Here are my results for the rest of my course:
Module 4
Case - Grade A
Professor comments:
Overall you did a good job with this Alex, you put forth a good effort at trying to bring some order to this assignment and I appreciate your hard work with this.
Dr. C
SLP - Grade A
Professor comments:
Good work here as well. You seem to understand what hypotheses are all about and the basics of hypothesis testing and did a good job overall.
Dr. C
Discussion question:
Scatterplots provide graphical indication of relationships, whether they be linear, non-linear or nonexistent. Correlations are numerical summary measures that indicate the strength of relationships between pairs of variables. A correlation is very useful, but it has an important limitation. Do you know what this limitation is and why this limitation exist?
Guess what are the rho values (correlation coefficients) for the following plots? Possible values for these plots are 0, +/- 0.4, +/- 0.6, +/- 0.8, +/- 0.99.
Discussion answer (Grade A):
Do you know what this limitation is and why this limitation exist?The limitation of correlation is that it depends both of the variables to be random as well as observation independent. The variable will need a normal distribution in the test of the null hypothesis (H0). "Both correlation and regression assume that the relationship between the two variables is linear." (Bewick, et al, 2003)
Ref: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC374386/
Guess what are the rho values (correlation coefficients) for the following plots? Possible values for these plots are 0, +/- 0.4, +/- 0.6, +/- 0.8, +/- 0.99.
a. p=1
b. p=-1
c. p=.4
d. p=-.8
e. p=.8
f. p=.-8
g. p=.4
h. p=0 (Wikipedia, 2011)
Ref: http://en.wikipedia.org/wiki/File:Correlation_examples2.svg
Module 5
Case - Grade A-
Professor comments:
In this assignment you were asked to take a look at the ‘Howie’s bakery’ data and to explore descriptive statistics and draw some scatter plots. You were also asked to run a multiple regression and to consider the t-statistics, coefficients and r-square and to discuss whether the regression was successful. You were also asked comment on ways to improve the regression and to consider exploring time-series.
I think in this assignment there was hope you would perform a multiple regression analysis rather than just a simple regression but overall you seem to have a grasp on the fundamental concepts.
Dr. C
SLP - Grade A
Professor comments:
In this assignment you were asked to comment on several scenarios. One involved the lack of correlation between Abraham Lincoln’s educational background vs. typical backgrounds of past presidents. You were also asked to comment on the hypothesis “When political instability increases, the price of quality increases.” In the Toxic Shock Syndrome case you were asked to comment on causal implications. Finally, you were asked to comment on the ethics of running a bivariate regression between absenteeism and a series of Likert scale items, and you were asked to present the assumptions underlying regression.
In the case of Lincoln’s situation, it’s pretty obvious that he was a rather exceptional, self educated individual and thus, an outlier when compared to other presidents who typically have exceptional academic credentials. With regards to the ‘political instability’ hypothesis, there may be a correlation, but it seems to me there would be a number of intervening variables that would explain the relationship. As far as the TSS case, we’re looking at a pretty small sample here and there also may be a concern regarding the cause/effect relationship given that we are looking at cross sectional data. Finally, I don’t see an ethical problem playing with data post-hoc, and frequently bivariate data is used in regression even if it is not ‘normal’ per se. However, post-hoc findings do have to be reported as such.
You made some good observations in your discussion Alex!
Dr. C
Discussion question:
Three prisoners, A, B, and C, have applied for parole. The parole board has decided to release two of the three, and the prisoners know this but not which two. Prisoner A realizes that it would be unethical to ask the warden if he, A, is to be released, but thinks of asking for the name of one prisoner other than himself who is to be released. He thinks that if the warden says “B will be released,” his own chances have now gone down from 2/3 to 1/2, because either A and B or B and C are to be released. And so, A decides not to reduce his chances by asking. Explain why A is mistaken in his calculations.
Discussion answer (Grade A):
This is almost the same type of question as the "what's behind door number ..." in module 2 with the exception that the prisoner may be too scared to find out the answer as to ruin their chances of being released. The reasoning is that if they ask, the warden may consider this an act of misconduct versus if they stay silent, they may get released. There is also a chance that the reverse may occur if they ask and the prisoner will get the answer they are looking for. So there are morality issues that come to play as well.
The reason that the prisoner is wrong is because of the following:
"The incorrect solution, and the source of the prisoner’s “paradox”: Three possible pairs of prisoners may be released, with each of the three pairs equally likely: (A & B), (A & C), (B & C). Knowledge of B’s release eliminates the option (A & C), and the conditional probability that A will be paroled becomes P(A paroled|B paroled) = 1/3 / 1/3+1/3 = 1/2"
The correct answer comes from 4 possible solutions:
"(B&C, B) 1/6
(B&C, C) 1/6
(A&B, B) 1/3
(A&C, C) 1/3"
Overall, the door question and the prisoner question both have the same probability initially, however, the prisoner has not considered the warden's motives or choice in the decision.
Source: http://www.amstat.org/publications/jse/v13n2/carlton.html
Professor comments:
You made some very well thought out contributions to the thread this session Alex!! Best of luck in your future endeavors!!!
And that ends RES600.
I am transferring out of Trident University (TUIU) to go to University of Phoenix now because I do not feel that TUIU is for me. The school was good for my masters but for my phd, I feel that I need a more thorough understanding of what I am learning. At the moment, I feel that I am moreso learning a crash course in statistics. It's a pain so hopefully UoP will be more thorough so that I can get an actual understanding of what I am learning in school.
