Math 37 Questions and Answers

About the Discussion list

A Discussion list or "listserver" (mailing list) has been set up expressly for the use of students enrolled in Math 37 at UOP. Students are encouraged to use this list for anything related to the class that they see fit. Discussions relating to the topics of the class as well as discussions relating to homework, and other articles about statistics of interest are encouraged. This mailing list enables you to ask questions about the course outside of the normal class time and office hours. It is not intended to replace coming to office hours, but to increase the availability of answers and the spread of information.

Subscribe to the list You will need to enter a valid email address. You many use GroupWise or any other email address where you want to receive your mail. Enter your name and address and then click on the Subscribe button.
Previous postings to the listserv this semester
Submit a question to the list. (You must include a valid email address.)
Submit an anonymous question to the list (send to me and I will post it).

Previous Postings to the list (Spring 1999)

Question: I was doing the first lab and got to the second task which asks us to subscribe to the email list. I followed the directions and clicked on "subscribe to the list" link, but when it took me to the link there wasn't anywhere to put my email address. The Math 37 Subscription Page didn't come up looking like the example on page 28 of the lab manual.

Answer: Make sure you're in a newer version of Netscape, e.g. 4.0, to be able to run that Java applet

Question: I was looking at the homework for Friday and I came across a question asking about political identity. I was wondering what that was about?

Answer: By "Political identity" I mean which political party (e.g. Republican, Democrat, etc.) that they identify with. (BC)

Question: In an observational study, is there an explanatory variable? The definition of that variable is to actively change something. Since we do not actively change something in an observational study, does that mean there isn't one?

Answer: Yes. I'd say the definition of an explanatory variable is what we think is leading to or preceding the changes in the response variable. If we are able to actively change it ourselves, that means we are conducting an experiment and not an observational study. But there are cases when we can't control it (e.g. smoking) but we still think it's explaining the other variable (e.g. cancer). (BC)

Answer: That is probably asking whether that person is a democrate, republican, independent, etc. (KL)

Answer: I asked Dr. Chance today and she said that it was what political party you were for. It is not a numerical value. Hope it helps. (DD)

Question: For Lab 3, it doesn't ask for a full lab write up. So are we supposed to answer the questions in paragraph form or is there a certain format we have to follow?

Answer: Just answer the questions :) (Still with the aim of integrating the graphs among your answers...) (BC)

Question: How can you tell which numbers have the smallest possible deviation number? For instance, is there a formula or is it just trial and error?

Answer: Earlier I stated that we would be learning about regression lines and that sort of thing. I got a little ahead of myself. In this case, since we only have one variable, the best way is to think about how it is that we can have the least deviation from the mean. In class we talked about having a deviation of zero as the smallest. With the understanding that numbers can be duplicated, how would you arrive at zero? There are several answers to this question (11 to be exact!) (MB)

Question: For Lab 3, it doesn't ask for a full lab write up. So are we suppose to just answer the questions in paragraph form or is there a certain format that we have to follow. (WY)

Answer: In the back of the lab book...I think that it is the index or something like that...there is a section that tells everything that the lab write up is supposed to contain in it. It has the details of the discussions and analysis sections and that stuff. It is all stated. I am pretty sure that she wants all of the labs to follow that format. Hope this helps. (LT)

Answer: Actually, what's in the appendix applies to "full lab write-ups". Just so you know I'm not completely evil, not every lab this semester has to be in that format. The title of the lab will say (full) next to it if I want a full lab write-up. Lab 3 is not a full lab writeup, but Lab 4 is. Usually we do full lab write-ups when we have to collect data, then we need to talk about the data collection procedures etc. We will do this 3 more times this semester... (BC)

Question: for the lab 5 write-up, how do you do number 2, part d and e? and what kind of graphs are we supposed to do? (TC)

Answer: You are supposed to use the information given in the problem statement, e.g. Normal with mean 34 and SD 3.5. The number 95% on it's own should ring a bell... For part (e) you need to think "backwards", start with the .10 work back to a Z value and then to a mpg value... (BC)

Question: CURTAINS!!! Can anyone explain to me why the switching has a higher percentage of winning than staying??? (lab 6) I don't understand. (KL)

Can anyone explain to me the results of the stay column and the results of the switch column it part three of the lab? (CJ)

Answer: The reason that the probability of winning is higher in the switch is because there is only one way to win if you stay, but there are two ways to win if you switch.

For example: here are the ways to win:

goat -- goat -- car

2) goat -- car -- goat

3) car -- goat -- goat

now if you take the first curtain at all times, then 2 outta the 3 times the goat will be revealed... and the car will be the other curtain. but only one of those times do you win if you stay. That is why the probability of winning in switching is 2/3 and the probability of winning in staying is only 1/3. I hope that didn't confuse anyone... (FC)

Answer: first try making a graph of all the possible outcomes. (MB)

Question: In Lab 7 , it asks to compare the center of the graph to the center of the population. Does this mean we compare the medians or the means of two graphs? (WY)

Answer: In Lab 7, use the means. (BC)

Question: I'm kind of confused about the question 4 (a). A and B don't depend on each other (I think) because you can get one without getting the other. What you roll first would not affect what you roll next. B is not dependent on A, but A is dependent on B? Am I getting it down right? Please help me! (L)

Answer: Let's put it this way:

If A is dependent on B, then B is dependent on A

If A and B are independent, then B and A are independent.

