Professional Documents
Culture Documents
Chapter Outline
Analysis of Variance (ANOVA)
One-Way ANOVA
Two-Way ANOVA
LSD-test
Scheffé Test
Duncan’s Multiple Range Test
1
In this chapter we expand our discussion to a comparison of three or more means.
The procedure for selecting sample data for a comparison of several population
means is called the design of experiment, and the statistical procedure for
comparing the means is called an analysis of variance (ANOVA). The objective of
this chapter is to introduce some aspects of experimental design and the analysis
of data from such experiments using an analysis of variance.
Experiment: The process of collecting sample data is called an experiment.
Design: The plan for collecting the sample is called the design of
experiment.
Response The variable measured in the experiment is called the response
Variable: variable. Response variable is denoted by ‘ Y ’. (In this chapter,
all response variables will be quantitative variables).
Experimental The object upon which the response variable is measured is
Unit: called an experimental (or sampling) unit.
One-way ANOVA: One-way analysis of variance, since it contains one independent
variable.
Two-way ANOVA: Two-way analysis of variance, since it contains two independent
variables.
2
One Way Classification
Statistical Analysis.
Samples ( or Treatments)
Observation Total
A B C ... j ... K
1 Y11 Y12 Y13 . . . Y1j ... Y1k
2 Y21 Y22 Y23 . . . Y2j ... Y2k
3 Y31 Y32 Y33 . . . Y3j ... X3k
. . . . . .
. . . . . .
. . . . . .
i Yi1 Yi2 Yi3 . . . Yij ... Yik
. . . . .
. . . . . .
. . . . . .
r Yr1 Yr2 Yr3 . . . Yrj ... Yrk
Total (C j) C1 C2 C3 Cj Ck T.. = S Y
C2j S C2 j
j
Si Y2 S S Y2
ij i j ij
C 2
j
SST =
j
− CM = Q1
r
SSE = TSS – SST = Q2
These results are summarized in the following ANOVA TABLE.
ANOVA TABLE
Source of
d.f. SS MS F-ratio = Fc F-table
Variation
s2
SST V1 = k – 1 Q1 Q1 / k – 1 = s2t Fc =
t
2
Fa(v1,v2)
se
SSE V2 = n – k Q2 Q2 / n – k = s2e ----
TSS n–1 Q ---- ---- ----
Always One-Tail! 0 Ft
Example # 1: You’re a trainer for Microsoft Corp. Is there a difference in mean
learning times of 12 people using 4 different training methods? Use a = 0.05
M1 M2 M3 M4
10 11 13 18
9 16 8 23
5 9 9 25
Solution
Source of
d.f. SS MS F-ratio = Fc F-table
Variation
SST 3 348 116 11.6 4.07
SSE 8 80 10 ----
TSS 11 428 ---- ---- ----
0 4.07
Conclusion: Our calculated value 11.6 falls in the critical region, so we will
accept H1 and conclude there is a difference in mean learning times of 12
people using 4 different training methods.
6
Example # 2
Determinations are made on the yield using three methods of catalyzing a chemical process.
Method - I Method - II Method - III
4 6 9
Measurements
9 8 13
10 10 15
11 11 20
17 12 23
19 12
15
Do the methods differ significantly at the 5% level of significance?
Methods Total
Method - I Method - II Method - III
4 6 9
9 8 13
10 10 15
11 11 20
17 12 23
19 12
15
Cj 70 74 80 224
(i) H0 : m1 = m2 = m3
(ii) H1 : Means methods are different (claim)
(iii) Level of significance a = 0.05 s2t 7
(iv) Test Statistic F= 2
se
Computation:
( 224 )2
CM = = 2787.56
18
TSS = 3206 – 2787.56 = 418.44
C 2j 4900 5476 6400
SST = – CM = + + − 2787.56 = 91.40
j r j 6 7 5
SSE = 418.44 – 91.40 = 327.04
ANOVA TABLE
Source of
d.f. SS MS F-ratio = Fc F-table
Variation
SST 2 91.40 45.70 2.10 3.68
SSE 15 327.24 21.80 ----
TSS 17 418.44 ---- ---- ----
0 3.68
Conclusion: Our calculated value 2.10 falls in the acceptance region, so we will
accept Ho and conclude that mean methods are same.
