Lean Six Sigma Master Black Belt Practice Questions with Explanations

Free Lean Six Sigma Master Black Belt practice questions. 123 of them, each with the correct answer, a full explanation, and the reason every other option is wrong.

Master Black Belt practice questions

Question 1

You are about to run a t-test on shield thickness from 2 suppliers when you determine the data from one group is not normally distributed and cannot be transformed.

Your next step would be to?

  • A. Use the Shapiro-Wilk test
  • B. Proceed with the t-test
  • C. Use a non - parametric test
  • D. Discontinue the analysis
Show answer and explanation ▾

Correct answer: C

When data from one group is not normal and cannot be transformed, the assumptions of the t-test no longer hold. The usual next step is to switch to a non-parametric test that does not require normality.

Why the other options are wrong:

  • A. The Shapiro-Wilk test checks for normality, which has already been established here.
  • B. Proceeding with the t-test would ignore the broken normality assumption.
  • D. Stopping the analysis is unnecessary because a non-parametric alternative exists.

Question 2

A manufacturing test process has 3 parallel machines performing exactly the same test. The data from this test process can be assumed to be normally distributed and the variances within each machine are the same. To understand if there is a significant statistical difference in the average test value between machines, what test should be used?

  • A. Kruskal - Wallis
  • B. Chi-Square
  • C. ANOVA
  • D. Bartlett or Levene
Show answer and explanation ▾

Correct answer: C

Comparing average values across three machines, with normal data and equal variances, is the case for one-way ANOVA. It tests whether at least one machine's mean differs.

Why the other options are wrong:

  • A. Kruskal-Wallis is the non-parametric option for data that is not normal.
  • B. Chi-square tests association between categories, not differences in averages.
  • D. Bartlett and Levene test whether variances are equal, which is already assumed.

Question 3

Which of the following is a commonly used test that examines the association between multiple discrete variables?

  • A. Kruskal-Wallace Test
  • B. Shapiro-Wilkes Test
  • C. Student's t-Test
  • D. Chi-Square Test
Show answer and explanation ▾

Correct answer: D

The chi-square test examines whether two or more categorical variables are associated, using counts in each category.

Why the other options are wrong:

  • A. Kruskal-Wallis compares groups on a continuous measure without assuming normality.
  • B. Shapiro-Wilk tests whether data is normally distributed.
  • C. The Student's t-test compares two means of continuous data.

Question 4

Which of the following statistical procedures is appropriate when there is one continuous input variable variable (X) and one continuous output variable (Y)?

  • A. T-test
  • B. Chi-Square test
  • C. One-Way ANOVA
  • D. Correlation
Show answer and explanation ▾

Correct answer: D

Correlation measures the strength and direction of the linear relationship between two continuous variables, so it fits one continuous X and one continuous Y.

Why the other options are wrong:

  • A. A t-test compares means and needs a discrete X that defines one or two groups.
  • B. A Chi-Square test is used when both the X and the Y are discrete.
  • C. One-Way ANOVA compares means across the levels of a discrete X.

Question 5

Sigma Saving and Loans processes loans and leases from around the world. The CEO wants to know if the current cycle time for processing is less than 9.5 days on average. To test the claim that the average cycle time is less than 9.5 days, use:

  • A. A 1-sample T-test
  • B. A 2 sample T-test
  • C. A One-way ANOVA
  • D. A Chi-square test of means
Show answer and explanation ▾

Correct answer: A

A 1-sample t-test compares the mean of a single sample with a fixed target value, here 9.5 days, when the population standard deviation is not known.

Why the other options are wrong:

  • B. A 2-sample t-test compares the means of two groups, and there is only one group here.
  • C. One-way ANOVA compares the means of several groups, not one mean against a target.
  • D. A Chi-square test works on counts of discrete data and is not a test of means.

Question 6

Two different samples were pulled randomly from the same population. One sample is size n=10 and the other is size n=100. A two-sided confidence interval for the mean was calculated separately for each.

How will the intervals compare?

  • A. The confidence interval for size n=10 will be smaller.
  • B. The confidence interval for size n=10 will be larger.
  • C. The confidence intervals will be the same for both n=10 and n=100.
  • D. There is not enough information given.
Show answer and explanation ▾

Correct answer: B

The width of a confidence interval for the mean depends on the standard error, the standard deviation divided by the square root of n. A smaller sample gives a larger standard error, so the n=10 interval is wider.

Why the other options are wrong:

  • A. A smaller sample gives less precision, which widens the interval.
  • C. Sample size directly changes the standard error, so the two intervals differ.
  • D. Both samples come from the same population, so sample size alone settles the comparison.

Question 7

The purpose of a screening experiment using DOE is to?

  • A. Optimize the response by determining the best levels for the input factors
  • B. Separating the "vital few from the trivial many"
  • C. Comparing various levels for one factor
  • D. Finding a set of levels for the inputs that produce a robust product
Show answer and explanation ▾

Correct answer: B

A screening experiment tests many factors in few runs to find the small number that really drive the response, separating the vital few from the trivial many before deeper study.

Why the other options are wrong:

  • A. Finding the best factor levels is the goal of an optimization experiment that follows screening.
  • C. Comparing levels of one factor is a comparative experiment.
  • D. Finding settings that resist noise is the goal of a robust design experiment.

Question 8

What is the correct order of steps in an experiment (DOE)?

  • Design the experiment & plan data collection
  • Run the experiment and collect data
  • State the problem or objective
  • Analyze the results
  • Interpret the results`
  • A. 1, 3, 4, 2, 5
  • B. 3, 1, 2, 4, 5
  • C. 3, 1, 4, 2, 5
  • D. 3, 1, 2, 5, 4
Show answer and explanation ▾

Correct answer: B

An experiment starts by stating the problem or objective, then the design and data collection plan are made, the runs are carried out, the results are analyzed and finally interpreted: 3, 1, 2, 4, 5.

Why the other options are wrong:

  • A. The experiment cannot be designed before the problem or objective is stated.
  • C. Results cannot be analyzed before the experiment has been run.
  • D. Results are analyzed first and interpreted afterwards.

Question 9

The four basic objectives for experiments, which are a vital part of Lean Six Sigma, include Screening, Optimization, and which of the following?

  • A. Result, Comparison
  • B. Comparison, Robust Design
  • C. Reduced Variance, Parsimony
  • D. Comparison, Result
Show answer and explanation ▾

Correct answer: B

The four basic objectives of designed experiments are Comparison, Screening, Optimization and Robust Design.

Why the other options are wrong:

  • A. Result is not one of the four experimental objectives.
  • C. Reduced Variance and Parsimony are not the names of the remaining two objectives.
  • D. Result is not one of the four experimental objectives.

Question 10

Experimental error includes the ____________ in the experiment caused by uncontrolled and unknown nuisance factors. This is also called____________.

  • A. Noise/pure error
  • B. Noise/signal
  • C. Signal/noise
  • D. Pure error/signal
Show answer and explanation ▾

Correct answer: A

Experimental error is the noise in the response caused by uncontrolled and unknown nuisance factors. This unexplained variation is also called pure error.

Why the other options are wrong:

  • B. Signal is the effect of the factors being studied, not another name for noise.
  • C. Signal comes from the controlled factors, not from nuisance factors.
  • D. Signal is the opposite of error, so it cannot be another name for it.

Question 11

A Black Belt candidate has found that three process factors significantly affect process throughput and variability. He suspects that the influence of one of the factors is non-linear over the range of throughput. Which of the following tools should be used to determine the relationship between the factors and responses?

  • A. Screening factorial experiment
  • B. Response surface method (RSM) experiment
  • C. Multiple linear regression analysis
  • D. 2-Level Full Factorial experiment
Show answer and explanation ▾

Correct answer: B

A response surface experiment tests factors at more than two levels, so it can fit quadratic terms and model a non-linear relationship between a small number of significant factors and the responses.

Why the other options are wrong:

  • A. A screening experiment identifies important factors and cannot model curvature.
  • C. Multiple linear regression analyzes data that already exists and is not an experiment designed to estimate curvature.
  • D. A 2-level design has only two points per factor and cannot estimate curvature.

Question 12

A full factorial experiment is characterized by all of the following properties EXCEPT:

  • A. All factors in the experiment are controlled
  • B. Higher level interaction effects cannot be estimated
  • C. All combinations of the levels of the factors are run in the experiment
  • D. The 2-factor, 2-level design is the simplest full factorial experiment
Show answer and explanation ▾

Correct answer: B

A full factorial runs every combination of factor levels, so all main effects and all interactions, including the higher order ones, can be estimated. The statement that they cannot be estimated is the exception.

Why the other options are wrong:

  • A. Controlling all factors in the experiment is a property of a full factorial.
  • C. Running every combination of factor levels is the defining property of a full factorial.
  • D. Two factors at two levels, four runs, is the simplest full factorial.

Question 13

A Black Belt runs an experiment on 2 different shifts comparing two different methods for data entry. The Black Belt wants to learn if there is a difference between methods. She suspects that shifts have impact on the response. However, she is not interested in optimizing the shift.

What experimental design technique should the Black Belt use?

  • A. Repeats only
  • B. Replication and repeats
  • C. Run in standard order
  • D. Blocking
Show answer and explanation ▾

Correct answer: D

Blocking groups the runs by a known nuisance variable, here the shift, so its effect is separated from the comparison of the two methods without treating shift as a factor to optimize.

Why the other options are wrong:

  • A. Repeats give extra measurements within a run and do not remove the shift effect.
  • B. Replication and repeats estimate error but do not separate out the shift effect.
  • C. Standard order does nothing to account for a nuisance variable.

Question 14

A Black Belt is in the early stages of a problem solving effort. The Black Belt determines that the data gathered so far requires a DOE to gain additional insight and to understand which factors are important.

There are 8 factors and budget is limited.

What type of experiment is most appropriate?

  • A. Full factorial
  • B. Fractional factorial
  • C. Response surface methodology
  • D. One factor at a time
Show answer and explanation ▾

Correct answer: B

With 8 factors and a limited budget in the early stage of a study, a fractional factorial screens for the important factors using only a fraction of the runs a full factorial needs.

Why the other options are wrong:

  • A. A full factorial with 8 two-level factors needs 256 runs, which a limited budget does not allow.
  • C. Response surface methodology is for optimizing a few known important factors later on.
  • D. One factor at a time is inefficient and cannot detect interactions.

