Webinar : Presented by Jeff Witmer
Title: Regression to the Mean/The regression effect
Please sign up to register, link to register: [ https://www.causeweb.org/cause/webinar/teaching/2017-10 | https://www.causeweb.org/cause/webinar/teaching/2017-10 ]
Abstract: Regression to the mean, also known as "the regression effect," is an important but sometimes overlooked topic in introductory statistics. We will discuss the regression effect and how to teach it. We will also consider a number of examples of the "regression fallacy," in which people who are ignorant of the regression effect make up ad hoc (and sometimes very misleading) explanations for what they see in data.
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Logistics: The webinar will be conducted using the GoToWebinar software platform. A computer with internet access is all you need. GoToWebinar offers audio participation through your computer microphone.
All registered webinar attendees will receive a confirmation email generated by the GoToWebinar system upon registering. This email includes a link to enter the webinar . Keep this confirmation email as you will use this link to enter the webinar – you will also be sent a reminder with the link two hours before the webinar begins. Once you leave the webinar , you cannot re-enter. If you have not used GoToWebinar before, please review the information below. The webinar will be recorded and the archived version will be available on-line within a few days following the presentation, if you are unable to attend.
For PC-based participants:
* Internet Explorer 7.0 or newer, Mozilla Firefox 4.0 or newer or Google Chrome 5.0 or newer. JavaScript must be enabled.
* Windows 8, 7, Vista, XP or 2003 Server.
* Cable modem, DSL, or better Internet connection.
* Dual-core 2.4GHz CPU or faster with 2GB of RAM or more.
* Participants wishing to connect to audio using VoIP will need a fast Internet connection, a microphone and speakers (or USB headset).
For Mac-based participants:
* Safari 3.0 or newer, Firefox 4.0 or newer or Google Chrome 5.0 or newer. JavaScript must be enabled.
* Mac OS X 10.6 – Snow Leopard or newer.
* Intel processor with 1GB of RAM or more.
* Cable modem, DSL, or better Internet connection.
* Macs have built-in speakers and a microphone with ambient noise reduction that will work well for VoIP.
The CAUSE Cartoon Caption Contest for October is now taking entries (with deadline of October 20 )
The Consortium for the Advancement of Undergraduate Statistics Education is happy to announce our seventeenth Cartoon Caption Contest. Each month a cartoon, drawn by British cartoonist John Landers, will be posted for you and your students to suggest statistical captions. (note the cartoons are posted the month before with submissions due by the 20th)
The next cartoon and the entry rules for the contest ending October 20th are at
[ https://www.causeweb.org/cause/caption-contest/october/2017/submissions | https://www.causeweb.org/cause/caption-contest/october/2017/submissions ]
The best captions will be posted on CAUSEweb and the winner(s) will receive their choice of a coffee mug or t-shirt imprinted with the cartoon or free registration to eCOTS 2018.
Enjoy.
September Results: We had 21 submissions for the September caption contest that featured a cartoon showing two patients in a hospital with one looking very sick and the other looking quite healthy. Charts at the front of their beds show a downward trend for the healthier looking patient and an upward trend for the sicker looking patient. The winning caption for the cartoon was
“Charts not starting at 0 make doctors lose patience!” written by Larry Lesser from The University of Texas at El Paso. The judges chose this caption for it’s play on words and its ability to start a discussion on the importance of appropriate axis labels. Another caption that also noticed the lack of a scale on the charts read simply “Label your axes!” and was submitted by Kyle Falbo of the College of the Redwoods. Two other honorable mentions noticed the disconnect between the charts and the condition of the patients: “Nurse, were these charts doctored?” by Larry Lesser, and “Looks like a bad case of Regression to the Mean” by an anonymous contributor. A final honorable mention went to Chantal Larose of Eastern Connecticut State University for her caption “On average, our patients show no improvement over time.”
Thanks to everyone who submitted a caption and congratulations to our Winners!
Police Data Challenge - Pre-register by October 20 and Submit by November 3
ASA's public education campaign, [ http://thisisstatistics.org/ | This Is Statistics ] , in collaboration with the Police Data Initiative, has launched the [ http://thisisstatistics.org/policedatachallenge/ | Police Data Challenge ] -a national contest for high-school and undergraduate students to put their statistical and data visualization skills to work creating safer communities. Data sets from metropolitan police departments in Baltimore, Cincinnati, and Seattle are available for participants to peruse in formulating analyses and recommending innovative solutions to enhance public safety. Teams of 2–5 high-school or college undergraduate students in the U.S. and Canada can submit an entry. What’s required? A short essay describing the team's process and a presentation of their analysis and recommendations via PowerPoint. Awards will be given in the categories of Best Overall Analysis, Best Visualization, and Best Use of External Data. Submissions are due by Friday , November 3 .
We hope your undergraduate or high school students enjoy participating!
Thanks,
Rebecca
Rebecca Nichols
Director of Education
American Statistical Association
Promoting the Practice and Profession of Statistics ®
732 North Washington Street
Alexandria, VA 22314-1943
[ callto:(703) 684-1221 | (703) 684-1221 ] , Ext. 1877
Fax: [ callto:(703) 684-3768 | (703) 684-3768 ]
[ blocked::http://www.amstat.org/ | www.amstat.org ]