11.2 T-test for a difference of independent sample means

With a second sunny day in a row, the obvious choice was to teach the students to make a better paper aircraft for distance of flight. 

Shi Ann's plane can be seen at the extreme upper left

The darts tended to embed into the soft soil

Although hard to see, the planes generally flew further

On the ground data values were entered using the Tripltek tablet




Calculations were then made on the Smartboard in A204. For 11.2 only the p-value was calculated. Cohen's d will come up in 11.3. The p-value is less than our alpha of 0.05 therefore we reject the null hypothesis of no difference in the sample means. The new design flew statistically significantly further. Note that the test remains a two-tail test to retain the relationship with confidence interval hypothesis testing. This course takes the position that if one cannot get a significant difference on a two-tailed test, then moving to one-tailed test is an error. One tailed tests should not be done. 

The type 3 in the TTEST function reflects that the standard deviations are different. 


Friday I tried to approach the pooled standard deviation by showing how a triangle with a leg of 20 and another leg of 21 "add up to" a length of 29. Not sure this helped.

Comments

Popular posts from this blog

Plotting polar coordinates in Desmos and a vector addition demonstrator

Setting up a boxplot chart in Google Sheets with multiple boxplots on a single chart

Traditional food dishes of Micronesia