Blog Post May 4

            Each of the two articles are very closely related to Chapter 4 of Freakonomics. Each of the three sources takes a close look at the relationship between legalized abortion and the declining crime rates. Freakonomics along with the Levitt’s article, to no surprise, tell a very similar story. Each of these works was written by Levitt’s which describes the similarities in both context and structure. The third article which is a critique of Levitt’s study finds some problems with the methodology of Levitt’s paper.

           Like Freakonomics in the Levitt article, finds that legalizing abortion has lead to a declining crime rate. His explanation for this assumption is that a decreasing number of unwanted births will lead to less criminal activity in the future. The reasoning behind this explanation is that children who are born against the will of their mother will receive less adequate parenting (prenatal care, single family homes, experiencing poverty) resulting in an increased probability of criminal activity in their future. (This assumption supported in the readings) This is described in Levitt’s paper as the possibility that abortion has a disproportionate effect on the births of those who are most at risk of engaging in criminal behavior. (Levitt, Donohue, 381) Levitt’s most sound description for the relationship between decreased criminal activities in relationship to legalized abortion is the overall decrease in cohort size following the legalization. 

        Within his study Levitt uses three different regressions to look at different relationships between abortion and crime. This first regression takes a look at abortion and crime on the national scale. Levitt is able to divide post-abortion legalization children into cohorts by age. This is what allows him to see the gradual effect the law has as time goes by. To do this he was able to design an index that reflected the effect of all previous abortions on crime in a particular year. (Levitt  and Donohue, 394) He then tested early legalized states against the rest of the U.S. at a particular year.  Finally he uses a panel analysis to relate state abortion rates after Roe v. Wade to state level changes in crime over the period from 1985 through 1997. (Levitt and Donohue, 400) For each of these test he found that they were significantly relevant for his hypothesis that legalized abortion has lead to declining crime rates.

        Foote and Goetz look at the same relationship but find some potential problems in the way Levitt’s was analyzing data. Their thesis was set to evaluate if the declining crime rate was due to other factors besides abortion. Their first problem was that Levitt’s was missing a key set of regressors because of a computer coding error. Also the Levitt’s regressions do not model arrests in per capita terms. After correcting for the errors they found that the same regressions were not significant.

        In my opinion, the Levitt’s article still holds weight. The Foote and Goetz article  helps make this conclusion for me. Depending on how you manipulate the data you will get a different answer. In my opinion the Levitt ideas make absolute sense. The assumption that poverty stricken, single mothers, who may not be great parents may influence children to develop criminal habitats makes absolute logical sense to me.

Summary
My paper topic is looking directly at quantitative look at the relationship between spending on education and student performance. To look at this study I believe it is most valuable to evaluate government spending on education and SAT scores. I believe these two variables are public knowledge making them easily accessible and quality measurements. I also believe evaluating the SAT scores will be a good measurement of U.S. student placement in higher education. Within my study I have developed a few variables that may affect the relationship between education spending and high school education performance. The first variable is family income. As we have discussed in class family wealth is a factor in the ability of a student to perform. My second variable is education enrollment rates, which are good indicators of schooling quantity. Unfortunately I tried to find a variable that would affect education quality but found that this is encompassed in the SAT score. My main goal in this study is not only to evaluate the performance following education spending but also to look at the ability of children to get into colleges. I believe this study is very valuable in evaluating not only U.S. government economic policy but also fiscal policies within the education spectrum.

Writing Assignment 8

Summary: The main public policy understanding in the poor's fertility decision is based on their access to contraception. Lately studies have shown that other social norms such as family dynamics, and economic considerations also play a key role. A collossal determinant is the treatment of women. It seems that social policy aimed at eliminating discrimination against women can be a huge step in influencing population.
The stat that I decided to create a regression analysis on is the birth rate for teens in America. " In the United States, which has one of the highest teen pregnancy rates in the developed world, there are 4.5 births per 100 adolescent women."
I would hypothesis that teen pregnancy is negatively correlated with family income and amount of education. I believe these two factors are key determinents in both sexual activity and sexual prevention. To test this hypothesis I would regress teen pregnancy rates on the amount of education per pregnant teen and the family income per pregnant teen. I would also regress these dependent variable by themselves to see the relationship.
The dummy variable I would add is the presence of sexual education in their school system. Pregnant teens with sex ed would get a 1 while teens without sex ed would get a 0. I would expect my dummy variable to tell me that more teens without sexual education courses get pregnant. I expect my dummy variable for sexual education courses to be negative.
Finally my regression equation would look something like this Teenbrthrates=X+X1education+X2familyincome+Ei

MONEY BALL EXTRA CREDIT


Billy Beane uses on base percentage as his statistical analysis to build his team. While this statistic is very useful for a team like the Oakland A's, who do not have a lot of money to spend, it is not very practical. In my opinion while the movie does make sense I think this is an isolated incident. The Oakland A's did not do very well the following year which leads me to believe it was a one time thing. My reasoning behind this is that baseball is not completely random. This theory violates the classical linear regression because baseball is not random. There is skill behind all of the statistics assumed in the movie.

