Welcome to Statistics

1
Welcome to Statistics!

• Be Sci 210 02 Behavioral Science Statistics
• TTh 800 940
• Andrea M. Karkowski

2
Statistics in Our Lives
3
Government Reports 69 Deaths among Viagra Users

• By Jonathan Wright (Reuters)

4
During the survey period more than 3.6
million outpatient prescriptions were dispensed
for Viagra, which was hailed on release in March
as the long-awaited wonder drug for many men
5
Fifty-one of the 69 patients had one or more of
the factors associated with cardiovascular
diseases or cerebrovascular disease, such as
hypertension, smoking, or obesity
6
Lou Gehrigs Disease Claims Gulf Veterans 28
Cases Have Researchers Taking a Closer Look

• By David Brown
• Washington Post
• August 3, 1999

7
Persian Gulf War veterans seem to have a
slightly higher incidence of the fatal
neurological disease ALS, according to a
six-month study of selected veterans health
records.
8
In recent months, the VA and the Department of
Defense identified 28 cases of ALS among the
697,000 men and women who took part in the
campaign against Iraq As with many rare
diseases, ALS sometimes seems to occur in
clusters.
9
Epidemiological studies predict there should be
27 cases.
10
Two Years Ago Ohio Raised the Tax on Cigarettes

• West Virginia obtained a record revenue from
  cigarette taxes of 1.7 million.
• People against Ohios increased tax claim that
  WVs record revenue is from Ohioans crossing the
  boarder to purchase cigarettes.
• What additional statistical information is needed
  to evaluate that claim?

11
Want to live longer? Toss back a few cocktails.
Alcohol in moderation may extend life span,
researchers find.

• Reuters
• December 12, 2006

12
Moderate drinking may lengthen your life, while
too much may shorten it, researchers in Italy
report. Their conclusion is based on pooled data
from 34 large studies involving more than 1
million people and 94,000 deaths.
13
According to the data, drinking a moderate
amount of alcohol up to four drinks per day in
men and two drinks per day in women reduces the
risk of death from any cause by roughly 18
percent.
You should ask As compared to what?????
14
The articles answer Heavy drinkers, or men
that drink MORE than four drinks per day and
women that drink MORE than two drinks per day.
15
True or False Can you learn to detect lies?

• By Guy Gugliotta
• July 12, 1999

16
A group of 23 federal law officers principally
CIA agents could spot the liars 73 percent of
the time.
17
A group of 107 clinical psychologists with an
abiding interest in lying research scored 67.5
percent, while 43 Los Angeles County sheriffs
personnel with reputations as good interrogators
scored 66.7 percent.
18
For the general population Average accuracy
is 54 percent, and two-thirds get between 50 and
59 percent.
19
How do you spot a liar lack of coherence in
answers words that dont fit the facial
expression, and gestures that dont fit the
voice
20
The Myth, The Math, The Sex

• By Gina Kolata
• The New York Times
• August 12, 2007

21
Everyone knows men are promiscuous by nature.
Its part of the genetic strategy that evolved to
help men spread their genes far and wide. The
strategy is different for a woman, who has to go
through so much just to have a baby and the
nurture it. She is genetically programmed to want
just one man who will stick with her and help
raise their children.
22
Surveys bear this out. In study after study and
in country after country, men report more, often
many more, sexual partners than women One
survey, recently reported by the federal
government, concluded that men had a median of
seven female sex partners. Women had a median of
four male sex partners. Another study, by British
researchers, stated that men had 12.7
heterosexual partners in their lifetimes and
women had 6.5.
23
But there is just one problem It is logically
impossible for heterosexual men to have more
partners on average than heterosexual women Dr.
David Gale even provided a proof
24
By way of dramatization, we change the context
slightly and will prove what will be called the
High School Prom Theorem. We suppose that on the
day after the prom, each girl is asked to give
the number of boys she danced with. These numbers
are then added up giving a number G. The same
information is obtained from the boys, giving a
number B.Theorem G BProof Both G and B
are equal to C, the number of couples who danced
together at the prom. Q.E.D.
25
Sex survey researchers say they know that Dr.
Gale is correct. Men and women in a population
must have roughly equal numbers of partners. So,
when men report many more than women, what is
going on and what is to be believed?Ronald
Graham, a professor of mathematics and computer
science at the University of California, San
Diego, agreed with Dr. Gale. After all, on
average, men would have to have more partners
than women, raising the question where all those
extra partners might be
26
What are some explanations?

