ASQ CSSGB Practice Questions with Explanations

Free ASQ CSSGB practice questions. 48 of them, each with the correct answer, a full explanation, and the reason every other option is wrong.

CSSGB practice questions

Question 1

The use of station warning lights, tool boards and jidohka devices in the application of Lean accomplish which of these principles?

  • A. Pilferage Minimization
  • B. Visual Factory
  • C. Management Awareness
  • D. Operator Attentiveness
Show answer and explanation ▾

Correct answer: B

Station warning lights, tool boards and jidoka devices make the state of work visible at a glance. That is the visual factory idea in lean.

Why the other options are wrong:

  • A. Pilferage minimization is about preventing theft, not making work status visible.
  • C. Management awareness is not the lean principle these visual tools serve.
  • D. Operator attentiveness is not the principle that visual tools and jidoka devices implement.

Question 2

A Lean Principle that addresses efficiency by the process worker is called ____________________?

  • A. Visual Factory
  • B. Supervising
  • C. Training
  • D. Standardizing
Show answer and explanation ▾

Correct answer: D

Standardizing is the lean principle about making work efficient for the process worker by defining one best-known way of doing a task.

Why the other options are wrong:

  • A. Visual factory makes status visible, which is not the same as the worker efficiency principle.
  • B. Supervising is not a named lean principle for worker efficiency.
  • C. Training supports standard work but is not the principle the question names.

Question 3

While management of a company must set the stage for all improvement efforts, which of these 5S’s is primarily driven by management?

  • A. Straighten
  • B. Sort
  • C. Shine
  • D. Sustain
Show answer and explanation ▾

Correct answer: D

Sustain, the fifth S, is the step that needs management leadership, because holding the gains depends on audits, discipline and reinforcement from management.

Why the other options are wrong:

  • A. Straighten is carried out by the people who work in the area.
  • B. Sort is done by the work team removing what is not needed.
  • C. Shine is the cleaning step the team performs daily.

Question 4

As part of a Visual Factory plan __________ cards are created and utilized to identify areas in need of cleaning and organization.

  • A. Kanban
  • B. Kaizen
  • C. Poke-Yoke
  • D. WhoSai
Show answer and explanation ▾

Correct answer: A

Kanban means signal card, and it is the only card based visual signal among these choices. In a Visual Factory, cards posted in the workplace show at a glance where action is needed.

Why the other options are wrong:

  • B. Kaizen is continuous improvement activity, not a type of card.
  • C. Poke-Yoke is mistake proofing, not a card used to flag areas.
  • D. WhoSai is not a recognized lean tool.

Question 5

The use of Kanbans work best with pull systems for determining the timing of which products or services are produced.

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

Correct answer: A

Kanban signals what to make and when, which is how a pull system works: production starts only when downstream demand triggers it.

Why the other options are wrong:

  • B. The statement is correct, because kanban and pull systems work together.

Question 6

When a Belt applies the practice of Poka-Yoke to a project challenge she is attempting to make certain the activity is _______________ .

  • A. Well documented
  • B. Removed from the line
  • C. Mistake proofed
  • D. Highly visible
Show answer and explanation ▾

Correct answer: C

Poka-Yoke is the lean practice of mistake proofing: the activity is designed so that an error cannot be made, or is caught at once before it becomes a defect.

Why the other options are wrong:

  • A. Documentation records a process but does not stop errors from happening.
  • B. Poka-Yoke improves an activity where it is, it does not remove it from the line.
  • D. Making work highly visible is the aim of visual management, not Poka-Yoke.

Question 7

The Lean Principle action in the 5S approach that deals with having those items needed regularly at hand and those items need less regularly stored out of the way is known as ___________.

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

Correct answer: D

Sorting separates items by how often they are needed, so that what is used regularly stays in the work area and what is used less often is stored out of the way.

Why the other options are wrong:

  • A. Shining is cleaning the work area and equipment.
  • B. Standardizing sets common rules and routines for keeping the first steps in place.
  • C. Sustaining is the discipline of holding the gains over time.

