4:544

4:54 < Goodness of Fit (MMs) Goodness of Fit Often, we will want to test how well a real-world scenario fits a distribution. We can do this by using a 1'2- distribution with a goodness-of-fit test. It works very much like other hypothesis tests. Generally: We make a claim about what we think the distribution is, Then we collect data and compare our results to what we expected, And finally, we decide whether our model is a good fit for the population. M&M's candies typically come in six assorted colors, as listed below. How do you think the colors are distributed? Uniformly? More blue, less green? All yellow? Fill in the table below with your proportions you predict will describe the distribution of the various colors of M&M's. Color: Proportion: Red Oran e Yellow Green Blue Brown Open your bag of M&M's. Don't count each color yet! What is your sample size? Based on this sample size and your predicted distribution, determine the count of each color you would EXPECT to have in your sample, and fill in the appropriate column. Now, count each color in your bag and fill in the appropriate column in the table. Color Expected Observed Fre uenc Frequency Red Orange Yellow Green Blue Brown Now let's see how a sample of M&M's will fit your predicted distribution. Using the table above and StatCrunch, perform a Goodness-of-Fit Test with a significance level 2.5%. Hypotheses Condition Calculate Dashboard Calendar To Do Notifications Inbox