8
Example # 3: A researcher wishes to try three different techniques to lower the blood pressure
of individuals diagnosed with high blood pressure. The subjects are randomly assigned to three
groups; the first group takes medication, the second group exercises, and the third group follows a
special diet. After four weeks, the reduction in each person’s blood pressure is recorded. At a = 0.05,
test the claim that there is no difference among the means. Apply Scheffé test if applicable.
Medication Exercise Diet
10 6 5
12 8 9
9 3 12
15 0 8
13 2 4
Groups Total
Medication Exercise Diet
10 6 5
12 8 9
9 3 12
15 0 8
13 2 4
59 19 38 116
Cj
C2 j 3481 361 1444 5286
2 719 113 330
SY ij 1162
Means 11.8 3.8 7.6
(i) H0 : m1 = m2 = m3 (claim)
(ii) H1 : At least two means are not equal. (Means are different)
(iii) Level of significance a = 0.05
(iv) Test Statistic s2t 9
F=
s2e
Computation:
( 116 )2
CM = = 897.07
15
TSS = 1162 – 897.07 = 264.93
SST = ( 5286 / 5 ) – 897.07 = 160.13
SSE = 264.93 – 160.13 = 104.80
ANOVA TABLE
Source of
d.f. SS MS F-ratio = Fc F-table
Variation
SST 2 160.13 80.07 9.17 3.89
SSE 12 104.80 8.73 ----
TSS 14 264.93 ---- ---- ----
0 3.89
Conclusion: Our calculated value 9.17 falls in the critical region, so we will
reject Ho and conclude that at least two means are different.
Now Scheffé test is applicable.
10
a. For X1 versus X2
Fs = = = 18.33
b. For X2 versus X3
Fs = = = 4.14
c. For X1 versus X3
Fs = = = 5.05
11
Example # 4: A state employee wishes to see if there is a significant difference in
the number of employees at the interchanges of three state toll roads. The data are
shown. At a = 0.05, can it be concluded that there is a significant difference in the
average number of employees at each interchange? Apply Scheffé test if applicable.
Pennsylvania Greensburg Bypass/ Beaver Valley
Turnpike Mon-Fayette Expressway Expressway
7 10 1
14 1 12
32 1 1
19 0 9
10 11 1
11 1 11
State toll roads Total
PT GB/M Exp BV Exp
7 10 1
14 1 12
32 1 1
19 0 9
10 11 1
11 1 11
Cj 93 24 35 152
C2 j 8649 576 1225 10450
(i) H0 : m1 = m2 = m3
(ii) H1 : At least two means are not equal. (Claim)
12
(iii) Level of significance a = 0.05 s2t
(iv) Test Statistic F = s2
Computation:
( 152 )2
CM = = 1283.56
18
TSS = 2424 – 1283.56 = 1140.44
SST = ( 10450 / 6 ) – 1283.56 = 458.11
SSE = 1140.44 – 458.11 = 682.33
ANOVA TABLE
Source of
d.f. SS MS F-ratio = Fc F-table
Variation
SST 2 458.11 229.06 5.04 3.68
SSE 15 682.33 45.47 ----
TSS 17 1140.44 ---- ---- ----
0 3.68
Conclusion: Our calculated value 5.04 falls in the critical region, so we will
reject Ho and conclude that at least two means are different.
Now Scheffé test is applicable.
13
a. For X1 versus X2
𝑋ത1 − 𝑋ത2 2 15.5 − 4 2
Fs = = = 8.72
2 1 1 1 1
𝑆 𝑒 + 45.47 × +
𝑛1 𝑛2 6 6
b. For X2 versus X3
𝑋ത2 − 𝑋ത3 2 4 − 5.8 2
Fs = = = 0.21
1 1 1 1
𝑆2𝑒 + 45.47 × +
6 6
𝑛2 𝑛3
c. For X1 versus X3
𝑋ത1 − 𝑋ത3 2 15.5 − 5.8 2
Fs = = = 6.21
2 1 1 1 1
𝑆 𝑒 + 45.47 × +
𝑛1 𝑛3 6 6
Since only the Fs test value for part a is greater than F/ value, 7.36, the
14
Example # 5:
ANOVA table shown below. Test the hypothesis that three-treatment means are equal at a = 0.05.
ANOVA TABLE
SOV d.f. SS MS Fc
SST 546 273
SSE 97.36
TSS 11 1422.25 ---- ----
(i) H0 : m1 = m2 = m3
H1 : At least two means are not equal.
