You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
20 lines
1.1 KiB
Markdown
20 lines
1.1 KiB
Markdown
6 years ago
|
# Chapter 1
|
||
|
|
|
||
|
|
## Take aways
|
||
|
|
|
||
|
|
This chapter uses a fictitious first day as a Data Scientist to illustrate some social science research. In particular it illustrates the need to ask good questions of the data you have.
|
||
|
|
|
||
|
|
For example. We had data on inter office friendships and professional interests from there we could determine:
|
||
|
|
|
||
|
|
- Average number of friends
|
||
|
|
- Total count of inter office friendships
|
||
|
|
- A suggested list of people a person could befriend based on mutual friend ships (Friends of a friend)
|
||
|
|
- A suggested list of people a person could befriend based on mutual interest.
|
||
|
|
|
||
|
|
|
||
|
|
These examples continued with salary data, tenure, and paid accounts
|
||
|
|
|
||
|
|
- At first we could explore with a visualization the relationship of salary and tenure. Resulting in us determining that individuals with longer tenures tend to earn more.
|
||
|
|
- Since tenure was widely distributed it was necessary to bucket individuals by tenure to get an average salary for a tenure range.
|
||
|
|
- This lead to the insight that data scientists with more than 5 years of experience earn 65% more than a data scientist with two years or less experience.
|