4 - Analyze - Intro To Hypothesis Testing

Analyze Phase
Introduction to Hypothesis Testing

Hypothesis Testing (ND)
Welcome to Analyze
“X” Sifting
Inferential Statistics Hypothesis Testing Purpose
Tests for Central Tendency

Intro to Hypothesis Testing
Tests for Variance
Hypothesis Testing ND P1
ANOVA
Hypothesis Testing ND P2
Hypothesis Testing NND P1
Hypothesis Testing NND P2
Wrap Up & Action Items
LSS Green Belt v11.1 MT - Analyze Phase 2 © Open Source Six Sigma, LLC
Six Sigma Goals and Hypothesis Testing
Process Capability of Process Before
LSL USL
P rocess Data Within
LS L 100.00000 Ov erall
Target *
USL 120.00000 P otential (Within) C apability
S ample M ean 108.65832 Cp 1.42
S ample N 150 C PL 1.23
S tDev (Within) 2.35158 C PU 1.61
S tDev (O v erall) 5.41996 C pk 1.23
C C pk 1.42
O v erall C apability
Pp 0.62
PPL 0.53
PPU 0.70
P pk 0.53
C pm *
Process Capability of Process After

96 100 104 108 112 116 120
O bserv ed P erformance E xp. Within P erformance E xp. O v erall P erformance
LSL USL
P P M < LS L 6666.67 P P M < LS L 115.74 P P M < LS L 55078.48
P rocess Data Within
PPM > USL 0.00 PPM > USL 0.71 P P M > U S L 18193.49
LS L 100.00000 Overall
P P M Total 6666.67 P P M Total 116.45 P P M Total 73271.97
Target *
USL 120.00000 P otential (Within) C apability
S ample M ean 109.86078 Cp 2.14
S ample N 100 C PL 2.11
C PU 2.17
Hypothesis Testing helps us S tDev (Within)
S tDev (O v erall)
1.55861
1.54407 C pk 2.11
C C pk 2.14
make fact-based decisions about O v erall C apability

Pp 2.16
PPL 2.13
whether there are different PPU
P pk
2.19
2.13
C pm *
population parameters or if the
differences are simply due to 102 105 108 111 114 117 120
expected sample variation. O bserv ed P erformance

P P M < LS L 0.00
E xp. Within P erformance
P P M < LS L 0.00
E xp. O v erall P erformance
P P M < LS L 0.00
P P M > U S L 0.00 P P M > U S L 0.00 P P M > U S L 0.00
P P M Total 0.00 P P M Total 0.00 P P M Total 0.00
Purpose of Hypothesis Testing
The purpose of appropriate Hypothesis Testing is to integrate the Voice of the

Process with the Voice of the Business to make data-based decisions to
resolve problems.
Hypothesis Testing can help avoid high costs of experimental efforts by using
existing data. This can be likened to:
– Local store costs versus mini-bar expenses.
– There may be a need to eventually use experimentation but careful data
analysis can indicate a direction for experimentation if necessary.
The probability of occurrence is based on a pre-determined statistical

confidence.
Decisions are based on:

– Beliefs (past experience)
– Preferences (current needs)
– Evidence (statistical data)
– Risk (acceptable level of failure)
The Basic Concept for Hypothesis Tests
Recall from the discussion on classes and cause of distributions that a

data set may seem Normal yet still be made up of multiple distributions.
Hypothesis Testing can help establish a statistical difference between

factors from different distributions.
0.8
0.7
0.6
0.5
freq
0.4
0.3
0.2
0.1
0.0
-3 -2 -1 0 1 2 3
x
Did my sample come from this population? Or this? Or this?

Significant Difference
Are the two distributions “significantly” different from each

other?
How sure are we of our decision?
How do the number of observations affect our confidence in

detecting population Mean?
 
Sample 1 Sample 2
Detecting Significance
Statistics provide a methodology to detect differences.