Question 8

SPC on the outputs is more preferred than SPC on the inputs when implementing SPC for your process.

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

Correct answer: B

SPC is most effective on the process inputs, because controlling the inputs prevents defects before the output is made. Charting only the outputs detects a problem after it has already happened.

Why the other options are wrong:

  • A. Output charts only detect problems after the fact, so they are not preferred over input charts.

Question 9

Significant variation in process performance is a consequence of several causes that can be classified using which of the terminologies shown. (Note: There are 2 correct answers). (Choose two.)

  • A. Common
  • B. Random
  • C. Uneducated
  • D. Special
  • E. Vital
Show answer and explanation ▾

Correct answer: A, D

Process variation is classified as common cause, the natural variation built into the process, or special cause, variation from a specific assignable event outside the normal process.

Why the other options are wrong:

  • B. Random describes the behavior of common cause variation but is not one of the two cause classifications.
  • C. Uneducated is not a classification of variation.
  • E. Vital describes the few most important inputs, not a type of cause of variation.

Question 10

When it comes to Control one of the most effective means of eliminating defects is to _________________ .

  • A. Train personnel often and thoroughly
  • B. Keep a Six Sigma project going on the process at all times
  • C. Design defect prevention into the product
  • D. Have each process consist of no more than five steps
Show answer and explanation ▾

Correct answer: C

Designing defect prevention into the product removes the opportunity for the defect to occur, which is stronger than any control that relies on people detecting or avoiding errors.

Why the other options are wrong:

  • A. Training helps, but people can still make errors, so it is weaker than prevention by design.
  • B. A permanent project on the process is not a control method.
  • D. Limiting a process to five steps does not by itself eliminate defects.

Question 11

A periodic time frame can be used to arrange for Control Limit and Center Line calculations with good SPC implementation in a process.

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

Correct answer: A

With good SPC practice, Control Limits and the Center Line are calculated from process data and can be reviewed and recalculated on a planned periodic schedule so the chart keeps reflecting the current process.

Why the other options are wrong:

  • B. Recalculating limits on a planned periodic basis is an accepted part of SPC practice.

Question 12

The data on SPC charts are typically constructed such that they have the most recent data point on the right hand side.

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

Correct answer: A

SPC charts plot data in time order from left to right, so the most recent point sits on the right hand side and trends over time are easy to see.

Why the other options are wrong:

  • B. Control charts are time ordered left to right, so the newest point is on the right.

Question 13

Which statement(s) describe an undesirable situation when implementing SPC?

  • A. The lower Control Limit for the R chart is equal to zero
  • B. Attempt to use SPC for tracking transaction times at a warehouse
  • C. A process is in Statistical Control before implementation of SPC
  • D. The Control Limits are wider than the customer specification limits
Show answer and explanation ▾

Correct answer: D

Control Limits wider than the customer specification limits mean the natural variation of the process is larger than the customer allows, so even a stable process will produce out of specification output.

Why the other options are wrong:

  • A. A lower Control Limit of zero on an R chart is normal for small subgroup sizes.
  • B. SPC can be applied to transactional measures such as warehouse transaction times.
  • C. A process already in Statistical Control is a good starting point for SPC.

Question 14

If a process has Outliers which pair of charts is most preferable if subgroups will exist for the Continuous Data?

  • A. Individual, Moving Range
  • B. Xbar-R Charts
  • C. Xbar-S Charts
  • D. nP and P Charts
Show answer and explanation ▾

Correct answer: B

Xbar-R Charts are the standard pair for Continuous Data collected in subgroups: the Xbar chart tracks the subgroup averages and the R chart tracks the range within each subgroup.

Why the other options are wrong:

  • A. Individual and Moving Range charts are for single readings, not subgrouped data.
  • C. Xbar-S Charts are typically chosen for larger subgroup sizes.
  • D. nP and P Charts are for attribute data, not Continuous Data.

Question 15

After a Belt has put data through the smoothing process which chart would be used to look for trends in the data?

