Lean Six Sigma - Green Belt
The focus of Six-Sigma is on Common Cause variation.
False
The greater is the variation of data around the mean, the smaller will be the standard deviation.
False
The two types of data that can be monitored include discrete and attribute.
False
The type of chart used in a control process is dependent only on the type of data - continuous or discrete - collected.
False
The variation of the distribution of sample means is the same as the standard deviation of the population.
False
X-Bar charts are used for small samples where the data are discrete.
False
A process has a mean of 150 and a standard deviation of 10. Six sigma limits will be 110 and 210.
False 150+6(10) = 210 and 150 - 6(10)=90
A sample is taken from the output of a process. The likelihood that this sample will fall within two standard errors of the mean from the process mean is 68%.
False 95 Percent
The difference between what a customer expects and what a process delivers is expressed by the difference between the LSL and USL.
False The difference is expressed by the distance between the LSL and LCL, and the distance between the USL and UCL
The Beta error associated with monitoring and controlling the radiation level in a nuclear reactor plant should be large.
False It needs to be a small as possible.
The number of good units or pieces produced divided by the number of total units that originally started through the process is measured by the _______ .
First Time Yield
When the Z Value is +3 which of the following best represents the area under the normal curve to the right of the Z value?
Less than 1 percent
Which of the following is the non-parametric equivalent of the Two Sample t Test?
Mann-Whitney
An R-Chart is used to ensure that the output from a process produces consistent results.
True
An alpha error detects a process change that is not present while a beta error fails to detect a process change that is present.
True
An alternative hypothesis states what it is you are trying to prove.
True
An organization wishes to determine if the variation across five processes is different. A multi-vari analysis could be used.
True
As long as a process is in-control and performing as expected there is not much that can be done to eliminate common cause variation short of redesigning the entire process.
True
As the sample size increases, any single sample result is very likely to be closer to the population mean.
True
As the sample size increases, the standard error of the mean will decrease.
True
Barring any accidents, the length of time it takes to commute to work is subject to common cause variation.
True
Brainstorming sessions are held to identify problems before they occur.
True
Consider that a P-Chart has been designed and is now ready for use. When entering sample results into the chart, the results will be entered in the form of a percentage. One result for online product return process, for example, would be 15% or 0.15.
True
Suppose it is critically important to determine when a process mean has shifted. In this situation the UCL and LCL should be set such that the Beta error is small.
True
If the Z value is zero, then _____ percent of the area under the curve is less than the mean.
50
The sample size is 25. The mean is 75. The estimate of the standard deviation is 10. What is the LCL?
69
The Pareto Principle states that _______ percent of the problem can be traced to _______ percent of the possible causes.
80, 20
For most processes, the X-bar chart is the only chart needed to monitor process output.
False
If a process mean is monitored, it is not necessary to monitor its variance.
False
In a Multi-Vari chart, the range of variation is expressed along the "X" or horizontal axis.
False
The Poisson distribution focuses on the number of discrete occurrences over a defined interval.
True
The Taguchi loss function suggests that the loss to the organization gets larger and larger as process variation migrates farther and farther away from the mean.
True
The UCL and LCL can be placed anywhere on the control chart. However, their position will determine the likelihood of the errors that can be made when determining whether or not the process is in-control or out-of-control.
True
The average range of a sample is needed to compute the UCL and LCL for an X-Bar chart.
True
The average standard deviation of preliminary samples is used as the center line of the S-Chart
True
The data or observations that fall under the normal distribution between plus one and plus two standard deviations from the mean is 13.6 percent.
True
The length of the vertical lines in a Multi-Vari chart represents the range of a specific variable over a specific period such as one week.
True
The long time average of rejects from a process is 2%. When constructing a P-Chart, and assuming that the process has not changed, the 2% could be used as the centerline of the control chart.
True
To determine the center line or target of a control chart, several samples must be taken, then the mean of each sample computed, and finally the mean of these means calculated.
True
Using the special properties of the normal distribution, 99.7 percent of the data in a normal distribution will fall between the mean and plus or minus three standard deviations from the mean.
True
When Sample sizes are larger than 25 an X-Bar chart is used to monitor process means.
True
While RPN is a quantitative measure, FMEA is primarily a qualitative tool.
True
Choosing the right chart depends upon the ___________ .
type of data, sample size, type of statistic
A t distribution is used to determine probabilities when the sample is small.
True
A process starts with 100 good pieces and 85 are produced. The FTY is ___ .