(ii) Level of significance a = 0.05
(iii) Test Statistic
s2t
F= 2
se
(iv) Computation:
ANOVA TABLE
Test the hypothesis that three varieties of potatoes are not different in the yielding capabilities. Let
a =0.05.
Answer # 1
Critical region: F F 0.05 (2, 9 ) = 4.26
SSE = 217.75
SST = 54.50
Fcal = 1.13
Question # 2: A consumer magazine rated dishwashers as excellent, very good, good, and
poor. A researcher wishes to see if the average prices for the four group differ. At a = 0.05, is
there a difference in the average prices of the machines rated?
Excellent Very good Good Poor
Answer # 2
49 49 44 58
SSE = 1967.71
42 44 57 54
SST = 349.62
47 50 34 64
S
Fcal = 1.18
76 58 48 60 C 2
j
69 79 50 53 SST = – CM
58 - - 64 rj
j
- - - 52
- - - 42
16
Question # 3: The grade point averages of students participating in university MBA
majors (Marketing, Finance, HRM) are compared. The data are shown below.
Majors
Marketing Finance HRM
3.2 3.8 2.6
2.6 3.1 1.9
2.4 2.6 1.7
2.4 3.9 2.5
1.8 3.3 1.9
At a =0.05, can one conclude that there is a difference in the mean GPA of the three majors?.
Question # 4: A researcher wishes to see whether there is any difference in the weight
gains of athletes following one of three special diets. Athletes are randomly assigned to three
groups and placed on the diet for six weeks. The weight gains (in pounds) are shown here. At a =
0.05, can the researcher conclude that there is a difference in the diets?
Treatments Total
Blocks R2i S Y2ij
A B C ... j ... k (Ri) j
Total (C j) C1 C2 C. 3 Cj Ck SY SR
T2..i
C2 j S C2 j
S S Y2ij
S Y2ij i j
i
R
rxk 2
n = Total number of observations.
i
TSS = Yij − CM = Q − CM = Q2
2 i
SSB =
i j c
C 2
j
SST =
j
− CM = Q1 SSE = TSS – SST – SSB = Q3
r
Scheffé Test
Similarly, when the null hypothesis of equal means is rejected by F – test after
ANOVA, then we can apply Scheffé – test.
In order to conduct this test compare, one must compare the means two at a time, using
all possible combinations of means. For example, if there are three means, the following
comparisons must be done:
Fs =
20
Where X i and X j are the means of the samples being compared, ni and nj are the
respective sample sizes, and s2e is the SSE (within-group variance).
To find the critical value F/ for the Scheffé test, multiply the Critical Value for the F test
by k – 1:
F/ = (k – 1)(C.V)
There is a significant difference between the two means being compared when Fs is greater
than F/.
(i) H0 : m1 = m2 = m3 = m4
H1 : At least two varieties means are not equal.
(ii) Level of significance a = 0.05
s2
(iii) Test Statistic F = t
s e2
1
21
(iv) Computation:
2 5 4 1 12 144 46
2 3 3 1 9 81 23
4 6 6 2 18 324 92
1 4 2 3 10 100 30
Total (C j) 9 18 15 7 49 649 191
C2j 81 324 225 49 679
S Y2ij 25 86 65 15 191
i
2 x 1.01
Now LSD = t0.025, (9) x
4
= 2.26 x 0.71
= 1.60
Arranging the means of the four varieties of wheat in ascending order of magnitude,
and drawing a line under pair of adjacent means that are not significantly different,
we have
V4 V1 V3 V2
1.75 2.25 3.75 4.50
23
Example # 7: A medical researcher wishes to test the four diets and the time of day on
the sodium level in a person’s blood. Analyze the data shown in the tables below using a two-
way ANOVA at a = 0.05. The sodium content is measured in milliequivalents per liter. Apply LSD
test to locate significant difference between the various pairs of diet types if applicable.
Diet type
7:00 am ~ 7:00 pm
I II III IV
4.0 4.8 5.0 4.6
Time
4.8 5.0 5.2 4.6
4.0 4.8 5.6 5.0
(i) H0 : m1 = m2 = m3 = m4
H1 : At least two diet means are not equal.
/
(ii) H0 : m1 = m2 = m3
/
H1 : At least two-time means are not equal.
