CSSGB

End of Section Knowledge Checks

Section IV – Analyze Phase
CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

A. Exploratory data analysis
1. Multi-vari studies
Select appropriate sampling plans to create multi-vari study charts and interpret the
results for positional, cyclical, and temporal variation. (Create)

Q1: New Question

During a process study, a green belt is assigned to collect data at 5 measurement
points within a sample. Five samples are picked each day for five consecutive days.
Choose an appropriate graphical method from the choices below to display within
sample, sample to sample and time variation in a single graph.

A. Correct Answer- Multivari chart

B. Distractor 1- Xbar R Chart
C. Distractor 2- Box Plot
D. Distractor 3- Pareto Chart

Feedback:
A. Correct – Multivari chart allows the data to present positional (within
sample), cyclical (sample to sample) and temporal (over time)
B. Incorrect – Xbar R chart can present variations from sample to sample and
variation with time. Within sample variation is not presented.
C. Incorrect – Only sample to sample variation is presented in a single box
plot.
D. Incorrect – A Pareto chart is used to present the frequency of causes that
impact variation. Measurement data is not presented.
Reference:
Munro, Roderick, et al. The Certified Six Sigma
Green Belt Handbook, ASQ Quality Press, 2008.
Quality Tool Box, Second editionReference page #: Chapter 16: A. Exploratory
Data Analysis 225-226

Quality Tool B ox, Pages 356-359

Reviewer #1 Comments: Answer A is correct as Multivari charts are a tool for

analyzing cyclical, temporal, and positional variation.
I feel the question could be better worded. Also the last sentence is not a complete
sentence. For example an improvement would be:
During a process study, a green belt is assigned to collect data at 5 measurement
points within a sample. Five samples are picked each day for five consecutive days .
Choose an appropriate graphical method to display within sample, sample to
sample and time variation.
In the initial question, the last sentence is not a complete sentence. Writer: (Edited)

ASQ Learning Offerings 2 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

A. Exploratory data analysis
1. Multi-vari studies
Select appropriate sampling plans to create multi-vari study charts and interpret the
results for positional, cyclical, and temporal variation. (Create)

Q2: New Question

As a process improvement professional, you are asked to collect data to conduct a
multivari study for a process parameter. How do you plan your study?
A. Correct Answer: Collect data on variations from within part, part to part, over time
B. Distractor 1: Collect data from various manufacturing locations
C. Distractor 2: Collect data from multiple machines
D. Distractor 3: Collect data from multiple processes

Feedback:
A. Correct – To conduct a multivari study, one should first collect data from
within part (positional), part to part (Cyclical), and over time (temporal).
B. Incorrect – This will include many sources of variation and will not be useful
to understand the variation from a given process.
C. Incorrect – This will include many sources of variation and will not be useful
to understand the process variation. Machine to machine variation will add
to the complexity as an additional source.
D. Incorrect – This will include many sources of variation and will not be useful
to understand the process variation. Adding variations from other processes
will not be helpful to the study.
Reference:
Munro, Roderick, et al. The Certified Six Sigma
Green Belt Handbook, ASQ Quality Press, 2008.
Quality Tool Box, second edition

Reference page #: Pge 226- Procedure for Multi-Vari Sampling Plan

Quality Tool Box, Pages 356-359

Reviewer #1 Comments: I agree that A is the correct answer to this question. I do

not have any other comments.

ASQ Learning Offerings 3 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

A. Exploratory data analysis
2. Correlation and linear regression
Describe the difference between correlation and causation. Calculate the correlation
coefficient and linear regression and interpret the results in terms of statistical
significance (p-value). Use regression models for estimation and prediction. (Evaluate)

Q3: New Question

As an experimenter which of the following tools would you use to verify correlation
and causation of an experimental data?
A. Correct Answer: Scatter diagram and Ishikawa diagram
B. Distractor 1: Control chart and histogram
C. Distractor 2: Pareto Chart and Ishikawa diagram
D. Distractor 3: Scatter diagram and histogram