.85
A Six Sigma process is one that produces products or delivers services such that there are fewer than ___ defects per million opportunities.
3.4
According to the central limit theorem, the distribution of sample means can not be represented as a normal distribution.
False
An X-Bar chart for samples greater than 25 can also be used to plot proportions.
False
The probabilities associated with tossing a coin would be represented by a ________________ distribution.
Binomial
The probability of getting exatly two heads in two tosses of a coin would be called a _________________ probability.
Binomial
Attribute data is the same as Continuous data.
False
The number of defects per unit sampled is measured by ______ .
DPU
The management plan designed to manage a quality control process is called a _________________ .
Elements of a Control Plan
There is only one source of variation.
False There are two sources: common and special
F distributions are used to compare __________________ .
Variances
Two statistics that define a normal distribution are the ________ and the ___________ .
mean, standard deviation
The purpose of a null hypothesis is to ____________________________ .
state the opposite of what it is you are trying to prove
A sample of 25 customer orders is selected. It is found that two orders had errors. In the first of these, two errors were found. In the second, three errors were found. No other errors were encountered. The DPU is _____ .
.20 Two orders had errors. In the first, two errors were found. In the second three were found. This is a total of 5 errors in 25 orders or 5/25 = .20 defects per unit
A company places a order from its supplier for 100 units. When the order arrives a sample of 10 units is selected and tested. Each test measures four dimensions of the units. They include color consistency, quality, included instructions and packaging. Of the ten units sampled there are a total of three problems found in two of the units. One unit had a single problem and other had two problems. The DPU is ___________ .
0.3 The total number of problems is 3, so the DPU is 3/10 or .30 defects per unit.
The Z table tells us that the area under the normal curve to the left of Z value of 1.5 is 0.9332. The area under the normal curve between plus and minus Z=1.5 is _______ .
0.8664 1.000 - 0.9332 = 0.0668. Then subtract this from 0.9332. ). The answer is .9332 - 0.0668 = 0.8664
A process has three steps. The yield on each step is 0.9. What is the yield (RTY) of the process?
0.9 X 0.9 X 0.9
The average wait time at call center is 7 minutes with a standard deviation of 1 minute. A customer has called and has waited 10.5 minutes. What is the Z value?
3.5
The mean of a process is 45. if the standard deviation of the process is 2, a process that has been set to Six (6) Sigma would produce 99.9997 percent of its parts between _____ and ____ .
33 to 57 45+6(2)=57 and 45-6(2) =33
Suppose samples are taken at periodic intervals from an operational process. What is the likelihood that a range of plus or minus one standard deviations (in this case it is more correct to say one standard error of the mean) will capture a single sample from the process?
68 Percent Notice that we are using the special properties of the normal distribution.
If 100 items are sampled and one unit is found to have three defects, another is found to have one defect, and a third is found to have four defects. The remaining units have no defects. How many "Defects" have been found?
8
There are 40 possible defects that can occur when manufacturing a single part. Suppose a sample of 30 parts is taken. The total number of defects found in the sample is 10. What is the DPMO?
8,333 10/((40)(30)) X 1,000,000 = 8,333
In a chip manufacturing process, 5,000 chips began the process. At the end, only 4,500 pass quality control tests. The FTY would be ___ percent.
90 Percent
A Z value of ±2 will include _____ percent of the observations under the normal distribution.
95
If 100 pieces are started but 95 acceptable pieces finish, the First Time Yield would be ______ percent.
95
For which of the following is ANOVA appropriate?
A hypothesis test when the means of more than two samples must be compared
Chi Square tests are used when count data is presented in a _______________ table.
Contingency
The ratio of good parts produced at the end of a process divided by the number of parts that were started is called ______ .
FTY
A Six Sigma designed process should have no defects per Unit.
False
Both the LSL and USL must be positioned above the Mean (center line) or the process is out-of-control.
False
Every process should set Six Sigma as a quality goal.
False
In a normal distribution most of the data items cluster toward the ends or tails of the distribution.
False
Ninety-five percent of the observations in a normal distribution fall above the mean.
False
Only one defect can occur in one unit.
False
R-Charts are used to monitor process means.
False
RPN is an abbreviation for Reject Priority Number
False
RPN measures the cost of defectives.
False
Several employees at a distribution center have reported ill and have been replaced by temporary help. The number of errors have increased. This would be considered common cause variation.
False
The distribution of sample means will only be symmetrical if the population is symmetrical.
