Chapter Six Non Parametric Statistics

Chapter Six
Non Parametric Statistics

Hypothesis Testing Procedures
Hypothesis
Hypothesis
Testing
Testing
Procedures
Procedures
Parametric
Parametric Nonparametric
Nonparametric
Wilcoxon
Wilcoxon Kruskal-Wallis
Rank Kruskal-Wallis
RankSum
Sum H-Test
H-Test
Test
Test
One-Way
One-Way
ZZTest
Test ttTest
Test ANOVA
ANOVA
Many More Tests Exist!
Parametric Test Procedures
1. Involve Population Parameters (Mean)
2. Have Stringent Assumptions

(Normality)
3. Examples: Z Test, t Test, F test

Nonparametric Test Procedures
1. Do Not Involve Population Parameters

Example: Probability Distributions, Independence
2. Data Measured on Any Scale (Ratio or

Interval, Ordinal or Nominal)
3. Example: Wilcoxon Rank Sum Test

Advantages of Nonparametric Tests
1. Used With All Scales

2. Easier to Compute
3. Make Fewer Assumptions
4. Need Not Involve
Population Parameters
5. Results May Be as Exact
as Parametric Procedures
Disadvantages of Nonparametric Tests
1. May Waste Information

Parametric model more efficient
if data Permit
2. Difficult to Compute by
hand for Large Samples
3. Tables Not Widely Available
Popular Nonparametric Tests
1. Sign Test
2. Wilcoxon Rank Sum Test
3. Wilcoxon Signed Rank Test

4. Spearman’s correlation Test
Sign Test
Sign Test
1. Tests One Population Median, 
2. Corresponds to t-Test for 1 Mean
3. Assumes Population Is Continuous
4. Small Sample Test Statistic: # Sample Values Above

(or Below) Median
5. Can Use Normal Approximation If n  10

Sign Test Concepts
Make null hypothesis about true median
Let S = number of values greater than median
Each sampled item is independent
If null hypothesis is true, S should have binomial

distribution with success probability .5
Sign Test Example
You want to know what people feel about Wednesday

night drama. You’ve asked 7 people to rate the drama
on a 5-point scale (1 = terrible,…, 5 = excellent) The
ratings are: 2 5 3 4 1 4 5.
At the .05 level, is there evidence that the median

rating is at least 3?
Sign Test Solution
H0: P-Value:
Ha:
=
Test Statistic:
Decision:
Conclusion:
Sign Test Solution
H0:  < 3 P-Value:

Ha:  > 3
=
Test Statistic:
Decision:
Conclusion:
Sign Test Solution
H0:  <3 P-Value:

Ha:  > 3
 = .05
Test Statistic:
Decision:
Conclusion:
Sign Test Solution
H0:  < 3 P-Value:

Ha:  > 3
 = .05
Test Statistic:
S=2
(Ratings 1 & 2 Are Decision:
Less Than  = 3:
2, 5, 3, 4, 1, 4, 5) Conclusion:
Is observing 2 or more a
small prob event?
Sign Test Solution
H0:  < 3 P-Value:

Ha:  > 3 P(S  3) = 1 - P(S  2)
 = .05 = 1-0.1641 = 0.8339
Test Statistic: (Binomial Table, n = 7, p
S=2 = 0.50)
Less Than  = 3:
2, 5, 3, 4, 1, 4, 5) Conclusion:
small prob event?
Sign Test Solution
H0:  <3 P-Value:

Ha:  > 3 P(S  3) = 1 - P(S  2)
 = .05 = 1-0.1641 = 0.8339
S=2 = 0.50)
Reject at  = .05
Less Than  = 3:
2, 5, 3, 4, 1, 4, 5) Conclusion:
small prob event?
Sign Test Solution
H0:  < 3 P-Value:

Ha:  > 3 P(S  3) = 1 - P(S  2)
 = .05 = 1-0.1641 = 0.8339
S=2 = 0.50)
(Ratings 1 & 2 are < Decision:
Reject at  = .05
 = 3:
2, 5, 3, 4, 1, 4, 5) Conclusion:
Is observing 2 or more a There is enough evidence
small prob event? for Median > 3
Wilcoxon Rank Sum Test
1.Tests Two Independent Population Probability