My main argument for this movie is that for one year it worked. No other team, even after people began using Beane's theory of OBS has ever been able to perform like the A's that year. In our regressions we assume everything is random. In baseball skill makes these variables not random. While OBP is a good indication for batters we also have to look at pitchers. A team will never be able to acquire an extremely good pitcher without giving the money necessary for their skill. Also when looking at OBP you are also evaluating walks. This in itself is very random. No one looks for walks but instead might have a "good eye". While my argument that skill effects the randomness of baseball I do believe Beane developed a very useful measurement. The only problem is that every team began to use the same technique.

My article was "Santorum's shifting views on education" from fox news. While some of this article focused on the presidential primary, the body of the article concentrated on government spending for public education. Santorum shows that he thinks less spending should be used for public schooling inferring that the education multiplier is not adequate. He instead believes we should be looking for support from parents in enhancing public schools in America. While Santorum's ideas may seem to be politically based it does ask a very good question. How much can increased government expenditures increase SAT performance if parents and students are not motivated?

This question present a kink in my analysis. It would be very useful for me to look into what motivated the parents and students. Is it purely a social phenomenon or can specific government finances shift this motivation. For example spending on new facilities can increase the motivation for the students to do better. From out readings in Chapter 4 it is safe to assume spending cannot directly influence motivation.

To help answer this question maybe we should look into non-financial based help the government can support. For example advocating a public service announcement instilling the importance of higher education. Finally there seems to be no one answer to the dilemma presented by Santorum. At the end of the day many districts are stuck in a poverty trap. Poor facilities combined with low motivation, due to unseen benefits, keeps many school districts under performing.

http://www.foxnews.com/us/2012/03/01/spin-meter-santorums-shifting-views-on-education/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+foxnews%2Fnational+%28Internal+-+US+Latest+-+Text%29

Chapter 4 of the book focuses on problems within schooling systems around the world. The chapter focuses on the problem within a supply and demand framework. The main question or problem asserted in the chapter is why schools continue to fail. The chapter has two very different answers to this question. On the supply side some argue the government is responsible for providing children with better schooling. This will require increased spending for local governments. Many of the problems within the schools are poor faculty, poor resources and poor facilities. The supply argument believes that the government should be willing to provide these necessities. On the demand side of the argument the chapter states that parents do no lobby for quality education because they do not see a direct benefit from schooling. “When the benefits of education become high enough, enrollment will go up, without the state having to push it.” Pg 76 Parents expect both too much and too little from schooling. For example passing college will get you a good job while passing high school can get you nowhere now-a- days. The parents do not see the benefit of their investment in their child’s education. Either their child reaps the benefits sometime in the future or they default on the investment. From the chapter it is obvious that neither problem is the sole cause. Each must be evaluated in order to solve the problem of poor schooling.

I found an article on the poverty ridden educational infrastructure in Mumbai India. My article was in the Wall Street Journal and is called “India Journal: The Basic Shortages that Plague our Schools”. http://blogs.wsj.com/indiarealtime/2012  This article deals directly with the supply side of the argument presented in the book. The Indian government is unable or unwilling to fun adequate educational infrastructure. Many of the overcrowded cities have very poor facilities with, in one case, no books to be used.  While the past decade has seen an increase in efforts to increase infrastructure many critics argue that more of an emphasize should be put on learning outcomes.  Some good statistics that the author uses is a survey showing 81,000 schools operate without chalkboards and 42,000 government schools operate without a building. These staggering numbers show the problems within the schools. Another good point made by the author is the loss in teacher and student motivation due to the poor schooling infrastructures. This plays into the chapters demand side argument.

  

Using www.bls.gov/nls I will be looking at the trends in housing sales in a specified area. I will be looking at the four areas of the U.S. (midwest, south, west, northeast) I will be tracking the housing sales over the time period during the financial crisis. My thesis is; Did areas located near the epicenter of the crisis (New York) have more  mortgage defaults than other areas.

The problem with accumulating data is I will have to distinguish what is mortgage defaults and what is sales due to relocation. My main objective will be to look at data for ears before and after the crisis.