False
The longer is the vertical line in a Multi-Vari chart the lower is the "position" variation.
False
The percent of the data or observations between minus one standard deviation from the mean and minus three standard deviations from the mean is 60 percent.
False
The range is used as an estimate of the standard deviation when sample sizes are greater than 25
False
The selection of an appropriate chart is not dependent on the size of the sample that is taken from the process.
False
The special properties of the normal distribution apply to a distribution whether or not it is normally distributed.
False
There are three types of data including continuous, discrete and digital.
False
There is no difference between DPU and DPMO.
False
When sample sizes are greater than 25, it is still necessary to correct for bias when calculating the UCL and LCL.
False
When the shortcut is used to create the control chart, the advantage is that the likelihood of incurring an Alpha or Beta error is reduced to zero.
False
A restaurant quality control system has been designed as a Five Sigma process. Its goal is to ensure that food is free of contaminants. If 10,000 meals are served each month about 100 meals would have traces of contaminants.
False A Five-Sigma designed process would have 230 defects per million opportunities (see chart in the chapter). To determine the number of defects when 10,000 meals are served use the following calculations. 230/1,000,000 X 10,000. This is the same as X=(230/1,000,000) (10,000) = 2.3 meals per month
Alpha and Beta errors need only be considered when creating charts that monitor means not charts that monitor variation.
False Alpha and Beta errors need be considered when monitoring means and variation.
An S-Bar chart is used to monitor process variation regardless of the size of the sample.
False An R chart is used when the sample size is less than 12.
When the proportion is as low as 0.01 or 1 percent, it is not possible to use the equation presented in this chapter for determining control limits.
False As long as the sample size is greater than 500 the equation can be used.
The binomial distribution represents discrete processes, those that have more than two possible outcomes.
False Binomial is for discrete processes with only two outcomes.
If the X-Bar chart shows that a process is in-control it will not be necessary to review the S-Chart. The conclusion should be made that the process is in-control.
False Both charts must be considered. However, if the S-Chart is examined first and it suggests an out-of-control situation, the X-Bar chart does not have to be examined. In that situation the process is out-of-control.
The mean output of a process is the major factor in controlling that process and ensuring that it meets process objectives.
False Both mean and standard deviation, a measure of variation, are important. Some would argue that variation is more important than mean.
Special cause variation is always present. No process can avoid it.
False Common cause is always present.
Consider a help line for those who have just purchased a virus protection program for their computer. The variation in the length of time it takes for a caller to reach a representative - under the assumption that the process is behaving as expected - would be considered a Special Cause variation.
False Common cause variation
Data that can be measured on a continuum are called discrete data.
False Continuous data
The R-Chart is used when monitoring process variation for discrete data.
False Continuous data
An X-Bar chart is used to monitor process means when the data are discrete.
False Data must be continuous
When the population is not normally distributed, sampling theory cannot be used to monitor and control an operational process.
False For large size samples, the sample distribution of the mean is normal regardless of the shape of the population so sampling theory does apply under this condition.
If the S-Chart suggests that a process is out-of-control, it will be necessary to confirm this with the X-Bar chart before the conclusion can be made that the process is out-of-control.
False If the S-Chart records a sample beyond the control limits, then the process is considered to be out-of-control. It is not necessary to review the X-Bar chart.
In multiple regression, at most one independent variable can be significant.
False In a multiple regression model of five independent variables all five could be significant.
As the sample size decreases, the distance between the UCL and LCL in a P-Chart becomes smaller.
False It becomes larger.
Performance reporting involves preparing a book of control charts showing the performance of the process over time.
False It involves a periodic review of the control plan
When setting the UCL and LCL at plus and minus 2 standard deviations from the mean the likelihood of an alpha error is 95%.
False It is 5 percent
A process produces a sample mean outside the UCL and LCL but the process mean has not actually shifted. If corrective action is taken this represents a Beta error.
False It is an Alpha error that has occurred.
Consider a process that is in-control; both its mean and variance are performing as expected. If the UCL and LCL have been set at plus or minus 3 standard deviations from the mean. The likelihood of a sample mean falling outside these limits is 1.00 - .997 or .003 percent. Suppose this has occurred. The process is shut down and examined for problems. A beta error has occurred.
False It is an Alpha error that has occurred. The process is in-control but a sample result has occurred beyond the UCL or LCL. This is expected to occur 0.3 percent (less than one percent) of the time
FMEA is a process that attempts to correct process problems once they have occurred.