Distributions
2.Corresponds to t-Test for 2 Independent Means
3.Assumptions
Independent, Random Samples
Populations Are Continuous
4.Can Use Normal Approximation If ni  10
Procedure
1. Assign Ranks, Ri, to the n1 + n2 Sample Observations

If Unequal Sample Sizes, Let n1 Refer to Smaller-Sized Sample
Smallest Value = 1
2. Sum the Ranks, Ti, for Each Sample

Test Statistic Is TA (Smallest Sample)
Null hypothesis: both samples come from the same underlying distribution
Distribution of T is not quite as simple as binomial, but it can be computed
Wilcoxon Rank Sum Test : Example
You’re a production planner. You want to see if the

operating rates for 2 factories is the same. For factory
1, the rates (% of capacity) are 71, 82, 77, 92, 88. For
factory 2, the rates are 85, 82, 94 & 97. Do the factory
rates have the same probability distributions at the .10
level?
Solution
H0: Test Statistic:

Ha:
=
n1 = n2 =
Critical Value(s):
Decision:
Conclusion:
 Ranks
Solution
H0: Identical Distrib. Test Statistic:

Ha: Shifted Left or Right
=
n1 = n2 =
Critical Value(s):
Decision:
Conclusion:
 Ranks
Solution

 = .10
n1 = 4 n2 = 5
Critical Value(s):
Decision:
Conclusion:
 Ranks
Solution

 = .10
n1 = 4 n2 = 5
Critical Value(s):
Decision:
Reject Do Not Reject

Reject Conclusion:
12 28  Ranks
Wilcoxon Rank Sum Test Computation
Table
Factory 1 Factory 2
Rate Rank Rate Rank
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 85
82 82
77 94
92 97
88 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85
82 82
77 94
92 97
88 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85
82 82
77 2 94
92 97
88 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85
82 3 82 4
77 2 94
92 97
88 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85
82 3 3.5 82 4 3.5
77 2 94
92 97
88 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85 5
82 3 3.5 82 4 3.5
77 2 94
92 97
88 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85 5
82 3 3.5 82 4 3.5
77 2 94
92 97
88 6 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85 5
82 3 3.5 82 4 3.5
77 2 94
92 7 97
88 6 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85 5
82 3 3.5 82 4 3.5
77 2 94 8
92 7 97
88 6 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85 5
82 3 3.5 82 4 3.5
77 2 94 8
92 7 97 9
88 6 ... ...
Rank Sum
Table
Factory 1 Factory 2
Rate Rank Rate Rank
71 1 85 5
82 3 3.5 82 4 3.5
77 2 94 8
92 7 97 9
88 6 ... ...
Rank Sum 19.5 25.5
Solution

Ha: Shifted Left or Right T2 = 5 + 3.5 + 8+ 9 = 25.5
 = .10 (Smallest Sample)
n1 = 4 n2 = 5
Critical Value(s):
Decision:

Reject Conclusion:
12 28  Ranks
Solution

n1 = 4 n2 = 5
Critical Value(s):
Decision:
Do Not Reject at  = .10
Reject Conclusion:
12 28  Ranks
Solution

n1 = 4 n2 = 5
Critical Value(s):
Decision:
Do Not Reject at  = .10
Reject Conclusion:
There is No evidence for
12 28  Ranks
unequal distrib
Spearman’s Rank
Correlation Coefficient
Spearman’s Rank
1.Measures Correlation Between Ranks

2.Corresponds to Pearson Product Moment
3.Values Range from -1 to +1
Spearman’s Rank
1.Measures Correlation Between Ranks

