Commentators, particularly conservatives, often worry about non-marital childbearing. In this assignment, you will try to learn whether non-marital child-bearing has a causal effect on the birth weight of the baby—and if so, how large the effect is. You will use a dataset on births in NYC in 2010, described further below. Your assignment should take the form of a brief essay understandable to a policy analyst who understands regression but is focused on real world policy issues.
Although you will need to use multiple regression and interpret statistics: they are not the main focus of the assignment; the idea of control variables and what you learn with different controls is the focus. Remember that the goal is to learn if there is a causal effect of non-marital child-bearing on birthweight (and the magnitude of effect), but you want to be completely honest about how well the results at each state do or do not provide such evidence. Specific topics to be explored and covered in essay (Note: The essay does not necessarily have to be in this order. It might make more sense to start with an introduction, then theory (path diagrams), then the analysis (naïve effect, and then controls), and conclusion. 1. Naïve “effects” First examine the apparent effect of non-marital childbearing on birthweight using regression.
This will be the naïve analysis: the effects cannot be considered causal at this point. Put your results in the first column of your regression results table. (See below for details on results table.) Interpret your results. Be sure to address each of these statistical points: • What is the adjusted R2 (or just R2) from the regression and what does it tell you? • What (in prose) does the regression coefficient tell you? • What are the units of the regression coefficient? • Is the sign of the coefficient (+ or -) what you expected? • Is the coefficient statistically significant?
• Is it practically significant? But remember that the goal is to learn if there is a causal effect of non-marital child-bearing on birthweight. How much have you learned about the causal effect? 2. Theory and possible omitted variables bias Develop some theories of how birth weight and non-marital childbearing could be related. First, consider theories of how non-marital childbearing could causally drive low birth weight: what would plausible mechanisms be? Second, consider alternative theories that could explain the correlation but are inconsistent with non-marital childbearing causing low birth weight, such as common causes of both birth weight and non-marital childbearing. Create path diagrams for these theories (There should be presented as separate theories. But in the real world, both could be true.) In addition, think of other variables likely correlated with your independent variable and dependent variable. Why might they be related?
Remember that you want control variables that are related to both independent and dependent variables but are not intervening variables (not part of the mechanism)—complex or unknown common causes. Predict what bias omitting those common causes would create and why (Remember to try to predict the direction of bias, if possible.) Using the dataset, try to find the common cause variables or proxies for them in the dataset. You will likely not be able to get the common causes (including complex and unknown common causes) that you want, and so think of proxies for the control variables you want. Even with proxies, you may not be able to find the controls you want in the dataset.
That is the reality of this kind of research. You may wish to find additional alternative theories, as well as accepting the poor proxies available. 3. Better causal effect estimates with control variables Perform a regression or regressions with the same dependent variable and independent variable of interest but now using the control variables you selected (You may need to create some new dummy variables or transform some variables.) You should include at least two new specifications (i.e., models, separate multiple regressions), but no more than four new specifications at the absolute most. One of these should be your best specification, with all the relevant controls and no others. The results of each specification should go in separate columns of the results table. This table is similar to Table 13.1 in the book. Interpret the results of each regression as follows: • Interpret the new main coefficient of non-marital childbearing. o Interpret it in prose, including its units, similarly to how you learned in statistics. o Is the sign of the coefficient (+ or -) what you expected? o Is the implied association with birth weight statistically significant?
o Is it practically significant? • Most important: o What does the coefficient say about the causal effect of non-marital childbearing? o Try to interpret the coefficient of non-marital child-bearing as a causal effect. On the one hand, don’t forget the goal is a causal effect estimate. On the other hand, don’t over-claim. Using your theories, is there remaining bias in the coefficient of interest as a causal effect estimate? o What happened to the coefficient of interest compared to the naïve regression and other specifications with control variables? What does this say about bias in interpreting the naïve regression coefficient as a causal estimate? • Interpret the coefficients of each control variable, in at least one specification. (prose, units, sign, practical significance, statistical significance, reasonableness)
• What is the adjusted R2 (or just R2) from at least one of the additional regressions and what does it tell you? Again: In addition to the simple (naïve) regression in part 1, you should have at least two regressions (two specifications), one of which should be your final preferred regression. However, you should have, at the absolute most, four additional regressions, for a total of five. 4. Conclusions Look over all the results. Assume that your main interest is the effect of marriage on birthweight. You may have some interest in the effects of other interventions but it’s not the main focus. And you are not interested at all, in this essay, on other outcomes. What are the policy implications of your results? Make sure that your conclusions stem from your results. Try to be clear, organized and specific. Feel free to comment on any lessons you think you have learned from these analyses. Conclusions about what is not known are also useful. One paragraph should be sufficient, but you may somewhat say more if you wish.
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
