Q SCI 381: Introduction to Probability and Statistics (Section C) --- Autumn 2014

Class Email Archive

During the quarter, a series of emails will be sent to all students who are enrolled in the class. Each of these messages is also posted to this page for later reference.


Email Communication:

All course email is automatically sent to each student's UW email account. Thus, students who use a private commercial email account or a departmental email account are responsible for having their UW email forwarded to their other account. Also, for access to some course web pages you must use your UW NetID and UW password.


Subject: Introduction to Your Online Autumn 2014 Quarter Course -- Q SCI 381C

This is a preliminary message to let you know that you are registered for Q SCI 381C this autumn term. If you wish to become familiar with how this online class will be conducted, please review materials on the following web site:

http://faculty.washington.edu/bare/qs381

Also, complete the registration process to use the MyLab & Mastering web site and the eText book as discussed on the Course Content web page (see above link for more information).

There are no formal lectures for this course, but we will communicate frequently with you via email, the online discussion board and during our TA's office hours. It is your responsibility to become familiar with the course web site materials. However, we are here to assist you in learning the course material.

I will be back in touch as the beginning of autumn term approaches. Meanwhile, I hope you enjoy your break from classes.


Sent: Monday, September 15, 2014 9:03 AM
Subject: Text for Autumn Quarter

Before you purchase your text for autumn quarter read the following information:

http://faculty.washington.edu/bare/qs381/texts.html

We are using the 6th edition of the Larson & Farber text this quarter. No other edition is sufficient.


Sent: Monday, September 22, 2014 10:09 AM
Subject: Welcome to Q SCI 381 Section C

We wish to welcome you to the start of autumn quarter and to Q SCI 381C. We hope you have an enjoyable experience this term.

To access information about this course go to the following web site:

http://faculty.washington.edu/bare/qs381

All course information is posted on this web site, so please familiarize yourself with all of this information. Also, please note that additional study materials are available on the MyLab & Mastering web site which is accessible from the above course web site. Please register on the MyLab & Mastering web site so you can access this material. Free access to MyLab & Mastering via an access code comes with a purchase of the NEW textbook package purchased from the U Bookstore. See the textbook information on our course web page for more information before buying any books for this course.

The first hourly exam is on Wednesday, October 8th. As with all exams, it is closed book. Please bring a hand held scientific or graphing calculator, your Husky Card and your statistical formula/table card (sheets) to all exams (see next email for more information about the formula/table sheets which come with the purchase of a new textbook). When needed, we will provide statistical tables that are not included on your formula/table card.

If you wish to take the exams at an off-campus location such as UWB, UWT, a regional university, a community college or a city/county library you must let me know the name and email address of the exam proctor at your preferred location by the end of week one of the term. If you use the services of UW DRS, please read the instructions shown beneath the table on the following page:

http://faculty.washington.edu/bare/qs381/qs381.html

All homework is self-evaluated and is not turned in for grading. Exam dates are shown on the course web site and all exams MUST be taken on the dates shown. Our course web site will show the times for all exams and exam locations will be posted when available.

Please use your UW email account and UW Net ID when communicating with us. If you use an alternate email system please set your UW email account to automatically forward your email. You will not be able to use the class email list or the online discussion group unless logged into the UW system. Also, be careful when using the class email list as your response might be sent to the entire list instead of only me or the person you intended.

If you need assistance please plan on coming to MGH 084 during office hours of our graduate assistant.

If you have any questions please email me or our graduate assistant or post your question on our class discussion board.

Thanks and welcome aboard!


Sent: Monday, September 22, 2014 10:11 AM
Subject: On-line quizzes, tests, reserve materials, formula cards (sheets), MyLab Access

Access to MyLab & Mastering

The course id for access to MyLab & Mastering is bare05248. The web site is www.coursecompass.com. You should use this web site throughout this course.

On-line Quizzes and Tests

The quizzes and tests that you may access on-line through the MyLab & Mastering web site are for YOUR USE only. The same holds for the quizzes and chapter tests found in your text book. Do NOT send to me or our graduate assistant as they are not graded and they are not counted in computing your grade for the course.

Reserve Materials

The reserve copy of the Larson/Farber text (6th Edition) and a copy of the student's solution manual are available in closed stacks reserve in the OGUL. Online study aids, including an online version of the text, are also accessible from the MyLab & Mastering web site (multi--media or eBook tabs). No other edition of the text will suffice.

