« Top 3 R resources for beginners | Main | Social Network Analysis at New Frontiers in Computing 2013 »

May 15, 2013

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a010534b1db25970b01901c36c13b970b

Listed below are links to weblogs that reference Statistics vs Data Science vs BI:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Oh heck...it's a good thing the Red Wings and Blackhawks are in the first intermission. Speaking as an econometrician who has constructed more than 15,000 forecasting and predictive models for economics and applied marketing... "horse puckey." Get it? "Predictive" = forward looking "Forecasting" = forward looking. You can SAY that data science is forward looking, but a lot of it is merely pattern-based analysis applied in the hope that what has happened in the past will happen again. The only reason that statistics has a branding problem is that non-quant MBA Marketing majors are still scratching their heads over probability and statistics. A statistician with kilobytes of data is in academia, not the real world. No matter your preferred moniker, real world data is just as large and just as dirty. What is Linux doing in a tools list, btw? Should I point out that one of the current uses for mainframes right now is to run large numbers of simultaneous linux instances for web sites? Basically, if you think Data Scientist is sexy, then more power to ya - I hereby dub thee data scientist, but you should still be proud of statistician.

I turned off the wings game in the third, no need to stick around and get more depressed.

As a developer who does a lot of BI work, I find this comparison of Business Intelligence vs Data Science a bit one sided. I have used Cognos and Tableau for analysis and reporting. Also, a good BI developer should present the data in an easily understandable format which doesn't mean it HAS to be tables, unless of course that is what the client requested. While a lot of BI does tend to be reflective, looking back on events is a great way to estimate what the future will look like.

I do however agree, that "Data Scientist" is a sexier name than "Business Intelligence" but see one as the natural evolution of the other.

My wife doesn't like the term Data Science because while they (we?) may deal with data, they don't do anything close to science (under any reasonable definition of science).

Don't try to search for any objective justification for something that's pure marketing. You are all quant people, don't get trapped by non quant people rationality.

From decades of sad experience, Data Science is as regressive (!) wrt both Data and Science as NoSql is to the Relational Model, and even SQL databases. In both cases, a cabal of coders want to appropriate the aura of smarts without having to actually be smart. Kind of like Web 2.0: a meme created by purveyors, not thinkers or even practitioners.

A scientist presents facts to support a hypothesis. I see neither in this blog post (specifically in the comparison of data scientist and BI).

Business Intelligence is called that, in part, because it helps the entire business. A data scientist is focused on his/her dataset, whethet in GB or KB.

Here is how I see it. I was trained as an econometrician. While that gave me a great stats foundation- I also had the trouble seeing myself as a statistician.

Statisticians were not then nor typically now trained to work with business. They don't speak the same language (and vice versa). Most statistics grad programs that I have read spend more time on real analysis, rather than how to deal with a complex, data oriented business problems and communication.

This is the true difference I see in DS vs Statistician. A DS probably cannot do real analysis, but can put a business problem into context and work to solve it with data. A Statistician is the opposite.

Of course the above is a generalization- I certainly know Statisticians who have conquered the business world. And some data scientists who do not have any understanding of the underlying model.

So both areas need to converge to be both rigorous and useful.

Agree with the 'more power to ya' comment, also agree with the comments re marketing - much of this is fashionable labelling.

What I don't agree with is the negative view of BI - yes, the Balanced Scorecard emerged c1997 (and that's only part of true BI) but it's still the most powerful tool available to any forward-thinking operation. I believe the main reason it (and related BI) is seen as old-hat by many conservative operations is because they are always looking for a quick fix, and they think the latest fad is their silver bullet, even though we all know it is the same thing re-badged.

A comprehensive use of statistical analysis, data mining (or analysis or science - whatever you wish), proven BI tools (eg. BSC), business analysis, process mapping and analysis, etc, etc is the only way to create genuine performance improvements and assist a company who can see beyond the start-up costs.

Let's take the 'vs' out of the conversation, converge the lot and thank Professor Deming for lighting our way.

Thanks everyone for the great feedback and conversation. I'm definitely proud to be a statistician -- my point was more that the world at large seems to respond more favourably to the "data science" title as representing what we'd call "real" applied statistics. One things I've learned from conversations since this post is that "data science" seems to be having a branding problem of its own. All of the data scientists I've worked with have excellent statistics skills, but (ironically?) that could well be due to a nonrepresentative sample. Anyway, thanks again for the responses.

How does the term "data artist" compare with the other terms?

Well there is always a fascination towards newly coined terms and I guess the same is the case with data scientist. However, I feel a data scientist is a complete combination of a statistician and BI guy. He or she is the master chef of data!

what a crap article to prove a preconceived notion about the glamour of the name data science

how do to get to become a data scientist. which career path takes to to that step.

The comments to this entry are closed.


R for the Enterprise

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid

Search Revolutions Blog