False It is anticipatory not reactionary
When several samples are taken from the same population and at the same point in time, each sample mean should be exactly the same as the others.
False It is extremely unlikely or rare that two samples taken from the same population and at the same point in time share the same mean.
Failure Mode and Effects Analysis is primarily used in the inspection process after items have been manufactured. It assures that defective output is not passed on to the customer or client.
False It is intended to prevent problems from occurring not to correct them once they have occurred.
It is suspected that sales for a company are related to population density and personal income. Sales is considered to be the independent variable.
False It is the dependent variable.
When a sample is taken from an operational process, the sample mean is plotted on the R-Chart.
False It is the sample range that is plotted on the chart.
Using the conventional critical p-value, if the p value of a regression equaition coefficient is 0.0651 it can be considered significant.
False It must be less than 0.05 to be significant.
The center line in an R-Chart represents the process mean.
False It represents the average range determined from preliminary samples.
The shortcut method uses the same constant B(3) to calculate both the UCL and LCL.
False It uses B(3) and B(4)
If a thorough FMEA effort has been made, there would be no reason to establish a Six-Sigma control process.
False It will never be possible to eliminate all problems, and as a result a Six Sigma process would still be required to monitor process output.
A quality control system together with control charts has been set up to monitor the length of time in hours it takes for an order to be shipped to a customer. The type of data monitored in this system would be classified as attribute.
False It would be continuous
For large size samples, in the range of 200 or more, the distribution of sample means will not be closely clustered around the mean of the population.
False Large sample size results will be tightly clustered around the population mean.
The equation presented in this chapter can be used when the sample size times the average proportion is less than 5.
False Must be greater than 5.
Before plotting the mean of a single sample on a control chart, the sample mean must be divided by the square root of the sample size.
False No, just the sample mean is plotted on the control chart.
The length of time it would take to be seen at a health care clinic, that has otherwise not expected emergencies during the day, would be subject to special cause variation.
False Normal variation is considered common cause variation.
In multiple linear regression, with four independent variables, it is possible to estimate the position of the regression line using a scatter plot.
False Not possible for most people to draw a four dimensional graph.
In general, a process capability of 10 or greater is preferred.
False One or greater
X-Bar and R- Charts are used to establish a control system when the sample result is a continuous variable and the sample size is less than 25.
False Only when sample size is less than 12.
P-Charts are used only for small size samples when the data are continuous.
False P-Charts are used for discrete or attribute data only.
The UCL and LCL are set at plus and minus three standard deviations from the mean or center line of the process control chart.
False Plus and minus three standard errors of the mean.
In a X-Bar chart where the size of the sample is between 12 and 25, it will not be necessary to take several preliminary samples to establish a center line on the chart since most sample means for samples of this size will be the same.
False Preliminary samples are necessary. Sample means will vary.
It is not necessary to create an S chart when sample sizes are greater than 12.
False S charts are used when the sample size is greater than 12.
A sample can be larger than the population from which the sample is taken.
False Samples are a sub-set of the population and therefore always smaller than the population.
To determine the mean of a process for the purpose of establishing the center line on the control chart, it is necessary to take only one small sample of process output.
False Several samples must be taken, then the grand mean calculated.
Declaring whether or not you will vote in an upcoming election (yes or no) is an example of continuous data.
False Since you answer yes or no, it is considered discrete data.
When the regression equation is significant, all of the coefficients of the independent variables will also be significant.
False Some may not be significant. Significance is determined by the p value associated with the independent variables.
The LSL and USL must be contained within the LCL and UCL.
False The LCL and UCL must be contained within the LSL and USL.
The acronym DPMO means Defects Per Multiple Operations
False The acronym DPMO means Defects Per Million Opportunities
A Six Sigma designed process is one that produces products or delivers services such that there are fewer than 340 defects per million opportunities.
False The actual number of defects is 3.4 per million opportunities
When designing an X-Bar chart for intermediate size samples, it is necessary to compute an average value of the standard deviation from several preliminary samples to establish a center line for the chart
False The average value or grand mean of the preliminary sample means is used to establish the centerline.
In an X-Bar Chart when the sample size is between 12 and 25, the constant D(4) is used to determine both the UCL and LCL.
False The constant used is A(3).
To find the Range of a sample, the highest value in the sample is subtracted from the mean of the sample.
False The lowest value is subtracted from the highest value
The p-value test is used to determine if the intercept in the regression model is significant.
False The p-value test is used to determine if the coefficients of the independent variables are significant. There is no test for the intercept.