Formula/Table Card (Sheets)

If you bought a used text, a new book but not from the UW Bookstore, or use the online text, you may be missing the detachable Formula/Table Card (sheets) that you are allowed to use during all of our closed book exams. If so, you may print a copy from the MyLab & Mastering web site. Go to Chapter Contents (Tools for Success) on the MyLab & Mastering web site.

You should bring your hand held scientific or graphing calculator, your Husky card and all pages of the formula/table card (sheets) -- including pages that contain the statistical tables -- to the exams.

You may write any notes that you wish on the formula/table sheets, but only on the side of the paper that contains printed material. You cannot use your text or any old exams during the exams and you cannot share your formula/table card or anything else with anyone during any exam.

Online Discussion Board

We encourage you to utilize the online discussion board if you have questions. It is accessible from our class web page.

Purchasing MyLab & Mastering Access Separately from the Text

If you purchased a used copy of the text, or purchased a new copy of the text from a vendor other than the UW Bookstore, you probably did not purchase an access code needed to use the MyLab & Mastering web site for our course. If this is the case, the following instructions will allow you to purchase this access code online.

1. Go to the MyLab & Mastering web site and click the Get Registered button in the Student box.

2. Locate the credit card option you wish to use and proceed to purchase your access code online.

3. The course id is bare05248

See the Text Book web page for more information about purchasing access to MyLab & Mastering and/or the online text.


Sent: Monday, September 22, 2014 10:11 AM
Subject: Week One

I hope you found our course web page and are working on the homework assignment one as described on the Reading/Homework page. Homework One is a short assignment that covers some elementary introductory concepts in Chapter One and the first section of Chapter Two.

Our course web site is: http://faculty.washington.edu/bare/qs381

The MyLab & Mastering web site is: www.coursecompass.com.

To register for MyLab & Mastering the course id is bare05248.

All information about the course is available online. You are responsible for reading this information.

A word about the course-- Although this is a relatively easy course, you must spend considerable time with the material and you must keep up. Expect to spend at least ten hours per week studying the material. Once you fall behind it is very difficult to catch up. We cover one chapter per week and we have five hourly examinations spaced about every two weeks. Plan on spending a good amount of time reading and working problems each week. If you have questions, contact either our graduate assistant or me and come to our office hours for extra help. We are here to help you succeed, but you are ultimately responsible for learning the course material and showing your proficiency on the exams.

An online discussion board is also available for you to exchange questions and answers. We monitor this site every day and will participate as needed. We also answer email inquiries frequently during the day. Note you must be logged into the UW computer system to access the online discussion group.

Hourly Examination One is on Wednesday, October 8th and covers Chapters 1-2. A practice examination (with answers) is available on our course web site. In addition, the MyLab & Mastering web site contains chapter quizzes and practice tests that you may use to test your understanding of material presented in the text.

I suggest that if you need extra help you use the video lectures available from the MyLab & Mastering web site under "Chapter Contents".

Best wishes.


Sent: Wednesday, September 24, 2014 9:23 AM
Subject: Late Class Registrants

If you registered late for our class, please note that all of the information you need to succeed in this class is available online at:

http://faculty.washington.edu/bare/qs381

We post all class emails online so you can always look back and find something sent earlier in the quarter. There is also a discussion board available for you to post questions/answers.

MyLab & Mastering is an integral part of this course as it contains much information to help you succeed in the class. You should purchase access to this online resource. Access comes with the purchase of the new textbook package from the U Bookstore.

Only purchase the 6th edition of the text as earlier versions do NOT correlate with the material available on our web site.

Any questions let us know. Thanks and best wishes.


Sent: Wednesday, September 24, 2014 9:24 AM
Subject: Online Communication for Q SCI 381

All online communication for this course is done through the use of email, a class discussion board or the MyLab & Mastering web site.

The online discussion board is accessed by going to our course home page (http://faculty.washington.edu/bare/qs381) and selecting Online Discussion Board. This Board uses the UW Catalyst GoPost System.

From our course home page you can also access the MyLab & Mastering web site which provides videos, extra homework and test problems, answers to the assigned homework problems, etc. which will help you learn the material discussed in our text.

