We can also use `np.mean`

to calculate the percent of array elements that have a certain property.

As we know, a logical operator will evaluate each item in an array to see if it matches the specified condition. If the item matches the given condition, the item will evaluate as `True`

and equal 1. If it does not match, it will be `False`

and equal 0.

When `np.mean`

calculates a logical statement, the resulting mean value will be equivalent to the total number of `True`

items divided by the total array length.

In our produce survey example, we can use this calculation to find out the percentage of people who bought more than 8 pounds of produce each week:

>>> np.mean(survey_array > 8) 0.2

The logical statement `survey_array > 8`

evaluates which survey answers were greater than 8, and assigns them a value of 1. `np.mean`

adds all of the 1s up and divides them by the length of `survey_array`

. The resulting output tells us that 20% of responders purchased more than 8 pounds of produce.

### Instructions

**1.**

You’re running an alumni reunion at your local college. You have a list of the names of each person in attendance and the year that they graduated.

We’ve saved this list as a NumPy array to the variable `class_year`

.
Calculate the percent of attending alumni who graduated on and after 2005 and save your result to the variable `millennials`

.

**2.**

Print the value of `millennials`

to the terminal.