(iii) Level of significance a = 0.05
24
(vi) Computation:
( 57.4 )2
CM = = 274.56 TSS = 276.84 – 274.56 = 2.28
12
SSD = (828.28 / 3) – 274.56 = 1.53 SSTi = (1099.08 / 4) – 274.56 = 0.21
2 x 0.09
Now LSD = t0.025, (6) x
3
= 2.447 x 0.2449
= 0.5993
Arranging the means of the four diets in ascending order of magnitude, and drawing
a line under pair of adjacent means that are not significantly different, we have
D1 D4 D2 D3
4.27 4.73 4.87 5.27
Pairs which are significantly different are (D1, D2) and (D1,D3)
26
Example # 8: Perform the analysis of variance on the following data and analyze the
fertilizer means using Duncan’s Multiple Range test with a = 0.05 level of
significance.
Fertilizer
1 2 3 4 5 6
I 1 3 6 4 3 1
Variety II 1 4 4 8 5 1
III 3 6 7 8 4 3
IV 2 3 2 3 2 1
(i) H0 : m1 = m2 = m3 = m4 = m5 = m6
H1 : At least two fertilizer means are not equal.
(ii) Level of significance a = 0.05
s 2t
(iii) Test Statistic: F=
(iv) Computation:
s e2
Computation is presented in the following ANOVA table.
ANOVA TABLE
SOV d.f. SS MS F-ratio = Fc F-table
SST(Fertilizer) 5 51.83 10.37 6.61 2.90
SSV 3 28.50 9.50
SSE 15 23.50 1.57 ----
TSS 23 103.83 ---- ---- ---- 27
(v) Critical region: F F0.05 (5, 15) = 2.90
(vi) Conclusion: Our calculated F value 6.61 falls in
the critical region, so we will accept
H1 & conclude that fertilizer means
are significantly different.
Duncan’s Multiple Range test is 0 2.90
applicable.
𝑠2𝑒
Rp = qa (p, v) × where p = 2, 3, 4, ……..,k-1, k
𝑟
r = no of variety in each treatment = 4
v = error d.f. = 15 and k = number of treatment = 6
𝑠2𝑒 1.57
= = 0.6265
𝑟 4
The values of q0.05 (p, 15) for p = 2, 3, 4, 5, 6. taken from Duncan’s table of significant
𝑠2 𝑒
ranges and the least significant ranges, Rp obtained by multiplying q0.05 (p, 15) by ,
𝑟
𝑎𝑟𝑒 𝑠ℎ𝑜𝑤𝑛 𝑏𝑒𝑙𝑜𝑤
p q0.05 (p, 15) Rp
2 3.01 1.89 (R2) 3.01 × 0.6265 = 1.89
3 3.16 1.98 (R3)
4 3.25 2.04 (R4)
5 3.31 2.07 (R5)
6 3.36 2.11 (R6) 28
Arranging the means of the six fertilizers in increasing order of magnitude, we get
F1 F6 F5 F2 F3 F4
1.75 1.75 3.50 4.00 4.75 5.75
Comparing the differences between all pairs of means with the least significant
ranges, Rp , beginning with the largest (F4) against the smallest (F1) , we have
the following results:
4 versus 1 : 5.75 – 1.75 = 4.00 > 2.11 (R6)
4 versus 6 : 5.75 – 1.75 = 4.00 > 2.07 (R5)
4 versus 5 : 5.75 – 3.50 = 2.25 > 2.04 (R4)
4 versus 2 : 5.75 – 4.00 = 1.75 < 1.98 (R3)
4 versus 3 : 5.75 – 4.75 = 1.00 < 1.89 (R2)
3 versus 1 : 4.75 – 1.75 = 3.00 > 2.07 (R5)
3 versus 6 : 4.75 – 1.75 = 3.00 > 2.04 (R4)
3 versus 5 : 4.75 – 3.50 = 1.25 < 1.98 (R3)
3 versus 2 : 4.75 – 4.00 = 0.75 < 1.89 (R2)
2 versus 1 : 4.00 – 1.75 = 2.25 > 2.04 (R4)
2 versus 6 : 4.00 – 1.75 = 2.25 > 1.98 (R3)
2 versus 5 : 4.00 – 3.50 = 0.50 < 1.89 (R2)
5 versus 1 : 3.50 – 1.75 = 1.75 < 1.98 (R3)
5 versus 6 : 3.50 – 1.75 = 1.75 < 1.89 (R2)
6 versus 1 : 1.75 – 1.75 = 0.00 < 1.89 (R2)
29
Drawing a line under means which are not significant different (difference less than
Rpvalue)
F1 F6 F5 F2 F3 F4
1.75 1.75 3.50 4.00 4.75 5.75
The Pairs which are significantly different (whose differences are greater than the
corresponding least significant ranges Rpvalue) are:
(F1, F2), (F1,F3), (F1,F4), (F6,F2), (F6,F3), (F6,F4), and (F5,F4).