We do not use the UW Canvas system in this course.


Sent: Wednesday, September 24, 2014 9:25 AM
Subject: Statistical Calculation Aids

Although a hand held scientific or graphing calculator will suffice for all problems encountered in this class, MyLab & Mastering also includes an online statistical program STATCRUNCH which you may find useful. You access this program from the MyLab & Mastering website for our class by clicking the STATCRUNCH tab along the left-hand margin of the page or under Tools for Success. This program may be useful for carrying out statistical calculations related to other classes you may take here at the UW. There is also a mobile app for this software.

Other online stat tools are located on our class web site at:

http://faculty.washington.edu/bare/qs381/res.html

Remember, all information concerning how this class is run is available on our class web site at:

http://faculty.washington.edu/bare/qs381


Sent: Wednesday, September 24, 2014 9:25 AM
Subject: pdf of Formula and Tables

If you purchased a used copy of the text, a pdf of the statistical formulas and tables are attached. You may also find this on the MyLab web site at the Tools for Success tab.


Sent: Tuesday, September 30, 2014 1:48 PM
Subject: Week Two and Hourly Exam One

Homework Two (self-graded) should be worked this week. Hourly Examination One is scheduled for Wednesday, October 8th. It covers Chapters 1-2 and is closed book. See our class web site for more information:

http://faculty.washington.edu/bare/qs381/hexam.html

Bring all pages of your formula card sheets, including those pages that contain the statistical tables, (with any additional notes you wish to inscribed thereon); a hand held scientific or graphing calculator and your Husky Card to the exams. You cannot use your text, nor any printed materials other than the formula card and tables, during the exams and you can't share your formula card or calculator with others.

You may arrange to take the exams at a remote location such as UWT, UWB or a county/city library. If you live away from the Seattle campus and wish to do this, I need the email address and name of the exam proctor by the close of business three working days prior to the exam so I can make proper arrangements with the remote site.

A practice exam similar to Exam One (with answers) is available on our class web site. Our exams usually consist of short problems similar to the homework, true-false and sometimes multiple choice. You should also use the review exercises located at the end of each text chapter as a study guide. Additional review quizzes and practice tests are available on our MyLab & Mastering web site.

In Chapter Two we continue our study of graphical ways to summarize and present data; investigate numerical ways to measure central tendency; measure variation in a data set; and measure the position of a data point in a data set. We are also introduced to the bell-shaped distribution, Empirical Rule and Chebychev's Theorem.

If you have questions or concerns about the class send me or our TA a cheerful email.


Sent: Wednesday, October 08, 2014 4:28 PM
Subject: Week Three

This week we study the elementary rules of probability in preparation for the rest of the course. Some of these rules are very simple while others may perplex you for a while. While the math is simple -- the logic may be confusing. In Chapters Four and Five we will apply many of these rules to more applied situations as we move on to study inferential statistics.

Probability is an essential aspect of inferential statistics because when we sample a population and make an inference about the larger population, we can never be absolutely sure that our inference is correct. Thus, there is always a chance that an error is made. Probability theory allows us to quantify the magnitude of this error. Thus, it is a very important aspect of statistics.

If you have questions or concerns about the class send me an email. Also, please work as many problems as you have time for and utilize the MyLab & Mastering web materials.


Sent: Wednesday, October 15, 2014 4:59 PM
Subject: Week Four and Exam Two

In week four, our attention turns to Chapter Four and an investigation of discrete probability distributions. We first learn how to define a probability distribution and then how to calculate the mean (expected value), variance and standard deviation. Some distributions are so common we have attached names to them. Specifically, we study the binomial, geometric and Poisson distributions. For each, we can easily determine the mean and standard deviation using formulas shown in our text. Each of these distributions is defined in terms of one or two parameters.

As usual, work plenty of problems to ensure that you have grasped the basics being described. Please visit our MyLab & Mastering web site for additional study suggestions.

Hourly Examination Two takes place on Wednesday, October 22nd. It covers Chapters Three and Four of our text. Please bring your formula and table sheets (with any additional notes inscribed on the printed side of the page only), your Husky Card and a hand held scientific or graphing calculator. No other materials may be used during the exam. You can't share formula cards or calculators during the exam.

We will hold the exam at the times and location shown on the class web site.

