lecture8[1] - Introduction Week Eight. Goal for this week:...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Introduction Week Eight. Goal for this week: 1. Go over assignment for this week. 2. Chapter Eight Notes 3. Preview the Assignment for next week. Stat 601, B. D. McCullough, Fall 2009 Assignment, HS #7-4 Perform a two sample t-test on the means. Then test whether the variances are equal. Many persons would just test whether the means are equal and then conclude that FEL doesnt matter. BUT THERE IS A DEEPER PROBLEM! Stat 601, B. D. McCullough, Fall 2009 Assignment, HS #7-4 Randomization was not a consideration when the data were collected. Therefore we have to be careful about interpreting these results. We are not told on what basis projects were candidates for FEL and therefore were selected for an FEL. It is possible that non-FEL-eligible projects naturally are more variable! Worse, it is possible that only the easy (or hard) projects were assigned to FEL. So the difference that we attribute to FEL is really due to the fact that one group of projects was hard and the other was easy. This is called confounding . Stat 601, B. D. McCullough, Fall 2009 Assignment, HS #7-6 Use a 2x2 factorial design with two factors: wristband/no wristband, and light/heavy ball. What is the process measure? (number of pins) In what order do we conduct experiments? (random) Does a wristband work? Then test tight/loose wristband. Does light ball work better? Then test light/very light ball. What about shoes and shirt? Why not test them? Stat 601, B. D. McCullough, Fall 2009 Assignment, HS #7-6 What about experimental conditions? Do we always use the same lane? Do we re-oil the lane after every throw? Any other conditions to consider? Stat 601, B. D. McCullough, Fall 2009 Assignment, HS #7-7 What is the role of alloy formulation and current? Factors! What kind of design is this? 2x2 factorial design. What is the response? Weld strength. Put the data in a minitab worksheet. be sure to turn alloy into a numeric variable, alloy1. Also be sure to create an interaction variable. Stat 601, B. D. McCullough, Fall 2009 Assignment, HS #7-7 We need to characterize the interaction. How did we do it last time? Page 302 Figure 7.4 Alloy 1 Alloy 2 140 2.5, 4.7, 3.0 2.9, 3.5, 2.5 mean 3.4 2.97 200 8.6, 8.3, 9.4 0.9, 2.0, 2.0 mean 8.77 1.63 Draw the graph with response on the ordinate, arc current on the abscissa, and lines for alloy 1 and alloy 2. Stat 601, B. D. McCullough, Fall 2009 HS Ch 7 #7 Conclusions Increasing arc current increases weld strength more for Alloy 1 than for Alloy 2. High weld strength for Alloy 1 comes with using an arc current of 200. arc cur ent weld strength 140 200 2 4 6 8 10 3.4 8.77 A l oy 1 2.97 1.63 A l oy 2 Arc current has essentially no effect on Alloy 2 weld strength, hence it doesnt matter what arc current levels you use for Alloy 2....
View Full Document

Page1 / 33

lecture8[1] - Introduction Week Eight. Goal for this week:...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online