PART TWO

Using the Scales documentation for the 9-year FFCWS ( http://www.fragilefamilies.princeton.edu/documentation/year9/ff_9yr_scales_2013.pdf)

Select and code two scales that you think will be useful for the research question( How parental SES affect child mental health) you have selected, and that you think might be related. Describe why you chose your two scales, and hypothesize how you expect them to relate to each other. In this relationship, which is your Y variable (the outcome of interest) and which is your X variable (your predictor)? (4 points)

X: Income (mother and father income need to be separate)
? Mother:m5j1a
? Father: f5j1a
? X variable: Education cm5edu Mother's education
? cf5edu Father's education

? Y Variable: (Child mental health)
? p5q3cv
? P5L12b

? If using a binary outcome, code two scales that you may use as X variables, and expect to be correlated with your binary outcomes.

Construct histograms and a scatterplot of the values of your two scales, and compute the correlation between the two scales. (4 points)

Answer the following questions:
1) How many observations do you have for each of your scales? (2 points)
2) Describe the distribution of each scale that you observe in your histogram. (2 points)
a. How many observations are used in your histogram? Why did you lose observations from the "full" 4,898 in the dataset? (2 points)
b. Do you think the distribution you observed is representative of the general population of urban parents? Why or why not? (2 points)
3) Does the relationship observed in the scatterplot, and the correlation you've computed, support your hypothesis? (2 points)
4) What additional variables would you want to consider or control for in a more thorough analysis? Why are these important? (2 points) (Control for parents' mental health)
5) start constructing these measures