A practice exam (with answers) is available on our class web site. You should also use the review exercises located at the end of each text chapter as a study guide. Review quizzes and exams are also available on our MyLab & Mastering web site.

If you need assistance please contact our TA. Thanks and good luck.


Sent: Wednesday, October 15, 2014 4:59 PM
Subject: 2013 International Year of Statistics

The International Year of Statistics occurred in 2013, but you can still access the following web page for a lot of interesting applications of statistics.

Link


Sent: Wednesday, October 22, 2014 7:49 PM
Subject: Week Five

In week five, our attention turns to Chapter Five and an investigation of continuous probability distributions -- notably the normal distribution. This two parameter distribution is the most used of all distributions we encounter in our studies this quarter.

As usual, work plenty of problems to ensure that you have grasped the basics being described. You can also view videos available from our MyLab & Mastering web site if you need extra help.

The reason the normal is so widely used is due to the Central Limit Theorem. In many practical applications of statistics, we must draw a sample from a population. Based upon this sample, we intend to draw an inference about the larger - unsampled - population. Usually we only draw one sample of some given size. The Central Limit Theorem states that if the sample size is about 30 or larger, the distribution of sample means is approximately normal and becomes more normal as the sample size increases. See our class web site for some additional information on the normal distribution and the Central Limit Theorem.

Thus, even if we sample from a population where the characteristic we are measuring is not normally distributed, if the sample size is 30 or more, the distribution of the sample means is approximately normal and we can use the normal distribution to determine the needed probabilities to place confidence bounds around our estimate of the sample mean.

And, if the sample we draw is random, we also know that the mean of the distribution of sample means is equal to the population mean. Of course, for any one sample, it is very very unlikely (i.e., probability =~0) that the mean of our single sample exactly equals the population mean. But, using the techniques from the next chapter (Six) we are able to establish confidence intervals around our sample mean and these tell us how much confidence to place in our sample results.

The idea to remember at this point is that the normal distribution is very important in our future study of inferential statistics -- largely because of the Central Limit Theorem.

With these inspiring words I turn you over to Chapter Five. Best wishes.


Sent: Wednesday, October 29, 2014 2:15 PM
Subject: Week Six and Examination Three

This week we study Chapter 6. It is a long chapter wherein several new ideas are introduced. You learn how to construct confidence intervals about population means, proportions and standard deviations/variances based upon sample evidence drawn from random samples. In developing confidence intervals about population means, two procedures are introduced depending upon sample size (see flow chart on page 314). One item not explicitly mentioned on page 314 is that if the sample size (n) is < 30, the population standard deviation is unknown, and the population is normally distributed the t-distribution is used. You also are shown how to determine the appropriate sample size required for a given level of confidence and accuracy desired. The maximum error of estimate is also defined and is equal to 1/2 of the confidence interval. When setting a confidence interval for the standard deviation or variance, another continuous distribution is introduced -- the chi-square distribution. Unlike the z and t-distributions, the chi-square is not symmetrical. You also learn how to determine the confidence interval for a population proportion.

As usual, work plenty of problems to ensure that you have grasped the basics being described. MyLab & Mastering is available to provide additional problems and study materials.

Examination Three is scheduled for Wednesday, November 5th. It covers Chapters Five and Six of our text. Please bring your formula/table sheets (with any notes you care to add on the printed side of the page only), your Husky Card and a scientific or graphing calculator to the exam. No other materials are allowed to be used during the exam. You cannot share your formula card (sheets) with others.

We will hold the exam at the times and location shown on the class web page.

A practice exam (with answers) is available on our class web site. You should also use the review exercises located at the end of each text chapter as a study guide.

If you need assistance please contact our TA. Thanks and good luck.


Sent: Wednesday, November 05, 2014 9:38 PM
Subject: Week Seven

This week we study Chapter 7. It is a fairly long chapter wherein several new ideas are introduced. In section 7.1 you are exposed to the concepts that underlie hypothesis testing. This is a very important discussion that involves logic much more than statistics. In the remaining sections of the chapter you apply this logic to hypothesis tests involving population means, proportions and standard deviations based upon sample evidence drawn from random samples.

