Exam 1 - Statistical Review Flashcards
What are two types of numerical data?
1) Variables (continuous)
2) Attributes (discrete)
On a z-distribution how much percentage of the data are within 3, 2, and 1 standard deviation(s) of the mean?
3 - 99.7%
2 - 95%
1 - 68%
Type I error is the probability of seeing a difference when none exists (the alpha). This occurs when mistaking _____ variation for _____ variation.
Type I error is the probability of seeing a difference when none exists (the alpha). This occurs when mistaking common cause variation for special cause variation.
T or False. The t-value is the signal to noise ratio which can also be stated as the difference between the group means divided by the variability of the groups.
True.
What is the key to defining and solving problems?
Variation
_____ cause variation is inherent to the system, while _____ cause variation is assignable to a known root cause.
Common cause variation is inherent to the system, while special cause variation is assignable to a known root cause.
Specification limits are known as the voice of the customer. The difference between the upper and lower limits is called the ______.
Specification limits are known as the voice of the customer. The difference between the upper and lower limits is called the tolerance.
The voices of the process are the _____ limits.
The voices of the process are the control limits (+-3 sd).
The upper and lower control limits are known as the span, which constitutes ____ standard deviations.
The upper and lower control limits are known as the span, which constitutes 6 standard deviations.
The difference between the process mean and the target value is ____.
The difference between the process mean and the target value is bias.
T or F. Spec limits apply to individual measures, while control limits apply to process means.
TRUE. Spec limits apply to individual measures, while control limits apply to process means (except for individual charts.
T or F. Statistical control is the absence of common cause variation.
FALSE. Statistical control is the absence of SPECIAL cause variation.
What are the seven basic graphical tools of statistical process control (plus one) and their major uses?
1) Histogram - gives idea of central tendency/variation
2) Scatter diagram - reveals relastionships btwn variables
3) Radar graph - compares multivariable data
4) Flow chart - illustrates sequential material flow through time and space
5) Pareto chart - Prioritizes problems
6) Cause-and-effect (Fishbone) diagram - brainstorm potential sources of variation
7) Run chart - performance through time
8) Control charts - differentiate special from common cause variation