Monday Lab Sections
meet every Monday
– TA: Julia Kozlitina
Tracy Xu
lab project
counts 10% of grade
(2 lowest scores will be dropped)
NOTE:
– first lab is Monday – January 24
(no lab this Monday -- Martin Luther King Day)
Major Exams
3 Hour Exams
- (lowest exam score dropped)
- count 40% of grade
Final Exam
- cumulative
- counts 25% of grade
NOTE: Special arrangements can be
made for students with learning
disabilities
Project
Internet-based assignment using
statistics web sites
Counts 5% of your grade
Course Packs
Course Packs for this class can be
picked up at the bookstore.
Includes:
Reading Handouts
Homework Handouts
Lab Assignments
Lecture Handouts
Practice Exams
IMPORTANT
NO late work will be accepted
NO make-ups will be given for
- hour exams
- daily quizzes
- lab projects
Dates to Remember
January 19 -- Last day to Add/Drop
March 14-18 -- Spring Break
April 4 -- Last day to drop a course
TIPS
always attend class
- lecture
- Monday labs
keep up
- reading
- homework
treat college like an 8 to 5 job
Math Fundamentals
(Quantitative Reasoning)
Literacy
– usual context: reading
Quantitative Literacy (“numeracy”)
– increasingly important in our society
– H.G. Wells: “The ability to think statistically will
someday be as important to a good citizen as the
ability to read and write.”
– most people have inadequate “numeracy” skills
In this course we will learn to deal critically
with data and numerical arguments
– develop the ability to apply statistical ideas to
“real world” situations
Elements of Statistics
collecting quantitative information
– surveys, experiments, etc.
describing and presenting data
– data summaries, graphs, etc.
drawing conclusions
– Statistics is the science of decision
making in the face of uncertainty
Examples:
Data Collection Issues
– Bias
- How was the question asked?
– Well-designed study or anecdotal
evidence?
- How were subjects selected?
- Placebo used?
Interpretation
– Simpson’s Paradox
Should Government Spending
for Public Television be Cut?
2 questions on a poll of 1031 people
(1) “I would be disappointed if Congress cut its
funding for public television.”
(2) “Cuts in funding for public television are justified
as part of an overall effort to reduce federal
spending.”
% saying cuts ok: (1) 40%
(2) 52%
A placebo might be the cure
Knight-Ridder Wire
Everybody
knows
what
a
placebo is: A phony pill the
doctors give to hypochondriacs
to make them think they’re
getting medical treatment. All
too true, but according to recent
studies, the deceptive little
sugar pills work pretty well on
people who are ill, too.
Science digest reports that
expectations and belief play so
important a part in making a
person
well
that
patients
suffering from very real disabilities
such as
coughs,
colds,
ulcers, seasickness, vertigo, and
hay fever have enjoyed improvement in their conditions, and
sometimes have been cured
when treated with placebos.
Other research indicates that
placebos can at times constrict
the pupil of the eye, alter the
blood pressure, change the
heart or respiratory rates,
influence, digestion, change
body temp-erature, and even
effect fat levels in the blood and
the number of certain types of
blood cells.
Gender Bias - Graduate Division, UC Berkeley
8442 men applied, 44% were admitted
4321 women applied, 35% were admitted
Is it correct to say the university prefers men to
women?
Major
A
B
C
D
E
F
MEN
Number of Percent
applicants admitted
825
62
560
63
325
37
417
33
191
28
373
6
WOMEN
Number of Percent
applicants admitted
108
82
25
68
593
34
375
35
393
24
341
7
More Extreme -Hypothetical Example
What’s the
point?
Admit Deny Percent
Mathematics
Males
Females
English
Males
Females
Combined
Males
Females
80
9
1
20
81
29
20
1
9
80
29
81
80%
90%
10%
20%
74%
26%
Combined -- university
seems to favor men.
However, for each dept.% male applicants
admitted is less than that
for females.
Gender is confounded
with major.
(Simpson’s Paradox)
Uses of Statistics
Government
Census Bureau, Food and Drug Administration
Business
Market Research, Economic Forecasting, Quality Control
Social Sciences
Psychological Testing, Housing Quality, Educational Effectiveness
Medicine
Cancer Research, Drug Trials, fMRI
Natural Sciences
Global Warming, Nuclear Monitoring, Air Quality
Law
Employment Discrimination Cases, Tax Audits