Sie sind auf Seite 1von 82

Chapter 15

Analysis of Variance

15.1 Introduction
Analysis of variance compares two or more populations of interval data. Specifically, we are interested in determining whether differences exist between the population means. The procedure works by analyzing the sample variance.

15.2 One Way Analysis of Variance


The analysis of variance is a procedure that tests to determine whether differences exits between two or more population means. To do this, the technique analyzes the sample variances

One Way Analysis of Variance


Example 15.1
An apple juice manufacturer is planning to develop a new product -a liquid concentrate. The marketing manager has to decide how to market the new product. Three strategies are considered
Emphasize convenience of using the product. Emphasize the quality of the product. Emphasize the products low price.

One Way Analysis of Variance


Example 15.1 - continued
An experiment was conducted as follows:
In three cities an advertisement campaign was launched . In each city only one of the three characteristics (convenience, quality, and price) was emphasized. The weekly sales were recorded for twenty

One Way Analysis of Variance


Con ce vn

Week ly sales

55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 Week 55 5 5ly 5 5 55 5 sales 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5

55 5 66 6 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 Weekl 55 5 55 5 y sales 55 5 55 5 55 5

Qa u lity

Price
55 5 66 6 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5

See file Xm15 -01

One Way Analysis of Variance


Solution
The data are interval The problem objective is to compare sales in three cities. We hypothesize that the three population means are equal

Defining the Hypotheses Solution


H0: 1 = 2= 3 H1: At least two means differ To build the statistic needed to test the hypotheses use the following notation:

Notation
1
First observation, first sample Second observation, second sample X11 x21 . . . Xn1,1 n 5

endent samples are drawn from k populations (treat


2
X12 x22 . . . Xn2,2

k
X1k x2k . . . Xnk,k

x5
Sample size Sample mean

n5 x5

nk xk

X is the response variable. The variables value are called responses.

Terminology
In the context of this problem
Response variable weekly sales Responses actual sale values Experimental unit weeks in the three cities when we record sales figures. Factor the criterion by which we classify the populations (the treatments). In this problems the factor is the marketing strategy. Factor levels the population (treatment) names. In this problem factor levels are the marketing trategies.

The rationale of the test statistic

Two types of variability are employed when testing for the equality of the population means

Graphical demonstration: Employing two types of variability

30

25

x5= 5 5
20

x5= 5 5
16 15 14 11 10 9

20 19

x5= 5 5

x5= 5 5 x5 5 = 5
12 10 9 7

x5 5 = 5

A small variability within The sample means are the same as before 1 the samples makes it easier the larger within-sample variability but Treatment 1 Treatment 2 Treatment 3 Treatment 1 TreatmentTreatment 3 2 to draw a conclusion about the it harder to draw a conclusion makes population means. about the population means.

The rationale behind the test statistic I


If the null hypothesis is true, we would expect all the sample means to be close to one another (and as a result, close to the grand mean). If the alternative hypothesis is true, at least some of the sample means would differ. Thus, we measure variability between sample means.

Variability between sample means

The variability between the sample means is measured as the sum of squared distances between each mean and the grand mean.

This sum is called the Sum of Squares for Treatments In our example treatments are SST represented by the different
advertising strategies.

Sum of squares for treatments (SST)

SST= nj (xj x)
j=5

There are k treatments The size of sample The mean of sample j j Note: When the sample means are close to one another, their distance from the grand mean is small, leading to a small SST. Thus, large SST indicates large variation between sample means, which supports H1.

Sum of squares for treatments (SST)


Solution continued Calculate SST
x5= 6 66 x5 = 6 65 x5 = 6 65 6. 6 65 . 65 . SST = nj (xj x)5
20(577.55 - 613.07)2 + n5 5+ n5x5+ ...+ nkxk + 20(653.00 - 613.07)2 + x X= + 20(608.65 - 613.07)2 = n5+ n5+ ...+ nk = 57,512.23
j=5 k

he grand mean is calculated by =

Sum of squares for treatments (SST)

Is SST = 57,512.23 large enough to reject H0 in favor of H1? See next.