Question 15

Following is a confounding pattern of a DOE:

  • A =BCD
  • B =ACD
  • C =ABD
  • D =ABC
  • AB=CD
  • AC=BD
  • AD = BC

Based on this structure, what is the resolution of the design:

  • A. III
  • B. IV
  • C. V
  • D. None, it is a full factorial
Show answer and explanation ▾

Correct answer: B

Main effects are aliased with three-factor interactions and two-factor interactions are aliased with each other. That confounding pattern defines a Resolution IV design.

Why the other options are wrong:

  • A. In Resolution III, main effects are confounded with two-factor interactions.
  • C. In Resolution V, two-factor interactions are not confounded with each other.
  • D. A full factorial has no confounding at all.

Question 16

In a sequential experimentation strategy, RSM experiments are most likely to be conducted at which of the following phases?

  • A. As a follow-up experiment to earlier experiments to help find optimum settings or if curvature was detected
  • B. In the initial conjecture phase
  • C. After making confirmation runs, especially if an acceptable linear model with high R-squared has been obtained
  • D. As a first experiment before spending 25% of resources
Show answer and explanation ▾

Correct answer: A

Response surface experiments come late in a sequence, after screening and factorial work have found the important factors. They are used to locate optimum settings or to model curvature that earlier experiments detected.

Why the other options are wrong:

  • B. The initial phase is for screening many factors, not for response surface work.
  • C. A good linear model confirmed by confirmation runs gives no reason to add a response surface experiment.
  • D. A first experiment should screen factors with few runs, not fit a response surface.

Question 17

Which of the following best completes this statement? Factors are ___________ and responses are ____________.

  • A. Input variables, output variables
  • B. Output variables, input variables
  • C. Outputs, inputs
  • D. A and C only
Show answer and explanation ▾

Correct answer: A

In a designed experiment, factors are the input variables that are deliberately changed, and responses are the output variables that are measured.

Why the other options are wrong:

  • B. This reverses the two terms.
  • C. Factors are inputs and responses are outputs, so this is reversed.
  • D. Option C is reversed, so the combination does not hold.

Question 18

Which experimental design has main effects confounded with second order interactions?

  • A. Resolution IV (Resolution 4)
  • B. Resolution V (Resolution 5)
  • C. Resolution III (Resolution 3)
  • D. Resolution II (Resolution 2)
Show answer and explanation ▾

Correct answer: C

In a Resolution III design, main effects are not confounded with each other but are confounded with two-factor interactions.

Why the other options are wrong:

  • A. In Resolution IV, main effects are clear of two-factor interactions.
  • B. In Resolution V, main effects and two-factor interactions are clear of each other.
  • D. A Resolution II design would confound main effects with each other.

Question 19

What is the recommended run order for an experiment?

  • A. Standard order
  • B. Random order
  • C. Center points last
  • D. It doesn't matter
Show answer and explanation ▾

Correct answer: B

Running an experiment in random order spreads the influence of unknown and time-related nuisance variables across all the runs, so they do not bias the estimated effects.

Why the other options are wrong:

  • A. Standard order lets time-related changes line up with the factor settings.
  • C. Center points are spread through the experiment so that drift can be detected.
  • D. Run order matters because lurking variables can change during the experiment.

Question 20

In a DOE analysis, residual plots are examined to accomplish which of the following?

  • A. Make precise predictions
  • B. Improve model fit
  • C. Determine which effects to keep in the model
  • D. Validate model assumptions
Show answer and explanation ▾

Correct answer: D

Residual plots check the assumptions behind the model: residuals should be normally distributed, independent and have constant variance. If the plots show otherwise, the conclusions of the analysis are in doubt.

Why the other options are wrong:

  • A. Predictions come from the fitted model equation, not from the residual plots.
  • B. Residual plots diagnose the model and do not themselves improve the fit.
  • C. Effects are kept or removed on the basis of their p-values.

Question 21

A 3^2 experiment means that we are considering:

  • A. Two levels for three factors.
  • B. Two dependant variables and three independent variables.
  • C. Three levels of two factors.
  • D. Two continuous variables and three attribute variables.
Show answer and explanation ▾

Correct answer: C

In factorial notation the base is the number of levels and the exponent is the number of factors. A 3^2 experiment has two factors at three levels each, 9 treatment combinations.

Why the other options are wrong:

  • A. Two levels for three factors is written 2^3.
  • B. The notation gives levels and factors, not counts of dependent and independent variables.
  • D. The notation says nothing about the data type of the variables.

Question 22

Which of the following is FALSE regarding FMEA?

  • A. The focus is reaction
  • B. Failure may or may not have occurred yet
  • C. It is initiated prior to any design activity
  • D. It is an assessment of potential failure modes
Show answer and explanation ▾

Correct answer: A

FMEA is a preventive tool: it looks for potential failure modes and acts before they occur. Saying its focus is reaction describes the opposite approach, so that statement is false.

Why the other options are wrong:

  • B. FMEA covers potential failures, so the failure may or may not have happened yet.
  • C. FMEA is started early, ahead of the design work, so that the risks it finds can shape the design.
  • D. Assessing potential failure modes is exactly what an FMEA does.

Question 23

The definition of Severity is best described by the following statement:

  • A. It is the likelihood for a failure mode to occur
  • B. It is an assessment of how serious the effect will be to the customer if the failure mode occurs
  • C. It is an assessment of the process' ability to contain a problem and keep it from reaching the customer
  • D. It is an assessment only of the impact on the process or business
Show answer and explanation ▾

Correct answer: B

Severity rates how serious the effect of a failure mode is for the customer if the failure occurs. It is one of the three FMEA ratings, along with occurrence and detection.

Why the other options are wrong:

  • A. The likelihood that a failure mode occurs is the Occurrence rating.
  • C. The ability to catch a problem before it reaches the customer is the Detection rating.
  • D. Severity is judged from the effect on the customer, not only on the process or business.

Question 24

Which of the following statements best summarizes the purpose for conducting an FMEA?

  • A. To develop cause-effect diagrams for process or product design.
  • B. To capture field failure data and track appropriate corrective actions to failure modes
  • C. To prioritize failures modes by severity, occurrence, and detection
  • D. To proactively anticipate failure modes and their risks for subsequent action
Show answer and explanation ▾

Correct answer: D

The purpose of an FMEA is to anticipate how a product or process could fail, assess the risk of each failure mode and drive action before the failure happens.

Why the other options are wrong:

  • A. Cause and effect diagrams are a separate tool and not the purpose of an FMEA.
  • B. Tracking field failures is reactive, while FMEA is done before failures occur.
  • C. Prioritizing by the three ratings is a step inside the FMEA, not its overall purpose.

Question 25

Potential Ca uses are listed on FMEA forms because:

  • A. Individual failure modes may have multiple causes
  • B. Different causes may have different monitors or controls to detect those occurrences
  • C. Individual causes may cause multiple modalities of failure
  • D. All of the Above
Show answer and explanation ▾

Correct answer: D

Causes are listed separately because one failure mode can have several causes, one cause can lead to several failure modes, and each cause can have its own controls for detection. All three statements hold.

Why the other options are wrong:

  • A. This is true, but it is only one of the three reasons listed.
  • B. This is true, but it is only one of the three reasons listed.
  • C. This is true, but it is only one of the three reasons listed.

Question 26

After developing a Process map (listing steps and functions), the first step in developing an FMEA is:

  • A. Brainstorming possible failure modes with team members
  • B. Assigning scores for SEVERITY, OCCURRENCE and DETECTION
  • C. Calculating RPN - Risk Priority Number
  • D. Prioritizing failure modes for action or resolution
Show answer and explanation ▾

Correct answer: A

Once the process steps and functions are mapped, the team brainstorms the ways each step could fail. Effects, causes, ratings and priorities are all built on that list of failure modes.

Why the other options are wrong:

  • B. Scores can only be assigned after the failure modes, effects and causes are listed.
  • C. RPN is calculated from the three scores, which come later.
  • D. Prioritizing is done after the RPN values are known.

Question 27

The BEST source of data for assigning an Occurrence Rating in an FMEA is:

  • A. Estimates from Industry Experts
  • B. Simulated failure rates from reliability models
  • C. Best guesses by those preparing FMEA
  • D. Historical failure data
Show answer and explanation ▾

Correct answer: D

The Occurrence rating reflects how often a cause or failure happens, so actual historical failure data from the process or similar processes is the best basis for it.

Why the other options are wrong:

  • A. Expert estimates are opinions and rank below actual recorded failures.
  • B. Simulated rates are model estimates and rank below actual recorded failures.
  • C. Best guesses are the weakest basis for a rating.

Question 28

The most common source of error in generating an FMEA is:

  • A. Assigning wrong SEVERITY scores
  • B. Missing Failure modes during brainstorming
  • C. Estimating probabilities incorrectly for OCCURRENCE
  • D. Scoring easily detectable failure modes as high DETECTION instead of low.
Show answer and explanation ▾

Correct answer: B

A failure mode that the team never lists receives no rating, no priority and no action, so missing failure modes during brainstorming is the most common and most damaging error in an FMEA.

Why the other options are wrong:

  • A. A wrong severity score still leaves the failure mode on the form where it can be reviewed.
  • C. A wrong occurrence estimate still leaves the failure mode on the form where it can be corrected.
  • D. A wrong detection score still leaves the failure mode on the form where it can be corrected.

Question 29

DFMEA should be:

I. Initiated no later than design concept finalization II. Updated continually as changes occur III. Completed before product drawings and specifications are released in final form¡±

  • A. I only
  • B. II only
  • C. III only
  • D. I, II, and III
Show answer and explanation ▾

Correct answer: D

A design FMEA is started by the time the design concept is finalized, kept up to date as changes occur, and completed before the drawings and specifications are released. All three statements apply.

Why the other options are wrong:

  • A. Statement I is correct, but II and III also apply.
  • B. Statement II is correct, but I and III also apply.
  • C. Statement III is correct, but I and II also apply.

Question 30

Actions listed on FMEA forms are:

  • A. Corrective actions to repair damage to products, which have experienced the associated failure modes.
  • B. Actions that are to be taken after failures occur
  • C. Actions that can be taken to reduce the associated RPN
  • D. Required for all failure modes
Show answer and explanation ▾

Correct answer: C

Recommended actions on an FMEA form are preventive steps aimed at lowering the Risk Priority Number by reducing severity, occurrence or detection ratings.

Why the other options are wrong:

  • A. Repairing damaged product is a reaction after failure, not an FMEA recommended action.
  • B. FMEA actions are taken before failures occur.
  • D. Actions are prioritized for the higher risk failure modes, not required for every one.

