Chance WorkshopGroup PresentationsDartmouth CollegeJuly 1998

I. **Black Smokers and Nicotine--An Absorbing Issue**

Melissa Cass, Carol Janik, Josephine Rodriguez, Steve Terry,

Beth Walters

II.** Non-Cents: A Simulation Activity**

Kal Godbole, Leona Mirza, Alice Richardson, Mark Rizzardi

III. **Clinical Trials**

Charlotte Buffington, Angela Hare, Ellen Musen, Nancy Roper

IV. **Therapeutic Touch**

Eli Brettler, Rita Kolb, Raj Prasad

V. **Chasing the Home Run Record**

Michael Dutko, Richard Iltis, Steve Samuels, Linda Thiel,

John Wasik

VI. **Where Have All the Boys Gone?**

Steve Givant, Eunice Goldberg, Ellen King, Barbara Stewart

**I. Black Smokers and Nicotine--An Absorbing Issue**

Melissa Cass, Carol Janik, Josephine Rodriguez, Steve Terry, Beth Walters

**Workshop Handouts**

Studies suggest black absorb more nicotine: Link to higher cancer rate is hinted. The Boston Globe, 8 July 1998, p. By Richard Saltus.

Studies show that black smokers absorb more nicotine. The Valley News, 8 July 1998, pB1. By John Schwartz.

Black smokers retain more nicotine, ABC News Online, July 10,
1998.

http://www.abcnews.com:80/sections/living/DailyNews/

[links updated daily--it does not appear that archives can be accessed!]

Discussion questions based on above articles.

Additional References

The group also distributed copies of the following three articles from
the July 8 issue of JAMA:

Caraballo, Ralph S. et. al., Racial and ethnic differences in serum cotinine levels of cigarette smokers. Journal of the American Medical Association 1998, 280: 135-139.

Prez-Stable, Elisio J. et. al., Nicotine metabolism and intake in black and white smokers. Journal of the American Medical Association 1998, 280: 152-156.

(Editorial), Pharmacogenetics and ethnoracial differences in smoking. Journal of the American Medical Association 1998, 280: 179-180.

**II. Non-Cents: A Simulation Activity**

Kal Godbole, Leona Mirza, Alice Richardson, Mark Rizzardi

**Workshop Handouts**

Exceprt from article in Milwaukee Journal, May 1992.

Worksheets for classroom simulation activities and discussion questions.

**III. Clinical Trials**

Charlotte Buffington, Angela Hare, Ellen Musen, Nancy Roper

**Workshop Handouts**

Magazine ad for ZYRTEC (allergy medication) & discussion questions

Quiz on terms from medical experiments

Additional References

Jessica Utts, Seeing Through Statistics. Chapter 5 "Experiments
and Observational Studies"

IV. **Therapeutic Touch**

Eli Brettler, Rita Kolb, Raj Prasad

Workshop Handouts

Hallelujah! Science looks at prayer for friend and fungus. The
New York Times on the Web, 5 April 1998. By Jeff Stryker.

http://search.nytimes.com/search/daily

Rosa, Linda et. al., A close look at therapeutic touch. Journal of the American Medical Association, 1998, 279, 1005-1010.

Chance class discussion handout to accompany above

Additional References

Eli Brettler posted the group's handout to his web page, including links to the articles online!

V. In Pursuit of the Home Run Record

Michael Dutko, Richard Iltis, Steve Samuels, Linda Thiel, John Wasik

Workshop Handouts

McGwire gets better, and a record looks more vulnerable. The
New York Times, 9 July 1998, A1. By Buster Olney

Discussion questions based on the above article

Additional References

VegasINSIDER online sports betting site report on wagers regarding
MdGwire

SportingNews online statistics on McGwire

Addendum.

Players who have hit 30 home runs before
the All-star game.

Plaryes who hit 30 or more homeruns after
the All-star game.

Side-by-side boxplots of McGwire's home run distances,
home and away

and time series plot of home run distances, with home/away plotting
symbols

VI. Where Have All the Boys Gone?

Steve Givant, Eunice Goldberg, Ellen King, Barbara Stewart

Workshop Handouts

¥ Alpert, Mark. Where have all the boys gone? Scientific American,
July 1998 issue online:

http://www.sciam.com/1998/0798issue/0798scicit3.html

¥ Discussion questions based on the article

Additional References

¥ Online JAMA abstract. Davis. D.L. et. al., Reduced ration of
male to female births in several industrial countries. A sentinel health
indicator? Journal of the American Medical Association 1998, 279: 1018-1023.