The rationale behind test statistic II


Large variability within the samples weakens the ability of the sample means to represent their corresponding population means. Therefore, even though sample means may markedly differ from one another, SST must be judged relative to the within samples variability.

Within samples variability


The variability within samples is measured by adding all the squared distances between observations and their sample means.

e three cities).

This sum is called the our example this is the m of all squared differences Sum of Squares for Error tween sales in city j and the mple mean of city j (over all SSE

Sum of squares for errors (SSE)


Solution continued Calculate SSE

5 s5 = 5,66 5 s 5 = 5 6,5 s5 = 5 5.5 5 6.5 5 ,66 5 5 ,55 5

SSE = (xij x j ) 5 j =5 + -1)s2i2=5 (n3 -1)s32

nj

= (n1 - 1)s12 + (n2

= (20 -1)10,774.44 + (20 -1)7,238.61+ (20-1)8,670.24 = 506,983.50

Sum of squares for errors (SSE)

Is SST = 57,512.23 large enough relative to SSE = 506,983.50 to reject the null hypothesis that specifies that all the means are equal?

The mean sum of squares


To perform the test we need to calculate the mean squares as follows:
Calculation of MST Mean Square for SST Treatments

Calculation of MSE
Mean Square for Error

M ST =

k 5 5,5 .5 55 5 5 = 55 = 6,5 .5 65 5 5

MSE =

SSE nk 55 5.5 5,55 5 = 55 5 = 5 5.5 ,55 5

Calculation of the test statistic


MST F= MSE 5,555 5 5. 5 = 5 5 .5 ,55 5 =5 5 .5

Required Conditions: 1. The populations tested are normally distributed. with the following degrees of freedom: 2. The variances of all the populations tested arev1=k -1 and v2=n-k equal.

The F test rejection region


And finally the hypothesis test:

H0: 1 = 2 = =k H1: At least two means differ

MST Test statistic: F= MSE


R.R: F>F,k-1,n-k

The F test
Ho: 1 = 2= 3 H1: At least two means differ
MT S F= ME S 5,66 5 5 6.5 = 5 5.5 ,55 5 =5 5 .5

R.R.: F > Fstatistic 65 55 5 5 .5 Test ,k5n k = F56,F= MST/ MSE= , . 5 ,5

3.23 3.23 > 3.15, there is sufficient Since


evidence to reject Ho in favor of H1, and argue that at least one of the mean sales is different than the

The F test p- value


Use Excel to find the p-value
Statistical fx = .0467
66 . 66 . 6 66 . 6 66 . 6 55 . 5 5 5 -555 . 5 5 5 5

FDIST(3.23,2,57)

p Value = P(F>3.23) = .0467

Excel single factor ANOVA


Xm15-01.xls
Anova: Single Factor SUMMARY Groups Convenience Quality Price

Count 5 5 5 5 5 5

Sum Average Variance 555 55 6 66 5 5 55 6. 6 55. 5 666 66 5 55 5. 5 66. 6 6 66 555 55 5 55 5. 5 66. 6 6 66

ANOVA Source of Variation Between Groups Within Groups Total

SS 555 55 666 666 555 555

df 5 5 5 5 5

MS 555 55 55 55

P-value 55 . 5 55 5 . 55

F crit 55 . 5

SS(Total) = SST + SSE

15.3 Analysis of Variance Experimental Designs


Several elements may distinguish between one experimental design and others.
The number of factors.
Each characteristic investigated is called a factor. Each factor has several levels.

Two - way ANOVA Two factors


Response

One - way ANOVA Single factor


Response

Treatment 3 (level 1) Treatment 2 (level 2) Treatment 1 (level 3)

Level2

Factor B

Level 1

Level 3 Level2 Level 1 Factor

Independent samples or blocks


Groups of matched observations are formed into blocks, in order to remove the effects of unwanted variability. By doing so we improve the chances of detecting the variability of interest.