Question 31

Select the correct definition for the role of Project Champion.

  • A. Has influence and often authority over the process and connecting organizations
  • B. Owns the process being changed and accountable for final results
  • C. Manages Sector change strategy
  • D. Manages and oversees multiple LSS projects
Show answer and explanation ▾

Correct answer: B

The Project Champion owns the process being changed and is accountable for the final results of the project.

Why the other options are wrong:

  • A. Influence and authority over the process and connecting organizations describes the Sponsor.
  • C. Managing sector change strategy is not the Project Champion role.
  • D. Overseeing multiple LSS projects is not the definition of the Project Champion.

Question 32

What percent of data falls between 2 standard deviations on either side of the mean for a normal distribution?

  • A. 68%
  • B. 92%
  • C. 95%
  • D. 99.73%
Show answer and explanation ▾

Correct answer: C

For a normal distribution, about 95% of the data falls within 2 standard deviations on either side of the mean.

Why the other options are wrong:

  • A. About 68% falls within 1 standard deviation of the mean.
  • B. 92% does not correspond to a whole number of standard deviations.
  • D. 99.73% falls within 3 standard deviations of the mean.

Question 33

What is the purpose of performing a Stakeholder Analysis?

  • A. Analyze the strength of commitment needed for project stakeholders vs. where they are currently
  • B. Analyze the risk for the project
  • C. Analyze the completeness of charter for the project
  • D. Analyze the methodology chosen for the project
Show answer and explanation ▾

Correct answer: A

A stakeholder analysis compares each stakeholder's current level of commitment with the level the project needs, so the team can plan how to close the gaps.

Why the other options are wrong:

  • B. Project risk is assessed with risk analysis tools.
  • C. Checking the charter for completeness is a charter review, not a stakeholder analysis.
  • D. The choice of methodology is not what a stakeholder analysis examines.

Question 34

What is the primary advantage of the Median over the Mean?

  • A. The Median is a better description of the average
  • B. The Median is easier to compute
  • C. The Median is less sensitive to outliers
  • D. The Median is easier to understand
Show answer and explanation ▾

Correct answer: C

The median depends only on the middle position of the ordered data, so a few extreme values do not move it. The mean uses every value and is pulled toward outliers.

Why the other options are wrong:

  • A. The median is not a better description of the average in general.
  • B. Ease of computation is not the main advantage of the median.
  • D. Ease of understanding is not the main advantage of the median.

Question 35

Electrical devices are evaluated as conforming or not conforming to specifications. The appropriate data type is:

  • A. Nominal discrete data
  • B. Continuous data
  • C. Ordinal data
  • D. Variable data
Show answer and explanation ▾

Correct answer: A

Conforming or not conforming puts each device into one of two categories with no natural order and no measured value. That is nominal discrete, or attribute, data.

Why the other options are wrong:

  • B. Continuous data is measured on a scale, which a pass or fail result is not.
  • C. Ordinal data has ranked categories, and these two categories are not a ranking scale.
  • D. Variable data is measured data, another name for continuous data.

Question 36

A team has just completed one of the DMAIC steps. As part of the DMAIC methodology, they have generated solutions to the problem.

What phase of the DMAIC process have they most likely just completed?

  • A. Analyze
  • B. Improve
  • C. Control
  • D. Measure
Show answer and explanation ▾

Correct answer: B

Generating, selecting and testing solutions to the problem is the work of the Improve phase of DMAIC.

Why the other options are wrong:

  • A. Analyze identifies and verifies root causes.
  • C. Control sustains the gains after the solution is in place.
  • D. Measure establishes the baseline performance of the process.

Question 37

Which of the following has the team role of being a creative thinker who challenges conventional ways of doing things?

  • A. Functional Team Expert
  • B. IT Representative
  • C. Assumption Buster
  • D. Champion
Show answer and explanation ▾

Correct answer: C

The Assumption Buster is the team member who thinks creatively and challenges the conventional ways of doing things, so the team does not accept old assumptions without question.

Why the other options are wrong:

  • A. The Functional Team Expert contributes knowledge of the process area.
  • B. The IT Representative contributes knowledge of systems and data.
  • D. The Champion supports the project and removes barriers.

Question 38

Which of the following is NOT a correct statement about Value Analysis?

  • A. 80% of most processes are non-value added work
  • B. Value added work physically changes the inputs and the customer is willing to pay for it, or requires it
  • C. Purpose is to design out non-essential work that consumes critical time and energy
  • D. Value added work is all the work necessary to deliver the product
Show answer and explanation ▾

Correct answer: D

Work is value added only when it changes the product or service and the customer is willing to pay for it or requires it. Much of the work needed to deliver a product does not meet that test, so the statement is not correct.

Why the other options are wrong:

  • A. Value analysis teaches that the large majority of most processes is non-value added work.
  • B. This is the standard definition of value added work.
  • C. Removing non-essential work that consumes time and energy is the purpose of value analysis.

Question 39

What is the most powerful factor in achieving success in Six Sigma in an organization?

  • A. Strategic Integration
  • B. Business Process Framework
  • C. A consistent, disciplined team meeting schedule
  • D. Committed Leadership
Show answer and explanation ▾

Correct answer: D

Committed leadership is the strongest driver of Six Sigma success, because leaders set priorities, provide resources and hold the organization accountable for results.

Why the other options are wrong:

  • A. Strategic integration depends on leaders who commit to it.
  • B. A business process framework is a supporting element.
  • C. A regular meeting schedule helps a team but does not drive organization-wide success.

Question 40

Which of the following is most sensitive to outliers in the data?

  • A. Median
  • B. Percentiles
  • C. Variance
  • D. Mean
Show answer and explanation ▾

Correct answer: D

The mean is calculated from every value in the data set, so a single extreme value pulls it toward the outlier.

Why the other options are wrong:

  • A. The median depends on the middle position and resists outliers.
  • B. Percentiles depend on rank order and resist outliers.
  • C. Variance describes the spread around the mean rather than the central value that an outlier pulls.

Question 41

A team has just developed an operational definition and has developed a plan for understanding current performance.

What is the DMAIC phase in which the team is currently working?

  • A. Define
  • B. Measure
  • C. Analyze
  • D. Improve
Show answer and explanation ▾

Correct answer: B

Writing operational definitions and planning data collection to understand current performance are Measure phase activities.

Why the other options are wrong:

  • A. Define sets the problem, scope and goals in the project charter.
  • C. Analyze looks for root causes using the data already collected.
  • D. Improve develops and tests solutions.

Question 42

Determine the process improvement goal which would require the usage of the DMADV methodology.

  • A. Improve Process Efficiency & Speed
  • B. Develop New Processes
  • C. Variation & Defect Reduction
  • D. Special Cause Elimination
Show answer and explanation ▾

Correct answer: B

DMADV (Define, Measure, Analyze, Design, Verify) is the Design for Six Sigma roadmap, used when a new process or product has to be developed.

Why the other options are wrong:

  • A. Efficiency and speed gains in an existing process are addressed with Lean methods.
  • C. Variation and defect reduction in an existing process uses DMAIC.
  • D. Special cause elimination is handled with structured problem solving.

Question 43

The person who ultimately selects Little y's and vital x's, has influence and often authority over the process and connecting organization is the:

  • A. Champion
  • B. Sponsor
  • C. LSS Leader
  • D. Master Black Belt
Show answer and explanation ▾

Correct answer: B

The Sponsor has influence and often authority over the process and connecting organizations, and ultimately selects the little y's and vital x's for the project.

Why the other options are wrong:

  • A. The Champion owns the process being changed and is accountable for final results.
  • C. The LSS Leader guides the deployment and does not make this selection.
  • D. The Master Black Belt coaches Belts and provides technical expertise.

Question 44

A team has just completed one of the DMAIC steps. As part of the DMAIC methodology, they now have an understanding of the root causes of the problem(s).

What phase of the DMAIC process are have they most likely just completed?

  • A. Analyze
  • B. Improve
  • C. Control
  • D. Measure
Show answer and explanation ▾

Correct answer: A

The Analyze phase ends when the team has identified and verified the root causes of the problem.

Why the other options are wrong:

  • B. Improve produces solutions, which come after root causes are understood.
  • C. Control sustains the improved process.
  • D. Measure provides the baseline data, not the root causes.

Question 45

Which of the following terms refers to a trial implementation of the proposed design or process change on a reduced scale?

  • A. Production Run
  • B. Pilot
  • C. Design of Experiments
  • D. Comparative Method
Show answer and explanation ▾

Correct answer: B

A pilot is a trial implementation of the proposed design or process change on a reduced scale, used to confirm results and find problems before full rollout.

Why the other options are wrong:

  • A. A production run is full scale operation.
  • C. Design of Experiments studies the effect of factors on a response.
  • D. A comparative method compares alternatives and is not a small scale implementation.

Question 46

Determine the process improvement goal which would require the usage of the FORD-8D methodology.

  • A. Improve Process Efficiency & Speed
  • B. Develop New Processes
  • C. Variation & Defect Reduction
  • D. Special Cause Elimination
Show answer and explanation ▾

Correct answer: D

The Ford 8D method is a structured team problem solving process for containing a specific problem, finding its root cause and preventing recurrence, which fits special cause elimination.

Why the other options are wrong:

  • A. Efficiency and speed gains are addressed with Lean methods.
  • B. Developing new processes uses DMADV.
  • C. Variation and defect reduction uses DMAIC.

Question 47

Scatter diagrams are most useful for which of the following?

  • A. Showing that output factor (Y) is normally distributed
  • B. Graphically showing the relationship between two variables
  • C. Determining whether a process is in control
  • D. Proving root causes of problems
Show answer and explanation ▾

Correct answer: B

A scatter diagram plots pairs of values for two variables, which shows graphically whether and how they are related.

Why the other options are wrong:

  • A. Normality is checked with a histogram or a normal probability plot.
  • C. Process control is judged with a control chart.
  • D. A scatter diagram shows association and cannot prove a root cause.

Question 48

An MSA study is generally performed during which of the DMAIC phases?

  • A. Define phase
  • B. Measure phase
  • C. Analyze phase
  • D. Control phase
Show answer and explanation ▾

Correct answer: B

Measurement System Analysis is done in the Measure phase, so the team knows the data can be trusted before baseline performance is established and analyzed.

Why the other options are wrong:

  • A. Define sets the problem and scope before data collection begins.
  • C. Analyze relies on data that has already been validated.
  • D. Control comes after the improvement is in place.

