lecture6[1]

# lecture6[1] - Introduction Week Six Goal for this week 1...

This preview shows pages 1–8. Sign up to view the full content.

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

View Full Document

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

View Full Document

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

View Full Document

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Introduction Week Six. Goal for this week 1. Hand back exams. 2. Present solutions to exam. 3. homework problems 4. hypothesis testing 5. Chapter Six 6. Preview homework problems. 7. Chapter Six Notes. 8. Chapter Seven Notes. Stat 601, B. D. McCullough, Fall 2009 Hand back exams Stat 601, B. D. McCullough, Fall 2009 Homework Problems, HS Ch 5 # 4 ◮ Errors in orders: Sample size of 50 is low for discrete data. Selecting the 50 largest PO’s will bias the data towards large PO’s. Also, one month of data will miss any quarterly or yearly cycles in this process. ◮ Internet leasing satisfaction: Use of 5 major metropolitan areas will bias the data by missing rural and suburban areas. Non-response may be a huge issue in an electronic survey. ◮ Billing accuracy: Sample size is OK, but data is all collected from 4:00-5:00pm, missing the rest of the data. The invoices processed at this time may be different from invoices processed in the morning. Two weeks of sampling will miss potential monthly, quarterly, and annual cycles in the billing process. Stat 601, B. D. McCullough, Fall 2009 Homework Problems, HS Ch 5 # 4 ◮ HR sampling: This process should be unbiased, unless there is some recurring pattern in the alphabetized list that has a cycle of 50 (very unlikely). It is not clear if the resulting sample size will be appropriate or not. ◮ HMO cycle time: The sample size is small if variation is of interest. In addition, we will only know about the process of the Chicago facility. If they are different relative to data collection, they are likely different in other ways. ◮ Bank satisfaction survey: Including this in statements could impact the results (people who were pleasantly surprised by their statement will likely give a better evaluation that those who were negatively surprised), and clearly non-response is a critical problem. Stat 601, B. D. McCullough, Fall 2009 Homework Problems, HS Ch 5 # 5 The average of OTR cycle time appears to be steady over time, but there are three special causes that can be identified on an individuals and moving range chart. This chart is appropriate because we have continuous data that is not sub-grouped, i.e.; it comes in one data point at a time. Stat – Control Charts – Variables Charts for Individuals – Individuals Stat – Control Charts – Variables Charts for Individuals – Moving Range Stat 601, B. D. McCullough, Fall 2009 Homework Problems, HS Ch 5 # 5 How do we delete these three observations from the dataset? Two ways. (1) Data – Delete Rows (if you know the row numbers) (2) Data – Subset Worksheet (if you know the condition for the observations you want dropped, e.g. , X > 10. Stat 601, B. D. McCullough, Fall 2009 Homework Problems, HS Ch 5 # 5 A histogram of cycle time is skewed, but we must be very careful here, because we know there are special causes in the data. If we remove the three identified special causes, the resulting histogram is still somewhat skewed. This is not surprising for cycle time data.is still somewhat skewed....
View Full Document

{[ snackBarMessage ]}

### Page1 / 33

lecture6[1] - Introduction Week Six Goal for this week 1...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online