Models of Fixed and Random Effects


Fixed effects
If all possible levels of a factor are included in our analysis we have a fixed effect ANOVA. The conclusion of a fixed effect ANOVA applies only to the levels studied.

Random effects
If the levels included in our analysis represent a random sample of all the possible levels, we have a random-effect ANOVA. The conclusion of the random-effect ANOVA

Models of Fixed and Random Effects.


In some ANOVA models the test statistic of the fixed effects case may differ from the test statistic of the random effect case. Fixed and random effects - examples
Fixed effects - The advertisement Example (15.1): All the levels of the marketing strategies were included Random effects - To determine if there is a difference in the production rate of 50 machines, four machines are randomly selected and there production recorded.

15.4 Randomized Blocks (Twoway) Analysis of Variance


The purpose of designing a randomized block experiment is to reduce the within-treatments variation thus increasing the relative amount of between treatment variation. This helps in detecting differences between the treatment means more easily.

Randomized Blocks
Block all the observations with some commonality across treatments

Treatment 4 Treatment 3 Treatment 2 Treatment 1

Block3

Block2

Block 1

Randomized Blocks
Block all the observations with some commonality across treatments
Treatment Block
5 5 . . . b 5 5 k Block mean X55X55. . . X5 k x[B] 5 X55X55 X5 k x[B]5

Xb5 Xb5

Xbk

Treatment mean

x[T] x[T]5 5

x[T]k

x[B]b

The sum of square total is Recall. For the independent partitioned into three sources of samples design we have: variation
Treatments Blocks Within samples (Error) SS(Total) = SST + SSB + SSE SS(Total) = SST + SSB + SSE
SS(Total) = SST + SSE

Partitioning the total variability

Sum of square for treatments of square for blocks of square for error Sum Sum

Calculating the sums of squares


Formulai for the calculation of the sums of squares
Treatment Block
5 5 ([ ) . + 5 5 5 5 k x B]5 X + SS (Total ) = ( x5 X ) + ( x5 X ) + ... + ( x5 X ) + ( x5 X ) 5 5 5 5 . 5 .... + ( X 5 X ) 5+ ( x5 X ) 5+ ... = k k k(x[B] ) X k b Xb5 Xb5 Xbk
5 5 5 k Block mean k(x[B] ) X + 5 X55X55. . . X5 k x[B] 5 X55X55 X5 k 5 x[B]5

SSB=

Treatment mean

x[T] x[T]5 5

x[T]k

SST

) = b(x[T]5 X + b(x[T]5) X + ...+ b(x[T]k ) X

Calculating the sums of squares


Formulai for the calculation of the sums of squares
Treatment
5 5 k Block mean 5 X55X55. . . X5 k x[B] 5 5 5 X x[ xX) + k SSE=5 5 x[T]5 x[X55X555 x[T]55B]5+ [B]5 ... (x 5 B]5+ X) + (x5 . (x5 x[T]5 x[B]5+ X)5+ (x5 x[T]5 x[B]5+ X)5+ ... 5 . 5 . k (x5 x[T]k x[B]5+ X)5+ (x5 x[T]k x[B]5+ X)5+ ... k b Xb5 Xb5 Xbk

Block

SSB=
k(x[B] ) X + 5 5 k(x[B] ) X + 5 k(x[B] ) X k
5 5

Treatment mean

x[T] x[T]5 5
5 5

x[T]k

SST

) = b(x[T]5 X + b(x[T]5) X + ...+ b(x[T]k ) X

Mean Squares
To perform hypothesis tests for treatments and blocks we need Mean square for treatments Mean square for blocks Mean square for error = SST MST k 5
SSB MSB= b 5

SSE M = SE nkb+5

Test statistics for the randomized block design ANOVA


Test statistic for treatments

MST F= MSE
Test statistic for blocks

MSB F= MSE

The F test rejection regions


Testing the mean responses for treatments

F > F,k-1,n-k-b+1
Testing the mean response for blocks

F> F,b-1,n-k-b+1

Randomized Blocks ANOVA Example


Example 15.2
Are there differences in the effectiveness of cholesterol reduction drugs? To answer this question the following experiment was organized:
25 groups of men with high cholesterol were matched by age and weight. Each group consisted of 4 men. Each person in a group received a different drug. The cholesterol level reduction in two months was recorded.