Question 49

When performing a Measurement System Analysis, σtotal = 5. What is the required measurement unit of the measurement tool in order to have acceptable discrimination?

  • A. 3 or less
  • B. 4.5 or less
  • C. 5 or less
  • D. 3.5 or less
Show answer and explanation ▾

Correct answer: A

Acceptable discrimination requires a measurement unit of at most one-tenth of the six sigma spread of total variation. Six times 5 is 30, and one-tenth of 30 is 3.

Why the other options are wrong:

  • B. 4.5 is larger than one-tenth of the six sigma spread.
  • C. 5 equals the total standard deviation and is too coarse.
  • D. 3.5 is larger than one-tenth of the six sigma spread.

Question 50

The ________________ the discrimination, the __________________ the changes that can be detected by the measurement tool.

  • A. Higher/higher
  • B. Higher/smaller
  • C. Smaller/smaller
  • D. None of the above
Show answer and explanation ▾

Correct answer: B

Discrimination is the ability of a measurement tool to tell small differences apart. The higher the discrimination, the smaller the change the tool can detect.

Why the other options are wrong:

  • A. Higher discrimination improves the detection of small changes, not large ones.
  • C. Lower discrimination means a coarser tool that misses small changes.
  • D. One of the listed pairs completes the statement correctly.

Question 51

With Measurement Systems Analysis (MSA), we want to determine how much of the variation in our data is due to:

  • A. Variation in our measurement system
  • B. Manufacturing Equipment
  • C. Part to part variation
  • D. Lot to lot variation
Show answer and explanation ▾

Correct answer: A

Measurement Systems Analysis determines how much of the observed variation comes from the measurement system itself, so it can be separated from the real variation in the process.

Why the other options are wrong:

  • B. Equipment variation is part of the process variation, not the subject of MSA.
  • C. Part to part variation is the real process variation that MSA separates the measurement error from.
  • D. Lot to lot variation is process variation, not measurement variation.

Question 52

% R & R can be expressed as:

  • A. % productivity divided by run rate
  • B. % repeatability divided by production (rate)
  • C. Sigma of the measurement system divided by total sigma
  • D. Variance of the measurement system divided by total variance
Show answer and explanation ▾

Correct answer: C

% R&R is the standard deviation of the measurement system divided by the total standard deviation, expressed as a percentage.

Why the other options are wrong:

  • A. Productivity and run rate are not part of a gauge study.
  • B. Production rate is not part of the % R&R calculation.
  • D. A ratio of variances gives the percent contribution, which is a different measure.

Question 53

If you have a gauge study with a % R&R of 25% what should you do?

  • A. Pull the tool from production.
  • B. This is fine-go ahead.
  • C. This is not acceptable for critical measurements, you should look into improvements in the measurement process.
  • D. This is only an issue if you fail calibration.
Show answer and explanation ▾

Correct answer: C

A % R&R under 10% is acceptable and one over 30% is unacceptable. A result of 25% sits in the marginal band between them, so it is not good enough for critical measurements and the measurement process should be improved.

Why the other options are wrong:

  • A. A marginal 25% result calls for improving the measurement process, not for removing the tool from production.
  • B. A 25% result is above the 10% level at which a gauge is accepted without concern.
  • D. Calibration addresses bias, while % R&R measures repeatability and reproducibility, so passing calibration does not settle it.

Question 54

Three different operators perform MSA on 10 parts. They measure the same 10 parts twice.

What would the Total # of readings be?

  • A. 30
  • B. 60
  • C. 120
  • D. 10
Show answer and explanation ▾

Correct answer: B

The total number of readings is operators times parts times trials. Three operators each measuring 10 parts twice gives 3 x 10 x 2 = 60 readings.

Why the other options are wrong:

  • A. 30 counts only one trial per operator and leaves out the second measurement.
  • C. 120 would need four trials per part, or six operators, not the study described.
  • D. 10 is the number of parts, not the number of readings.

Question 55

A measurement system is said to have a consistent bias if:

  • A. The standard deviation of the system is consistent across the measurement range
  • B. The difference between the average reading of the measurement tool and reference value is the same across the operating range of the tool operation
  • C. The standard deviations of the reading given by the measurement tool when measuring a standard is the same across the operating range of the tool operation
  • D. All tools of the same type give the same variance reading
Show answer and explanation ▾

Correct answer: B

Bias is the difference between the average reading and the reference value. When that difference stays the same across the operating range of the tool, the bias is consistent, which is what a linearity study checks.

Why the other options are wrong:

  • A. A consistent standard deviation describes uniform precision, not bias.
  • C. The spread of readings on a standard describes repeatability, not the offset from the reference value.
  • D. Agreement in variance between tools of the same type says nothing about the offset of a tool from the reference value.

Question 56

The variance of your measurement system is 25. The total standard deviation is 50.

What is your % R & R?

  • A. 15%
  • B. 50%
  • C. 10%
  • D. 22%
Show answer and explanation ▾

Correct answer: C

% R&R compares standard deviations, not variances. A measurement system variance of 25 gives a standard deviation of 5, and 5 divided by the total standard deviation of 50 is 10%.

Why the other options are wrong:

  • A. 15% does not result from dividing 5 by 50.
  • B. 50% comes from dividing the variance of 25 by 50, which mixes a variance with a standard deviation.
  • D. 22% does not result from dividing 5 by 50.

Question 57

What type of knowledge is gained by doing an MSA?

  • A. Determine the size of the measurement variability
  • B. Determine if the measurement system is stable over time
  • C. Determine if the measurement system is capable of making the required measurements
  • D. All of the above
Show answer and explanation ▾

Correct answer: D

A measurement system analysis quantifies the size of the measurement variability, shows whether the system is stable over time, and shows whether it is capable of making the required measurements. All three listed outcomes apply.

Why the other options are wrong:

  • A. The size of the measurement variability is only one of the three things an MSA reveals.
  • B. Stability over time is only one of the three things an MSA reveals.
  • C. Capability to make the required measurements is only one of the three things an MSA reveals.

Question 58

A measurement system has sufficient discrimination if:

  • A. A measurement system should not discriminate
  • B. The measurement unit is at most one-tenth of the six sigma spread of the total process variation
  • C. You can get output to the second decimal point
  • D. If the tool can be easily calibrated
Show answer and explanation ▾

Correct answer: B

Discrimination is the smallest change a measurement tool can detect. The rule of thumb is that the measurement unit should be no more than one-tenth of the six sigma spread of the total process variation.

Why the other options are wrong:

  • A. A measurement system must discriminate, because it has to tell different parts apart.
  • C. Reading to two decimal places is only sufficient if that unit is small relative to the process spread.
  • D. Ease of calibration relates to bias, not to the smallest change the tool can detect.

Question 59

When assessing gauge performance, you should look at the following (Choose the best answer).

  • A. Repeatability
  • B. Reproducibility
  • C. Repeatability and reproducibility
  • D. Improve variation of process being measured
Show answer and explanation ▾

Correct answer: C

Gauge precision has two components: repeatability, the variation when one operator measures the same part repeatedly, and reproducibility, the variation between operators. Both must be assessed together.

Why the other options are wrong:

  • A. Repeatability alone leaves out the variation between operators.
  • B. Reproducibility alone leaves out the variation within a single operator and gauge.
  • D. Improving the process being measured is a separate activity and does not assess the gauge.

Question 60

Given repeatability σRPT =4 and reproducibility σRPD = 3, calculate precision σMS

  • A. σMS = 3
  • B. σMS = 7
  • C. σMS = 25
  • D. σMS = 5
Show answer and explanation ▾

Correct answer: D

Measurement system precision combines repeatability and reproducibility as the square root of the sum of their squares. With values of 4 and 3, the square root of 16 plus 9 is the square root of 25, which equals 5.

Why the other options are wrong:

  • A. A value of 3 is only the reproducibility term and ignores repeatability.
  • B. A value of 7 simply adds the two terms, but they combine as the square root of the sum of squares.
  • C. A value of 25 is the sum of squares and leaves out the square root.

Question 61

The following is the best description for the purpose of a process map:

  • A. It is a tool used only after a Six Sigma project is completed
  • B. It is a tool to convert customer requirements to a set of measurable technical requirements with target values.
  • C. It is a tool for identifying alternative solutions or concepts
  • D. It is a tool used to identify all major steps, outputs, and inputs to look for potential causes of a problem.
Show answer and explanation ▾

Correct answer: D

A process map lays out the major steps of a process with their inputs and outputs. Seeing the whole flow lets the team look for the steps and inputs that could be causing the problem.

Why the other options are wrong:

  • A. A process map is built early in a project, not only after it is completed.
  • B. Converting customer requirements into measurable technical requirements with targets describes Quality Function Deployment.
  • C. Identifying alternative solutions or concepts is the job of concept generation and selection tools.

Question 62

In a regression analysis, model assumptions are validated using which of the following?

  • A. The ANOVA table
  • B. A plot of X versus predicted/fitted Y
  • C. Chi-square statistic
  • D. Residual diagnostics
Show answer and explanation ▾

Correct answer: D

The regression assumptions are about the errors: normally distributed, independent and with constant variance. They are checked with residual diagnostics such as a normal probability plot and plots of residuals against fitted values and run order.

Why the other options are wrong:

  • A. The ANOVA table tests whether the model is significant, not whether its assumptions hold.
  • B. A plot of X against fitted Y simply redraws the fitted line and shows nothing about the errors.
  • C. The chi-square statistic is not the tool used to check regression assumptions.

Question 63

For a multiple regression model, the R2 is the parameter that:

  • A. Represents the fraction of total variation explained by the model
  • B. Explains the strength of only one of the inputs in the model
  • C. Explains if the input factor causes the output response
  • D. Quantifies the slope of the line in the model
Show answer and explanation ▾

Correct answer: A

R-squared is the model sum of squares divided by the total sum of squares. It is the fraction of the total variation in the response that the regression model explains.

Why the other options are wrong:

  • B. R-squared describes the whole model, not the strength of a single input.
  • C. R-squared measures fit and says nothing about whether an input causes the response.
  • D. The slope is given by the regression coefficients, not by R-squared.

Question 64

When evaluating residuals from a regression model, a Black Belt discovers that she has outliers in the data.

What is best course of action for the outliers?

  • A. Ignore them. They should not impact the model.
  • B. Omit them.
  • C. Explore the source of the outlier.
  • D. Transform the independent variable and re-fit the model.
Show answer and explanation ▾

Correct answer: C

An outlier may be a recording error, a measurement problem or a real and informative process event. The source has to be explored first, because only the cause tells you whether to correct, keep or remove the point.

