Table of Contents

• 1: Do those shoes fit?
• 2: Dilbert's 3 day work week
• 3: The day is saved ... or not
• 4: What you observe vs expect
• 5: The Chi-squared Equation
• 6: Give it a try
• Detour
• 7: What's a p-value?
• 8: What's a Lookup Table?
• 9: What's a "df"?
• Completing the Chi-squared Test
• 10: Interpreting a Chi-squared
• 11: The answer, finally
• 12: Summary
• 13: What about 90% M/F sick days?
• 14: What about 50% M/F sick days?
• Chi-squared in Biology
• 15: Testing for a dihybrid ratio
• 16: Review and Words of Wisdom

The day is saved ... or not

Apparently we have saved the day ... 40% of sick days SHOULD fall on Monday or Friday, which means that employees are not abusing the system.

But wait. What if next year, the evil pointy-haired boss finds that 42% of sick days fell on Monday or Friday? Proof positive, in his view, that employees are out to get him.

Let's be Dilbert for a minute. How could we confirm or disprove the evil pointy-haired boss' claim? Clearly 42% is more than 40% -- but how much is too much? Do the extra 2% just represent the natural "slop" around 40%?

Or, what if next year 90% of sick days fell on Monday or Friday? Would that make you think that Dilbert was wrong, and sick-days were not random? What about 50% of sick days on Monday or Friday?

When you use statistics, you can do one of two things. First, where your data is fairly definitive, you can calculate a numerical score that simply confirms common sense. Secondly (and more interestingly), when your data falls into a 'grey area', statistics can allow you to make a decision one way or the other. Accordingly, what we expect out of statistics is the following:

• If 40.1% of sick days are Monday or Friday: statistics confirms that this most likely fits the random sick day model (not a grey area).
• If 50% of sick days are Monday or Friday: statistics helps me to make a decision about this "grey" area.
• If 90% of sick days are Monday or Friday: statistics confirms that this most likely doesn't fit the random sick day model (not a grey area).