From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Expanding our definition of level of measurement

Expanding our definition of level of measurement

From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Start my 1-month free trial

Expanding our definition of level of measurement

“

- [Instructor] When Stevens first proposed level of measurement back in the '40s, he was focused on statistics, specifically survey style data. Since we're doing machine learning, we really have to expand our notion of level of measurement and include some additional categories. First, we have to focus more carefully on potential ID fields, not in the same way that the IT team does when they're building the data warehouse, but thinking creatively about fields that may help us link the current dataset with other datasets during data integration, things like street address, for instance, grabbing neighborhood level detail using zip code, then some fairly obvious ones, like telephone number and email can connect us with other data, but then also credit card number. Not because we're going to check a customer's credit score, I suppose in the case of banking, that might be relevant, but generally using it as a proxy. For…

Contents