Why the other options are wrong:

  • A. Outliers can pull the fitted line and inflate the error, so they cannot be ignored.
  • B. Omitting a point without knowing its cause can throw away real information.
  • D. Transforming the independent variable addresses curvature, not unexplained outliers.

Question 65

In Logistic Regression, an Odds Ratio of 1 means?

  • A. One group is more likely than the other
  • B. Both groups are very likely
  • C. No association between groups - they are equally likely
  • D. There is a significant relationship between groups
Show answer and explanation ▾

Correct answer: C

An odds ratio compares the odds of the event in one group with the odds in another. A value of 1 means the odds are identical, so there is no association between group and outcome.

Why the other options are wrong:

  • A. One group being more likely gives an odds ratio above or below 1, not equal to 1.
  • B. An odds ratio compares groups and does not say how likely the event is in absolute terms.
  • D. A significant relationship shows up as an odds ratio different from 1.

Question 66

How do you check that a regression model is valid?

  • A. Residuals are normally distributed
  • B. Residuals are independent
  • C. Residuals have constant variation
  • D. All of the above
Show answer and explanation ▾

Correct answer: D

A regression model is valid when its residuals are normally distributed, independent of each other and have constant variation. All three conditions must be checked, so every listed item applies.

Why the other options are wrong:

  • A. Normality of the residuals is only one of the three checks.
  • B. Independence of the residuals is only one of the three checks.
  • C. Constant variation of the residuals is only one of the three checks.

Question 67

Which of the following statements is true regarding correlation and simple regression?

  • A. A positive correlation is more significant than a negative correlation
  • B. R2 = the amount of variance explained by the regression model
  • C. A statistically significant correlation confirms a causal relationship between the X and Y variables
  • D. They are the best statistical methods when you have discrete X and Y variables
Show answer and explanation ▾

Correct answer: B

R-squared is the proportion of the variation in Y that the regression model explains. In simple regression it equals the square of the correlation coefficient.

Why the other options are wrong:

  • A. The sign of a correlation shows direction only, so positive is no more significant than negative.
  • C. Correlation, even when statistically significant, does not confirm causation.
  • D. Correlation and regression are meant for continuous X and Y variables, not discrete ones.

Question 68

What action should be taken if the residuals from a regression model show a pattern over time?

  • A. It is acceptable to use the results as they exist
  • B. Transform the input variable
  • C. Transform the output variable
  • D. Track down the assignable cause of the time trend I pattern
Show answer and explanation ▾

Correct answer: D

A pattern in the residuals over time means the errors are not independent and something outside the model changed during data collection. The assignable cause of the time trend has to be tracked down before the model can be trusted.

Why the other options are wrong:

  • A. A time pattern breaks the independence assumption, so the results cannot be used as they stand.
  • B. Transforming the input variable addresses curvature, not a trend over time.
  • C. Transforming the output variable addresses unequal variance or non-normality, not a trend over time.

Question 69

A Black Belt is working a project on battery life. The Black Belt finds that the process is unstable, but wants to do a capability analysis. The Black Belt should:

  • A. Continue on with the capability analysis to gain more insight
  • B. Work to identify the special causes before continuing on with the capability analysis
  • C. Transform the data
  • D. Find and remove the data points (outliers) that caused it to appear unstable. If the process is capable without the outliers, the outliers can simply be discarded.
Show answer and explanation ▾

Correct answer: B

Capability indices only predict future performance when the process is stable. With an unstable process the special causes must be identified and addressed first, then the capability analysis can be carried out.

Why the other options are wrong:

  • A. Capability figures from an unstable process are not reliable predictions.
  • C. Transforming data addresses non-normality, not instability.
  • D. Discarding points without finding their cause hides the instability instead of removing it.

Question 70

A Black Belt is reviewing a stable, normal process that has a Cp = 3 and a Cpk = 1. In order to achieve a 6 sigma process, what must be done?

  • A. Reduce variation only
  • B. Shift the mean only
  • C. Reduce variation and shift the mean
  • D. Not enough information given
Show answer and explanation ▾

Correct answer: B

By the usual convention a six sigma process has a Cp of 2.0 and a Cpk of 1.5. A Cp of 3 shows the variation is already small enough, and the Cpk of 1 shows the process is off center, so only the mean has to be shifted.

Why the other options are wrong:

  • A. The Cp of 3 already exceeds the level required, so variation is not the problem.
  • C. Variation needs no reduction because the Cp of 3 already exceeds the level required.
  • D. Cp and Cpk together give enough information to see that centering is the only issue.

Question 71

A black belt needs to determine if a process is stable.

What technique should be used?

  • A. Capability Chart
  • B. Pareto Chart
  • C. Control Chart
  • D. Histogram
Show answer and explanation ▾

Correct answer: C

A control chart plots data in time order against a center line and statistically calculated control limits. It separates common cause from special cause variation, which is how process stability is determined.

Why the other options are wrong:

  • A. A capability chart compares the process with its specifications and assumes stability has already been shown.
  • B. A Pareto chart ranks categories by frequency and shows nothing about behavior over time.
  • D. A histogram shows the distribution of the data but not its time order.

Question 72

You have been asked to monitor daily production yields to determine if the process yield is in statistical control.

Select the control chart best suited for this purpose.

  • A. C-chart
  • B. U-chart
  • C. X and R chart
  • D. P-chart
Show answer and explanation ▾

Correct answer: D

Yield is a proportion of good or defective units out of the total produced, which is attribute data. The P chart tracks a proportion and handles daily production volumes that vary.

Why the other options are wrong:

  • A. A C chart tracks the count of defects in a constant sample size, not a proportion.
  • B. A U chart tracks defects per unit, not the proportion of units that pass or fail.
  • C. X-bar and R charts are for continuous measurement data in subgroups.

Question 73

Lean Six Sigma process I product improvement suggests that the measure we use in our business is directly derived from which of the following sources?

  • A. Voice of Technology
  • B. Voice of Finance
  • C. Voice of the Customer
  • D. The Six Sigma Project Team
Show answer and explanation ▾

Correct answer: C

Lean Six Sigma defines quality from the customer's point of view. The measures a business uses are derived from the Voice of the Customer, which is translated into critical requirements and then into measurable metrics.

Why the other options are wrong:

  • A. Technology is a means of delivering what customers need, not the source of the measures.
  • B. Financial measures follow from serving the customer and are not the source of process measures.
  • D. The project team translates customer needs into measures but is not the source of them.

Question 74

The most important process metrics are often determined by what the customer tells us is important to them.

Which of these sequences describes the way customer requirements are converted to the customer-based process metrics?

  • A. Voice of the customer>critical to quality metrics>critical customer requirements
  • B. Voice of the customer>critical customer requirements>critical to process metrics
  • C. Critical customer requirements>critical to quality metrics>critical to process metrics
  • D. Voice of the customer>critical customer requirements>critical to quality metrics
Show answer and explanation ▾

Correct answer: D

Customer statements are gathered first as the Voice of the Customer. They are refined into critical customer requirements, which are then converted into measurable critical to quality metrics.

Why the other options are wrong:

  • A. Critical customer requirements come before the critical to quality metrics, not after them.
  • B. Critical to process metrics are internal measures, not the customer-based metrics asked about.
  • C. The sequence has to begin with the Voice of the Customer.

Question 75

Which of the following are true regarding the Central Limit Theorem?

  • A. Sample averages are normally distributed as sample size gets larger.
  • B. Sample averages converge on the population mean.
  • C. Sample standard deviations are dependent upon the standard deviation of the population from which it was sampled.
  • D. All of the above
Show answer and explanation ▾

Correct answer: D

The Central Limit Theorem says the distribution of sample averages approaches normal as sample size grows, is centered on the population mean, and has a spread equal to the population standard deviation divided by the square root of the sample size. All three statements follow from it.

Why the other options are wrong:

  • A. Normality of sample averages is only one of the three results.
  • B. Sample averages centering on the population mean is only one of the three results.
  • C. The link to the population standard deviation is only one of the three results.

Question 76

Suppose an X-bar / S Chart revealed that the variation of a process was consistent over time (consistent standard deviation, consistent mean) but a significant proportion of outcomes fell outside the customer requirements.

Which of the following conclusions can best be made about the process?

  • A. The process is in control but has poor capability
  • B. The process variation is out of control
  • C. Special or assignable causes are affecting the process
  • D. The process mean needs to be reduced
Show answer and explanation ▾

Correct answer: A

A consistent mean and standard deviation over time mean the process is in statistical control. Outcomes falling outside the customer requirements mean it does not meet specification, which is poor capability.

Why the other options are wrong:

  • B. The standard deviation is consistent over time, so the variation is in control.
  • C. A stable mean and standard deviation show only common cause variation.
  • D. Nothing given shows that lowering the mean, as opposed to raising it or reducing spread, would fix the problem.

Question 77

A ___________ is used primarily to track the stability of the average value of a metric of interest.

  • A. NP Chart
  • B. Xbar-R Chart
  • C. I-MR Chart
  • D. C Chart
Show answer and explanation ▾

Correct answer: B

The X-bar chart in an Xbar-R pair plots subgroup averages, so it tracks the stability of the average value of a continuous metric. The R chart alongside it tracks the variation within subgroups.

Why the other options are wrong:

  • A. An NP chart tracks the number of defective units, not an average value.
  • C. An I-MR chart plots individual readings, not subgroup averages.
  • D. A C chart tracks the count of defects, not an average value.

Question 78

For her injection molding project a Belt needed to track the percentage of defectives of a particular sample set so she used a ___________ to display the data?

  • A. Individual Chart
  • B. C Chart
  • C. Xbar Chart
  • D. P Chart
Show answer and explanation ▾

Correct answer: D

The P chart plots the proportion or percentage of defective units in each sample. It is the attribute chart designed for tracking percent defective.

Why the other options are wrong:

  • A. An Individual chart is for single continuous measurements, not a percentage of defectives.
  • B. A C chart tracks the count of defects, not the percentage of defective units.
  • C. An Xbar chart tracks subgroup averages of continuous data.

Question 79

Control Charts were developed by Dr. Shewhart to track data over time. To detect Special Cause variation the Control Charts use which of these?

  • A. Data shift analysis
  • B. Outlier analysis methods
  • C. Center Line and Control Limits
  • D. None of the above
Show answer and explanation ▾

Correct answer: C

A Shewhart control chart compares each plotted point with a center line and control limits calculated from the process data. Points outside the limits, or non-random patterns around the center line, signal special cause variation.

