StatMate

StatMate is an ideal companion to Prism and InStat. StatMate complements these programs by providing a collection of 8 statistical calculators to help you design statistically sound experiments and to help you interpret results. StatMate comes with a short but comprehensive manual that explains the uses of each calculator, and how to interpret the results.

Plan an experiment
Three of StatMate's calculators help you properly design and set up your experiments, so that your results are statistically sound.

• Test advisor - StatMate helps you pick the right statistical test to use. Tell it the type of data you are collecting, and the kind of comparison you want to make. StatMate suggests a test, and tells you which GraphPad programs support that test.
  
• Calculate sample size - How many subjects (data points) do you need? Naturally, the answer is "it depends". It depends on how large a difference you are looking for, how much your data vary, and on how willing you are to risk mistakenly finding a difference by chance or mistakenly missing a real difference. StatMate helps you see the tradeoffs, so you can pick an appropriate sample size for your experiment- To avoid subtle biases, it is best to randomly assign subjects to treatments. StatMate will do this for you. Simply tell it how many treatment groups you need, and how many subjects you want in each group. StatMate randomly assigns each subject to a group.

Interpret statistical results
The remaining calculators provide you with tools to gain further insight into your own or published results.

• Calculate confidence intervals - While many publications focus on P values, results are often easier to interpret when expressed as a confidence interval. StatMate calculates a confidence interval of a proportion, count, or mean. You specify whether you want 90%, 95% or 99% confidence.
  
• Calculate power - Just because a study reaches a conclusion that the results "are not statistically significant" doesn't mean that the treatment was ineffective. It is possible that the study missed a small effect due to small sample size and/or large scatter. StatMate calculates the power of a test to detect various hypothetical differences.
  
• Binomial or Poisson distributions - Knowing the overall success rate (binomial) or average number of events per time or objects per volume (Poisson) in a population, you might be interested in predicting the likelihood of observing a particular outcome. StatMate calculates the probabilities and reports the distribution of results.

  Bayesian calculations - When reviewing the results of an experiment or laboratory test, you need to interpret the statistical information in the context of what you knew before you did the experiment (based on theory and previous data). StatMate performs the Bayesian calculations required to do this.

Simple to use
StatMate is completely self-explanatory. View the StatMate screen to see how it works. Choose a calculator by clicking one of eight tabs at the top of the screen. Answer the questions at the top of the screen, and click the Calculate button. StatMate reports your results in the bottom half of the window. It's that simple.

There is no demo version of StatMate available, but you may download the StatMate manual (415 kB) as a pdf file onto your computer to learn about StatMate. You will need a copy of Adobe Acrobat (readily available) on your system to view or print this file.

System requirements
StatMate runs under Windows 3.1 or Windows 95, 98, NT 4, Me, 2000. It occupies less than 1 MB of space on your hard drive.

GraphPad Software, Inc., 5755 Oberlin Drive, #110, San Diego, CA 92121. Tel: 858-457-3909; Fax: 858-457-8141.