Other Codes of Conduct

Loading...

En provenance du cours de University of Washington

Communicating Data Science Results

125 notes

University of Washington

Communicating Data Science Results

125 notes

Cours 3 sur 4 dans Specialization Data Science at Scale

Important note: The second assignment in this course covers the topic of Graph Analysis in the Cloud, in which you will use Elastic MapReduce and the Pig language to perform graph analysis over a moderately large dataset, about 600GB. In order to complete this assignment, you will need to make use of Amazon Web Services (AWS). Amazon has generously offered to provide up to $50 in free AWS credit to each learner in this course to allow you to complete the assignment. Further details regarding the process of receiving this credit are available in the welcome message for the course, as well as in the assignment itself. Please note that Amazon, University of Washington, and Coursera cannot reimburse you for any charges if you exhaust your credit. While we believe that this assignment contributes an excellent learning experience in this course, we understand that some learners may be unable or unwilling to use AWS. We are unable to issue Course Certificates for learners who do not complete the assignment that requires use of AWS. As such, you should not pay for a Course Certificate in Communicating Data Results if you are unable or unwilling to use AWS, as you will not be able to successfully complete the course without doing so. Making predictions is not enough! Effective data scientists know how to explain and interpret their results, and communicate findings accurately to stakeholders to inform business decisions. Visualization is the field of research in computer science that studies effective communication of quantitative results by linking perception, cognition, and algorithms to exploit the enormous bandwidth of the human visual cortex. In this course you will learn to recognize, design, and use effective visualizations. Just because you can make a prediction and convince others to act on it doesn’t mean you should. In this course you will explore the ethical considerations around big data and how these considerations are beginning to influence policy and practice. You will learn the foundational limitations of using technology to protect privacy and the codes of conduct emerging to guide the behavior of data scientists. You will also learn the importance of reproducibility in data science and how the commercial cloud can help support reproducible research even for experiments involving massive datasets, complex computational infrastructures, or both. Learning Goals: After completing this course, you will be able to: 1. Design and critique visualizations 2. Explain the state-of-the-art in privacy, ethics, governance around big data and data science 3. Use cloud computing to analyze large datasets in a reproducible way.

À partir de la leçon

Privacy and Ethics

Big Data has become closely linked to issues of privacy and ethics: As the limits on what we *can* do with data continue to evaporate, the question of what we *should* do with data becomes paramount. Motivated in the context of case studies, you will learn the core principles of codes of conduct for data science and statistical analysis. You will learn the limits of current theory on protecting privacy while still permitting useful statistical analysis.

Motivation: Barrow Alcohol Study5:57

Barrow Study Problems4:59

Reifying Ethics: Codes of Conduct6:50

ASA Code of Conduct: Responsibilities to Stakeholders4:43

Other Codes of Conduct6:51

Examples of Codified Rules: HIPAA3:30

Rencontrer les enseignants

Bill Howe
Director of Research
Scalable Data Analytics

[MUSIC]

In these last few we'll just put up here, they are, I'd argue, less relevant for

the barrel example, which I haven't highlighted anything but

responsibilities to research team colleagues.

Inform colleagues from other disciplines about relevant aspects of statistical

ethics, making sure that people aren't sort of using your methods,

and you haven't conveyed some of these ethical considerations as well.

You know them, but you don't convey them,

that lack of communication itself is a lack of ethics.

Generally promote the proper use of statistics, you

have an obligation to convey the proper use of things, when given the opportunity.

Reciprocally respect the epical obligations

of people from outside your field.

If they have a professional code of conduct that they need to follow as

an industrial engineer, or as as in other engineering fields, you need to understand

that and the professional reporting,

and the last one here is compromising statistical validity for expediency.

You don't wanna trade off quality for efficiency.

Understanding that, of course, approximations are appropriate.

So overall what I like about these is that, many of them are not the bland,

obvious, generic ethical considerations that you at the highest level.

Many of them really do have specific teeth related to statistical analysis.

Responsibilities to other statisticians, or statistics practitioners.