  • A. Moving Average Chart
  • B. Multi-Vari Chart
  • C. X bar Chart
  • D. Pareto Chart
Show answer and explanation ▾

Correct answer: A

A Moving Average Chart plots averages of successive data points, which smooths out short term noise and makes underlying trends in the data easier to see.

Why the other options are wrong:

  • B. A Multi-Vari Chart displays sources of variation, not smoothed trends.
  • C. An X bar Chart plots subgroup averages without smoothing across time.
  • D. A Pareto Chart ranks categories by frequency and does not show trends over time.

Question 16

A Belt concludes a Lean Six Sigma project with the creation of a Control Plan. At what point can the Control Plan be closed?

  • A. Never, a Control Plan is a living document
  • B. As soon as the Champion signs off
  • C. Within 30 days of the LSS project review team meeting
  • D. After the project has been presented at the recognition event
Show answer and explanation ▾

Correct answer: A

A Control Plan is a living document. It stays with the process owner and is updated whenever the process, its measurements or its reaction plans change, so it is never closed.

Why the other options are wrong:

  • B. Champion sign off closes the project, not the Control Plan.
  • C. There is no 30 day point at which a Control Plan ends.
  • D. A recognition event has no bearing on the life of the Control Plan.

Question 17

When analyzing a data set we frequently graph one metric as a function of another. If the slope of the Correlation line is -2.5 we would say the two metrics are ___________ correlated?

  • A. Positively
  • B. Not
  • C. Negatively
  • D. None
Show answer and explanation ▾

Correct answer: C

A negative slope means that as one metric increases the other decreases, which is a negative correlation. A slope of -2.5 shows the output falls 2.5 units for each unit increase in the input.

Why the other options are wrong:

  • A. A positive correlation needs a positive slope.
  • B. A nonzero slope shows the two metrics are related.
  • D. None does not describe a line with a clear negative slope.

Question 18

Multiple Linear Regressions (MLR) is best used when which of these are applicable? (Note: There are 3 correct answers). (Choose three.)

  • A. Non-linear relationships between the inputs X’s and output Y
  • B. Uncertainty in the slope of the linear relationship between an X and a Y
  • C. Relationships between Y (output) and more than one X (Input)
  • D. Preventing the use of a Designed Experiment if unnecessary
  • E. We assume that the X’s are independent of each other
Show answer and explanation ▾

Correct answer: C, D, E

Multiple Linear Regression models one output Y against more than one input X and assumes the X's are independent of each other. Because it can work from existing process data, it can avoid running a Designed Experiment that is not needed.

Why the other options are wrong:

  • A. Non-linear relationships call for a non-linear model, not a linear one.
  • B. Uncertainty about the slope between one X and one Y is a simple regression question.

Question 19

Fractional Factorial designs for an experimental approach are used when ____________ about the multiple metric interaction in a process.

  • A. Much is known
  • B. Little is known
  • C. We don’t care
  • D. Data exists
Show answer and explanation ▾

Correct answer: B

Fractional Factorial designs run only part of the full set of combinations, which makes them efficient screening experiments when little is known about which factors and interactions matter.

Why the other options are wrong:

  • A. When much is known, screening with a fractional design is not the need.
  • C. Not caring about interactions is not a basis for choosing an experimental design.
  • D. The existence of data does not determine the choice of a Fractional Factorial design.

Question 20

A Belt will occasionally do a quick experiment referred to as an OFAT which stands for ____________________.

  • A. Only a Few Are Tested
  • B. Opposite Factors Affect Technique
  • C. One Factor At a Time
  • D. Ordinary Fractional Approach Technique
Show answer and explanation ▾

Correct answer: C

OFAT stands for One Factor At a Time: a single factor is changed while all others are held constant. It is quick, but it cannot reveal interactions between factors.

Why the other options are wrong:

  • A. Only a Few Are Tested is not what OFAT stands for.
  • B. Opposite Factors Affect Technique is not a recognized term.
  • D. Ordinary Fractional Approach Technique is not a recognized term.