From the analysis, we see that the LSD method and Duncan’s Multiple Range Test
produce the same conclusion.
30
Example # 9: Four observers determine the moisture content of samples of a powder,
each man taking a sample of six consignments. Perform the ANOVA on
the following data and discuss whether there is any difference between
consignments & observers. a =0.05.
ANOVA TABLE
SOV d.f. SS MS Fc
SSC
SSO 13.13
SSE 13.12
TSS 35.96 ---- ----
(i) H0 : mc1 = mc2 = mc3 = mc4 = mc5 = m c6 H1 : At least two consignment means are not equal.
/ /
H0 : mo1 = mo2 = mo3 = mo4 H1 : At least two observers means are not equal.
(ii) Level of significance a = 0.05
s2o
(iii) Test Statistic s2c F2 = 2
F= s
(iv)
1
s2e e
Computation:
ANOVA TABLE
SOV d.f. SS MS Fc
SSB 781.15
SSR
SSE 229.63
TSS 1449.42 ---- ----
At a = 0.05, test the null hypothesis that
(i) There are no differences in the mean gains in weight for the different breeds of cattle.
(ii) There are no differences between rations.
(iii) Apply LSD-test on breeds if applicable. Breeds means are B1 = 48, B2 = 47.8, B3 = 29.5, &
B4 = 35
Question # 6: The analysis of variance produced the ANOVA table shown below:
SOV d.f. SS MS Fc
Treatments 3 28.2
Blocks 5 13.80
Error 34.1
(i) Complete the ANOVA table.
(ii) Do the data provide sufficient evidence to indicate a difference between the treatment
means? Test using a =0.01
32
Critical region: F F 0.01 (3, 15 ) = 5.42
Fc = 4.14
Question # 7: Consider aspirin brands H0: µA = µB = µC = µD = µE
(i) Complete the following ANOVA table with α = 0.05
Source of Variation d.f. SS MS FC
Treatments (aspirin) 49.15
Blocks 4 46.67
Error 2.34
Total 280.71
(ii) Apply LSD test if applicable.
(iii) Means for aspirin brands A, B, C, D, & E are 6.84, 6.46, 13.12, 7.96, & 4.92
respectively. Indicate which pairs of means for aspirin brands differ significantly.
(iv) Pick up the best aspirin brand.
Question # 8: Four blocks I, II, III, and IV, & four varieties of
wheat A, B, C, & D are given below:
A B C D
I 17 27 15 25
II 28 26 16 22
III 22 25 11 14
IV 17 18 18 19
(a) Perform the ANOVA to test at a = 0.05, the differences in the yields of varieties & in blocks.
(b) What would have been the result if no blocking had been done.
(a) Critical region: F F 0.05 (3, 9 ) = 3.86 SSV = 168, SSB = 72, TSS = 392
F1 = 3.32 F2 = 1.42, SSE = 152 33
(b) Critical region: F F 0.05 (3, 12 ) = 3.49 SSE = 224, TSS = 392, F = 3.00
Question # 9: Perform the analysis of variance on the following data and analyze the
fertilizer means using Scheffé test with a = 0.05 level of significance.
Fertilizer
A B C D E F
I 2 6 12 8 6 4
Variety II 2 8 8 16 10 2
III 6 12 14 16 8 6
IV 4 6 4 6 4 2
Question # 10: Each of five varieties of corn is planted in three plots in a large field. The
respective yields, in bushels per acre, are indicated below
V1 V2 V3 V4 V5
46.2 49.2 60.3 48.9 52.5
(a) Test whether the differences among the average yields are statistically significant at 5%
level of significance.
(b) Use Duncan’s Multiple Range Test to make comparisons between pairs of means.
