Deck 29: Introduction to Statistics and Sampling

Question 1

Write a simple, valid statement that correctly uses the terms survey, population, and sample.

Answer 1

Example: You decide to survey a small sample of the population of widgets produced on the assembly line.

Question 2

Explain the difference between a population parameter and a sample statistic.

Answer 2

A population parameter refers to the particular datum for the whole population, whereas a sample statistic refers to the datum only for the particular sample used.

Question 3

Suppose the university wants to find out from the whole student body whether they support adding a swimming pool to the recreation center. Name the type of sampling for each item and discuss each in terms of bias (say more than "biased" or "not biased").

A) A notice in the student newspaper invites all students to email their opinions.
B) Telephone at random 20 groups of five names in an alphabetical list of the student body.

Answer 3

A) This is a self-selected (or voluntary) sample because only those interested enough to respond will be represented. This sample could represent a bias because those interested in swimming would be more likely to respond.
B) This is a cluster sample. Although there may be no bias in how the names are distributed in the telephone book, it is possible that those students without telephones might have different opinions. If there are a lot of students without telephones, the sample could be biased.

Question 4

In a (fictitious) group project, two interviewers shared the data collection for one sample. Each stood by the campus bookstore entrance and asked people who went into the bookstore whether they thought the bookstore's prices were too high.
Interviewer #1's results: 33 said "yes"; 27 said "no"
Interviewer #2's results: 15 said "yes"; 25 said "no"

A) Name the type of sample the interviewers used and discuss it in terms of bias.
B) Combing the data in their samples together, give a confidence interval that has a "pretty good" confidence level for the population parameter.
(Note: Part B uses 32.2.)

Question 5

Why would one gather data on just a part of a population rather than the whole population? Give at least two reasons.

Question 6

Give an example of a stratified random sample.

Question 7

The university is thinking of changing from a semester to a quarter system. I want to know what students think of this, so I go to the campus coffee shop during final exam week and ask this question of the first 40 students I see: "Do you want to change to a quarter system where you have final exams three times a year, or do you want to stay with the present semester system with final exams only twice a year?" Give three reasons to be suspicious of my results.