Feedback:
A. Correct – A scatter diagram provides a pictorial relationship between an
independent variable and a dependent variable. An Ishikawa diagram
provides cause and effect relationship. The combination of the two tools is
key to understanding the relationship and degree of relationship.
B. Incorrect – A control chart is useful for understanding process behavior and
a histogram is useful for understanding the spread and direction of shift in
the process. These two tools help understand process variations.
C. Incorrect – The Pareto chart is useful to prioritize causes of failures and the
Ishikawa diagram provides cause and effect relationships. The combination
of these tools helps to understand causes and prioritize improvement
efforts.
D. Incorrect – A scatter diagram provides a pictorial relationship between an
independent variable and a dependent variable. A histogram is useful for
understanding the spread and direction of shift in the process. The
combination of these tools would help to understand the relationship
between variations and the variation of each variable.
Reference:
Reference used Quality Tool Box second edition
Reference page #: Quality Tool Box Pages: 471, 247, 155, 292, 376 (tools listed
in the problem)
Reviewer #1 Comments: I do not like how the question is worded but I agree that
the answer should be A. It seems like the question should say “…verify correlation
and causation, respectively as you are not looking for one tool to verify both but
rather two tools, one for correlation and one for causation.
Writer: (Question rephrased)

ASQ Learning Offerings 4 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

A. Exploratory data analysis
2. Correlation and linear regression
Describe the difference between correlation and causation. Calculate the correlation
coefficient and linear regression and interpret the results in terms of statistical
significance (p-value). Use regression models for estimation and prediction. (Evaluate)

Q4: New Question

A study is conducted which compares the amount of time children spend watching
television to their scores on a reading proficiency test. With the following results,
what is the correlation coefficient and what is the maximum amount of television a
child could watch and still expect to achieve a score of seventy percent on the
reading proficiency test?

TV Time In minutes Vs Test Score%

30 80
45 85
60 75
120 60
240 50

A. Correct Answer -0.943 and 98.97 minutes

B. Distractor 1: 0.89 and 98.97 minutes
C. Distractor 2: No correlation, Value not predictable
D. Distractor 3: -0.943 and 0.89 minutes

Feedback:
A. Correct –

Correlation coefficient can be from -1 to +1.

Create a table similar to 16.7 of Green belt handbook Page 240 to calculate

ASQ Learning Offerings 5 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

correlation coefficient.
Formula for regression equation:

Applying the above correlation coefficient equation and regression

calculations Pearson correlation of TV Time and Test Score = -0.943
TV Time = 488.1 - 5.559 Test Score

488.1 - 5.559 X 70= 98.97 minutes (See also the plot of regression line with dotted
line from 70% X axis corresponding to 98.97 minutes in the Y axis

Analysis of Variance

Source DF SS MS F P
Regression 1 26265.4 26265.4 24.21 0.016
Error 3 3254.6 1084.9
Total 4 29520.0

Fitted Line Plot

TV Time = 488.1 - 5.559 Test Score
250
S 32.9371
R-Sq 89.0%
R-Sq(adj) 85.3%
200

150
TV Time

100

50

0
50 60 70 80 90
Test Score

A.
2
B. Incorrect – 0.89 is the Coefficient of determination R . The following
equations are needed for this scenario:

ASQ Learning Offerings 6 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

Formula for regression equation:

C. Incorrect –The data sets have high correlation; therefore, the value is
predictable. The following equations are needed for this scenario:

Formula for regression equation:

D. Incorrect – Correlation coefficient and predicted time to achieve 70% score

ASQ Learning Offerings 7 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

required. The following equations are needed for this scenario:

Formula for regression equation:

Reference: Munro, Roderick, et al. The Certified Six Sigma

Green Belt Handbook, ASQ Quality Press, 2008
Reference used
Reference page #: 233-240
Reviewer #1 Comments: I am not able to determine where the calculation for the
70% test score and the TV time of 98.97 minutes. I agree with the number for the
regression line.
Writer comments: Good Catch. This was missing in the Answer Key. I have added
how the 98.97 minutes is arrived.

