MARCH 2009
IN THIS ISSUE:
- Experimental Biology, April 19-21
- Ambiguous results?
- Making sense of ANOVA
- Prism Magic
- Setting up Prism in a teaching lab
CURRENT VERSIONS
CHECK FOR UPDATES
Check here
QUICKCALCS
Free online calculators for radioactivity calculations, detecting outliers, t tests, ANOVA post tests, and more.
CONTACT US
GraphPad Software
2236 Avenida de la Playa
La Jolla, CA 92037
USA
Visit us in the Big Easy!
Stop by and see us at the Experimental Biology 2009 meeting April 19-21 in New Orleans. We'll be in booth #1535. GraphPad founder and CEO Harvey Motulsky and company president Paige Searle will be there. Stop by and let them know what you'd like to see in Prism 6. Or just bring them some beignets from Cafe du Monde!
High R2, but useless results?
When you first look at this log(dose) vs. response curve, it looks
fine. The curve comes close to the points, resulting in a high R2 value (0.997). But, in fact, the results reported with this fit are
useless. How can you tell? Learn to diagnose a useless fit and find
out how to salvage this analysis.
More articles by Dr. Motulsky:
- When plotted on semi-log axes, why is an exponential decay curve not always linear?
- Why does fitting data to allosteric EC50 shift model sometimes lead to 'ambiguous' results?
- Fitting radioligand binding to receptor dimers
- Understanding the '# of subjects at risk' page of survival analysis
- The advantages of centered polynomial regression
Making sense of ANOVA
Analysis of Variance (ANOVA) is a powerful technique in statistics, but one that can be quite confusing and is often abused. If you want to really understand ANOVA, this clearly written book will help. It explains ANOVA from the perspective of comparing the fit of alternative models, rather than the more conventional approach of dividing variation into its components. It has plenty of math, but explains all concepts in plain English.
Designing Experiments and Analyzing Data: A Model Comparison Perspective, Second Edition. by Scott E. Maxwell, Harold D. Delaney ISBN: 0805837183. More details.
Click on the graphs below to learn how to create an error envelope instead of error bars, and how to format individual pairs of data points on a before-after graph.
Use Prism Magic to unify your graphs
We're not promoting mixing science with the black arts -- Prism's MaGiC (Make Graphs Consistent) tool is the fastest and surest way to make all the graphs in a project have the same colors, symbols, fonts, axes, etc. Select the graphs you want to format, click on the Magic wand, and Shazaam!, your graphs are formatted and ready to publish.
Setting up Prism in a teaching lab
Prism is a great tool for teaching students the principles of biostatistics, curve fitting, and graphing. Many universities install Prism on all the computers in their teaching labs, but there can be a potential problem. By default, Prism remembers which files you have recently opened or closed, user-defined equations you have created, color schemes you have made, autocomplete text, and more. While this can be a huge timesaver for an individual researcher, in a teaching lab you want each student to make his or her own choices, not have them automatically remembered from the last person who used the computer. To make Prism start up exactly the same each time, write-protect the preference file. Learn how.
GraphPad Prism, InStat and StatMate are trademarks of GraphPad Software Inc.