You might want to start by finding P(A), P(B), P(C)

Then see if P(A knowing B occurred)=P(A), if so, then A and B are independent... So if someone tells you "I'm looking at my die and it's even", what probability would you assign to that die being a 6? Is that different then if you just rolled a die with no other information? (BC)

Question: On practice problem 2 part a why did the book use .3707 as the probability instead of .6293? Isn't 1/3 a positive number so why did they look up a negative number in the table? Also, how do you do parts b and c? (SP)

Answer: Great questions. Part (a) wants to find P(X>105) so I want the area above 105 which corresponds to the area above 1/3 in Table A. Table A gives you the probability below, so you subtract from one to get the probability above. Remember to sketch your curve and shade the area of interest. Since 105 is above the mean, the area beyond 105 should be less than half.

(b) uses the facts we learned about the sampling distribution of x-bar. You might want to review those properties, e.g. p. 400. Once you know how to find the mean and SD of the x-bar, and that it follows a Normal distribution, you can standardize the observation to find P(X-bar > 105) the same way... (BC)

Question: In the lab writeup for Lab 7, can someone explain to me what we are suppose to do for number 8? (WY)

Answer: For each graph (A-E or at least the three you didn't pick earlier) suggest a population shape, a sample size, and a number of samples that could lead to that picture for the sampling distribution. In fact, you might be able to just change one of those characteristics at a time. So you'll say things like, to get this kinda shape for the sampling distribution my population would've had to look like .... and my sample size would've had to be around .... (BC)

Question: writeup #6, I do not find an example where the Central Limit Theorem did not apply. Do this implies that I've done something wrong in the lab? (LR)

Answer: Basically, did you ever find a sampling distribution that wasn't quite normal? (BC)

Question: For the Possible Extension Activities on Hw6,I would like to know how to do these extension activities. Is there a certain way of working the problems? Or do I just answer the questions? (SY)

Answer: The extension activities are VERY open, whatever you find of interest. Just discuss something you found interesting and, MOST IMPORTANTLY, what you thought about it. For example, let semester a student went with her grandmother to bingo night over Thanksgiving. At first she was dreading it, but then she actually found herself thinking about some of the statistical issues we discussed in class. For the ones I suggested in HW 6, reading the article and telling me what you think about the issue would suffice. Some more details and older examples can be found http://www.uop.edu/cop/math/chance/coursenotes/extension.html (BC)

Question: when are these extension activities and the extra credit things do?how many of each can we do? (TC)

Answer: The extension activity is due by the end of the semester though I would advise you to get them done soon (and go ahead and turn them in). See your syllabus for some extra credit opportunities. (BC)

Question: I don't understand how to figure this problem out. I tried to use Table C. (Is that right?) But the numbers did not go high enough to look up the probability. Can anyone help me please? (L)

Answer: You want to use the normal approximation to the binomial... we won't be using Table C, we'll hope n is large enough that we can keep using Table A... (BC)

Question: How do you get the homework solutions from the web page? I forgot what the login name and password is.

Answer: The username is: math37; The password is: tiggers (BC)

Question: When I was working on HW6, I was skimming through the chapter and saw that we were going to study the section of Binomial Formulas and Binomial Coefficients. When we get to this section, which I'm sure we will really soon, can you please spend a little more time on it, giving a lot of examples and details. I think it would take a while to synthesize these concepts. The Binomial Coefficients calculates the number of ways of obtaining success? Would the Binomial Probability be similar to what we learned in class today, like confidence intervals? (I don't mean it's the same, but same correlation? I hope I'm not confusing you. Ok, what I mean is that the binomial probability calculates the probability that k is one of the possible success values. So this would be similar to the probability that xbar is within 1 STD or 2STDs of the mu if those values within the specified STD's are considered as success? Am I making sense here? Well, I just don't wa! nt to! be too late in asking these questions. Hope this helps everyone else too. (SY)

Answer: The book might be but *we* aren't... It's not worth spending the time on the calculations, we assume a normal distribution instead and stick with those. Yes, it gives an exact probability for the count of successes, whereas the normal distribution gives us an approximation. Not exactly, but you had the computer calculate and exact binomial probability for you in lab yesterday... If X is Binomial, then you can calculate P(X=k) for any number of k successes. You could ask questions like this. You could also ask questions like, if the probability the interval contains the parameter is .95, then what's the probability that 5 out of 5 intervals each capture the parameter. This would be a binomial calculation, though not one we'll do. (BC)

Question: Can someone help me on the above question? 1. what does " the scale is balanced" mean?(hw1)

Answer: A balanced scale is "unbiased", it will read close to the target weight and not by consistently off in one direction or the other.

Question: 2.what does 'the election was too close to call" mean?hw2(6.20b)

Answer: The results are inconclusive and I don't feel comfortable saying one candidate will definitely win. You may have seen on election night, they count returns (votes) for a while and at some point they are willing to predict who will win even though all the votes haven't been cast yet.