To find the UCL and LCL for an X-Bar chart when the sample size is less than 12 the standard deviation of the sample must be computed.
False The range is used to estimate the variation associated with the chart.
The number of returns received by an online retailer is an example of a situation where a P-Chart would be appropriate.
False The reason that it would be inappropriate is because the data are continuous.
When the X-Bar chart is completed and then used to monitor process output, only the range is plotted on the control chart.
False The sample mean is plotted.
When determining the UCL and LCL for proportions, it is not necessary to consider the size of the sample.
False The size of the sample is required when computing the UCL and LCL. Refer to the Formula in this chapter.
The standard deviation of the population is 50, the sample size is 25. The standard error of the mean is 2.
False The standard error of the mean is 50 divided by the square root of 25 which is 5. So the standard error of the mean is 10.
If the RTY is 80 percent, the likelihood that a process will suffer a failure or reject is 80 percent.
False There is an 80 percent likelihood that the unit will be processed without any failures or rejects.
It is not possible for process capability to be greater than 60.
False There is no limit.
Non-parametric tests are considered more powerful in differentiating between accepting or rejecting a null hypothesis than a parametric test.
False They are less powerful.
R-Charts are only used when sample sizes are greater than 12.
False They are only used when sample sizes are less than 12.
P-Charts are used for continuous data.
False They are used for discrete data.
P-Charts can also be used to monitor continuous variables.
False They are used to monitor discrete variables
If the Beta error is relatively small the Alpha error will also be relatively small.
False They move in the opposite direction. If Alpha is large, Beta will be small.
Multi-Vari charts also uncover why special cause variation occurs.
False They only help expose the presence of special cause variation they cannot determine its sources.
In a multiple linear regression model, the p-value for the regression line and the p-values for all of the coefficients of the independent variables will be the same.
False They will be different
Six Sigma limits imply that outcomes beyond six standard deviations on either side of the mean will never occur.
False They will occur but rarely.
Multi-Vari Charts display the root cause of a problem
False This is not the purpose of a multi-vari chart
The control document needs to be sufficiently complete such that it does not need changes over time.
False This is the point of the step called Corrective Actions.
It is not unreasonable to set surgical procedures at a hospital to a standard of Three Sigma.
False Three Sigma would generate 66,800 defects or errors per million opportunities, but Six Sigma would generate 3.4 defects per million opportunities. Given the risk of surgery and consequences, a Six Sigma design strategy should be followed.
An S-Bar chart is appropriate for monitoring variation when samples of less than 5 items are taken.
False To use an S-Bar chart the sample size must be greater than 12.
To compute process capability you must have the UCL and LCL.
False USL and LSL
Once the X-Bar chart has been designed for samples between 12 and 25, a chart that monitors the variation in process output would not be necessary since the sample size is larger than 12.
False Variation must always be monitored regardless of the sample size.
The Alpha error is considered too large. It can be made smaller by raising the LCL and lowering the UCL. In other words, it can be made smaller by decreasing the distance between the UCL and LCL on the control chart.
False When the alpha error is considered too large then the limits would be widened by raising the UCL and lowering the LCL
The sample size is included when calculating the standard deviation of a proportion.
False You only need the average value of the proportion.
Which of the following is not one of the seven elements of a control plan.
Financial Analysis
The value of "a" in the regression equation represents which of the following?
Intercept with the Y axis
Which of the following is the non-parametric equivalent of the One sample t Test?
One Sample Sign
Which of the following tests require that the distribution from which samples are taken is normal or close to normally distributed?
One Sample t
You need to determine whether or not a process has met its objectives. A sample is taken and must be compared to a target. Which of the following tests is most appropriate? Assume normality and large sample size.
One Sample t Test
If you need to determine the probability that exactly five calls will arrive at the call center between 8 and 9 AM today, what probability distribution could be used to provide the answer.
Poisson
Which of the following is not a Six Sigma control chart?
Q-Chart
Which of the following represents the likelihood that a process will complete all of the required steps without any failures or rejects.
RTY
Which of the following is an appropriate null hypothesis for an ANOVA test?
Sample means are equal
The value of "b" in the simple linear regression equation represents _____________________ .
The increase in Y attributed to a unit increase in X
Which of the following is an appropriate null hypothesis for a Two Sample t Test?