Randomized Blocks ANOVA Example


Solution
Each drug can be considered a treatment. Each 4 records (per group) can be blocked, because they are matched by age and weight. This procedure eliminates the variability in cholesterol reduction related to different combinations of age and weight.

Randomized Blocks ANOVA Example


ANOVA Source of Variation Rows Columns Error Total SS 55. 5 55 6 66 6. 55. 5 55 55. 5 55 df 5 5 5 5 5 5 5 MS 5 55 5. 5 6. 6 66 5. 5 55 F P-value F crit 5 . 5 55 5 55 . 55 55 . 5 55 . 5 55 5 . 55 55 . 5

Treatments

Block b-1 K-1 MST / MSEMSB / MSE s

onclusion: At 5% significance level there is sufficient evidenc infer that the mean cholesterol reduction gained by at leas o drugs are different.

Chapter 15 continued of Analysis

Variance

15.5 Two-Factor Analysis of Variance Example 15.3


Suppose in Example 15.1, two factors are to be examined:
The effects of the marketing strategy on sales.
Emphasis on convenience Emphasis on quality Emphasis on price

The effects of the selected media on sales.


Advertise on TV Advertise in newspapers

Attempting one-way ANOVA


Solution
We may attempt to analyze combinations of levels, one from each factor using one-way ANOVA. The treatments will be:
Treatment 1: Emphasize convenience and advertise in TV Treatment 2: Emphasize convenience and advertise in newspapers . Treatment 6: Emphasize price and advertise

Attempting one-way ANOVA


Solution The hypotheses tested are:
H0: 1= 2= 3= 4= 5= 6 H1: At least two means differ.

Attempting one-way ANOVA


Solutio one of six cities sales are recorded In each nfor ten
weeks. In each city a different combination of marketing emphasis and media usage is employed.

City1

Convnce Price TV Paper

City2 City3 City4 City5 City6


Convnce Quality Quality Price Paper TV Paper TV

Attempting one-way ANOVA


Solutio n City1
City6
Convnce Price TV Paper

City2
Convnce Paper

City3
Quality TV

City4 City5
Quality Price TV

Xm15-03

Paper

The p-value =.0452. We conclude that there is evidence that differences exist in the mean weekly sales among the six cities.

Interesting questions no answers


These result raises some questions:
Are the differences in sales caused by the different marketing strategies? Are the differences in sales caused by the different media used for advertising? Are there combinations of marketing strategy and media that interact to

Two-way ANOVA (two factors)


The current experimental design cannot provide answers to these questions. A new experimental design is needed.

Two-way ANOVA (two factors)


Factor A: Marketing strategy Convenience Quality Price Factor B: Advertising media TV City 1 sales City3 sales City 4 sales City 5 sales City 6 sales

City 2 Newspapers sales

Are there differences in the mean sales caused by different marketing strategies?

Two-way ANOVA (two factors)


Test whether mean sales of Convenience, Quality, and Price significantly differ from one another. H0: Conv.= Quality =
Calculations are based on the sum of square for factor A Price SS(A)

H1: At least two means differ

Two-way ANOVA (two factors)


Factor A: Marketing strategy Convenience Quality Factor B: Advertising media TV City 1 sales City 3 sales City 4 sales Price City 5 sales City 6 sales

City 2 Newspapers sales

Are there differences in the mean sales caused by different advertising media?