Why the other options are wrong:

  • A. Data shift analysis is not the mechanism built into a control chart.
  • B. Outlier analysis methods are not the mechanism built into a control chart.
  • D. The center line and control limits are the correct mechanism, and they are listed.

Question 80

Common and ______________ Cause Variation are the focus of Statistical Process Control.

  • A. Uncommon
  • B. Ordinary
  • C. Special
  • D. Selective
Show answer and explanation ▾

Correct answer: C

Statistical Process Control separates two kinds of variation. Common cause variation is inherent in the process, and special cause variation comes from specific, identifiable events outside the normal process.

Why the other options are wrong:

  • A. Uncommon is not the term used in Statistical Process Control.
  • B. Ordinary is another way of describing common cause variation, not its counterpart.
  • D. Selective is not a type of variation in Statistical Process Control.

Question 81

Special Cause Variation falls into which two categories?

  • A. Natural & Unnatural
  • B. Short Term & Long Term
  • C. Assignable & Pattern
  • D. Attribute & Discreet
Show answer and explanation ▾

Correct answer: C

Special cause variation is grouped into assignable causes, which can be traced to a specific event and removed, and pattern causes, which show up as non-random patterns in the data such as trends, shifts or cycles.

Why the other options are wrong:

  • A. Natural variation is another name for common cause variation, so this pair does not divide special causes.
  • B. Short term and long term describe time frames for capability, not categories of special cause.
  • D. Attribute and discrete describe data types, not categories of variation.

Question 82

Range Charts are the technique used to determine if Special Causes are occurring within the subgroups of the ______________.

  • A. Histograms
  • B. SPC Charts
  • C. NP Charts
  • D. Pareto Charts
Show answer and explanation ▾

Correct answer: B

A Range chart plots the range of each subgroup on an SPC chart. It shows whether the variation within the subgroups is stable or whether special causes are acting inside them.

Why the other options are wrong:

  • A. Histograms show a distribution and have no subgroups plotted over time.
  • C. NP charts track the number of defective units and do not use a Range chart.
  • D. Pareto charts rank categories and have no subgroups.

Question 83

If the production is for higher volume and monitoring and the Mean and variability is to be monitored for four machines producing product and the characteristic to be monitored is Variable Data, which SPC Chart is best to be selected?

  • A. Xbar-R Chart
  • B. Individual-MR Chart
  • C. NP Chart
  • D. CUSUM Chart
Show answer and explanation ▾

Correct answer: A

With high volume production of variable data, measurements can be taken in rational subgroups. The Xbar chart then monitors the mean and the R chart monitors the variability.

Why the other options are wrong:

  • B. An Individual-MR chart is for low volume situations where only one reading is available at a time.
  • C. An NP chart is for attribute data, not variable data.
  • D. A CUSUM chart is designed to detect small shifts in the mean and does not monitor variability alongside it.

Question 84

When a Belt Poka-Yoke's a defect out of the process entirely then she should track the activity with a robust SPC system on the characteristic of interest in the defect as an early warning system.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

The statement is false. Poka-Yoke removes the possibility of the defect, so there is nothing left for SPC to detect. SPC is an early warning tool for characteristics that can still vary, and adding it here would be wasted effort.

Why the other options are wrong:

  • A. Once a defect has been designed out of the process, SPC on that characteristic adds cost without adding protection.

Question 85

Following the completion of a LSS project the Belt not only creates a Control Plan he also develops a _____________ so those involved in the process know what to do when the critical metrics move out of spec.

  • A. Response Plan
  • B. Call List
  • C. Chain-of-Command
  • D. Defect Analysis Plan
Show answer and explanation ▾

Correct answer: A

A Response Plan accompanies the Control Plan. It tells the people in the process exactly what action to take, and who takes it, when a critical metric moves out of specification or out of control.

Why the other options are wrong:

  • B. A call list names people to contact but does not say what action to take.
  • C. A chain of command shows reporting lines, not the response to a metric going out of specification.
  • D. Defect Analysis Plan is not the standard companion document to a Control Plan.

Question 86

The Control Limits width varies if the sample size varies for which type of chart?

  • A. P Charts
  • B. NP Charts
  • C. Xbar-R Charts
  • D. Time Series Charts
Show answer and explanation ▾

Correct answer: A

P chart control limits are calculated from the subgroup size. When the sample size changes from one subgroup to the next, the limits widen for small samples and narrow for large ones.

Why the other options are wrong:

  • B. NP charts require a constant sample size, so their limits stay fixed.
  • C. Xbar-R charts are normally run with a constant subgroup size and fixed limits.
  • D. Time Series Charts have no control limits.

Question 87

Which of these elements are not included in Implementation plans?

  • A. Work breakdown structure
  • B. Risk management plans
  • C. Cost/Benefit ratios
  • D. Planned audits of work completion
Show answer and explanation ▾

Correct answer: C

An implementation plan covers how the solution will be rolled out: the work breakdown, the risks and the checks on completed work. Cost/benefit ratios belong to the earlier step of selecting and justifying the solution.

Why the other options are wrong:

  • A. A work breakdown structure is a standard part of an implementation plan.
  • B. Risk management plans are a standard part of an implementation plan.
  • D. Planned audits of work completion are a standard part of an implementation plan.

Question 88

Upon completion and validation of an improvement to a process a Belt and the Project Team create a Control Plan that contains which of these?

  • A. Standard operating work description of the process change
  • B. Description of the monitoring system in place to assure continued compliance
  • C. Summary of the targeted critical metrics for process performance measurement
  • D. All of the above
Show answer and explanation ▾

Correct answer: D

A Control Plan documents how the improved process is run and held in place. It includes the standard work for the change, the monitoring system that assures continued compliance, and the critical metrics used to measure performance.

Why the other options are wrong:

  • A. The standard operating work description is only one part of the Control Plan.
  • B. The monitoring system description is only one part of the Control Plan.
  • C. The summary of critical metrics is only one part of the Control Plan.

Question 89

How many experimental runs exist in a Full Factorial and fully randomized design for 4 factors with 2 replicates for the Corner Points and no Center Points? The factors in the experiment are only at 2-levels.

  • A. 10
  • B. 32
  • C. 256
  • D. 64
Show answer and explanation ▾

Correct answer: B

A 2-level full factorial with 4 factors has 2 to the power of 4, which is 16 corner point combinations. Running 2 replicates of each gives 16 x 2 = 32 experimental runs.

Why the other options are wrong:

  • A. 10 is far fewer than the 16 combinations needed for even one replicate.
  • C. 256 is 4 to the power of 4, which would be four factors at four levels.
  • D. 64 would need four replicates, or six factors in a single replicate.

Question 90

If an experiment has 5 factors and no replicates for a 2-level Experimental Design with 16 experimental runs which statement is incorrect?

  • A. The Experimental Design is half-fractional
  • B. The Main Effects are confounded with only 4-way interactions
  • C. The Main Effects for the 5 factors are not aliased or confounded but the 2-way interactions are confounded with the 3-way interactions
  • D. The experiment has 8 experimental runs with the first factor at the high level
Show answer and explanation ▾

Correct answer: C

Five factors in 16 runs is a half fraction of resolution V. Its main effects are confounded with 4-way interactions and its 2-way interactions with 3-way interactions, so the claim that the main effects are not aliased or confounded is the incorrect statement.

Why the other options are wrong:

  • A. 16 runs is half of the 32 runs in the full design, so this statement is correct.
  • B. Main effects are confounded with 4-way interactions in this design, so this statement is correct.
  • D. Each factor is at the high level in half of the 16 runs, which is 8, so this statement is correct.

Question 91

Fractional Factorial, ____________and Response Surface Method are types of planned experiments.

  • A. Multi-Vari Analysis
  • B. Baldridge Channels
  • C. One Factor at a Time or OFAT
  • D. Factorial Design
Show answer and explanation ▾

Correct answer: D

Planned experiments deliberately set factor levels in a structured design. Full Factorial Design sits alongside Fractional Factorial and Response Surface Method as the main types.

Why the other options are wrong:

  • A. Multi-Vari Analysis is a passive study of existing process variation, not a planned experiment.
  • B. Baldridge Channels is not a type of designed experiment.
  • C. One Factor at a Time is the unstructured approach that designed experiments replace.

Question 92

Relative to a Design of Experiments the term ________________ refers to variables being a linear combination of each other.

  • A. Mirror Image
  • B. Directly Parallel
  • C. Collinear
  • D. None of the above
Show answer and explanation ▾

Correct answer: C

Variables are collinear when one can be written as a linear combination of the others. Their effects then cannot be separated in the analysis.

Why the other options are wrong:

  • A. Mirror Image is not the term for variables that are linear combinations of each other.
  • B. Directly Parallel is not a Design of Experiments term.
  • D. Collinear is the correct term and it is listed.

Question 93

Which statement(s) are incorrect about Fractional Factorial Designs?

  • A. A Half Fractional Design for 5 factors has the same number of experimental runs as a Full Factorial Design for 4 factors assuming no repeats or replicates or Center Points
  • B. Quarter Fractional experiments can exist for those with 4 factors
  • C. Resolution V design is desired while controlling costs of experimentation
  • D. Half Fractional experiments do not exist for those designs with only 2 factors
Show answer and explanation ▾

Correct answer: C

Higher resolution needs more runs. A Resolution V design keeps main effects and 2-way interactions clear but is costly to run, so it is not the design chosen when the aim is to control the cost of experimentation.

Why the other options are wrong:

  • A. A half fraction of 5 factors has 16 runs, the same as a full factorial of 4 factors, so this statement is correct.
  • B. A quarter fraction of a 4-factor design can be constructed, so this statement is correct.
  • D. Halving a 2-factor design leaves too few runs to separate the two factors, so this statement is correct.

Question 94

If in an experiment all possible variable pairs sum to zero the design is Orthogonal.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

The statement is true. A design is orthogonal when, for every pair of factor columns, the products of the coded levels sum to zero. Each effect can then be estimated independently of the others.

Why the other options are wrong:

  • B. Every pair of columns summing to zero is exactly the condition that defines an orthogonal design.

Question 95

Which Experimental Design typically is most associated with the fewest number of input variables or factors in the design?

  • A. Fractional Factorial design
  • B. Full Factorial design
  • C. Simple Linear Regression
  • D. Response Surface Design
Show answer and explanation ▾

Correct answer: D

Response Surface Designs are used late in experimentation to optimize the few factors that earlier screening showed to matter. Because they add extra levels to model curvature, they are normally run with only a small number of factors.