(a) Critical region: F F 0.05 (4, 10 ) = 3.48, F(calculated) = 6.39 Reject Ho
𝑠2 𝑒 p = 2, 3, 4, 5. q0.05 (p, 10) = 3.15, 3.30, 3.37, 3.43. Rp = 5.91(R2), 6.19(R3), 6.33(R4), 6.44(R5) 34
(b) s2e = 10.57, = 1.877
𝑟
The Pairs which are significantly different (whose differences are greater than the corresponding least significant ranges Rpvalue)
are: (V4, V2), (V4,V3), (V1,V3), and (V5,V3).
Question # 11 Perform an ANOVA upon the following data and test whether there
is any significant evidence of difference due to the drugs and diets
at 5% level of significance.
.
Diet
Drug
1 2 3 4
A 46.5 62 41 45
C 50 40 25.6 28.5
The Pairs which are significantly different (whose differences are greater than the corresponding least significant ranges Rpvalue) 35
are: (D3, D2), (D3,D1), (D4,D2), and (D4,D1).
1. ANOVA for Simple Linear Regression & Test of Ho: b = 0
Example # 10
Given n = 12, SY = 1700, SXY = 109380,
SY2 = 246100 and the estimated
regression line is Ŷ = – 179.42 + 5.03 X.
Test the hypothesis that b 0 at a = 0.05
by applying the analysis of variance
technique.
2. ANOVA for Multiple Regression & Testing Hypothesis about the b Parameters
Example # 11
Consider the following set of data:
Y 4 6 5 8 12 13 19
X1 1 1 2 3 4 5 7
X2 2 4 3 4 8 8 11
(a) Find the estimated regression equation Ŷ = a + b1X1 + b2X2
(b) Obtain the ANOVA table and test the hypothesis that there
is no association between either regressor and the
dependent variable. Use the 0.01 level of significance.
37
(a) The equation of the estimated multiple linear regression is Ŷ = a + b1X1 + b2X2,
where a, b1 and b2 are the least-squares estimates of a, b1 and b2 respectively.
(b) Ho: b1 = b2 = 0 (Overall model is not useful or none of the regressors is significant)
H1: At least one of the model parameters, b1 or b2 differs from zero.
a = 0.01 (Overall model is useful)
F = Reg SS / SSE
( SY) 2
TSS = S(Y - Y) = SY −
2 2
= 815 – (67)2/7 = 173.71
n
Reg SS = S(Ŷ - Y) 2 = aSY + b1SX1Y + b2SX2Y - (SY) / n
2
39
Question # 12
The data given below represent the heights (X) and the weights (Y) of five men. We selected the
heights in advance and then observed the weights of a random group of men having the
selected heights.
X 60 62 65 70 72
Y 130 135 158 170 185
(a) Find the estimated regression equation Ŷ = a + bX.
(b) Obtain the ANOVA table and test the hypothesis that Regression model is useful at a = 0.05.
Ans: Ŷ = - 137.87 + 4.46X, TSS = 2157.2, SSE = 73.26,
Reg SS = 2083.94, Critical region: F F 0.05 (1, 3 ) = 10.13
Question # 13
Consider the following set of data:
Y 12 10 9 13 20
X1 2 2 3 4 4
X2 1 1 0 0 3
(a) Find the estimated regression equation Ŷ = a + b1X1 + b2X2
(b) Obtain the ANOVA table and test the hypothesis H0: b1 = b2 = 0. Use the 0.05 level of
significance.