The two means come from the same population
A SIPOC chart can be useful in identifying input variables for use in the Multi-Vari chart
True
A company is interested in the response to a promotional item given to fans who attended a Miami Dolphins game. They have sampled 100 fans and asked them to write a review. Those who attended the game would be considered the population.
True
A health clinic with four locations administers a patient satisfaction questionnaire across all clinics. The objective is to determine if patient satisfaction differs across these locations. A two-tailed test would be appropriate.
True
A manufacturer of headphones finds that an increasing number of phones have failed to pass the final quality control check before they are shipped to a distributor. This suggests Special Cause variation.
True
A more common standard for the design of a monitoring process is to choose a Four Sigma standard
True
A multi-vari analysis is visual and uses no statistics to compare output variation.
True
A one-tailed test is used if deviations of the sample mean in only one direction from the target or population needs to be considered.
True
A relationship between the divorce rate in New Jersey and the per capita consumption of coffee in that state has an R-squared of 0.75. This is an example of accidental correlation.
True
A sample of 30 customers rate the quality of a product as satisfactory or unsatisfactory. The data in the sample would be considered discrete.
True
A single sample mean is not likely to fall near the tails of the population distribution since a sample is unlikely to include only results from one end of the population. In most situations a sample may include a few values relatively far from the population mean with most of the values selected nearer the center of the population.
True
A t-Test is used to determine the difference between two samples when the sample sizes are small.
True
A two-tailed test is used if deviations of the sample mean in either direction from the target or population need to be considered.
True
Alpha and beta errors help to establish where the UCL and LCL will be positioned on the control chart.
True
Customer expectations are expressed using the concept of service levels.
True
Customer satisfaction registered on a scale from one to ten is an example of continuous data. For example, a customer could rate satisfaction as 6.5
True
Data for the length of time it takes to undergo a routine pre-operative physical in a hospital is an example of continuous data.
True
Defects Per Unit (DPU) is the number of defects in a sample divided by the number of units sampled.
True
Discrete data can be summarized as a proportion such as the percent of customers that would recommend a website to a friend.
True
FMEA is a proactive approach that examines what could go wrong with a process, product or service before the design is finalized.
True
For large size samples, the distribution of sample means is symmetrical and centered on the population mean.
True
Four identical machines in a manufacturing process are operated over three shifts. A multi vari chart would be appropriate to study the variability of each machine and then to determine which machine contributes most of the output variation
True
If process capability is 10, this suggests the process is very capable of meeting customer expectations as expressed by the LSL and USL.
True
If the Alpha error is relatively large, the Beta error will be relatively small.
True
If the calculated p-value is 0.005 as compared to a critical value of 0.01 the null hypothesis should be rejected.
True
If the sample size is 100 and the average value of the proportion is 0.06, the equation presented in this chapter can be used to establish control limits.
True
In Multi-Vari analysis, a graph is created that displays the possible sources of variation affecting process output.
True
In a Six Sigma designed process half of the outcomes will occur above the mean of the process.
True
In the RPN calculation the letter "O" represents the frequency with which the problem is likely to occur.
True
Many distributions describing natural processes such as IQ (Intelligence Quotient) or height can be classified as normal distributions.
True
Monitoring good and defective parts produced by a machine is an example of collecting discrete data.
True
Multi-Vari charts also display output means.
True
Multi-Vari charts can combine variables. One example is displaying the relationship between call duration, location and time of day on the same chart.
True
Non-parametric tests do not rely on the assumption of normality in the underlying population distribution.
True
Normal distributions have special properties that specify the percent of observations or data that fall within a specified number of standard deviations from the mean.
True
Once the control chart for variation has been designed, the standard deviation of a sample is plotted on the chart.
True
Once the type of control chart has been determined, the center line must be established.
True
One hundred samples are taken from the same population. The sample means are grouped and these groups are displayed as a histogram. A distribution of sample means can be obtained by drawing a smooth curve through the top of the bars in the histogram.
True
One reason why small size samples of less than 25 items or units is taken is because the cost of sampling a larger number would be expensive.
True
Preliminary samples are necessary to determine the center line of the X-Bar chart when sample sizes are between 12 and 25.
True
Preliminary samples are necessary to find the target or center line in a control chart.
True
Process capability depends upon the difference between the USL and LSL.
True
Process capability is determined by three variables; the LSL, USL and standard deviation of the process.
True
Process capability measures how close a process is running to its capability or service limits.
True
Process means, which measure the capability of a process, can not be changed by simply changing the UCL and LCL.
True
R-Charts are used to monitor process variation by plotting the range of items collected in a sample.
