connected code, question and answer in challenge

Rephrased introduction, corrected code snippet and extended answer for Looping Over DataFrame challenge, following #347 (comment)

Author: NPalopoli

_extras/extra_challenges.md

@@ -8,31 +8,39 @@ permalink: /extra_challenges/
 
 A collection of challenges that have been either removed from or not (yet) added to the main lesson. 
 
-> ## Looping Over Dataframe
+> ## Looping Over DataFrame
 >
 > (Please refer to lesson `06-loops-and-functions.md`)
 >
 > The file `surveys.csv` in the `data` folder contains 25 years of data from surveys,
-> starting from 1977. We can load the data and print all the years surveyed using a `for` loop:
+> starting from 1977. We can extract data corresponding to each year in this DataFrame
+> to individual CSV files, by using a `for` loop:
 > 
 > ~~~
 > import pandas as pd
 > 
 > # Load the data into a DataFrame
 > surveys_df = pd.read_csv('data/surveys.csv')
 > 
-> # Loop through a sequence of years and print the year
+> # Loop through a sequence of years and export selected data
 > start_year = 1977
 > end_year = 2002
 > for year in range(start_year, end_year+1):
->     print(year)
+>
+>     # Select data for the year
+>     surveys_year = surveys_df[surveys_df.year == year]  
+>
+>     # Write the new DataFrame to a CSV file
+>     filename = 'data/surveys' + str(year) + '.csv' 
+>     surveys_year.to_csv(filename)
 > ~~~
 > {: .language-python}
 >
 > What happens if there is no data for a year in a sequence? For example,
-> imagine we used `1976` as the year in `surveys_df[surveys_df.year == year]`
+> imagine we used `1976` as the `start_year`
 >
 > > ## Solution
-> > An empty file with only the headers  
+> > We get the expected files for all years between 1977 and 2002,
+> > plus an empty `data/surveys1976.csv` file with only the headers.  
 > {: .solution}  
 {: .challenge}