Ans: Ŷ = 3.88 + 2.09X1 + 2.65X2, TSS = 74.8, Reg SS = 70.69, SSE = 4.11, Fc = 17.20,
Critical region: F F 0.05 (2, 2 ) = 19.00, Accept H0 40
Homework
EXERCISES. (Elementary Statistics, Bluman, 4th Edition)
41
q(0.05)
Significant Ranges for Duncan’s Multiple Range Test
q 0.05 (p, v)
p
2 3 4 5 6 7 8 9 10 20
1 18.00 18.00 18.00 18.00 18.00 18.00 18.00 18.00 18.00 18.00
2 6.09 6.09 6.09 6.09 6.09 6.09 6.09 6.09 6.09 6.09
3 4.50 4.50 4.50 4.50 4.50 4.50 4.50 4.50 4.50 4.50
4 3.93 4.01 4.02 4.02 4.02 4.02 4.02 4.02 4.02 4.02
5 3.64 3.74 3.79 3.83 3.83 3.83 3.83 3.83 3.83 3.83
6 3.46 3.58 3.64 3.68 3.68 3.68 3.68 3.68 3.68 3.68
7 3.35 3.47 3.54 3.58 3.60 3.61 3.61 3.61 3.61 3.61
8 3.26 3.39 3.47 3.52 3.55 3.56 3.56 3.56 3.56 3.56
9 3.20 3.34 3.41 3.47 3.50 3.52 3.52 3.52 3.52 3.52
10 3.15 3.30 3.37 3.43 3.46 3.47 3.47 3.47 3.47 3.48
11 3.11 3.27 3.35 3.39 3.43 3.44 3.45 3.46 3.46 3.48
12 3.08 3.23 3.33 3.36 3.40 3.42 3.44 3.44 3.46 3.48
13 3.06 3.21 3.30 3.35 3.38 3.41 3.42 3.44 3.45 3.47
14 3.03 3.18 3.27 3.33 3.37 3.39 3.41 3.42 3.44 3.47
15 3.01 3.16 3.25 3.31 3.36 3.38 3.40 3.42 3.43 3.47
16 3.00 3.15 3.23 3.30 3.34 3.37 3.39 3.41 3.43 3.47
17 2.98 3.13 3.22 3.28 3.33 3.36 3.38 3.40 3.42 3.47
18 2.97 3.12 3.21 3.27 3.32 3.35 3.37 3.39 3.41 3.47
19 2.96 3.11 3.19 3.26 3.31 3.35 3.37 3.39 3.41 3.47
20 2.95 3.10 3.18 3.25 3.30 3.34 3.36 3.38 3.40 3.47
30 2.89 3.04 3.12 3.20 3.25 3.29 3.32 3.35 3.37 3.47
40 2.86 3.01 3.10 3.17 3.22 3.27 3.30 3.33 3.35 3.47
42
q(0.01)
Significant Ranges for Duncan’s Multiple Range Test
q 0.01 (p, v)
p
2 3 4 5 6 7 8 9 10 20
1 90.00 90.00 90.00 90.00 90.00 90.00 90.00 90.00 90.00 90.00
2 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00
3 8.26 8.50 8.60 8.70 8.80 8.90 8.90 9.00 9.00 9.30
4 6.51 6.80 6.90 7.00 7.10 7.10 7.20 7.20 7.30 7.50
5 5.70 5.96 6.11 6.18 6.26 6.33 6.40 6.44 6.50 6.80
6 5.24 5.51 5.65 5.73 5.81 5.88 5.95 6.00 6.00 6.30
7 4.95 5.22 5.37 5.47 5.53 5.61 5.69 5.73 5.80 6.00
8 4.74 5.00 5.14 5.23 5.22 5.40 5.47 5.51 5.80 5.80
9 4.60 4.86 4.99 5.08 5.17 5.25 5.32 5.36 5.40 5.70
10 4.48 4.73 4.88 4.96 5.06 5 .13 5.20 5.24 5.28 5.55
11 4.39 4.63 4.77 4.86 4.94 5.01 5.06 5.12 5.15 5.39
12 4.32 4.55 4.68 4.76 4.84 4.92 4.96 5.02 5.07 5.26
13 4.26 4.48 4.62 4.69 4.74 4.84 4.88 4.94 4.98 5.15
14 4.21 4.42 4.55 4.63 4.70 4.78 4.83 4.87 4.91 5.07
15 4.17 4.37 4.50 4.58 4.64 4.72 4.77 4.81 4.84 5.00
16 4.13 4.24 4.45 4.54 4.60 4.67 4.72 4.76 4.79 4.94
17 4.10 4.30 4.41 4.50 4.56 4.63 4.68 4.73 4.75 4.89
18 4.07 4.27 4.38 4.46 4.53 4.59 4.64 4.68 4.71 4.85
19 4.05 4.24 4.35 4.43 4.50 4.56 4.61 4.64 4.67 4.82
20 4.02 4.22 4.33 4.40 4.47 4.53 4.58 4.61 4.65 4.79
30 3.89 4.06 4.16 4.22 4.32 4.36 4.41 4.45 4.48 4.65
40 3.82 3.99 4.10 4.17 4.24 4.30 4.34 4.37 4.41 4.59 43