Open for Comment: Proposed "Data Science Code of Professional Conduct"

We are seeking comments and suggestions on a proposed "Data Science Code of Professional Conduct".

Data science is an independent profession. Data scientists have a higher calling than just technical skills. We have a duty to use data science to make life, business and government better.

Six (6) key reasons why data science needs a voluntary code of conduct:

1. to establish data science as a true science and profession distinguished from data analysts, statisticians, actuaries and mathematicians - helps data scientists obtain new clients because clients reassured about contract to follow code, use scientific methods and protect clients confidential information;

2. to protect data scientists from unscrupulous clients who may desire to misuse data science;

3. to protect clients from data scientist charlatans and fraudulent data science;

4. to protect confidential information;

5. preempt imposed regulation - to have a system of self regulation good enough to persuade law-makers to avoid regulating data science;

6. to militate against data science leading to dystopia.

Review and Download Code @ http://bit.ly/YbsjXR

Please add comments in the reply section or email comments, suggestions and corrections to: [email protected]

Tags: Code, Conduct, Data, Ethics, Professional, Rules, Science, of

0 members like this

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by Michael Walker on March 8, 2013 at 9:22am: I suggest data science is a profession and requires a robust and durable code of conduct considering the great damage data science malpractice can do to society. The best and brightest industry and government leaders rely on data science to make decisions and are unable to differentiate between good and bad data science.

For example, consider that data strongly suggests the majority of financial quants are not adding any value. If you had put your money into hedge funds in 2005 and distributed it across the whole universe of hedge funds, you would have roughly the same amount of money today. By contrast, if you had simply bought the existing universe of publicly traded stocks and bonds and reinvested interest and dividends, your portfolio would have grown by half.

Perhaps if the financial quants had a code of conduct the majority would not have produced valueless models that created a false view of reality. The illusion of understanding produced bad decisions by policy makers and financial heads of state. This contributed to the financial melt-down.

Likewise, data science malpractice can produce an illusion of understanding that can blow up firms and create bad public policy.

Hedge Funds Underperform Stock and Bond Indices by Huge Margin

See: http://bit.ly/J2r2Oa

Comment by Vincent Granville on March 7, 2013 at 9:35am: Have you checked the code of conduct of the American Statistical Association? The Digital Analytics Association also has one. Both organizations offer certifications. DSC offers an apprenticeship specifically to become a real data scientist. You might also want to read our article about fake data science.

Ultimately, it is the responsibility of the data science users to do his due diligence before paying for data science services. A university degree is not guarantee of competency. The same applies to anything - lawyer, doctor, SEO expert, when buying a car, a vacation trip or a house.