Why the other options are wrong:

  • A. Fractional Factorial designs are used to screen a large number of factors.
  • B. Full Factorial designs commonly handle more factors than a Response Surface Design.
  • C. Simple Linear Regression is an analysis method, not an Experimental Design.

Question 96

The method of Steepest Ascent guides you toward a target inside the original inference space.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

The method of Steepest Ascent moves the experiment step by step along the direction of greatest improvement, away from the original design region and toward the optimum. It therefore leads outside the original inference space, so the statement is false.

Why the other options are wrong:

  • A. Steepest Ascent moves the experiment beyond the original inference space, not toward a target inside it.

Question 97

Situations where standardized work needs to be incorporated include all of these except _____________________.

  • A. Machines continually operating to reduce the labor cost per piece
  • B. Lack of a system to assure proper inventory levels at repair stations
  • C. Changeover instructions incomplete
  • D. Process flow for the same product assembly taking various cycle time for completion
Show answer and explanation ▾

Correct answer: A

Standardized work fixes the best known method so tasks are done the same way every time. Running machines continually to lower labor cost per piece is a production scheduling choice that leads to overproduction, not a gap in work methods.

Why the other options are wrong:

  • B. A missing system for keeping proper inventory levels at repair stations is a gap that a standard method closes.
  • C. Incomplete changeover instructions call for documented, standardized steps.
  • D. Varying cycle times for the same assembly show the work is not being done one standard way.

Question 98

The Lean toolbox includes all of these items except ________________.

  • A. Mistake Proofing
  • B. Visual Factory
  • C. Design of Experiments
  • D. Inventory Management
Show answer and explanation ▾

Correct answer: C

Design of Experiments is a statistical Six Sigma tool used in the Improve Phase to model how inputs affect an output. It is not one of the Lean tools, which focus on waste, flow and workplace organization.

Why the other options are wrong:

  • A. Mistake Proofing is a Lean tool that prevents errors from being made or passed on.
  • B. Visual Factory is a Lean tool that makes the status of the workplace obvious at a glance.
  • D. Inventory Management is part of the Lean toolbox because excess inventory is a form of waste.

Question 99

Questions that can be best answered by a Visual Factory include all of these except _____________________.

  • A. Are downtime issues easily noted?
  • B. Can extra inventory be seen easily?
  • C. Are unneeded tools or supplies easily noted?
  • D. Are setups optimized for lower scrap levels?
Show answer and explanation ▾

Correct answer: D

A Visual Factory makes conditions visible at a glance, such as downtime, extra inventory and unneeded tools. Whether setups are optimized for lower scrap needs data analysis and experimentation, not a visual check.

Why the other options are wrong:

  • A. Downtime is made obvious by visual signals such as status lights and boards.
  • B. Extra inventory stands out when storage locations and limits are visually marked.
  • C. Unneeded tools or supplies are easy to spot in a visually organized workplace.

Question 100

If a Six Sigma project was to reduce repair station inventory and the team found the inventory was creeping up over time which Lean tools should be considered in the Control Phase to reestablish and sustain the project success?

  • A. Review the Visual Factory to assure inventory in excess of desired visible
  • B. Improve the lighting to assure adequate visibility
  • C. Analyze data from supplier deliveries
  • D. Reword the standardized work instructions to use active verbs and not passive phrases
Show answer and explanation ▾

Correct answer: A

Inventory creeps up when nobody can see that it has passed the agreed level. Reviewing the Visual Factory so that any inventory above the desired amount is immediately visible lets the team react and hold the gain.

Why the other options are wrong:

  • B. Better lighting does not show whether inventory has passed its limit.
  • C. Analyzing supplier delivery data does not create an ongoing control on repair station inventory.
  • D. Rewording work instructions does not make excess inventory visible.

Question 101

When a Belt implements an improvement that is automated thus requiring no particular understanding for use he has applied which Lean tool?

  • A. Mistake Proofing
  • B. Kaizen Event
  • C. 5S
  • D. None
Show answer and explanation ▾

Correct answer: A

Mistake Proofing builds the correct outcome into the process so it does not depend on the knowledge or attention of the person using it. An automated improvement that needs no particular understanding is a clear example.

Why the other options are wrong:

  • B. A Kaizen Event is a short team improvement activity, not an automated safeguard.
  • C. 5S organizes and maintains the workplace and does not automate the task.
  • D. A Lean tool has been applied here, so None does not fit.

Question 102

Kaizens or Kaikakus and Six Sigma projects are intended to create incremental process improvements versus breakthrough, significant improvements.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

Kaizen means small, continuous, incremental improvement, but Kaikaku means radical change and Six Sigma projects are aimed at breakthrough improvement. Grouping all three as incremental makes the statement false.

Why the other options are wrong:

  • A. Kaikaku and Six Sigma projects target breakthrough improvement, not incremental change.

Question 103

Which of these items contribute to what is necessary for successful Kaizen events?

  • A. Analysis tools
  • B. Management support
  • C. Operator support
  • D. All of these answers are correct
Show answer and explanation ▾

Correct answer: D

A Kaizen event succeeds when the team has analysis tools to find the waste, management support to free up time and resources, and operator support to make and keep the changes. All three are needed.

Why the other options are wrong:

  • A. Analysis tools are only one of the listed needs.
  • B. Management support is only one of the listed needs.
  • C. Operator support is only one of the listed needs.

Question 104

Kanban establishes a means of monitoring production, conveyance and delivery information such that efficient flow is established. The method used by Kanban is to require a _________ before anything moves.

  • A. Sign-off
  • B. Signal
  • C. Bell to ring
  • D. Work order
Show answer and explanation ▾

Correct answer: B

Kanban is a pull system in which nothing is produced or moved until a signal, such as a card or an empty container, shows that the next step needs it. This keeps flow matched to actual demand.

Why the other options are wrong:

  • A. A sign-off is an approval step, not the trigger Kanban uses.
  • C. A ringing bell is not the method Kanban defines for authorizing movement.
  • D. A work order pushes work by schedule, which is the opposite of the Kanban pull method.

Question 105

When a Belt decides to use written procedures and visual controls to improve the consistency of the tasks that must occur in the process he is improving he has utilized the _________________ activity of 5S.

  • A. Sustaining
  • B. Sorting
  • C. Standardizing
  • D. Straightening
Show answer and explanation ▾

Correct answer: C

Standardizing is the 5S activity that sets written procedures and visual controls so tasks are carried out the same way each time. It locks in the gains from the earlier 5S activities.

Why the other options are wrong:

  • A. Sustaining is about the discipline to keep following the standards over time.
  • B. Sorting removes items that are not needed in the work area.
  • D. Straightening arranges the needed items so each has a set place.

Question 106

Fractional Factorial designs are used to reduce the ____________ because the number of runs has been lowered.

  • A. Time and cost of experiments
  • B. Number of people involved
  • C. Number of data measurement points
  • D. Output summary
Show answer and explanation ▾

Correct answer: A

A Fractional Factorial design runs only a chosen fraction of the full set of treatment combinations. Fewer runs mean less time and lower cost for the experiment.

Why the other options are wrong:

  • B. The number of people involved is not what a Fractional Factorial design reduces.
  • C. The design reduces the number of runs, not the measurements taken in each run.
  • D. The output summary is not what the design is meant to reduce.

Question 107

Fractional Factorial Designs are used to analyze factors to model the output as a function of inputs if Hypothesis Testing in the Analyze Phase was inadequate to sufficiently narrow the factors that significantly impact the output(s).

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

When Hypothesis Testing in the Analyze Phase leaves too many candidate factors, a Fractional Factorial design tests them together in a small number of runs. The results show which factors matter and model the output as a function of the inputs.

Why the other options are wrong:

  • B. Fractional Factorial designs are used for exactly this purpose, so the statement is not false.

Question 108

A Factorial Experiment based on a Level 2 Design with 6 factors would require 16 runs to fully assess the interactions.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

A 2 level design with 6 factors needs 2 to the power of 6, which is 64 runs, to assess all interactions. Sixteen runs would only be a quarter fraction of that design, so the statement is false.

Why the other options are wrong:

  • A. Sixteen runs is a fraction of the 64 runs needed to fully assess the interactions.

Question 109

A Full Factorial experiment using a 3 level 3 factor approach has been proposed to test the viability of an extrusion machine experiment.

How many treatment combinations will this approach involve?

  • A. 6
  • B. 9
  • C. 27
  • D. 54
Show answer and explanation ▾

Correct answer: C

The number of treatment combinations in a Full Factorial experiment is the number of levels raised to the power of the number of factors. With 3 levels and 3 factors that is 3 x 3 x 3 = 27.

Why the other options are wrong:

  • A. 6 comes from adding 3 and 3, not from levels raised to the number of factors.
  • B. 9 is 3 squared, which covers only two factors at 3 levels.
  • D. 54 is double the 27 combinations the design requires.

Question 110

Screening experiments are the proper choice when a Belt is faced with the situation of highly Fractional Factorial Designs.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

Screening experiments use highly Fractional Factorial Designs to test many factors in very few runs. Their purpose is to find the few factors that matter before running a more detailed experiment.

Why the other options are wrong:

  • B. Highly Fractional Factorial Designs are the standard form of a screening experiment, so the statement is not false.

Question 111

A valid mathematical Regression represents all of the characteristics shown except ___________________________.

  • A. The residuals when plotted follow a Normal Distribution
  • B. The sum of the residuals is zero
  • C. All of the standardized residuals will be within ±3 Standard Deviations
  • D. Most standardized residuals are within ±2 Standard Deviations
Show answer and explanation ▾

Correct answer: A

Fitting a regression does not make its residuals Normal. Normality is an assumption that has to be checked on a plot of the residuals, while the other statements describe what the residuals of a sound fitted model look like.

Why the other options are wrong:

  • B. Least squares fitting with an intercept makes the residuals sum to zero.
  • C. A standardized residual beyond 3 Standard Deviations is treated as an outlier, so a sound model is expected to keep them inside that range.
  • D. In a sound model the large majority of standardized residuals are expected within 2 Standard Deviations.

Question 112

When a Belt conducts a Linear Correlation Analysis and finds that as an X increases the Y also increase then he has proven a ______________ correlation.

  • A. Negative
  • B. Positive
  • C. Monomial
  • D. Single alignment
Show answer and explanation ▾

Correct answer: B

A positive correlation means the two variables move in the same direction: as X increases, Y increases as well. The Correlation Coefficient is greater than zero in this case.

