Hypothesis Tests Flashcards
A hypothesis test that determines whether a statistically significant difference exists between the average of a normally distributed continuous data set and a standard. It provides a way to determine if there is truly a difference between the standard and a particular data set mean or whether the difference is due to random chance.
An example would be testing whether a supplier, that has guaranteed an average 12-ounce fill rate on their beverages, is performing as promised.
One-Sample T-Test
A hypothesis test that determines whether a statistically significant difference exists between the averages of two independent sets of normally distributed continuous data. It is useful for determining if a particular strata or group could provide insight into the root cause of process issues.
An example would be if Location A has average sales of $3,567 per month whereas Location B has average sales of $3,843 per month and you want to determine if Location B truly has greater averages sales or the difference is just due to random chance.
Two-Sample T-Test
A hypothesis test that determines whether a statistically significant difference (aka variance) exists between the averages of two or more independent sets of normally distributed continuous data. It is useful for determining if a particular strata or group could provide insight into the root cause of process issues.
An example would be if Assembly Line A products weigh an average of 10.4 pounds, Assembly Line B products weigh an average of 9.2 pounds and Assembly Line C products weigh an average of 11 pounds and you want to determine if any of the 3 lines truly has a different average from the others or if the difference is just due to random chance.
One Way ANOVA
A hypothesis test that determines whether a statistically significant difference exists between the variance of two or more independent sets of non-normally distributed continuous data. It is useful for determining if a particular strata or group could provide insight into the root cause of process issues.
An example would be if Assembly Line A has cycle times with a variance of 2 minutes where as Assembly Line B has cycle times with a variance of 3 minutes and you want to determine if Line A truly has less variation or if the difference is just due to random chance.
Levene’s Test
A hypothesis test that determines whether a statistically significant difference exists between the median of a non-normally distributed continuous data set and a standard. It provides a way to determine if there is truly a difference between the standard and a particular data set median or whether the difference is due to random chance.
An example would be testing whether a call center, that has guaranteed a median hold-time of 1 minute, is performing as promised.
One-Sample Sign Test
A hypothesis test that determines whether a statistically significant difference exists between the medians of two independent sets of non-normally distributed continuous data. It is useful for determining if a particular strata or group could provide insight into the root cause of process issues.
An example would be if Pizza Delivery Person A has median delivery time of 15 minutes whereas Pizza Delivery Person B has median delivery time of 17 minutes and you want to determine if Person B is truly slower or if the difference is just due to random chance.
Mann-Whitney Test
A hypothesis test that determines whether a statistically significant difference exists between the medians of two or more independent sets of non-normally distributed continuous data. It is useful for determining if a particular strata or group could provide insight into the root cause of process issues.
An example would be if Assembly Line A products have a median production cycle time of 10.3 minutes, Assembly Line B products have a median production cycle time of 9 minutes and Assembly Line C have a median production cycle time of 11.5 minutes and you want to determine if any of the 3 lines truly have different median cycle times from the others or if the difference is just due to random chance.
Mood’s Median Test
A hypothesis test that determines whether there is a correlation between two paired sets of continuous data. It is useful for determining if changes in Y can be attributable to a particular X. It produces a “prediction equation” that estimates the value of Y that can be expected for any given value of X within the range of the data set.
An example would be to test if rainfall and crop yield were correlated and then to calculate approximately how much water is required to achieve the desired yield.
Regression Test
A hypothesis test that determines whether there is a correlation between two or more values of X and the output, Y, of continuous data. It is useful for determining the level to which changes in Y can be attributable to one or more Xs. It produces a “prediction equation” that estimates the value of Y that can be expected for given values of one or more X values within the range of the data set.
An example would be to test if crop yield were correlated to both rainfall and fertilizer amount, and then to calculate approximately how much water and fertilizer are required to achieve the desired yield.
Multiple Regression Test
This was Dr. W. Edwards Deming’s guide for organizational leadership to better use their role to improve the effectiveness of any organization. These come from Dr. Deming’s book, “Out of the Crisis”
- Create constancy of purpose toward improvement of product and service, with the aim to become competitive and to stay in business, and to provide jobs.
- Adopt the new philosophy. We are in a new economic age. Western management must awaken to the challenge, must learn their responsibilities and take on leadership for change
- Cease dependence on inspection to achieve quality. Eliminate the need for inspection on a mass basis by building quality into the product in the first place.