Question 21

The actual experimental response data varied somewhat from what a Belt had predicted them to be. This is the result of which of these?

  • A. Inefficiency of estimates
  • B. Residuals
  • C. Confounded data
  • D. Gap Analysis
Show answer and explanation ▾

Correct answer: B

A residual is the difference between an observed response and the value the model predicted for it, so the gap between actual and predicted data is the residuals.

Why the other options are wrong:

  • A. Inefficiency of estimates is not the name for the gap between observed and predicted values.
  • C. Confounding is when effects cannot be separated from one another.
  • D. Gap Analysis compares current and desired performance, not observed and predicted data points.

Question 22

A Six Sigma tool that helps to screen factors by using graphical techniques to logically subgroup multiple discrete X's plotted against a continuous Y is known as a _____________Chart.

  • A. SIPOC
  • B. Multi-Vari
  • C. Box Plot
  • D. Whisker
Show answer and explanation ▾

Correct answer: B

A Multi-Vari Chart is a graphical screening tool that plots a continuous Y against logical subgroups of several discrete X's, showing which sources of variation matter most.

Why the other options are wrong:

  • A. SIPOC is a high level process map, not a chart of X's against Y.
  • C. A Box Plot shows the distribution of data for a group, not logically subgrouped multiple X's.
  • D. Whisker is part of a Box Plot, not a factor screening chart.

Question 23

A primary benefit of using a Multi-Vari Chart is it provides a visual presentation of two-way interactions.

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

Correct answer: A

A Multi-Vari Chart shows the response across several factors on one graph, so a pattern that changes from one level of a factor to another gives a visual indication of a two-way interaction.

Why the other options are wrong:

  • B. Showing how factors act together on one graph is a recognized benefit of the Multi-Vari Chart.

Question 24

Bias in Sampling is an error due to lack of independence among random samples or due to systematic sampling procedures.

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

Correct answer: A

Sampling bias is a systematic error that arises when samples are not independent or when the sampling procedure favors some units over others, so the sample does not represent the population.

Why the other options are wrong:

  • B. Lack of independence and systematic sampling procedures are recognized sources of sampling bias.

Question 25

To draw inferences about a sample population being studied by modeling patterns of data in a way that accounts for randomness and uncertainty in the observations is known as ____________________.

  • A. Influential Analysis
  • B. Inferential Statistics
  • C. Physical Modeling
  • D. Sequential Inference
Show answer and explanation ▾

Correct answer: B

Inferential Statistics uses sample data to draw conclusions about a population, modeling the patterns in the data in a way that accounts for randomness and uncertainty.

Why the other options are wrong:

  • A. Influential Analysis is not the name of this branch of statistics.
  • C. Physical Modeling builds a physical representation, not a statistical inference.
  • D. Sequential Inference is not the term for drawing conclusions about a population from a sample.

Question 26

For a Normal Distribution the Mean, Median and Mode are the same data point.

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

Correct answer: A

A Normal Distribution is symmetric and has a single peak, so the Mean, Median and Mode all fall at the same central point.

Why the other options are wrong:

  • B. The symmetry and single peak of the Normal Distribution place all three measures at the center.

Question 27

When two Inputs have an impact on the Output together yet seem to have no or little impact on their own this is called a/an ________________.

  • A. Interaction
  • B. Oddity
  • C. Coincidence
  • D. Impossibility
Show answer and explanation ▾

Correct answer: A

An interaction exists when the effect of one input on the output depends on the level of another input, so the two together can have an impact that neither shows on its own.

Why the other options are wrong:

  • B. Oddity is not a statistical term for a combined effect.
  • C. Coincidence implies no real relationship, while an interaction is a real combined effect.
  • D. The effect is real and measurable, so it is not an impossibility.

Question 28

Hypothesis Testing can save time and help avoid high costs of experimental efforts by using existing data.

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

Correct answer: A

Hypothesis Testing can be run on data that already exists, which lets a team check suspected causes statistically before spending time and money on new experiments.