– Examples might include differences in suppliers, shifts or
equipment.
– Two types of significant differences occur and must be well
understood…. practical and statistical.
– Failure to tie these two differences together is one of the most
common errors in statistics.
HO: The sky is not falling.
HA: The sky is falling.
Practical vs. Statistical
Practical Difference: The difference resulting in an improvement of

practical or economic value to the company.
– Example, an improvement in yield from 96 to 99 percent.
Statistical Difference: A difference or change to the process that

probably (with some defined degree of confidence) did not happen
by chance.
– Examples might include differences in suppliers, markets or servers.
We will see it is possible to realize a statistically

significant difference without realizing a
practically significant difference.
Detecting Significance
During the Measure Phase

Mean Shift
it is important the nature of
the problem be well understood.
In understanding the problem

the practical difference to be
achieved must match the
statistical difference.
Variation Reduction
The difference can be either a
change in the Mean or in the
variance.
Detection of a difference is then

accomplished using statistical
Hypothesis Testing.
Hypothesis Testing
A Hypothesis Test is an a priori theory relating to differences between

variables.
A statistical test or Hypothesis Test is performed to prove or disprove

the theory.
A Hypothesis Test converts the practical problem into a statistical

problem.
– Since relatively small sample sizes are used to estimate
population parameters there is always a chance of collecting a
non-representative sample.
– Inferential statistics allows us to estimate the probability of getting
a non-representative sample.
Don’t worry, it’s just a

Hypothesis Test!
DIE Example
We could throw a die a number of times and track how many times each face
occurred. With a standard die we would expect each face to occur 1/6 or
16.67% of the time.
If we threw the die 5 times and got 5 ones what would you conclude? How
sure can you be?
– Pr (1 one) = 0.1667 Pr (5 ones) = (0.1667)5 = 0.00013
There are approximately 1.3 chances out of 10,000 we could have gotten 5
ones with a standard die.
Therefore we would say we are willing to take a 0.013% chance of being

wrong about our hypothesis that the die was “loaded” since the results do not
come close to our predicted outcome.
Hypothesis Testing
DECISIONS
β n
Statistical Hypotheses
We have two alternatives for hypothesis:
– The “null hypothesis” Ho assumes there are no differences or

relationships. This is the default assumption of all statistical
tests.
– The “alternative hypothesis” Ha states there is a difference or

relationship.
P-value > 0.05 Ho = no difference or relationship

P-value < 0.05 Ha = is a difference or relationship
Making a decision does not FIX a

problem, taking action does.
Steps to Statistical Hypothesis Test
1. State the Practical Problem.

2. State the Statistical Problem.
a) HO: ___ = ___
b) HA: ___ ≠ ,>,< ___
3. Select the appropriate statistical test and risk levels.
a) α = .05
b) β = .10
4. Establish the sample size required to detect the difference.
5. State the Statistical Solution.
6. State the Practical Solution.
Noooot THAT practical

solution!
How Likely is Unlikely?
Any differences between observed data and claims made under H0

may be real or due to chance.
Hypothesis Tests determine the probabilities of these differences

occurring solely due to chance and call them P-values.
The alpha level of a test (level of significance) represents the

yardstick against which P-values are measured and H0 is rejected if
the P-value is less than the alpha level.
The most commonly used levels are 5%, 10% and 1%.
Hardly any P-value at

all!!!
Hypothesis Testing Risk
The alpha risk or Type 1 Error (generally called the “Producer’s

Risk”) is the probability we could be wrong in saying something is
“different.” It is an assessment of the likelihood the observed
difference could have occurred by random chance. Alpha is the
primary decision-making tool of most statistical tests.
Actual Conditions
Not Different Different
(Ho is True) (Ho is False)
Not Different Correct Type II

(Fail to Reject Ho)
Decision Error
Statistical
Conclusions
Different
Type 1 Correct
(Reject Ho) Error Decision
Alpha Risk
Alpha () risks are expressed relative to a reference distribution.

