Before any filtering is actually done, you will obviously need to know the exact string names that will be used. The Series value_counts method is an excellent way to get both the exact string name and number of occurrences of that value.

The isin Series method is equivalent to the SQL IN operator and accepts a list of all possible values that you would like to keep. It is possible to use a series of OR conditions to replicate this expression but it would not be as efficient or idiomatic.

The criteria for salary, criteria_sal, is formed by combining two simple inequality expressions. All the criteria are finally combined together with the pandas and operator, &, to yield a single boolean Series as the filter.