Why the other options are wrong:

  • B. Using existing data for Hypothesis Testing does reduce experimental time and cost.

Question 29

It is a Type II error if we decide to reject the Null Hypothesis when it is actually true.

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

Correct answer: B

Rejecting a Null Hypothesis that is actually true is a Type I error. A Type II error is the opposite: failing to reject a Null Hypothesis that is actually false.

Why the other options are wrong:

  • A. The situation described is a Type I error, not a Type II error.

Question 30

A Belt experienced an Alpha of .05 and a Beta of .10 and knew these are the most common risk levels when running a Statistical test.

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

Correct answer: A

An Alpha risk of .05 gives 95% confidence and a Beta risk of .10 gives 90% power, and these are commonly used risk levels for a statistical test.

Why the other options are wrong:

  • B. An Alpha of .05 and a Beta of .10 are conventional, widely used risk levels.

Question 31

Inferential Statistics is largely about Significance. There are both Practical and _______________ Significance to consider during an analysis of data in a Lean Six Sigma project.

  • A. Problematic
  • B. Impractical
  • C. Usable
  • D. Statistical
Show answer and explanation ▾

Correct answer: D

Statistical Significance asks whether a difference is unlikely to be due to chance, while Practical Significance asks whether the difference is large enough to matter to the business. Both must be considered.

Why the other options are wrong:

  • A. Problematic is not a type of significance.
  • B. Impractical is not a type of significance.
  • C. Usable is not a type of significance.

Question 32

The Central Limit Theorem helps us understand the ___________ we are taking and is the basis for using sampling to estimate population parameters.

  • A. Analysis
  • B. Kurtosis
  • C. Risk
  • D. Route
Show answer and explanation ▾

Correct answer: C

The Central Limit Theorem says sample means tend toward a Normal Distribution, which lets us quantify the risk of being wrong when a sample is used to estimate a population parameter.

Why the other options are wrong:

  • A. Analysis is too general and is not what the theorem quantifies.
  • B. Kurtosis measures the shape of the tails of a distribution.
  • D. Route is not a statistical concept.

Question 33

Hypothesis Tests determine the probabilities of differences between observed data and the hypothesis being solely due to _________ based on the result of the P-values.

  • A. Human error
  • B. Measurement error
  • C. Shift differences
  • D. Chance
Show answer and explanation ▾

Correct answer: D

A P-value is the probability of seeing a difference at least as large as the one observed if the Null Hypothesis is true, that is, a difference due solely to chance.

Why the other options are wrong:

  • A. A P-value does not measure human error.
  • B. Measurement error is assessed with Measurement System Analysis, not a P-value.
  • C. Shift differences may be the subject of a test but are not what the P-value measures.

Question 34

The Alpha level of a test (level of significance) represents the yardstick against which P-values are measured and the Null Hypothesis is rejected if the P-value is which of these?

  • A. Less than the Alpha level.
  • B. Greater than the Alpha level.
  • C. Greater than the Beta and Alpha level.
  • D. Less than one minus Alpha.
  • E. Less than the power of one minus Beta.
Show answer and explanation ▾

Correct answer: A

The Null Hypothesis is rejected when the P-value is less than the Alpha level, because a result that extreme would then occur by chance less often than the Alpha risk accepted for the test.

Why the other options are wrong:

  • B. A P-value greater than Alpha means we fail to reject the Null Hypothesis.
  • C. Beta is not part of the rejection rule.
  • D. One minus Alpha is the confidence level, not the rejection threshold.
  • E. Power, one minus Beta, is not the threshold for rejecting the Null Hypothesis.

Question 35

A 1-Sample t-test is used when you want to compare the Median of one distribution to a target value.

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

Correct answer: B

A 1-Sample t-test compares the Mean of one sample to a target value, not the Median. Medians are compared with non-parametric tests.

Why the other options are wrong:

  • A. The t-test is a test of the Mean, so the statement is not correct.

