Q&A on Assignment One: Quiz on Friday Sept. 19th
Q: It appears to me that there are two parts. The first, the computer part will account for 3 out of 10 mark. The second component consists of the quiz on sept.9, this will account for the remaining 7 marks. Is this correct? A: Yes. Please bring your printout to the quiz. You will be asked to hand it in. Q: i am a little perturbed about how i should go about studying for the quiz! is it more so a definition level or is it at a conceptual level or is it a compare and contrast, or is it just everything?!?! and i was also wondering if there is any reading in the text that i should be doing at the same time? A: For 3 points out of 10, you'll hand in the printout on Friday. For the other 7 points, you'll be answering questions very similar to those on the "conceptual part." A good way to prepare would be to write out answers to all the questions (including original examples of all the tests mentioned), and then make up and write out answers to similar questions that YOU make up. Of course you will not be asked to hand this in, but it is likely that what you write on the quiz will be very similar to what you have prepared. You should carefully re-read the class notes, but there is no need for the text, YET. IN THIS NEXT ONE THE QUESTIONS ARE PRECEDED BY > AND THE ANSWERS ARE NOT. > > Can you tell me if the answers I came up with for some of the questions in > Assignment 1 is ok? > > > 1)Answer: A study is being conducted to see if there is a relationship > between the amount of fat consumed and the person's blood pressure and eye > sight. Random sample of 50 people were given different amount of fat in > their diet, and after one month, their blood pressure (low, high) and eye > sight (good, poor) were recorded. Good; full marks. Anybody else who writes this one on the quiz gets a zero. > > 6) Answer: Its not possible for a study to be both experimental and > observational. In an observational study, values of all independent > variables are not randomly assigned. (the key word is 'all'). This answer is good too. I wish I had asked "Is it possible for a study with one independent variable to be both experimental and observational?" Here the answer would have been no, and it would have been less tricky. > > 7) Answer: Not necessarily. While it is possible that some people who > graduate from University have higher lifetime earnings, it is also > possible that they were physically very healthy and so they were able to > work for a longer period. It may be that they were employed (therefore > had earnings) way before graduating, and were able to effort University > education and graduate. > Here, I think you are borrowing the phrasing too directly from the class notes, and it hurts the answer. You should take it as true and given that people who graduate from university have higher lifetime earnings on average than those who do not (it really is true). Let me re-write your answer to make it clearer. While it may be that a university education causes some people to earn more money (and it probably does not hurt anyone's earnings), it might be that healthier individuals are able to work for a longer period over the course of their lives. It may be that these exceptionally healthy individuals also tend to be employed (and therefore have earnings) way before graduating, and so are more able to afford a University education and graduate. Note that in your alternative explanation, it is important to argue explicitly for a link between the confounding variable and BOTH the IV and DV. You did this and that's good. I must say that I don't think your alternative explanation is very plausible (at least not in North America), but that's no problem unless the alternative explanation is totally crazy -- and yours is not. > I don't know the answer to questions 4 and 8. Is there anyway you can outline the answer? Four is VERY tough, but how about using a median test? A hint for eight is it's a 5 by 2 table -- now YOU label the rows and columns (the entries in the 10 cells are frequency counts of cases).