In testing a hypothesis for population means, two procedures are introduced depending upon sample size. If the sample size is < 30 and the population is normally distributed the t-distribution is used. If n >= 30, we use the standard normal distribution. You also are shown how to calculate and interpret the P-value. When testing a hypothesis for the standard deviation or variance, we use the chi-square distribution. We also learn how to test a hypothesis for a population proportion for a large sample and use the standard normal distribution.

As usual, work plenty of problems to ensure that you have grasped the basics being described.

Thanks and best wishes.


Sent: Wednesday, November 05, 2014 9:38 PM
Subject: Hypothesis Testing Memo

Some additional information related to hypothesis testing can be reviewed at this link.


Sent: Friday, November 07, 2014 1:38 PM
Subject: Type I and Type II Errors

When we test a statistical hypothesis we always use sample-based information to draw our conclusions. Because we do not sample the entire population there is always a chance that we may draw an incorrect conclusion and, hence, make an error. This error can occur in two ways:

a) we may reject our null hypothesis when we shouldn't because it is true, or
b) we may not reject our null hypothesis when we should because it is false

We label the first error a Type I error and the second error a Type II error. It turns out that the probability of a Type I error is the level of significance (labeled alpha) that is established by the user. Usually this is set at 0.10, 0.05 or 0.01, but any value can be used. Thus, we might accept a 10%, 5% or 1% chance of a Type I error occurring. The level chosen depends on the consequence of rejecting the null hypothesis when it is true.

Our text does not explain how to calculate the probability of a Type II error (labeled beta). Suffice it to say that for a given sample size (n), as we decrease alpha we simultaneously increase beta. The probability of a Type II error is also referred to as the false negative rate. Another term often used in conjunction with a hypothesis test is the 'power of the test' which is equal to (1 - beta). Power is equal to the probability of rejecting the null hypothesis when we should because it is false.

Think of the legal analogy. If alpha is the probability of convicting an innocent person, then beta is the probability of letting a guilty person go free. The US system of justice usually sets alpha rather low, which in turn implies that beta is rather large. Hence, society feels that it would rather let a guilty person go free than send an innocent person to prison.


Sent: Wednesday, November 12, 2014 8:29 PM
Subject: Week Eight and Exam Four

This week we study Chapter 8. Material covered is a logical extension from Chapter 7. In section 8.1 you are exposed to a test involving two means based upon two large and independent samples. In sections 8.2 and 8.3 you test hypotheses concerning two means based upon small independent and dependent samples. Lastly, in section 8.4, we learn how to test a hypothesis involving two proportions. Note that a test for two standard deviations or variances is postponed until Chapter 10. As usual, work plenty of problems to ensure that you have grasped the basics being described.

Hourly Exam Four is scheduled for Wednesday, November 19th. It covers Chapters Seven and Eight of our text. Please bring your formula card (sheets) (with any notes you care to add on the printed side of the page only), your Husky Card and a hand held scientific or graphing calculator to the exam. You cannot use your text, nor any other materials, during the exams, and you can't share your formula card with others.

We will hold the exam at the times and location shown on the class web site. As usual, work plenty of problems to ensure that you have grasped the basics being described. MyLab & Mastering is available to provide additional problems and study materials.

A practice exam (with answers) similar to Exam Four is available on our class web site. You should also use the review exercises located at the end of each text chapter as a study guide.

If you need assistance please contact our TA. Thanks and good luck.


Sent: Wednesday, November 12, 2014 8:29 PM
Subject: NYTimes: Share of Homes With Guns Shows 4-Decade Decline

Check the 5th paragraph of this article. Although the data arent provided, this is a good example of a test we covered in Ch 8.4.

Share of Homes With Guns Shows 4-Decade Decline

The rate has fallen from an average of 50 percent in the 1970s to 35 percent in the 2000s, suggesting a shift in demographics even as the debate over gun control remains contentious.


Sent: Wednesday, November 12, 2014 8:31 PM
Subject: Statistics is Great

This is extra reading related to a practical use of statistics. The key message is in the last paragraph.

May 1, 2012

Everyone Should Learn Statistics

During jury duty, Kevin Carey sees an attorney playing fast and loose with numbers. And the guy has decent odds of getting away with it, too, given Americans mathematical illiteracy.