True
RTY represents the likelihood that a process will complete all of the required steps without a failure.
True
Regardless of the shape of the population distribution, the distribution of sample means will be normal when the sample size is large.
True
Rejects in a process are suspected to be related to the average number of hours of training offered to employees. The number of hours of training would be considered the independent variable.
True
Setting a Six Sigma standard for a process requires that the process has been designed to meet this standard. You can not take a process that was designed to meet a Three-Sigma standard and expect it to meet a Six Sigma standard without undertaking a serious re-engineering effort to improve the performance of the process
True
Setting the UCL and LCL requires that the costs associated with Beta and Alpha errors be balanced.
True
Several people working at a call center have called to report that they are sick with the flu and have been replaced by temporary help. Call waiting time has increased significantly. This would be considered special cause variation since it is not inherent in the process and is not predictable.
True
Sixty- Eight (assume rounding) percent of the observations or data in a normal distribution fall between the mean and plus or minus one standard deviation from the mean.
True
Special cause variation is of most concern in Six-Sigma. It is unexpected and can disrupt process output.
True
Special cause variation represents the category of variation that effective Six Sigma monitoring systems are expected to detect.
True
The S-Chart is similar in function to the R-Chart in that it monitors process variation.
True
The UCL and LCL will be positioned closer to the process mean or center line of the control chart as the estimate of the standard deviation gets smaller.
True
The average standard deviation of preliminary samples that was computed when establishing the control limits for the X-Bar chart for intermediate size samples can be used as the center line for the S-Chart.
True
The center line for an R chart is determined by taking several samples, determining the range for each of the samples and then computing the average of these ranges.
True
The centerline for a P-Chart would be determined by taking several preliminary samples, determining the proportion for each sample group, and finally computing the average of these group proportions.
True
The computed p-value from a Two Sample t Test measures how likely it would be to observe two samples whose means are that far apart.
True
The concept of Six Sigma implies near perfection.
True
The consequence of raising the UCL and lowering the LCL is that the Alpha error will decrease.
True
The data within a vertical bar of a Multi-Vari chart is referred to as "within variation" or "position variation."
True
The distribution of all samples that are taken and positioned on an X-Bar chart, when the sample sizes are larger than 25, will all fall in a symmetrical or bell-shaped pattern around the center line of the chart.
True
The first decision when selecting an appropriate control chart is whether or not the data are continuous or discrete.
True
The major problem with setting a Six Sigma standard for a process is the cost of achieving this level of quality.
True
The mean from a process has shifted but a sample result falls within the UCL and LCL, This is an example of a Beta error.
True
The mean of a sample is plotted on the X-Bar chart when sample sizes are greater than 25.
True
The mean of the means is called the "grand mean" and is used to establish the center line in an X-Bar Chart.
True
The mean represents the centralness of a distribution
True
The median statistic is used in non-parametric statistics because the median is less sensitive to extreme values found in a highly skewed distribution. So it is most appropriate for small sample sizes.
True
The more difficult it is to detect a problem the higher will be its RPN.
True
The normal or expected cause of variation is called Common Cause.
True
The number "3" in the formula, used to determine the UCL and LCL, represents the setting of these limits at three standard errors of the proportion on either side of the centerline.
True
The number of data points or observations beyond plus and minus six standard deviations from the mean would be much less than one percent.
True
The objective behind the construction and use of an R chart is to determine when the variation of a process is higher than expected.
True
The problem with designing a Six Sigma process is that it can be quite costly to operate with such high quality standards.
True
The range in an R-Chart is used to approximate the variation in sample means from one sample to the next.
True
The response plan identifies the steps that need to be taken should a problem occur when monitoring process performance.
True
The shortcut formula to determine the UCL and LCL on an X-Bar chart uses the constant A(2) as well as the range to establish their distance from the center line.
True
The special properties of the normal distribution used to make statements about population distributions can also be used to make statements about the distribution of sampling means. So, 68 percent of the samples means will fall within plus or minus one standard error of the mean from the mean of the distribution of sample means.
True
The standard deviation measures the variation of the population distribution but the standard error of the mean measures the variation of the distribution of sample means.
True
The standard deviation represents the extent of variation in a distribution.
True
The standard error of the proportion is similar in concept to the standard error of the mean.
True
The terms discrete and attribute data are used interchangeable.
True
The values D(4) and D(3) are used when determining the UCL and LCL for R-Charts.
True
There is a difference between the quality of what customers expect and the capabilities of the process that delivers these products and services.