This is when you're reviewing other people's papers, you're reviewing other

people's work, or you're doing performance reviews for other people in the company

that perform these skills to make sure you're being sort of fair and appropriate.

Assess the methods, not the individuals.

Be willing to sort of offer them training.

As I mentioned,

I think this code of conduct is pretty good because it's pretty complete.

It's mature, it's been thought about, and I think this is an interesting

stakeholder community that may not be the first thing you think of

when you think about a code of ethics, but it is important.

Nonetheless, especially if this term data science will persist and

become a strong community, making sure you're figuring out opportunities to help

each other out, is gonna be increasingly important, okay.

And responsibilities of employers, including organizations.

The first one I've highlighted here is recognize that the results of valid

statistical studies cannot be guaranteed to conform to the expectations of those

commissioning this study, and this is again, pretty obvious, but

it's also pretty crucial.

This is the source of the trouble over the years with

perceived or real conflicts of interest where you imagine the look of

company should be perfectly justified in funding a study.

But what you don't see is the publication bias in effect.

If they don't publicize results that didn't come out the way they want,

that's a nearly undetectable, but very strong way of steering

research to produce the outcome that they want, right?

They could fund 20 studies and only publish the one that came out in favor

of the results that helped their business.

Okay, so the fact that there's a section in here specifically for

employers and statisticians where or not they're reading it.

I approve of the fact that they include that in there on equal footing with some

of the other groups of stakeholders, and some of the others,

you can read them here, and I won't just read them out.

The second code of conduct

that I just quickly want to go through is the Data Science Association, and

this one is less mature than the ASA code of conduct, as I've mentioned.

So I'm just going to show you the top level categories and

let you click the link to read it yourself, and

then highlight a couple of differences of what it chooses to emphasize.

So here are the categories, the first one is really just terms and

definitions and you can take a look at them.

The second is the notion of competence,

which we saw emphasized pretty strongly in the ASA's code of conduct.

Number 3 is interesting because it calls out the tension between

legal requirements and what the client asks and also the ethical considerations.

And so we've seen this come up more, and

more often, where you may be in a position where

you need to make a decision about whether you're gonna collect the data or not.

Not because you think it might be released or misused by you, or

you're not planning on sorta publishing the result.

It's only for some internal usage, but it could be subpoenaed, and so

you need to consider that if you put your hands on that data at all,

you may be required in the future to give it up.

In which case you're either faced with violating the law, violating your client's

interests, or violating some ethical consideration, okay?

Communication with clients they call out as explicit in a category,

which I thought was interesting.

This is, you know, don't take the problem specification,

run away into a hole, do the work, and then come back out and dump the results.

You should be in constant communication as you work through the decisions.

Again, this is.

General enough where it seems obvious, but I think it's interesting to analyze,

to consider what they've chosen to emphasize and

what they have not chosen to emphasize.

Confidential information, we've seen, we've had that covered,

conflicts of interest.

Number 7 is more on conflicts of interest and confidentiality.

Number 8 is more of a generic thing on scientific integrity.

I think overall the point I want to make is that trying to break it down by topic

seems more difficult and results in kind of a muddier collection of rules than

breaking it down by the parties and how they might be harmed,

which is what the SA did.

Fine.

And then the last one I wanna mention is the Certified Analytics Professional.

I won't claim that either one of these last two again are necessarily indicative

of The organizations that present them aren't necessarily representing

the entirety of the science community, but they did take the time to sort of publish

these codes of conduct, and therefore, it's worthwhile to be aware of them.

Really, the only thing I wanna point out here is that they all, along with ASA,

emphasized the parties involved and how they might be harmed.

I think the ASA code of conduct did a better job of

being exhaustive about that and being clear about that but it's one important

take away that if you're going to think about codes of conduct.

Think about who has interest in your work and

how they might be harmed by your work one way or another.

What their interests are and how you can try to meet them.

[MUSIC]

Explorer notre catalogue

Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.

Coursera propose un accès universel à la meilleure formation au monde, en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.

© 2019 Coursera Inc. Tous droits réservés.

Télécharger dans l'App Store

Disponible sur Google Play