Two-way ANOVA (two factors)

st whether mean sales of the TV, and Newspaper nificantly differ from one another. H0: TV = Newspapers H1: The means differ
Calculations are based on the sum of square for factor B SS(B)

Two-way ANOVA (two factors)


Factor B: Advertising media Factor A: Marketing strategy Convenience Quality Price TV TV City 1 sales City 3 sales City 4 sales City 5 sales City 6 sales

City 2 Newspapers sales

Are there differences in the mean sales caused by interaction between marketing strategy and advertising medium?

Two-way ANOVA (two factors)


Test whether mean sales of certain cells are different than the level expected.
Calculation are based on the sum of square for interaction SS(AB)

Graphical description of the possible Graphical description of the possible relationships between factors A and B. relationships between factors A and B.

Difference between the levels of factor A, and between the levels of factor A Difference difference between the levels of factor B; no No difference between the levels of factor interaction M R Level 1 of factor B M R Level 1and 2 of factor B e e e e s s a p a p Level 2 of factor B n o n o n n s s e e Levels of factor A Levels of factor A 1 2 3 1 2 3

M R M R e e Interaction eNo difference between the levels of factor A. e s s a Difference between the levels of factor B p a p n o n n o n s e s e Levels of factor A Levels of factor A 1 2 3 1 2 3

Sums of squares
SS(A) = rb SS(B) = ra

(x[A]i x)5

(5(5 xconv. x) 5+ ( xquality x) 5+ ( x price x) 5 5 ){( }

i=5 b

(x[B]j x)5
b

(5)( 5 xTV x) 5+ ( x Newspaper x ) 5 5 ){( }

j=5
a

SS(AB = r )
a

i=5 j=5

(x[ABij x[A]i x[B]j + x)5 ]


r

SSE =

( xijk x[ AB ]ij ) 5

i =5 j =5 k =5

F tests for the Two-way ANOVA


Test for the difference between the levels of the main factors A SS(A)/(a-1) and B SS(B)/(b-1)
MS(B) MS(A) F= F= MSE MSE Rejection region: F > F,a-1 ,n-ab > F, b-1, n-ab F
SSE/(n-ab)

Test for interaction between MS(AB) factors A and B SS(AB)/(a-1)(b-1) F=


Rejection region:
(b-1),n-ab

MSE

F > F,(a-1)

Required conditions:
1. The response distributions is normal 2. The treatment variances are equal. 3. The samples are independent.

F tests for the Two-way ANOVA


Example 15.3 continued( Xm15-03)
Conve nce Qua y nie lit T V T V T V T V T V T V T V T V T V T V Ne spa r w pe Ne spa r w pe Ne spa r w pe Ne spa r w pe Ne spa r w pe Ne spa r w pe Ne spa r w pe Ne spa r w pe Ne spa r w pe Ne spa r w pe Price

55 5 66 6 66 6 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5

55 5 55 5 66 6 66 6 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5

55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5 55 5

F tests for the Two-way ANOVA


Example 15.3 continued
Test of the difference in mean sales between the three marketing strategies
H0: conv. = quality = price H1: At least two mean sales are different
ANOVA Source of Variation Sample Columns Interaction Within Total SS 5 5 55 55 . 5 5 55 55 . 55 . 5 55 555 . 5 5 55 555 . 5 5 55 df 5 5 5 5 5 5 5 MS 5 5 55 55 . 5 5 55 55 . 5 55 5. 55 . 5 55 F P-value 66 . 6 66 6 . 66 55 . 5 55 5 . 55 55 . 5 55 5 . 55 F crit 55 . 5 55 . 5 55 . 5

Factor A Marketing strategies

F tests for the Two-way ANOVA


Example 15.3 continued
Test of the difference in mean sales between the three marketing strategies
H0: conv. = quality = price H1: At least two mean sales are different MS(A)/MSE F = MS(Marketing strategy)/MSE = 5.33 Fcritical = F,a-1,n-ab = F.05,3-1,60-(3)(2) = 3.17; (p-value = . 0077)