ASQ Learning Offerings 8 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

A. Exploratory data analysis
2. Correlation and linear regression
Describe the difference between correlation and causation. Calculate the correlation
coefficient and linear regression and interpret the results in terms of statistical
significance (p-value). Use regression models for estimation and prediction. (Evaluate)

Q5: New Question

The P value for a regression model is 0.07. If the cut off for the Alpha risk is 0.05
and the assumed Beta risk is 0.10, the regression model is
A. Correct Answer: Not statistically significant since the P value is greater than Alpha
Risk of 0.05.
B. Distractor 1: Statistically significant since the P value is smaller than Beta Risk of
0.10.
C. Distractor 2: Statistically significant since the P value is greater than Alpha Risk of
0.05.
D. Distractor 3: Not Statistically significant since the P value is smaller than Beta Risk
of 0.10.

Feedback:
A. Correct – The p-value (0.07) is compared to Alpha Risk (0.05) and if the P
value is smaller than the Alpha risk, then is considered statistically
significant. In our example, since P value is greater than assumed alpha
value, the regression model is not statistically significant.
B. Incorrect – The p-value is compared to Alpha Risk (Not Beta Risk) and if P
value is smaller than the Alpha risk, it is considered statistically significant.
C. Incorrect – The p-value is compared to Alpha Risk and if P value is smaller
than the Alpha risk, it is considered statistically significant.
D. Incorrect – The p-value is compared to Alpha Risk (Not Beta Risk) and if P
value is smaller than the Alpha risk, it is considered statistically significant. .

Reference:

Reference used Munro, Roderick, et al. The Certified Six Sigma

Green Belt Handbook, ASQ Quality Press, 2008

Reference page #: 246

Reviewer #1 Comments: I agree that in concept the answer A is correct. However,
the answer is not stated properly. Since the p-value for the variable is a value which
is greater the alpha value we must conclude that a relationship between variables
does not exist. I do not agree with the significance conclusion.
Writer comments:
Nothing wrong with my earlier statement. However I expanded the statement in A
and the Feedback statements for better clarity.

ASQ Learning Offerings 9 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

ASQ Learning Offerings 10 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

B. Hypothesis testing
1. Basics
Distinguish between statistical and practical significance. Determine appropriate sample
sizes and develop tests for significance level, power, and type I and type II errors. (Apply)

Q6: New Question

A one-sample T study was performed on a supplier‘s lot:
T study μ = 10 vs ≠ 10 and Alpha Value=0.05
The mean of 10 samples tested is 9.85.
The 95% confidence interval of the population mean is 9.75, 9.95.
P-Value = 0.046.
A discussion with the process expert determined that the population mean can vary
from 9.7 to 10.1 without impacting the product performance. Choose the best
course of action:

A. Correct Answer: Reject the null hypothesis since it is statistically significant and
accept the lot since it is not practically significant
B. Distractor 1: Reject the null hypothesis since it is statistically significant and reject
the lot irrespective of practical significance.
C. Distractor 2: Fail to reject the null hypothesis since it is not statistically significant
and accept the lot since it is not practically significant.
D. Distractor 3: Fail to reject the null hypothesis since it is not statistically significant
and reject the lot irrespective of practical significance.
Feedback:
A. Correct – Since the P value is lower than the Alpha value, the null
hypothesis of μ = 10 vs ≠ 10 is rejected. However, due to the practical
significance that there is no impact to the product, the supplier’s lot is
accepted. NOTE: If the analyst had consulted with the engineer to
understand practical significance beforehand, one tail hypothesis could
have been performed for ≥ 9.7 and ≤10.1
B. Incorrect – The practical significance should be determined before deciding
to accept or reject the lot.
C. Incorrect – This hypothesis conclusion incorrect. Statistical conclusion is to
reject hypothesis.
D. Incorrect – Hypothesis conclusion incorrect. Practical significance ignored.

Reference:

Reference used Munro, Roderick, et al. The Certified Six Sigma

Green Belt Handbook, ASQ Quality Press, 2008
Reference page #: P 248-250
Reviewer #1 Comments: I agree that A is the correct answer.