Question 36

When a Belt is analyzing sample data she should keep in mind that 95% of Normally Distributed data is within +/- 2 Standard Deviations from the Mean.

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

Correct answer: A

For Normally Distributed data, about 95% of values fall within plus or minus 2 Standard Deviations of the Mean.

Why the other options are wrong:

  • B. The 95% within 2 Standard Deviations figure is the standard rule for Normal data.

Question 37

The Standard Deviation for the distribution of Means is called the ___________ and approaches zero as the sample size reaches 30.

  • A. Standard Error
  • B. Mean Deviation
  • C. Mean Spread
  • D. Mean Error
Show answer and explanation ▾

Correct answer: A

The Standard Deviation of the distribution of sample Means is the Standard Error. It equals the Standard Deviation divided by the square root of the sample size, so it shrinks as the sample size grows.

Why the other options are wrong:

  • B. Mean Deviation is the average absolute distance of values from the Mean.
  • C. Mean Spread is not a statistical term.
  • D. Mean Error is not the name for the Standard Deviation of sample Means.

Question 38

According to a manager it takes an average weekday commute of 39 minutes with a Standard Deviation of 7 minutes for the employees to get to work when they use their personal vehicles for their office commute while management set a policy of not more than 40 minutes for their daily one-way commute. A survey conducted one day on 70 employees showed an average of 34 minutes commuting time using the metro public transportation system with a Standard Deviation of 21 minutes. For the employees choosing to increase their chances to come on time using personal transportation their variation should be reduced to ___________?

  • A. 1 minute
  • B. 6 minutes
  • C. 3.5 minutes
  • D. Eliminate it to 0.0 minutes
Show answer and explanation ▾

Correct answer: C

With a Standard Deviation of 3.5 minutes, the 40 minute limit sits about 0.29 Standard Deviations above the 39 minute Mean, the same margin the metro has with 6 minutes over 21, so the on time probability matches the metro.

Why the other options are wrong:

  • A. 1 minute is not the reduction that brings personal vehicles level with the metro.
  • B. At 6 minutes the margin to the 40 minute limit is still smaller than the metro's.
  • D. Eliminating variation completely is not achievable for a commute.

Question 39

According to a manager it takes an average weekday commute of 39 minutes with a Standard Deviation of 7 minutes for the employees to get to work while they use their personal vehicles for their office commute while the management set a policy of not more than 40 minutes for their daily one-way commute. A survey conducted one day on 70 employees showed an average of 34 minutes commuting time using the metro public transportation system with a Standard Deviation of 21 minutes. If the Standard Deviation is uncontrollable then the other option to increase the probability of coming in on time via personal vehicles to work could be ____________?

  • A. Increase the average time of commute
  • B. Maintain the average time of commute and change route to work
  • C. Reduce average commute time to work by departing earlier
  • D. Change policy at work and request for flexible times based on location
Show answer and explanation ▾

Correct answer: C

If the Standard Deviation cannot be changed, the only other lever is the Mean. Lowering the average commute time moves the whole distribution further below the 40 minute limit, which raises the probability of arriving on time.

Why the other options are wrong:

  • A. Increasing the average pushes more commutes over the limit.
  • B. Keeping the same average leaves the probability unchanged.
  • D. Changing the policy does not change the probability of meeting the commute time.

Question 40

Which of the following is used to test the significance for the analysis of a Variance Table?

  • A. t Test
  • B. F Test
  • C. Chi Square Test
  • D. Acid Test
Show answer and explanation ▾

Correct answer: B

An Analysis of Variance table is tested with the F Test, which compares the variation between groups to the variation within groups.

Why the other options are wrong:

  • A. The t Test compares one or two Means and is not the test in a variance table.
  • C. The Chi Square Test is used for counts and proportions.
  • D. Acid Test is not a statistical test.

Question 41

Non-parametric testing is done when which of these are applicable? (Note: There are 3 correct answers). (Choose three.)

