Plotting relationship between 2 data points where one data point is a boolean

Question

I am working with the titanic survivors data set.

I have the data as a DataFrame and I can create 1D visualizations such as histograms, and also see the correlations by calling data.corr().

I would like to create a scatter plot to represent the correlation between 'age' and 'survived'. I can't figure out how to plot this data because 'survived' is effectively an integer of 0 or 1 (died or lived, respectively)

If I do something like:

titanic_data.plot(x='Age', y='Survived', style='o')

I get a plot that looks like this:

What I would like is a plot that somehow takes the average survival rate by age and created something more like this:

Answer 1

You can precalculate the survival rate (probability) and plot a bar plot:

import seaborn as sns

x = sns.load_dataset('titanic')

bins = np.linspace(0, 100, 11)
labels = bins[1:]

# let's group all ages by bins (10, 20, 30, ..., 100)    
rpt = (x.groupby(pd.cut(x.age, bins, labels=labels))
        .survived.mean()*100
      ).dropna().to_frame('survival_rate')


rpt.plot.bar(rot=0, width=0.8, alpha=0.5, figsize=(12, 10))

calculated data:

In [84]: bins
Out[84]: array([   0.,   10.,   20.,   30.,   40.,   50.,   60.,   70.,   80.,   90.,  100.])

In [85]: labels
Out[85]: array([  10.,   20.,   30.,   40.,   50.,   60.,   70.,   80.,   90.,  100.])

In [86]: rpt
Out[86]:
      survival_rate
age
10.0      59.375000
20.0      38.260870
30.0      36.521739
40.0      44.516129
50.0      38.372093
60.0      40.476190
70.0      23.529412
80.0      20.000000