Avg interview scores by location

Suppose your team is looking to see how interview scores may differ by location (while normalizing for differing roles) at your company.

You're given the below tables, detailing candidate roles/locations as well as associated interview scores. Using these tables, write a SQL query to return the average interview scores by role/location, and then provide the team with next steps outlining how you might test to see if the interview scores do indeed differ by location (you can assume the actual tables will have enough n-count for common statistical tests).

Table: candidateInfo

name	role	location
john_smith	SWE	New York
sarah_daniels	Analyst	San Francisco
tim_cook	Analyst	San Francisco
lisa_perelli	Data Scientist	New York
jenny_west	Data Scientist	San Francisco
karl_tran	SWE	New York
tammy_turner	Analyst	San Francisco

Table: candidateInterviews

Each row represents a unique interview/candidate as well as the candidate's score (an integer from 1-5, with 5 representing the highest possible score).

candidate_name	interview_score
john_smith	4
sarah_daniels	5
tim_cook	4
lisa_perelli	2
jenny_west	2
karl_tran	4
tammy_turner	3

You can view/query the data in an interactive SQL editor here.