  • A. When the traditional t tests don’t produce the results we need
  • B. A Hypothesis Test for the Median of the population is in question
  • C. It does not require data to come from Normally Distributed populations
  • D. They look at the Median rather than the Mean of populations
  • E. When there are no parameters to measure in the process
Show answer and explanation ▾

Correct answer: B, C, D

Non-parametric tests do not require the data to come from Normally Distributed populations, and they work with the Median rather than the Mean, so they are the choice when a hypothesis about a population Median is in question.

Why the other options are wrong:

  • A. A test is chosen for the type of data, not because another test gave an unwanted result.
  • E. Non-parametric does not mean the process has no parameters to measure.

Question 42

The Mann-Whitney Test is used to test if the Means for two samples are different.

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

Correct answer: B

The Mann-Whitney Test is a non-parametric test that compares the Medians of two samples, not the Means. A 2-Sample t-test is the test for two Means.

Why the other options are wrong:

  • A. Mann-Whitney works on Medians, so the statement is not correct.

Question 43

Contingency Tables are used to perform which of these functions?

  • A. Illustrate one-tail proportions
  • B. Analyze the "what if" scenario
  • C. Contrast the Outliers under the tail
  • D. Compare more than two sample proportions with each other
Show answer and explanation ▾

Correct answer: D

A Contingency Table arranges counts by category and, with the Chi Square test, is used to compare more than two sample proportions with each other.

Why the other options are wrong:

  • A. Illustrating one-tail proportions is not the purpose of a Contingency Table.
  • B. What if analysis is not what a Contingency Table does.
  • C. Contingency Tables do not deal with Outliers.

Question 44

Skewed, or Mixed, Distributions occur when data comes from several sources that are supposed to be the same yet are not.

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

Correct answer: A

Combining data from several sources that should be the same but are not gives a mixed distribution that departs from the symmetric Normal shape, which is the situation the statement describes.

Why the other options are wrong:

  • B. Pooling sources that differ is a recognized cause of these non-Normal distributions.

Question 45

The perfect sample size is the minimum number of data points required to provide exactly 6% overlap or risk if one wants a 95% confidence level.

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

Correct answer: B

Sample size is worked out from the confidence level, the variation in the data and the precision needed. There is no rule tying a perfect sample size to exactly 6% overlap.

Why the other options are wrong:

  • A. A fixed 6% overlap is not how sample size is determined for 95% confidence.

Question 46

Hypothesis Tests determine the probabilities of differences between observed data and the hypothesis being solely due to chance. This is determined based on the result of the ______________.

  • A. Random acts
  • B. P-values
  • C. Standard Deviations
  • D. R-values
Show answer and explanation ▾

Correct answer: B

The P-value is the probability that a difference as large as the one observed would occur by chance alone if the Null Hypothesis were true, so it is the result used to make the decision.

Why the other options are wrong:

  • A. Random acts is not a statistical result.
  • C. Standard Deviations measure spread, not the probability that a difference is due to chance.
  • D. R-values measure the strength of a correlation.

Question 47

It is a Type I error if we reject the Null Hypothesis when it is actually true.

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

Correct answer: A

A Type I error is rejecting a Null Hypothesis that is actually true. Its probability is the Alpha risk.

Why the other options are wrong:

  • B. Rejecting a true Null Hypothesis is the definition of a Type I error.

Question 48

Having an Alpha of .05 and a Beta of .10 are the most common risk levels when running a Statistical test.

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

Correct answer: A

An Alpha of .05 corresponds to 95% confidence and a Beta of .10 to 90% power, and these are commonly used risk levels for a statistical test.

Why the other options are wrong:

  • B. An Alpha of .05 and a Beta of .10 are conventional, widely used risk levels.

Get the complete CSSGB bank

These 48 questions are roughly 33% of the bank. The full pack has 144 real exam questions, each with the same depth of explanation, plus a questions-only PDF for timed practice and free updates forever.

View the full ASQ CSSGB question bank →

About the CSSGB exam

These 48 ASQ CSSGB questions are free here, and as the free sample PDF.

Back to blog