Supporting Statistical Analysis for Research

4.1 Preparatory exercises

The skills in these exercise are used in the exercises at the end of the discourses of this chapter. Take a moment and complete these to confirm that you are prepared for this chapter. If these exercises are difficult, review the prior chapters.

1. Import the PSID.csv data set.

   The following is used at the RStudio prompt to enter Python mode.

   library(reticulate)
   repl_python()

   The remainer is Python code.

   from pathlib import Path
   import pandas as pd
   import plotnine as p9

   psid_path = Path('..') / 'datasets' / 'PSID.csv'
   psid = pd.read_csv(psid_path)
   
   print(psid.dtypes)

   Unnamed: 0      int64
   intnum          int64
   persnum         int64
   age             int64
   educatn       float64
   earnings        int64
   hours           int64
   kids            int64
   married        object
   dtype: object

2. Plot earnings verse hours.

   print(
       p9.ggplot(psid, p9.aes(x='hours', y='earnings')) + 
       p9.geom_point() +
       p9.theme_bw())

   <ggplot: (143591241869)>

   

3. Make a boxplot of earnings with separate boxplots for each married status.

   print(
       p9.ggplot(psid, p9.aes(x='married', y='earnings')) + 
       p9.geom_boxplot() +
       p9.theme_bw())

   <ggplot: (143591233586)>

   

4. Make a horizontal boxplot of earnings with separate boxplots for each married status.

   This should be the same plot as in the prior example only the earnings are displayed on the horizontal axis.

   This is useful when there are many boxplots or the category names are long.

   print(
       p9.ggplot(psid, p9.aes(x='married', y='earnings')) + 
       p9.geom_boxplot() +
       p9.coord_flip() +
       p9.theme_bw())

   <ggplot: (-9223371893263501845)>

   

5. Do all of the categories of married make sense?

   The NA/DF and no histories would make more sense being combined into a single set of NA observations.

6. Plot earnings verse kids.

   print(
       p9.ggplot(psid, p9.aes(x='kids', y='earnings')) + 
       p9.geom_point() +
       p9.theme_bw())

   <ggplot: (143591209501)>

   

7. What can be learned from this plot?

   There appears to a number of observations that have a kids value of over
   90. These are likely a code for NA.

   This would be more informative if earnings were displayed as a boxplot for each number of kids.