- End the practice of awarding business on the basis of price tab. Instead, minimize total cost. Move toward a single supplier for any one item, on a long-term relationship of loyalty and trust.
- Improve constantly and forever the system of production and service, to improve quality and productivity and thus constantly decrease costs.
- Institute training on the job.
- Institute leadership. The aim of supervision should be to help people and machines and gadgets to do a better job. Supervision of management is in need of overhaul, as well as supervision of production workers.
- Drive out fear, so that everyone may work effectively for the company.
- Break down barriers between departments. People in research, design, sales and production must work as a team, to foresee problems of production and in use that may be encountered with the product or service.
- Eliminate slogans, exhortations, and targets for the work force asking for zero defects and new levels of productivity. Such exhortations only create adversarial relationships, as the bulk of the causes of low quality and low productivity belong to the system and thus lie beyond the power of the work force.
- Eliminate work standards (quotas) on the factory floor. Substitute leadership
- Eliminate management by objective. Eliminate management by numbers, numerical goals. Substitute leadership - Remove barriers that rob the hourly worker of their right to pride of workmanship. The responsibility of supervisors must be changed from sheer numbers to quality.
- Remove barriers that rob people in management and engineering to their right to pride of workmanship. This means, inter alia, abolishment of the annual or merit rating and of management by objectives.
- Institute a vigorous program of education and self-improvement.
- Put everybody in the company to work to accomplish the transformation. The transformation is everybody’s job.
The 14 Points for the Transformation of Management
This represents a list of basic process improvement tools and techniques. The list is generally attributed to Kaoru Ishikawa, a follower of Dr. Edwards Deming, who is also famous for popularizing the Fishbone or Ishikawa Diagram. In an effort to reduce the complexity of Statistical Process Control, and make it more accessible for the average worker, he compiled a shortlist of simple but powerful Lean Six Sigma tools.
The list includes:
1. Cause & Effect Diagram (aka Fishbone Diagram)
2. Checksheet
3. Control Chart
4. Histogram
5. Pareto Chart
6. Scatter Diagram (aka Scatter Plot)
7. Stratification (often replaced with Flow Chart)
7 Basic Quality Tools
These are quality guru Philip Crosby’s recipe for long-term process improvement. His opinion was that these steps were the responsibility of management but involved the people who did the work. These steps provided guidelines as well as a method for communicating his Four Absolutes.
Step 1: Management Commitment Step 2: Quality Improvement Team Step 3: Quality Measurement Step 4: Cost of Quality Evaluation Step 5: Quality Awareness Step 6: Corrective Action Step 7: Establish an Ad Hoc Committee for the Zero Defects Program Step 8: Supervisor Training Step 9: Zero Defects Day Step 10: Goal Setting Step 11: Error Cause Removal Step 12: Recognition Step 13: Quality Councils Step 14: Do It Over Again
14 Steps to Quality Improvement
These were developed by quality guru Philip Crosby as a way to promote the idea increased quality did not mean increased cost. Quality and cost were not in competition which he expanded on in his best-seller, “Quality Is Free.”
- Quality is defined as conformance to requirements
- The system for causing quality is prevention, not appraisal
- The performance standard must be Zero Defects
- The measurement of quality is the Price of Nonconformance
The Four Absolutes of Quality
A hypothesis test that determines whether a statistically significant difference exists between the variances of two or more independent sets of normally distributed continuous data. It is useful for determining if a particular strata or group could provide insight into the root cause of process issues.
An example would be if Assembly Line A product weights have a variance of 1 gram, Assembly Line B product weights have a variance of 2 grams and Assembly Line C product weights have variance of 2.5 grams and you want to determine if any of the 3 lines truly has less/more variation than the others or if the difference is just due to random chance.
Bartlett’s Test
A hypothesis test that determines whether a statistically significant difference exists between the variance of two independent sets of normally distributed continuous data. It is useful for determining if a particular strata or group could provide insight into the root cause of process issues.
An example would be if Assembly Line A has product weights with a variance of 1 pound whereas Assembly Line B has product weights with a variance of 2 pounds and you want to determine if Line A truly has less variation or the difference is just due to random chance.
F-Test (aka Test for Two Variances)
