Week 6 of Notes
Week 6 of Notes 400
Popular in Research in Political Science
Popular in Public Relations
This 4 page Class Notes was uploaded by Dora Notetaker on Friday October 2, 2015. The Class Notes belongs to 400 at University of Alabama at Birmingham taught by John E McNulty in Summer 2015. Since its upload, it has received 38 views. For similar materials see Research in Political Science in Public Relations at University of Alabama at Birmingham.
Reviews for Week 6 of Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/02/15
PSC 4002D Research in Political Science Dr McNulty Set of Notes 6 Week of Sept 28 TueSept29 Dr McNulty gave us a handout at the beginning of class about Steps in Calculating a Standard Deviation We went over the 1st problem set that was due today Here are the questions from the problem set look on canvas for the articles 1 Identify one hypothesis in the article Note The article may contain more than one hypothesis but focus on a single hypothesis in answering this question and those below 2 What is one independent variable discussed in the article What are at least two possible values of the independent variable 3 What is one dependent variable discussed in the article What are at least two possible values of the dependent variable 4 How would you model the proposed hypothesis using the variables X and Y nclude a coefficient for slope and a constant for the Yintercept 5 Can one know with certainty whether the hypothesis in the article is true Why or why not Be specific How do we formally handle this conundrum 6 What is one piece of evidence either supporting or refuting the hypothesis If no evidence is given in the article what is one piece of evidence that could be used to support or refute the hypothesis 7 In no more than two to three paragraphs assess the evidence That is explain why the evidence is or is not convincing Consider each of the steps involved in making a persuasive causal claim Here are the answers to the first problem set using the first article Which talked about children s behavior problems if they are older than their classmates 1 Extreme behavior in children when the age of the child is more than the age of the child s classmates 2 Independent variable Age of the child or age of the classmates but essentially age 3 Dependent variable Behavior a Don t forget this can be hard to measure b 80 on the test don t put Dependent variable 5 demerits per every year of age 0 because then you d be defining a relationship not a variable d so stick to variables 4 X Age Y Demerits Equation Demerits 1 Age3 s 5 No we can never be certain because there will always be error and some things are just simply random So the answer here will always be no 6 A piece of evidence against the hypothesis is that bad behavior compared to age only happened to white children in the study not black children 7 Here you have to use the 4 C s to assess the causal claim a Correlation There is a correlation between age and behavior b Clock Age is what affects behavior so age comes before behavior c Causal Mechanism An explanation as to why the correlation is happening The article does not give a good causal mechanism d Controlling for Confounds Outlying variables that could be affecting the child s behavior besides age So a confound can be unstable home life mean siblings or bad teachers e Lastly you have to say if this article is making a good causal claim He said in his opinion none of the articles were No the article does not make a good causal claim because it does not explain why the correlation is happening and it does not control for confounds Thursdav Oct1 2015 We spent some more time going over the first problem set which he will give back to us next week Reviewing central tendencies 1 Mean Average the sum divided by the number of cases 2 Median the middle value if it is an even number it is the average of the 2 middle values 3 Mode the most frequently occurring value Range 0 The variation between the maximum and minimum values Quartiles o A way of splitting up data o It is needed because one group can be outliers 0 Example When measuring income there may be great variation between the richest and the poorest Very wealthy Middle class Very poor o The gray area is called the interquartile range Standard Deviation o SPSS will calculate this for you but Dr McNulty wants us to know what calculations the computer is doing 0 The steps to calculating the standard deviation is on the handout he gave us Example 0 If the mean of the ages of everyone in our class and everyone in an elementary school class is 20 M 20 The ages For elem school 66302830 For PSC 400 21 2220201 1 Calculate the mean a ES 1005 20 b PSC400 1005 20 2 For each case subtract the mean from the value of the case to get the mean deviation a PSC 400 i 21201 ii 22 202 iii 2020O iv 2020O v 1720 3 b ES i 6 20 14 ii 6 20 14 iii 3020 10 iv 28 208 v 30201O 3 Square each these values mean deviations a PSC 400 i 12 1 ii 224 iii 020 iv 020 v 32O b ES i 142 196 ii 142 196 iii 102 100 N 8264 V 1021OO c Why do we square it Because it gets rid of any negative numbers and it puts extra weight on the bigger numbers which allows us to set apart outliers 4 Calculate the sum of squared mean deviations a PSC 400 i 14OO9 14 ES ii 19619610064100 656 5 Divide that sum by number of cases N minus 1 a PSC 400 i 144 35 b ES i 656 51 164 c The value you get is called a variance 6 Take the square root of the variance a PSC400 i 1513 b ES i 16 4128 Standard Deviation for the ages of P80 400 187 St Dev for an elementary school class 128