Why the other options are wrong:

  • A. A negative correlation means Y decreases as X increases.
  • C. Monomial is an algebra term, not a type of correlation.
  • D. Single alignment is not a term used to describe correlation.

Question 113

A valid Multiple Linear Regression (MLR) is characterized by all of these except _______________________.

  • A. It is an assumption that the X's (inputs) are not correlated to each other
  • B. The X's (inputs) are assumed to be independent of each other
  • C. MLR is conducted based on a deliberate form of experimentation
  • D. The Residuals from MLR analysis have to be Normally Distributed
Show answer and explanation ▾

Correct answer: C

Multiple Linear Regression is normally run on historical or observed process data, where the inputs vary on their own. Deliberate, planned changes to the inputs are the mark of a Designed Experiment, not of MLR.

Why the other options are wrong:

  • A. MLR assumes the inputs are not correlated with each other.
  • B. Independence of the inputs is a standard MLR assumption.
  • D. Normally Distributed Residuals are a standard requirement of a valid MLR.

Question 114

It would be more likely than not for a Belt conducting a Regression Analysis to find that the _______________________.

  • A. r2 value is smaller than the absolute value of r
  • B. Correlation Coefficient equals r2
  • C. Coefficient of Determination is less than r2
  • D. Correlation Coefficient equals r divided by 2
Show answer and explanation ▾

Correct answer: A

The Correlation Coefficient r lies between -1 and 1, and the Coefficient of Determination is r squared. Squaring a number whose absolute value is below 1 gives a smaller number, so r squared is smaller than the absolute value of r in almost every real analysis.

Why the other options are wrong:

  • B. The Correlation Coefficient is r, not r squared.
  • C. The Coefficient of Determination is r squared, so it cannot be less than r squared.
  • D. The Correlation Coefficient is r itself, not r divided by 2.

Question 115

When a Belt properly analyzes the results of an experiment he must examine the Residuals in expectation of finding all of the following except _________.

  • A. Some Residuals higher than others
  • B. Some Residuals lower than others
  • C. All Residuals within 2 Standard Deviations of the Mean
  • D. Residuals will represent a Linear Regression
Show answer and explanation ▾

Correct answer: D

Residuals are what is left after the model has explained the data, so they should be scattered randomly around zero with no pattern. Residuals that follow a line show the model has missed part of the relationship.

Why the other options are wrong:

  • A. Residuals naturally vary, so some will be higher than others.
  • B. Residuals naturally vary, so some will be lower than others.
  • C. Checking that residuals stay close to the mean, without extreme values, is part of residual analysis.

Question 116

Which of these might contribute to similar distributions having Unequal Variance?

  • A. Extreme tails
  • B. Outliers
  • C. Multiple Modes
  • D. All of the above
Show answer and explanation ▾

Correct answer: D

Variance measures how far values spread from the mean. Extreme tails, outliers and multiple modes each add spread to one data set, so any of them can make otherwise similar distributions show Unequal Variance.

Why the other options are wrong:

  • A. Extreme tails are only one of the listed contributors.
  • B. Outliers are only one of the listed contributors.
  • C. Multiple Modes are only one of the listed contributors.

Question 117

A Belt working in a supply chain environment has to make a decision to change suppliers of critical raw materials for a new product upgrade. The purchasing manager is depending on the Belt's effort requiring that the average cost of an internal critical raw material component be less than or equal to $4,200 in order to stay within budget. Using a sample of 35 first article components, a Mean of the new product upgrade price of $4,060, and a Standard Deviation of $98 was estimated. The Alternative Hypothesis in the above example is?

  • A. The Standard Deviation is equal to $300.
  • B. The Mean is less than $4,320.
  • C. The Mean is equal to $4,060.
  • D. The Mean is less than $4,200.
  • E. The Mean is greater than $ 4,200.
Show answer and explanation ▾

Correct answer: E

The requirement that the average cost be less than or equal to $4,200 contains the equality, so it is the Null Hypothesis. The Alternative Hypothesis is the opposite claim, that the Mean is greater than $4,200.

Why the other options are wrong:

  • A. The hypothesis is about the Mean cost, not a Standard Deviation of $300.
  • B. $4,320 is not a value given in the scenario.
  • C. $4,060 is the sample Mean, not the value being tested.
  • D. A Mean below $4,200 sits inside the budget requirement that forms the Null Hypothesis.

Question 118

A Belt working in a supply chain environment has to make a decision to change suppliers of critical raw materials for a new product upgrade. The purchasing manager is depending on the Belt's effort requiring that the average cost of an internal critical raw material component be less than or equal to $3,600 in order to stay within budget. Using a sample of 42 first article components, a Mean of the new product upgrade price of $3,200 and a Standard Deviation of $180 was estimated. Based on the data provided, the Z value for the data assuming a Normal Distribution is?

  • A. 1.11
  • B. 2.22
  • C. 4.30
  • D. 5.42
Show answer and explanation ▾

Correct answer: B

The Z value is the distance from the Mean to the limit divided by the Standard Deviation. Here that is (3,600 - 3,200) / 180 = 400 / 180 = 2.22.

Why the other options are wrong:

  • A. 1.11 is half of the correct value.
  • C. 4.30 does not result from 400 divided by 180.
  • D. 5.42 does not result from 400 divided by 180.

Question 119

A Belt working in a supply chain environment has to make a decision to change suppliers of critical raw materials for a new product upgrade. The purchasing manager is depending on the Belt's effort requiring that the average cost of an internal critical raw material component be less than or equal to $3,800 in order to stay within budget. Using a sample of 38 first article components, a Mean of the new product upgrade price of $3,680, and a Standard Deviation of $120 was estimated. In order to increase the Long Term Z value to 5, what is the maximum long term variation in pricing the Belt can accept for his upgraded critical raw material component?

  • A. $6
  • B. $12
  • C. $24
  • D. $48
Show answer and explanation ▾

Correct answer: C

Z is the distance from the Mean to the limit divided by the Standard Deviation. The distance is 3,800 - 3,680 = 120, so a Z of 5 needs a Standard Deviation of no more than 120 / 5 = $24.

Why the other options are wrong:

  • A. $6 would give a Z of 20, far tighter than the maximum variation allowed.
  • B. $12 would give a Z of 10, tighter than the maximum variation allowed.
  • D. $48 would give a Z of only 2.5.

Question 120

Sally and Sara sell flower pots at their garage sale. Sally motivates Sara mentioning that they will sell a minimum of 15 pots per day if the outside temperature exceeds 60o F. From a sample, whose population is assumed to follow a Normal Distribution, taken for 30 days at 60 degrees or more an average of 13.6 pots per day were sold with a Standard Deviation of 0.7 pots. For the sales accomplished above, what test would validate if they met their requirements?

  • A. F Test
  • B. Test for Equal Variance
  • C. Chi Square Test
  • D. One-Sample t-Test
Show answer and explanation ▾

Correct answer: D

One sample Mean, 13.6 pots per day, is being compared with a single target value of 15, and the Standard Deviation comes from the sample. That is the situation the One-Sample t-Test is built for.

Why the other options are wrong:

  • A. An F Test compares two variances, not a Mean against a target.
  • B. A Test for Equal Variance compares the spread of two or more groups.
  • C. A Chi Square Test is used for counts or a variance, not for comparing a Mean with a target.

Question 121

Sally and Sara sell flower pots at their garage sale. Sally motivates Sara mentioning that they will sell a minimum of 15 pots per day if the outside temperature exceeds 60o F. From a sample, whose population is assumed to follow a Normal Distribution, taken for 30 days at 60 degrees or more an average of 13.6 pots per day were sold with a Standard Deviation of 0.7 pots. The statistical Degrees of Freedom for this example are?

  • A. 1
  • B. 29
  • C. 30
  • D. 31
  • E. 2
Show answer and explanation ▾

Correct answer: B

For a One-Sample t-Test the Degrees of Freedom equal the sample size minus one. With 30 days of data that is 30 - 1 = 29.

Why the other options are wrong:

  • A. 1 is not the sample size minus one.
  • C. 30 is the sample size, before one is subtracted.
  • D. 31 adds one to the sample size when one should be subtracted.
  • E. 2 is not the sample size minus one.

Question 122

Sally and Sara sell flower pots at their garage sale. Martha motivates Rose mentioning that they will sell a minimum of 16 pots per day if the outside temperature exceeds 60o F. From a sample, whose population is assumed to follow a Normal Distribution, taken for 30 days at 60 degrees or more an average of 15.2 pots per day were sold with a Standard Deviation of 0.6 pots.

What is the Z value for this sales process?

  • A. 0.67
  • B. 1.13
  • C. 1.33
  • D. 2.66
Show answer and explanation ▾

Correct answer: C

The Z value is the distance between the requirement and the Mean divided by the Standard Deviation. Here that is (16 - 15.2) / 0.6 = 0.8 / 0.6 = 1.33.

Why the other options are wrong:

  • A. 0.67 is half of the correct value.
  • B. 1.13 does not result from 0.8 divided by 0.6.
  • D. 2.66 is double the correct value.

Question 123

The relationship between a response variable and one or more independent variables is investigated and modeled by use of __________________.

  • A. X-Y Matrix
  • B. Baldridge Assessment
  • C. Analysis of Variance (ANOVA)
  • D. Critical X's Definition
Show answer and explanation ▾

Correct answer: C

Analysis of Variance tests whether changes in one or more independent variables produce real differences in a response variable. It splits the total variation into the part explained by the inputs and the part left as error, which is the basis for modeling the relationship.

Why the other options are wrong:

  • A. An X-Y Matrix ranks inputs against outputs using team scores and does not model the relationship statistically.
  • B. A Baldridge Assessment evaluates an organization's overall performance excellence.
  • D. Critical X's Definition names the important inputs and does not model how they relate to the response.

That was 123 of 400.

The full Lean Six Sigma Master Black Belt pack has all 400 questions, each with the answer, the explanation and why the other options are wrong, plus a questions-only copy for timed runs. $39.00 USD, paid once, with free monthly updates and a pass-or-your-money-back guarantee.

Get the full pack

Not ready yet? Get the first 149 free as one PDF by email. Get the Half Pack →

About the Master Black Belt exam

These 123 Lean Six Sigma Master Black Belt questions are free here. The free PDF has 149, free by email.

They are drawn from the same bank as the full Master Black Belt pack, which has 400 questions in total.

Back to blog