Distributions include:
– t-distribution
The a-level is represented
– z-distribution by the clouded areas.
– 2- distribution Sample results in this area

lead to rejection of H0.
– F-distribution
Region of Region of
DOUBT DOUBT
Accept as chance differences
Hypothesis Testing Risk
The beta risk or Type 2 Error (also called the “Consumer’s Risk”) is
the probability we could be wrong in saying two or more things are
the same when, in fact, they are different.
Actual Conditions
Not Different Different
(Ho is True) (Ho is False)
Not Different Correct Type II

(Fail to Reject Ho)
Decision Error
Statistical
Conclusions
Different
Type 1 Correct
(Reject Ho) Error Decision
Beta Risk
Beta Risk is the probability of failing to reject the null hypothesis

when a difference exists.
Distribution if H0 is true
Reject H0
 = Pr(Type I error)
 = 0.05
H0 value
Accept H0 Distribution if Ha is true

= Pr(Type II error)
Critical value of test

statistic
Distinguishing between Two Samples
Recall from the Central Limit Theoretical Distribution

Theorem as the number of of Means
individual observations d When n = 2
d=5
increase the Standard Error S=1
decreases.
In this example when n = 2 we

cannot distinguish the
difference between the Means
(> 5% overlap, P-value > 0.05).
Theoretical Distribution
of Means
When n = 30 we can When n = 30
d=5
distinguish between the Means S=1
(< 5% overlap, P-value < 0.05)
There is a significant
difference.
Delta Sigma—The Ratio between d and S
Delta (d) is the size of the

difference between two Means or
one Mean and a target value.
Large Delta
Sigma (S) is the sample Standard d
Deviation of the distribution of
individuals of one or both of the
samples under question.
When d & S is large we do not
need statistics because the
differences are so large.
If the variance of the data is large it
is difficult to establish differences.
We need larger sample sizes to
Large S
reduce uncertainty.
We want to be 95% confident in all of our estimates!
Typical Questions on Sampling
Question: “How many samples should we take?”

Answer: “Well, that depends on the size of your delta and
Standard Deviation”.
Question: “How should we conduct the sampling?”
Answer: “Well, that depends on what you want to know”.
Question: “Was the sample we took large enough?”
Answer: “Well, that depends on the size of your delta and

Standard Deviation”.
Question: “Should we take some more samples just to be sure?”
Answer: “No, not if you took the correct number of samples the
first time!”
The Perfect Sample Size
The minimum sample

size required to
provide exactly 5%
overlap (risk) in order
to distinguish the
Delta.
Note: If you are 40 50 60 70
working with Non- Population

normal Data multiply
your calculated
sample size by 1.1
40 50 60 70
Hypothesis Testing Roadmap
Normal
Test of Equal Variance 1 Sample Variance 1 Sample t-test
Variance Equal Variance Not Equal
Two Samples Two Samples
2 Sample T One Way ANOVA 2 Sample T One Way ANOVA
Non Normal
Test of Equal Variance Median Test
Mann-Whitney Several Median Tests
Attribute Data
One Factor Two Factors

Two or More
One Sample Two Samples
Samples
One Sample Two Sample Chi Square Test

Proportion Proportion (Contingency Table)
Minitab:
Stat - Basic Stats - 2 Proportions Minitab:
If P-value < 0.05 the proportions Stat - Tables - Chi-Square Test
are different If P-value < 0.05 at least one
proportion is different
Chi Square Test

(Contingency Table)
Minitab:
Stat - Tables - Chi-Square Test
If P-value < 0.05 the factors are not
independent
Common Pitfalls to Avoid
While using Hypothesis Testing the following facts should be borne in

mind at the conclusion stage:
– The decision is about Ho and NOT Ha.
– The conclusion statement is whether the contention of Ha was upheld.
– The null hypothesis (Ho) is on trial.
– When a decision has been made:
• Nothing has been proved.
• It is just a decision.
• All decisions can lead to errors (Types I and II).
– If the decision is to “Reject Ho” then the conclusion should read “There
is sufficient evidence at the α level of significance to show that “state
the alternative hypothesis Ha.”
– If the decision is to “Fail to Reject Ho” then the conclusion should read
“There is not sufficient evidence at the α level of significance to show
that “state the alternative hypothesis.”
Summary
At this point you should be able to:

• Articulate the purpose of Hypothesis Testing
• Explain the concepts of the Central Tendency
• Be familiar with the types of Hypothesis Tests
IASSC Certified Lean Six Sigma Green Belt (ICGB)
The International Association for Six Sigma Certification (IASSC) is a Professional

Association dedicated to growing and enhancing the standards within the Lean Six Sigma
Community. IASSC is the only independent third-party certification body within the Lean Six
Sigma Industry that does not provide training, mentoring and coaching or consulting
services. IASSC exclusively facilitates and delivers centralized universal Lean Six Sigma
Certification Standards testing and organizational Accreditations.
The IASSC Certified Lean Six Sigma Green

Belt (ICGB) is an internationally recognized
professional who is well versed in the Lean
Six Sigma Methodology who both leads or
supports improvement projects. The Certified
Green Belt Exam, is a 3 hour 100 question
proctored exam.
Learn about IASSC Certifications and Exam options at…

http://www.iassc.org/six-sigma-certification/
LSS Green Belt v11.1 MT - Analyze Phase © Open Source Six Sigma, LLC

4 - Analyze - Intro To Hypothesis Testing

Uploaded by

Copyright:

Available Formats

4 - Analyze - Intro To Hypothesis Testing

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

4 - Analyze - Intro To Hypothesis Testing

Uploaded by

Copyright:

Available Formats

Analyze Phase

Introduction to Hypothesis Testing

Inferential Statistics Hypothesis Testing Purpose

Tests for Central Tendency

Hypothesis Testing NND P1

Hypothesis Testing NND P2

Wrap Up & Action Items

Process Capability of Process Before

Process Capability of Process After

make fact-based decisions about O v erall C apability

expected sample variation. O bserv ed P erformance

The purpose of appropriate Hypothesis Testing is to integrate the Voice of the

The probability of occurrence is based on a pre-determined statistical

Decisions are based on:

Recall from the discussion on classes and cause of distributions that a

Hypothesis Testing can help establish a statistical difference between

Did my sample come from this population? Or this? Or this?

Are the two distributions “significantly” different from each

How sure are we of our decision?

How do the number of observations affect our confidence in

Statistics provide a methodology to detect differences.

HO: The sky is not falling.

HA: The sky is falling.

Practical Difference: The difference resulting in an improvement of

Statistical Difference: A difference or change to the process that

We will see it is possible to realize a statistically

During the Measure Phase

In understanding the problem

Detection of a difference is then

A Hypothesis Test is an a priori theory relating to differences between

A statistical test or Hypothesis Test is performed to prove or disprove

A Hypothesis Test converts the practical problem into a statistical

Don’t worry, it’s just a

Therefore we would say we are willing to take a 0.013% chance of being

We have two alternatives for hypothesis:

– The “null hypothesis” Ho assumes there are no differences or

– The “alternative hypothesis” Ha states there is a difference or

P-value > 0.05 Ho = no difference or relationship

Making a decision does not FIX a

1. State the Practical Problem.

Noooot THAT practical

Any differences between observed data and claims made under H0

Hypothesis Tests determine the probabilities of these differences

The alpha level of a test (level of significance) represents the

Hardly any P-value at

The alpha risk or Type 1 Error (generally called the “Producer’s

Not Different Correct Type II

Alpha () risks are expressed relative to a reference distribution.

– 2- distribution Sample results in this area

Accept as chance differences

Not Different Correct Type II

Beta Risk is the probability of failing to reject the null hypothesis

Accept H0 Distribution if Ha is true

Critical value of test

Recall from the Central Limit Theoretical Distribution

In this example when n = 2 we

Delta (d) is the size of the

We want to be 95% confident in all of our estimates!

Question: “How many samples should we take?”