Amount of Data

• Quite obviously, the more data you have the more likely your stats are to be significant.
• Consequently, you want more data.
• If you run the prior tTest numbers through the calculator, as you increase the df (or numbers of data), the significance goes down.
• Note that it really is not significant as they are based on the same data.
• Also note that you really need to have a good idea what is going on.
• The ttest that we're using assumes a normal distribution.
• Since there isn't (it's uniform), the test is less viable.