Sent: Friday, November 21, 2014 2:14 AM
Subject: Week Nine

This week we study Chapter 9. This is a new subject that should be of general interest. In section 9.1 you are exposed to linear correlation and scatter plots. A t-test for testing the significance of a linear correlation coefficient is also introduced. In section 9.2 you learn about simple linear regression. Use of a statistical calculator comes in handy for estimating the intercept, slope and linear correlation coefficients. You may wish to use STATCRUNCH (available on MyLab & Mastering) to check your calculations. In section 9.3 you are exposed to ways to measure the goodness of fit of your regression line. Lastly, in section 9.4, you are very briefly introduced to multiple linear regression. STATCRUNCH - may be used to perform these calculations if numerical results are required.

As usual, work plenty of problems to ensure that you have grasped the basics being described.

If you need assistance please contact either our TA or me. Thanks and good luck.


Sent: Wednesday, November 26, 2014 8:44 AM
Subject: Week Ten and Hourly Exam Five

In week ten we study Chapter 10. In section 10.1 you are introduced to a chi-square test which allows you to determine whether a frequency distribution fits a claimed distribution. In section 10.2 you use a contingency table to find expected frequencies followed by how to use a chi-square distribution to test whether two variables are independent. In Section 10.3 you are introduced to the F-distribution. You learn how to interpret the F-distribution and use an F-table to find critical values. Then, you apply the F-distribution to perform a two-sample F-test to compare two variances. Lastly, in section 10.4 you learn how to test for differences between three or more means. This involves an F-test and uses a technique called the "analysis of variance." You learn how to setup and run a one-way analysis of variance test. This is an appropriate test if you have one independent variable (or factor) and you wish to test its mean effect against a single dependent variable. A two-way analysis of variance involves testing the mean effect of two independent variables against one dependent variable.

Hourly Exam Five is scheduled for Friday, December 5th. We will hold the exam at the times and location shown on the class web site.

The closed book exam covers Chapters Nine and Ten of our text. A practice exam is available on our class web site. As usual, work the review exercises and chapter quizzes at the end of each chapter. Please bring your formula card/sheets (with any notes you care to add on the printed side of the page only), your Husky Card and a hand held scientific or graphing calculator to the exam. We will provide copies of the F-table. You will not be asked to use Table 11 in Appendix B. No materials may be shared with anyone else at the exam and no other materials may be used during the exam.

If you need assistance please contact either our TA or me. Thanks and good luck.

There are NO makeups for Exam Five. You must take it before, or on, the scheduled date. Remember, we drop the lowest of your five hourly exam scores when computing your course grade.


Sent: Wednesday, November 26, 2014 8:50 AM
Subject: Class Evaluation Survey

An online student class evaluation form will be available for you to provide feedback on this class. The web site is available between 12/1 and 12/12, inclusively.

You will receive an email from the UW Office of Educational Assessment asking you to complete an anonymous course evaluation. You must login using your UW NETID to access the evaluation form. Thanks much, we appreciate this feedback so improvements can be made for future classes.


Sent: Sunday, November 30, 2014 4:12 PM
Subject: Q SCI 381 Final Exam

The FINAL EXAM will be given on Monday, December 8th. See our web site for the times and location for the exam. The room location for the Final Exam is different from the room used for our hourly exams. No makeups for this exam.

The final exam is closed book. Please bring your formula/table sheets (with any notes you care to add on the printed side of the page only), your Husky Card, and a hand held scientific or graphing calculator. We will furnish an F-table. It is a comprehensive two-hour (110 minutes) exam that is structured to be very similar to the five hourly exams and covers Chapters 1-10. The Final covers all material discussed in the course but stresses material related to inferential statistics. A practice final exam is available on our class web site.

You may not share any materials with others taking the exam and no other materials may be consulted during the exam.

Thanks and best wishes.


Sent: Wednesday, December 03, 2014 2:59 PM
Subject: Hourly Exam Five and Final Exam

Just a reminder that Hourly Exam Five is this Friday, Dec 5th and the Final Exam is Monday, Dec 8th. Please check the web site for the times and locations for both exams.

Since we drop the lowest of the five hourly exam scores when calculating your course gpa, you may decide to skip taking the fifth hourly exam on Friday and use it as your one dropped exam score. However, you MUST take the Final Exam to complete course requirements. Many thanks and best wishes.


To Return to:Q SCI 381, Center for Quantitative Science