¥ Vital Statistics. No. 89. Birth and Birth Rates: 1970-1992. (Photocopy
from US Statistical Abstract)

¥ Report: Environmental factors may support dip inmale birth rates.

http://www.sportserver.com/newsroom/ntn/health/033198/health19_4144_noframes.html

Addendum

The group sent us an extended write-up after the meeting. It is reproduced
on the following pages.

**Chance Workshop**
**July 1998**
**An Addendum to the Group Presentation:**

Ellen King

Eunice Goldberg

Barbara Stewart

Steven Givant

from Eunice Goldberg, in consultation with Ellen King

Article:

Albert, Mark

Where Have All the Boys Gone

Scientific American -July 1998

www.sciam.com/1998/0798issue/0798scicit3.html

**Other Articles and Sources:**

Tanner, Lindsey

Report: Environmental factors may support dip in male birth rates

Chicago, Associated Press, 1998

Davis, Gottlieb, Stamplinitzky

Reduced Ratio of Male to Female Births in Several Industrial Countries

JAMA Abstracts, April 1, 1998

U.S. National Center for Health Statistics: Vital Statistics of the United States (Birth rates by gender and ethnicity, infant mortality, ages of mothers and fathers, etc.)

JAMA, New England Journal of Medicine

**Recommended searches**

Information on male-female ratios in the rest of the world

Government policies that promote one gender over the other (China,
India, Middle East, Latin America, etc.)

Long range effects of unbalanced ratios

Environmental factors that may be gender biased.

War, etc.

**Recommended Activity--Population Simulation Problem**

Several versions of this problem have circulated:

One version is that a country prefers males and families keep having
children until a male is born. A second version says that females are preferred
and therefore families keep having children until they have a female.

Other versions say that a country does not want to take a chance on
having too many girls-- so once a girl is born, the family must quit having
children. The government does not care how many boys are born.

What can be learned?

1.The analysis of the article helps students become quantitatively literate. By asking questions and doing research, the students will learn to evaluate and make sense of data.

2. The problem helps students learn about issues of probability, simulation, and distributions of data.

3. There can be meaningful integration of subject areas: social studies and mathematics and/or science.

For the analysis of the article students could possibly: learn to define the issues being presented and investigate whether the data actually explains or enlightens

Think about ways to obtain needed information either through further
research or investigation

ask other questions about this issue that are compelling or need clarification
and should be investigated, ie. explain the relationship, if any, between
a declining ratio and a declining birth rate

interpret, analyze, question, and evaluate data: enough, convincing,
misleading, contradictory, etc.

think about what information is needed to inform the issue: demographics,
original studies, census data, etc.

learn to find backup data: internet, journals, newspapers, experts,
etc.

gain experience looking at large quantities of information and sifting
out what is important and/or meaningful

learn to ask "what ifs": what if the ratio became very disparate? what
if it continued for many generations? what if science changes the way babies
are reproduced?

etc.

By doing the problem students could:

generate their own data

have the experience of dealing with a data set that is messy yet still shows the theoretical patterns

look at measures of central tendency: mean, median, mode

look at spread: standard deviation and eyeballing the distribution

learn to simulate two-choice problems using different two choice simulations

learn the need for lots of data before patterns appear--learn that any one sample may not show the pattern-- or even a few samples may not--

compare the results of experimental simulations to the theoretical model of this problem

compare the variation between the number of boys or girls (depending on the version) in individual families-- and the average for many families.

look at the shape of the distributions for average number of boys vs. girls, and the distribution of the number of families with zero girls, 1 girl, 2 girls, 3 girls, etc.

compare real life to the simulation:

Is there a physical limit on the number of children one family can have?

Is the ratio really 50-50?

Can we simulate other ratios using these methods?

What do those other ratios say for future generations?

Are the simulations reasonable models?

etc.

Time frame: At least four hours of class time

**materials:**

(Copies of the article, questions presented in class, additional articles
and data can be included here)

I am adding a copy of the problem as I used it in a workshop. It is
similar to what Ellen said she did.

Males Preferred

Does it Work?

There are countries in the world where wives are considered to be a
failure unless they produce a male child. King Henry the VIII beheaded
several wives because they didn't produce a male heir.

There are still countries in the world that express a preference for
boys. A certain unnamed country prefers boys to girls and thinks it has
come up with a way to guarantee more boys. The leaders of the country developed
this family-planning scheme:

A couple will continue to bear children until a son is born, at which
time they will stop having children.