• Everyone is lying to be more consistent with what
  they believe to be true.
• Men and women define sexual partners differently.
• Men are counting their imaginary partners.
• NOTE This explanation was identified in the
  article.
• Researchers are not interviewing sex workers.
• Researchers are reporting the wrong statistic.
• Median rather than the mean and standard
  deviation.

27
Why do you need statistics?
28
10 Foundational Questions that Educated People
Should AskAdapted from Lutsky, NITOP 2007

• What do the numbers show?
• What do the numbers mean? Where are the numbers?
  Do the numbers support the conclusions?
• How representative is that?
• The mean, the median, and the mode are all
  averages, which one is being used? Is that the
  best measure for that data set?
• Compared to what?
• What is the frame of reference? What is the unit
  of measurement?
• Is the outcome statistically significant?
• Do the results differ from chance? What is the
  practical significance?
• What is the effect size?
• How substantial is the outcome?

29
10 Foundational Questions that Educated People
Should AskAdapted from Lutsky, NITOP 2007

• Are the results of a single study or of a
  literature?
• What is the source? Who sponsored it? What do
  other studies show?
• What is the research design?
• NO CAUSAL CLAIMS WITH CORRELATIONAL RESEARCH!
• How were the variables operationalized?
• How were the variables defined and measured? Is
  the measure both reliable and valid?
• Who is in the sample?
• Who is left out of the sample?
• What was controlled (and what was not
  controlled)?
• What other variables might be influencing the
  findings?

30
F(2, 27) 4.40, p lt .05X2(3, n 84) 12.01,
p lt .01t(24) 1.94, N.S.
31
Lets Look at the Syllabus
32
Suggestions for the ClassStudent Comments from
Actual Teaching Evaluations

• Go to every class work in groups ask
  questions.
• If you dont understand anything at all ask for
  it to be explained, because it all builds up like
  blocks, if your miss one thing youll be lost.
• Organize your notes so it is easy to find what
  youre looking for. Always go over each
  assignment you get back because that is where you
  learn the most.
• Dont skip class.
• Never miss class.

33
Research by Dillion (1998)

• When a student misses classes, a faculty member
  is less likely to
• Sponsor an internship, independent study, or
  research project for the student
• Write a letter of recommendation for the student
  and
• Think that the student is a serious student.

34
Regarding Peer LearningStudent Comments from
Actual Teaching Evaluations

• I enjoyed the way that it was not only lecture,
  there were examples and other things we did to go
  along with lecture. I also liked the in-class
  assignments, they helped me to put to use the
  things we learned.
• It has allowed me to further understand the
  material because I had to explain how to do this
  to others.
• A joint effort on the work was more relaxed
  learning environment.
• When I explain something to another person, it
  reinforced the concept in my head.

35
Questions
???
36
Why Statistics?

• To sell a concept
• To argue a point
• To assist in decision-making
• Evidence-based practice

37
Types of Statistics

• Descriptive
• Summarizes or describes data
• Expresses characteristics of a data set with one
  or a few numbers
• Inferential
• Attempts to account for chance
• Makes inferences about a population when
  information is obtained from a sample
• Population ?? Parameter
• Sample ?? Statistic

38
Research Designs

• Key Concept
• Statistics are completely dependent upon the
  research design used.

39
Research Designs

• Experimental and Quasi-Experimental
• Variables
• Independent, Grouping, and Subject
• Dependent
• Types
• Between Subjects
• Within Subjects
• Mixed
• Correlational

40
Research Designs

• Variables
• Qualitative
• Measures qualities
• Discrete measurements
• Quantitative
• Measures degree or how much
• Continuous measurements

41
Scales (Levels) of Measurement

• Key Concept
• Statistics are completely dependent upon the
  scale of measurement used.

42
Scales (Levels) of Measurement

• Nominal
• Ordinal
• Interval
• Ratio

43
Scales of MeasurementQuick Quiz

• The distance between Columbus and Cincinnati
• Classifying people as Protestant or Catholic
• How fast a rat responds to a stimulus
• Phone numbers
• Ranking a bunch of cookies from best to worst
• Number of calories consumed
• Classifying people as left or right handed
• Peoples preferences for political candidates
• Type of therapy a person recieved
• Zip codes

44
Be sure to read the first couple of chapters in
the text and review the scales of measurement!