[noelsusman]

In many ways it's just a new word for the same thing, but there's a few key differences. The main difference between traditional statistics and data science is strength in programming. Data scientists are also expected to be more well versed in statistical modeling than your average programmer or data analyst.

With that said it's not really a new thing, people have been doing data science for decades. The demand for people who can program and also do more complex statistical modeling has skyrocketed so I think that's why there's a new name for it now.

Part of the problem is that even with this definition there's a wide range of abilities present in data scientists. A long time computer programmer who has dabbled in statistics and a long time statistician who has dabbled in computer programming would both be data scientists even though they bring very different strengths to the table.