At 5% significance level there is evidence to infer that differences in weekly sales exist

F tests for the Two-way ANOVA


Example 15.3 - continued
Test of the difference in mean sales between the two advertising media
H0: TV. = Nespaper H1: The two mean sales differ
ANOVA Source of Variation Sample Columns Interaction Within Total SS 555 55 . 5 555 55 . 5 55 . 55 5 555 5 555 . 555 5 555 . df 5 5 5 5 5 5 5 MS 555 55 . 5 555 55 . 5 55 5 .5 55 . 55 5 F 66 .6 55 .5 55 .5 P-value 6 66 .66 5 55 .55 5 55 .55 F crit 55 .5 55 .5 55 .5

Factor B = Advertising media

F tests for the Two-way ANOVA


Example 15.3 - continued
Test of the difference in mean sales between the two advertising media
H0: TV. = Nespaper H1: The two mean sales differ MS(B)/MSE F = MS(Media)/MSE = 1.42 Fcritical = F,a-1,n-ab = F.05,2-1,60-(3)(2) = 4.02 (p-value = . 2387)

At 5% significance level there is insufficient evidence to infer that differences in weekly sales exist between the two advertising

F tests for the Two-way ANOVA


Example 15.3 - continued
Test for interaction between factors A and B
H0: TV*conv. = TV*quality ==newsp.*price H1: At least two means differ
ANOVA Source of Variation SS Sample 5 5 55 5 5. Columns 5 5 55 5 5. Interaction 5 5. 5 55 W ithin 5 5 5. 5 5 55 Total 5 5 5. 5 5 55 df 5 5 5 5 5 5 5 MS 5 5 55 5 5. 5 5 55 5 5. 5 55 5. 5 5. 5 55 F P-value F crit 66 . 6 66 6 . 66 55 . 5 55 . 5 55 5 . 55 55 . 5 55 . 5 55 5 . 55 55 . 5

Interaction AB = Marketing*Media

F tests for the Two-way ANOVA


Example 15.3 - continued
Test for interaction between factor A and B
H0: TV*conv. = TV*quality ==newsp.*price H1: At least two means differ MS(AB)/MSE

F = MS(Marketing*Media)/MSE = .09 Fcritical = F,(a-1)(b-1),n-ab = F.05,(3-1)(2-1),60-(3)(2) = 3.17 (p-value= .9171)

At 5% significance level there is insufficient evidence to infer that the two factors interact to affect the mean weekly sales.

15.7 Multiple Comparisons


When the null hypothesis is rejected, it may be desirable to find which mean(s) is (are) different, and at what ranking order. Three statistical inference procedures, geared at doing this, are presented:
Fishers least significant difference

Two means are considered different if the difference between the corresponding sample means is larger than a critical number. Then, the larger sample mean is believed to be associated with a larger population mean. Conditions common to all the methods here:
The ANOVA model is the one way analysis of variance The conditions required to perform the

15.7 Multiple Comparisons

This method builds on the equal variances t-test of the difference between two means. The test statistic is improved by using MSE rather than sp2. We can conclude that i and j differ (at % significance level if 5 i -5j| > LSD, where | LSD= t 5 MSE + ) ( ni nj

Fisher Least Significant Different (LSD) Method

df. = n k .

Experimentwise Type I error rate (E) (the effective Type I error)


The Fishers method may result in an increased probability of committing a type I error. The experimentwise Type I error rate is the probability of committing at least one Type I error at significance level of . It is calculated by E = 1-(1 )C where C is the number of pairwise comparisons (I.e. C = k(k-1)/2 The Bonferroni adjustment determines the required Type I error probability per pairwise comparison () , to secure a pre-determined overall E.

Bonferroni Adjustment
The procedure:
Compute the number of pairwise comparisons (C) [C=k(k-1)/2], where k is the number of populations. Set = E/C, where E is the true probability of making at least one Type I error (called experimentwise Type I error). 5 5 ( conclude (5 ) i and ) We can i j > t thatMSE + j differ (at /C% C ni nj significance level if

df. = n k .

