Transcript Document

Lesson 11 - 1

Inference about Two Means Dependent Samples

Objectives

Distinguish between independent and dependent sampling

Test claims made regarding matched pairs data

Construct and interpret confidence intervals about the population mean difference of matched pairs

Vocabulary

Robust – minor deviations from normality will not affect results

Independent – when the individuals selected for one sample do not dictate which individuals are in the second sample

Dependent – when the individuals selected for one sample determine which individuals are in the second sample; often referred to as

matched pairs

samples

Now What

Chapter 10 covered a variety of models dealing with one population

  

The mean parameter for one population The proportion parameter for one population The standard deviation parameter for one population

However, many real-world applications need techniques to compare two populations

Our Chapter 10 techniques do not do these

Two Population Examples

We want to test whether a certain treatment helps or not … the measurements are the “before” measurement and the “after” measurement

We want to test the effectiveness of Drug A versus Drug B … we give 40 patients Drug A and 40 patients Drug B … the measurements are the Drug A and Drug B responses

Two precision manufacturers are bidding for our contract … they each have some precision (standard deviation) … are their precisions significantly different

Types of Two Samples

An independent sample is when individuals selected for one sample have no relationship to the individuals selected for the other

Examples

 

50 samples from one store compared to 50 samples from another 200 patients divided at random into two groups of 100 each A dependent sample is one when each individual in the first sample is directly matched to one individual in the second

Examples

 

Before and after measurements (a specific person’s before and the same person’s after) Experiments on identical twins (twins matched with each other

Match Pair Designs

Remember back to Chapter 1 discussions on design of experiments: the dependent samples were often called matched-pairs

Matched-pairs is an appropriate term because each observation in sample 1 is matched to exactly one in sample 2

  

The person before



the person after One twin



the other twin An experiment done on a person’s left eye



the same experiment done on that person’s right eye

Terms

d-bar or d – the mean of the differences of the two samples

x 1 – x 2 = d 30 – 25 = 5 23 – 27 = - 4

s

d

is the standard deviation of the differenced data

Requirements

Testing a claim regarding the difference of two means using

matched pairs

Sample is obtained using simple random sampling

Sample data are matched pairs

Differences are normally distributed with no outliers or the sample size, n, is large (n ≥ 30)

Classical and P-Value Approach – Matched Pairs P-Value is the area highlighted t 0 -t α -|t 0 | |t 0 | -t α/2 t α/2 Critical Region Remember to add the areas in the two-tailed!

Test Statistic: t 0 d = -------- s d /√n Left-Tailed Reject null hypothesis, if P-value < α Two-Tailed Right-Tailed t 0 < - t α t 0 < - t α/2 or t 0 > t α/2 t 0 > t α t α t 0

Confidence Interval – Matched Pairs

Lower Bound: d – t α/2 · s d /√n Upper Bound: d + t α/2 · s d /√n t α/2 is determined using n - 1 degrees of freedom d is the mean of the differenced data s d is the standard deviation of the differenced data Note: The interval is exact when population is normally distributed and approximately correct for nonnormal populations, provided that n is large.

Two-sample, dependent, T-Test on TI

If you have raw data:

– –

enter data in L1 and L2 define L3 = L1 – L2 (or vice versa – depends on alternative Hypothesis)

L1 – L2 STO

L3

Press STAT, TESTS, select T-Test

– –

raw data: List set to L3 and freq to 1 summary data: enter as before

Example Problem

Carowinds quality control manager feels that people are waiting in line for the new roller coaster too long. To determine is a new loading and unloading procedure is effective in reducing wait time, she measures the amount of time people are waiting in line for 7 days and obtains the following data.

Day

Old New

Mon

11.6

10.7

Tue

25.9

28.3

Wed

20.0

19.2

Thu

38.2

35.9

Fri

57.3

59.2

Sat

32.1

31.8

Sat

81.8

75.3

Sun

57.1

54.9

Sun

62.8

62.0

A normality plot and a box plot indicate that the differences are apx normal with no outliers. Test the claim that the new procedure reduces wait time at the α=0.05 level of significance.

Example Problem Cont.

Requirements: seem to be met from problem info

Hypothesis H 0 : H 1 : Mean wait time the same (d-bar = 0, new-old) Mean wait time reduced (d-bar < 0, new-old)

• •

Test Statistic: t 0 d-bar - 0 = --------------------- s d /

n = -1.220, p = 0.1286

Critical Value: t c (9-1,0.05) = -1.860, α = 0.05

Conclusion: Fail to Reject H 0 : not enough evidence to show that new procedure reduces wait times

Summary and Homework

Summary

Two sets of data are dependent, or matched-pairs, when each observation in one is matched directly with one observation in the other

In this case, the differences of observation values should be used

The hypothesis test and confidence interval for the difference is a “mean with unknown standard deviation” problem, one which we already know how to solve

Homework

pg 582-587; 1, 2, 4-8, 12, 15, 18, 19

6) independent

HW Answers

8) dependent 12a) your task 12b) d-bar = -1.075

s d = 3.833

12c) Fail to reject H 0 12d) [-5.82, 3.67] 18) example problem in class