ASQ Learning Offerings 11 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

B. Hypothesis testing
1. Basics
Distinguish between statistical and practical significance. Determine appropriate sample
sizes and develop tests for significance level, power, and type I and type II errors. (Apply)

Q7: New Question

At the 90 percent confidence level, what is the minimum sample size that would
confirm the significance of a mean shift greater than 10 units per hour without
considering the beta error? (Historically, the standard deviation of the hourly output
is 25 units.)

A. Correct Answer: 17
B. Distractor 1: 24
C. Distractor 2: 1
D. Distractor 3: Not enough information to calculate sample size

Feedback:
A. Correct – Why is this answer correct?

2 2 2
n= (1.645) * (25) / 10 = 2.56 * 625/ 100= 16.91 approx 17.
B. Incorrect – The confidence interval is 90%

C. Incorrect – Standard deviation and mean shift are not correctly substituted.

D. Incorrect –Without considering Beta error, the necessary information is

available.

Reference:

Reference used Munro, Roderick, et al. The Certified Six Sigma

Green Belt Handbook, ASQ Quality Press, 2008

ASQ Learning Offerings 12 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

Reference page #: 252

Reviewer #1 Comments: I agree that answer A is correct and have no comment on
the question.

ASQ Learning Offerings 13 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

B. Hypothesis testing
1. Basics
Distinguish between statistical and practical significance. Determine appropriate sample
sizes and develop tests for significance level, power, and type I and type II errors. (Apply)

Q8: New Question

In the curve below, at an alpha risk at 10%, the Power of detecting a difference of
10 units is 0.5. Determine the beta risk for detecting a difference of 15 units at 10%
alpha risk.

A. Correct Answer: 0.25

B. Distractor 1: 0.75
C. Distractor 2: 90%
D. Distractor 3: 10%

Feedback:
A. Correct –On the graph, the difference of 15 intersects with 0.75 Power.
Beta is 1-Power which is 0.25.
B. Incorrect – 0.75 is the power. The question is looking for Beta risk.
C. Incorrect – 90% is the confidence level. The question is looking for Beta
risk
D. Incorrect –10% is the alpha risk. The question is looking for Beta risk

Reference:
Reference used Munro, Roderick, et al. The Certified Six Sigma

ASQ Learning Offerings 14 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

Green Belt Handbook, ASQ Quality Press, 2008

Reference page #: Page 249 Table 17.1

Reviewer #1 Comments: I agree that answer A is correct. No other issues with this
question.

ASQ Learning Offerings 15 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

B. Hypothesis testing
2. Tests for means, variances, and proportions
Conduct hypothesis tests to compare means, variances, and proportions (e.g., paired-
comparison t-test, F-test, analysis of variance (ANOVA), chi square) and interpret the results.
(Analyze)

Q9: New Question

A process improvement team is comparing the mean differences between two
processes.

Process A Process B
10 11
14 12
15 16
16 14
11 11

Assuming equal variances, analyze the data sets to verify if the null hypothesis
Process Mean A= Process Mean B is true at alpha risk of 0.05

A. Correct Answer: No difference between the means of Process A and B at Alpha

risk 5%.

B. Distractor 1: Process Means are different at Alpha risk 5%.

C. Distractor 2: Process means are different at significance level of 95%

D. Distractor 3: Insufficient information to test the hypothesis.

Feedback:
A. Correct – Use the following formula to calculate the test statistic and obtain the
critical value from the table.

ASQ Learning Offerings 16 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

MINITAB analysis:

Two-Sample T-Test and CI: Process A, Process B

Two-sample T for Process A vs Process B

N Mean StDev SE Mean

Process A 5 13.20 2.59 1.2
Process B 5 12.80 2.17 0.97

Difference = μ (Process A) - μ (Process B)

Estimate for difference: 0.40
95% CI for difference: (-3.08, 3.88)
T-Test of difference = 0 (vs ≠): T-Value = 0.26 P-Value = 0.798
DF = 8
Both use Pooled StDev = 2.3875

B. Incorrect – Use the formula for Two Sample T. Calculate Pooled standard
deviation using the formula.

C. Incorrect – Use the formula for Two Sample T. Calculate Pooled standard
deviation using the formula. Significance level 95% is Alpha of 5%
D. Incorrect – Adequate information is available to calculate test statistic.

