1 of 21

�Data Collection Activity�(adapted from the activity Seeing Red presented by Daren Starnes, Roxy Peck, and Celia Rowland)

To save and make a local (editable) copy, do: File, Make a copy. �

2 of 21

1st Step of Any Scientific Research?

Identify a question��

Scenario:

The Library is worried that books keep disappearing. They are investigating various security badges to add to the books.

Questions:

1) What proportion of the books are red?

2) What is the average number of books per shelf?

Will use red if proportion of red books is less than 0.2.

3 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual
Population
Variable
Var Type
Parameter
Statistic

4 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book
Population
Variable
Var Type
Parameter
Statistic

5 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population
Variable
Var Type
Parameter
Statistic

6 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library
Variable
Var Type
Parameter
Statistic

7 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable
Var Type
Parameter
Statistic

8 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color
Var Type
Parameter
Statistic

9 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color	# of books/shelf
Var Type
Parameter
Statistic

10 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color	# of books/shelf
Var Type	categorical
Parameter
Statistic

11 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color	# of books/shelf
Var Type	categorical	numerical(discrete)
Parameter
Statistic

12 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color	# of books/shelf
Var Type	categorical	numerical(discrete)
Parameter	p: true proportion of red books
Statistic

13 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color	# of books/shelf
Var Type	categorical	numerical(discrete)
Parameter	p: true proportion of red books	𝝻: true average number of books/shelf
Statistic

14 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color	# of books/shelf
Var Type	categorical	numerical(discrete)
Parameter	p: true proportion of red books	𝝻: true average number of books/shelf
Statistic	: proportion of red books in a sample

15 of 21

Identify

Identify...	Question #1�What proportion of the books are red?	Question #2 What is the average number of books per shelf?
Individual	book	shelf
Population	All books in Library	All bookshelves in Library
Variable	color	# of books/shelf
Var Type	categorical	numerical(discrete)
Parameter	p: true proportion of red books	𝝻: true average number of books/shelf
Statistic	: proportion of red books in a sample	: average number of books/shelf in sample

16 of 21

Approach:

How will we collect data? Experiment, sample survey?

Observational Study�
We are just recording information, not imposing any treatment.

We will take a sample, then calculate a statistic to try to estimate the parameter.

17 of 21

Data Collection

With your groups, make a well-defined plan that someone else could implement and that can be carried out within 5 minutes.

18 of 21

Calculations

Calculate your two statistics. Make a decision.�
Go around room and share values.�
Whose values are the closest?�

19 of 21

Calculations

UNKNOWN!

We only have as best estimate

true p =

?

20 of 21

Debrief

Sampling methods?

Each group briefly explain sampling method�
If someone’s estimate happens to be spot on does that make their method better than another person’s?

We want to think about the distribution of all possible values of the statistic using the sampling method. Without a random sample, we cannot know this distribution and therefore cannot attach an error to our estimate.

What makes one method better than another?�

21 of 21

Debrief

Possible Ambiguities

What possible ambiguities/questions arose that might affect your estimates?�
What do we mean by red?

What constitutes a book? �
What constitutes a bookshelf?