six sigma ttests(ppt 57)英文.pptVIP

t-Tests Comparing Two Groups Objectives Statistical Tests Hypothesis Testing Compares groups Is there sufficient statistical evidence to reject the null hypothesis? After collecting data, how do we TEST it? Several different tests Appropriate test depends on: Objective Type of data Here we will look at tests for variables Output (continuous Y) data with attributes Input (discrete X) data Example: Two Levels of X t-tests Use t-test when comparing: Sample mean to another sample mean – 2 Sample t-test Sample mean to a specific value (e.g., target, goal, or belief) – 1 Sample t-test Covered in the appendix 中国最大的资料库下载 Test is a Signal-to-Noise ratio that compares sample averages to what would be expected if H0 were true Analyze Roadmap - 2 Samples Analyze Roadmap – Comparing means of two Samples This is an example of a hypothesis test that compares two sample means. H0: ?Bob = ?Jane Ha: ?Bob ≠ ?Jane 20 samples from each operator Manipulating the Data It’s easier to get useful graphs and analyses when the data are all in the same column, i.e., A single column for the measurement data (Y) One column for each of the ‘identifiers’ (X) In our example, use the following columns: Person (with Bob and Jane as the values) Time (with all of Bob’s and Jane’s values) Reading across a row gives the time and operator who answered the call Minitab - Stacking Data Minitab - Stacking Data Verify Assumptions t-test is the Signal-to-Noise ratio used to evaluate this hypothesis t-test makes two assumptions Stable process Normal distribution Test still performs well with moderate departures from normality Analyze Roadmap – 2 Samples Checking Stability Use an I-MR chart to evaluate the stability of the data Since there are two groups of data, use the “Estimate Parameters BY Groups in…” option in Minitab Checking Stability Analyze Roadmap - 2 Samples Testing Shape Normality Check Analyze Roadmap - 2 Samples Graphs for X with 2 or More Levels Dotplots and boxplots compare sha