Fisher and Bonferroni Methods


Example 15.1 - continued Rank the effectiveness of the marketing strategies (based on mean weekly sales). Use the Fishers method, and the Bonferroni adjustment method Solution (the Fishers method) The sample mean sales were 577.55, 653.0, 608.65.5 55 5 5 = 5.5 x5 x5 = 55 55 5 5 . . 5 5 t 5 MSE + ) = ( Then, ni nj x5 x5 = 6 65 5 55 5.5 65 55 = 5 5 . .
5 (5 5 / 5 5 5 5 x5 x5 = 5 5 6 65 5.5 t.555 5 5 /5) + (55) 5.5 55 65 = 5 5 / . .

Fisher and Bonferroni Methods


Solution (the Bonferroni adjustment)
We calculate C=k(k-1)/2 to be 3(2)/2 = 3. We set = .05/3 = .0167, thus t.0167/2, 60-3 = 2.467 (Excel).

5 5 t 5 MSE + ) = ( ni nj x5 x5= 6 65 5 55 5.5 65 55 = 5 5 . .


x5 x5= 5 55 5 5 = 5.5 55 55 5 5 . . x5 x5= 5 5 6 65 5.5 5 6 5 5 /5) + (55) = 5.5 55 65 = 5 5 .6 6 5 (5 5 / 5 5 5 . . 5

Again, the significant difference is between 1 and 2.

Tukey Multiple Comparisons The test procedure:


Find a critical number as follows:

MSE = q (k, ) ng

k = the number of samples =degrees of freedom = n - k ng = number of observations per sample (recall, all the sample sizes are the same) = significance level q(k,) = a critical value obtained from the studentized range table

Tukey Multiple Comparisons


Select a pair of means. Calculate the xmax difference between the larger and xmin the smaller mean. If xmax xmin> there is sufficient evidence to conclude that max > min .

n of bove procedure with ng calculated as the harmonic meang = e sample sizes.

Repeat this procedure for each pair of samples. Rank the means if possible. the sample sizes are not extremely different, we can use the k

5 5 5 5 ... 5 k n+ n + + n

Tukey Multiple Comparisons


Example 15.1 - continued We had three populations (three marketing strategies). K = 3,
MSE 55 55 = q (k, ) = q.555 5 ( ,5) = 6.5 65 ng 5 5
Population Mean

Sample sizes were equal. n1 = n2 = n3 = 20, = n-k = 60-3Take q.05(3,60) from the table. = 57, MSE = 8894.
xmax xmin>

xmax xmin

Sales - City 1 City 1 vs. City 2: 653 - 577.55 = 75.45 577.55 Sales - City 2 City 1 vs. City 3: 608.65 - 577.55 = 31.1 653 Sales - City 3 City 2 vs. City 3: 653 - 608.65 = 44.35 698.65

Excel Tukey and Fisher LSD method


Xm15-01
Fishers LDS = .05
Multiple Comparisons LSD Omega Treatment Treatment Difference Alpha = 555 . Alpha = 555 . Convenience Quality -55 . 55 6. 6 66 6. 6 66 Price -55 .5 5. 5 55 5. 5 55 Quality Price 5. 5 55 5. 5 55 5. 5 55

= Bonferroni adjustments.05/3 = .0167


M u ltip le C o m p a riso n s LS D O m ega Tre a tm e n t Tre a tm e n t iffe re n c A lp h a = . 5 A lp h a = . 6 D e 55 5 5 66 C o n ve n ie n c eQ u a lit y -6 6 6 6 . 6 .6 66 5. 5 55 P ric e -5 5 .5 5. 5 55 5. 5 55 Q u a lity P ric e 5. 5 55 5. 5 55 5. 5 55

Das könnte Ihnen auch gefallen