Do you think this family-planning scheme will produce more boys or
more girls?

Explain how you made your decision.

(As presented by Eunice Goldberg at Barat College, 1998)

SAMPLE 1: Using coin tosses simulate the generation of 20 families. Let heads represent boys and tails represent girls.

FAMILY NUMBER

TALLY BOYS

TALLY GIRLS

# GIRLS

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

TOTAL BOYS:

TOTAL GIRLS:

SAMPLE 2: Using coin tosses simulate the generation of 20 families. Let heads represent boys and tails represent girls.

SAMPLE 3: Using THE RANDOM NUMBER TABLE to simulate the generation of 20 families. Let even numbers stand for boys and odd numbers stand for girls.

SAMPLE 4: Using THE RANDOM NUMBER TABLE to simulate the generation of 20 families. Let even numbers stand for boys and odd numbers stand for girls.

SAMPLE 5: Using a DIE simulate the generation of 20 families. Let 1,2, or 3 be a boy and 4, 5, and 6 be a girl.

SAMPLE 6: Using DICE simulate the generation of 20 families. Let 1,2, or 3 be a boy and 4,5, and 6 be a girl.

Answer the following questions:

1. How many families did you generate altogether?____________

2. How many families had zero girls?_______

3. How many families had exactly 1 girl?_____

4. How many families had exactly 2 girls?_____

5. How many families had exactly 3 girls?____

6. Do you see any patterns?_______

7. What number occurred the most often?____

8. What is the average number of girls in a family?_______

Notes to me:

Make charts:

1. Make a spread sheet of samples derived from coin tosses, random number tables, and tosses of die. Subject name on the side, type of sample across the top. (This can be done in EXCEL-- with a split screen and a graph so the graph will change as data is added in-- or else it can be done on the board where students make dots along the distribution values.)

2. Look at the shape of the data after each set of data has been input.
Does it gradually get a more normal shape?

Choices: Make separate plots for coins, random number tables, and die--
all should have similar shape--and put all on one plot to see how the shape
normalizes.

CONCEPT:

IN THE LONG RUN THE THEORETICAL DISTRIBUTION WILL START TO APPEAR.
INDIVIDUAL SAMPLES CAN BE ALL OVER THE PLACE. YOU MUST LOOK AT THE PATTERN
IN THE LONG RUN.

3. What is the overall mean?

4. Find the mean for each subject's six samples. Then find the mean
of those means. How does it compare to the overall mean? Look at the histogram
of that data. How does it compare to the overall histogram? Look at the
spread.

CONCEPT: THE MEAN OF THE MEANS SHOULD BE CLOSE TO THE OVERALL (POPULATION)
MEAN.

THE SPREAD SHOULD BE TIGHTER. LARGER SAMPLES YIELD TIGHTER, MORE PRECISE
DATA.

5. Find the mean for each of the six samples. How do the different
methods of sampling compare to each other?

CONCEPT: TWO CHOICE METHODS OF SIMULATION SHOULD REVEAL SIMILAR RESULTS.

What is the mean of the six samples?

How does it compare to the overall mean?

How does the graph compare to the graph of the overall data and the data of the means of the subjects?

CONCEPT: THE MEAN OF THE MEANS SHOULD BE SIMILAR TO THE OVER ALL MEAN. THIS GRAPH SHOULD BE EVEN TIGHTER BECAUSE THE SAMPLE SIZES ARE LARGER.

Make a chart of individual families.

How many times were there zero girls?

How many times was/were there 1 girl? 2 girls? 3 girls? 4 girls?

CONCEPT: ZERO GIRLS IN A FAMILY WAS THE MODE, AND EACH INCREASE OF A GIRL WAS HALF AGAIN AS MANY FAMILIES AS THE PREVIOUS NUMBER.

THE SHAPE OF THE GRAPH IS SKEWED AND ASYMPTOTIC-- UNTIL YOU REACH A POINT WHERE IT WOULD BE PHYSICALLY IMPOSSIBLE FOR A WOMAN TO HAVE THAT NUMBER OF CHILDREN.

What does the distribution of boys look like? What is the mean, median, and mode?

CONCEPT: WHAT DOES A GRAPH LOOK LIKE WITHOUT ANY SPREAD.

Compare that to the distribution of girls. Does this method produce more boys?

LOOK AT THE THEORETICAL MODEL.