Winter, 2010 Wednesday, Feb. 17Stat 217 – Day 19Two-sample t-tests (Topic 22)Example 1: Place one of two types of chips on your tongue and hold it to the roof of your mouth. Record how long it takes (in seconds) for the chip to melt completely – without any “encouragement” onyour part. On a separate piece of paper record:Chip Time (sec)Example 2 - Lab 3 Recap:We wanted to compare two groups on the number of letters remembered: Statistical Inference: Could this difference in means (16.76-9.47 = 7.29) have happened by the random assignment process alone?(a) What is the statistic in this study? (What symbols can we use the represent it?)(b) Define a corresponding population parameter.(c) State null and alternative hypotheses about this population parameter.(d) What should the sampling distribution of the statistic look like under the null hypothesis? How did we go about investigating this in Lab 3?Winter, 2010 Wednesday, Feb. 17(e) Here is one set of 1000 hypothetical random assignments of our scores: How would you label the horizontal axis? How would you determine the p-value?(f) Estimate a “z-score” based on this simulation.(g) Formal test statistic formula:Test of Significance Calculator:Technical conditions for this approach: (h) Interpret the p-value (what is it the probability of?). Then state your final decision in context.(i) Interpret the confidence interval (Hint: Remember what our parameter is)Back to Example 1: Is there a better way to conduct this study? Why is it better? How will we analyze the
View Full Document