Main Point Discussion Why do the effect-size calculations use standard deviation rather than standard error? (Add Main Point Discussion here - 200 words or more - Use examples, or charts, etc. to build on the Discussion Point) Discussion Response 1 Cheri Burton - Question 5 Explain random assignment and what it controls? Random assignment is a primary consideration of a scientific study. What a random assignment does is ensure that every study participant has a chance of being assigned to any level of the independent variable. (Nolan,2014). Random assignment is an essential component to internal validity, and allows the researcher to make a causal claim about effects that occur from variable intervention (Statistics, 2016). Random assignment occurs after participants have already been chosen for the study. There is no guarantee that the random groups will be equal only that the experimental results that are found are due to chance. In random assignment the experimenter controls circumstances for the members of the study (Nolan, 2014). So in choosing a wrong example of splitting participants up would be if there were 50 students that had been chosen. Twenty five were blonde and twenty five were brunette, and we were trying to see if drinking coffee before studying can get better quiz grades. So if we gave just the blonds coffee and not the brunettes, and then had them take quiz's this may be biased because we chose the groups based on hair color and each group regardless of hair color did not have a chance of getting coffee. So in a random assignment we may put numbers in a bowl from 1-50 and we have predetermined that 1--25 will be in group with coffee, and 26-50 will not get coffee. This would be a random assignment. I found interesting information on randomization while researching this. Randomization was first emphasized in the theory of statistical inference by Charles S. Pierce. He wrote "Illustrations of the Logic of Science" (1877-1878) and "A theory of Probable Inference" (1883) (Pierce,1883). His first study using this method was on weight perception. Isn't it crazy that he thought about randomization and experiment results over 100 years ago? (Add Discussion Response 1 here - 100 words or more - Add some feedback to build on the discussion) Discussion Response 2 Tara Zbikowski - Question 3 When is it appropriate to use a single-sample t test? In my own words, you would use a one-sample t test if you want to estimate the mean of a certain sample and then want to compare it to a specific target (population) when you do not actually know the standard deviation of the said population. You will be able to see if the mean of a group (sample) is different than a specific target. For an example, the manager of a delivery food restaurant collects the times that it takes to deliver the food to the customers. He/she does this randomly. They will then use the one-sample t-test to determine if their driver's delivery time is lower than a competitor's advertised delivery time. When using the test, you will calculate the differences between the mean of your sample (that you randomly chose) and the hypothesized mean that is relative to your sample. The smaller your sample size and larger the difference means the greater the chance that the population mean will differ. According to Chapter 9, page 227, "The logic of the single-sample t test is a model for other t tests that allow us to compare two samples--and all of the other more sophisticated statistical tests that will follow". (Add Discussion Response 2 here - 100 words or more - Add some feedback to build on the discussion)