2.Corresponds to Pearson Product Moment
3.Values Range from -1 to +1
4.Equation (Shortcut)
6 d 2
rs  1 

n n2 1 
Spearman’s Rank Correlation
Procedure
1.Assign Ranks, Ri , to the Observations

of Each Variable Separately
2.Calculate Differences, di , Between
Each Pair of Ranks
3.Square Differences, di 2, Between Ranks
4.Sum Squared Differences for Each Variable
5.Use Shortcut Approximation Formula
Spearman’s Rank
Correlation Example
You’re a research assistant for the ABC. You’re

investigating the relationship between a person’s attempts
at deception Subj. Deception Pupil
& % changes in their 1 87 10
pupil size. You ask
2 63 6
subjects a series of
questions, some of 3 95 11
which they must 4 50 7
answer dishonestly. 5 43 0
At the .05 level, what is the correlation coefficient?
Spearman’s Rank Correlation Table
Subj. Decep. R1i Pupil R2i di di2
Total

1 87 10
2 63 6
3 95 11
4 50 7
5 43 0
Total

1 87 4 10
2 63 3 6
3 95 5 11
4 50 2 7
5 43 1 0
Total

1 87 4 10 4
2 63 3 6 2
3 95 5 11 5
4 50 2 7 3
5 43 1 0 1
Total

1 87 4 10 4 0
2 63 3 6 2 1
3 95 5 11 5 0
4 50 2 7 3 -1
5 43 1 0 1 0
Total

1 87 4 10 4 0 0
2 63 3 6 2 1 1
3 95 5 11 5 0 0
4 50 2 7 3 -1 1
5 43 1 0 1 0 0
Total 2
Spearman’s Rank
Correlation Solution
n
6 d i
2
rs  1  i 1

n n 1 2

6 2 
 1 2

5 5 1 
 1  0.10
 0.90
End of Chapter 6
Thank you!

Chapter Six Non Parametric Statistics

Uploaded by

Copyright:

Available Formats

Chapter Six Non Parametric Statistics

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter Six Non Parametric Statistics

Uploaded by

Copyright:

Available Formats

Chapter Six

Non Parametric Statistics

1. Involve Population Parameters (Mean)

2. Have Stringent Assumptions

3. Examples: Z Test, t Test, F test

1. Do Not Involve Population Parameters

2. Data Measured on Any Scale (Ratio or

3. Example: Wilcoxon Rank Sum Test

1. Used With All Scales

1. May Waste Information

2. Wilcoxon Rank Sum Test

3. Wilcoxon Signed Rank Test

1. Tests One Population Median, 

2. Corresponds to t-Test for 1 Mean

3. Assumes Population Is Continuous

4. Small Sample Test Statistic: # Sample Values Above

5. Can Use Normal Approximation If n  10

Make null hypothesis about true median

Let S = number of values greater than median

Each sampled item is independent

If null hypothesis is true, S should have binomial

You want to know what people feel about Wednesday

At the .05 level, is there evidence that the median

H0:  < 3 P-Value:

H0:  <3 P-Value:

H0:  < 3 P-Value:

H0:  < 3 P-Value:

H0:  <3 P-Value:

H0:  < 3 P-Value:

1.Tests Two Independent Population Probability

1. Assign Ranks, Ri, to the n1 + n2 Sample Observations

2. Sum the Ranks, Ti, for Each Sample

You’re a production planner. You want to see if the

H0: Test Statistic:

H0: Identical Distrib. Test Statistic:

H0: Identical Distrib. Test Statistic:

H0: Identical Distrib. Test Statistic:

Reject Do Not Reject

H0: Identical Distrib. Test Statistic:

Reject Do Not Reject

H0: Identical Distrib. Test Statistic:

H0: Identical Distrib. Test Statistic:

1.Measures Correlation Between Ranks

1.Measures Correlation Between Ranks

1.Assign Ranks, Ri , to the Observations

You’re a research assistant for the ABC. You’re

Subj. Decep. R1i Pupil R2i di di2

Subj. Decep. R1i Pupil R2i di di2

Subj. Decep. R1i Pupil R2i di di2

Subj. Decep. R1i Pupil R2i di di2

Subj. Decep. R1i Pupil R2i di di2

Subj. Decep. R1i Pupil R2i di di2

You might also like