True
To detect even the smallest shift in a process mean it would be necessary to set the UCL and LCL close to the process mean.
True
To determine the center line for X-Bar charts, when the sample size is between 12 and 25, the grand mean of preliminary samples is calculated.
True
We can conclude from the shortcut formula that the range of the sample is used in estimating the standard error of the mean.
True
When designing an X-Bar chart for continuous data and where the sample size will be 14, the standard deviation is used to determine the UCL and LCL.
True
When determining process capability, the larger is the estimate of the standard deviation of the process, the lower is the process capability.
True
When establishing the UCL and LCL in a P-Chart, and under the assumption that these limits will be set at plus and minus three standard errors of the proportion from the centerline, only the average value for p and sample size n are needed.
True
When process variation is greater, we would expect that the LCL and UCL would be set farther apart.
True
When the DPMO is 6,210 this suggests a process that has been set to a standard of Four Sigma.
True
When the consequences for failing to meet high standards are significant, such as in many areas of health care, Six Sigma standards of product or process design are justifiable.
True
When the sample size is exactly 25, the shortcut and non-shortcut methods will generate very close to the same UCL and LCL.
True
When the sample size is large and discrete data will be collected, a P-Chart is used.
True
When the standard deviation is large, the spread of the data around a mean is larger than if the standard deviation were small.
True
X-Bar Charts, where sample sizes are less than 12, rely on the range of preliminary samples to establish the UCL and LCL. However, X-bar charts, where sample sizes are between 12 and 25, rely on the standard deviation of preliminary samples to establish these levels.
True
The process mean is 35. Sample size is 25. The average standard deviation is 5. The value of A3 is 0.606. The UCL will be 38.
True 35 + .606 (5) = 38.03. Rounding to 38
Common Cause variation is predictable.
True Common cause variation represents variation from the target that can be attributed to many random causes. It is predictable because random variations will always occur. Special cause variation, on the other hand, is not predictable and can be attributed to a special cause.
If the UCL and LCL are moved closer to the center line, the USL and LSL will not be affected.
True Customer's preferences do not change.
A multiple regression model that proves to be significant cannot confirm a causative relationship between the independent and dependent variables.
True Just because R-squared is high does not prove that there is a causative relationship between the dependent and independent variables. It merely suggests that a relationship exists, not necessarily a causative one.
You are interested in determining whether or not a training program has met expectations. The alternative hypothesis states that it has met expectations.
True Remember, the alternative hypothesis states what it is you are trying to prove. In this case you are trying to prove that the training was effective.
In a multiple regression model the number of independent variables is always greater than one.
True That is the definition of "multiple" regression.
In a simple linear regression model, the p-value for the regression line and p-value for the coefficient of the independent variable are the same.
True The p-value for the regression line tells us whether or not the regression line is significant. The p-value for the coefficient of the independent variable tells us if the coefficient is significant. In a simple linear regression model they are both the same. Go back to the example in the chapter and you will confirm that this is true.
It is not possible to eliminate all common cause variation in most processes.
True There is some variation in even the most carefully designed and controlled processes.
An X-Bar and S-Chart can be used for continuous data when sample size is greater than 25.
True This answer is correct.X-Bar and S-Charts are not only used when the sample size is greater than 12 and less than 25, but also when the sample size is greater than 25. The difference is that when the sample size is between 12 and 25 the shortcut formula is used to determine the UCL and LCL, but when it is over 25, the mean plus or minus three stand errors of the mean is used directly.
If correlation analysis finds that 82 percent of the variation (R-squared = .82) in the dependent variable can be explained by three independent variables, we can say that 18 percent of the variation is unexplained by this model.
True Unexplained = 1.00 - Explained
X-Bar and S-Charts are used when the data that is sampled is continuous and the sample size is greater than 12 but less than 25.
True X-Bar and S-Charts are not only used when the sample size is greater than 12 and less than 25, but also when the sample size is greater than 25. The difference is that when the sample size is between 12 and 25 the shortcut formula is used to determine the UCL and LCL, but when it is over 25, the mean plus or minus three stand errors of the mean is used directly.
When a Two Sample t Test is run, the results indicate that the calculated p-value is 0.06. Using the conventional critical p-value, the null hypothesis _____________.
cannot be rejected
Non-parametric tests are necessary when _________________________________ .
sample sizes are small, normality of the distribution cannot be assumed, the distribution from which the sample has been taken is highly skewed