Reference:
Reference used Munro, Roderick, et al. The Certified Six Sigma
Green Belt Handbook, ASQ Quality Press, 2008
Reference page #: 260
Reviewer #1 Comments: The question indicates that variances are equal which is
not true. The variance for group a is 6.71 while the variance for group b is 4.71.
Based on the calculated and critical t values of 0.26 and 2.306, respectively, the null
hypothesis is not rejected. I do not see any information in the CSSGB handbook
that calculates the confidence intervals for t values. (removed)
Writer comments:
Here is why it is equal Variances:
Test and CI for Two Variances: Process A, Process B

ASQ Learning Offerings 17 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

Method

Null hypothesis σ(Process A) / σ(Process B) = 1

Alternative hypothesis σ(Process A) / σ(Process B) ≠ 1
Significance level α = 0.05

Statistics

95% CI for
Variable N StDev Variance StDevs
Process A 5 2.588 6.700 (1.485, 7.422)
Process B 5 2.168 4.700 (1.066, 7.253)

Ratio of standard deviations = 1.194

Ratio of variances = 1.426

(F Table 0.05, Df1 =4 and DF2=4 is 6.39) Hand book Page 378

The calculated value is not greater than critical value.

Advanced MINITAB analysis:

95% Confidence Intervals

CI for StDev CI for Variance

Method Ratio Ratio
Bonett (0.478, 3.638) (0.229, 13.239)
Levene (0.210, 55.990) (0.044, 3134.860)

Tests

Test
Method DF1 DF2 Statistic P-Value
Bonett 1 — 0.21 0.645
Levene 1 8 0.17 0.694

Confidence Interval (CI) is going away in the revised BOK. So I am rewriting

the question with CI.

ASQ Learning Offerings 18 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

IV. Analyze Phase

B. Hypothesis testing
2. Tests for means, variances, and proportions
Conduct hypothesis tests to compare means, variances, and proportions (e.g., paired-
comparison t-test, F-test, analysis of variance (ANOVA), chi square) and interpret the results.
(Analyze)

Q10: New Question

A call center is studying the effect of a retraining initiative. Specifically, they want to
know if there has been an improvement in variance of the Average Handling Time
(AHT). Assuming a 95% confidence interval, what conclusion can be made on the
effectiveness of the retraining?
Before After retraining
Number of data points 10 10
Variance 100 64
A. Correct Answer: The effect of retraining has had no improvement.
B. Distractor 1: Reduced variation indicates that the retraining has been effective.
C. Distractor 2: Increased variation indicated that the retraining has had a detrimental
effect.
D. Distractor 3: There is not enough information to make a conclusion.

Feedback:
A. Correct – The Critical Value for Df1 and Df2 9 is 3.18, the test statistics is
1.56. There is no effect from retraining.
 Ratio of variances = 100/64 = 1.563
 Critical value is from the F Distribution table: 3.18 with Df1=9,
DF2=9
 Null hypothesis: σ(First) / σ(Second) = 1
 Alternative hypothesis: σ(First) / σ(Second) > 1

B. Incorrect – The test statistics are not greater than critical value; therefore,
we can only conclude that reduction of variation cannot be concluded.
C. Incorrect – Null hypothesis is a two tail test. We cannot make this
conclusion.
D. Incorrect – There is adequate information to calculate test statistics to
compare with the critical value.
Reference:

Reference used Munro, Roderick, et al. The Certified Six Sigma

Green Belt Handbook, ASQ Quality Press, 2008
Reference page #: 264
Reviewer #1 Comments: I agree that the correct answer is A based on the F test
statistic.

ASQ Learning Offerings 19 of 20 10/06/14

CSSGB Knowledge Check Questions IV. Analyze Phase

ASQ Learning Offerings 20 of 20 10/06/14