On data literacy by Marek Danis
•Download as PPTX, PDF•
0 likes•1,465 views
- Marek Danis is an experienced data scientist and trainer who has worked for Schlumberger Oilfield Services and the Digital Transformation Team. He has a MSc from Texas A&M University Mays Business School and specializes in QHSE (Quality, Health, Safety and Environment) analytics. - Marek runs his own consulting company in Austria focusing on QHSE analytics and using data science to decrease risk and increase business outcomes. He has developed a strong understanding of data analytics applications in corporate environments.
Report
Share
On data literacy by Marek Danis
- 1. On data literacy Marek Danis, MSc
- 2. Marek Danis, MSc Marek is an experienced data scientist and trainer with strong business outcome focus. Previously, Marek worked for more than 10 years for Schlumberger Oilfield Services around the world in several technical, engineering and training positions (including 3 years as a trainer of technical as well as corporate culture related content in Russia and UAE), and later with the Digital Transformation Team. Marek began his a Data Analytics education at the Texas A&M University Mays Business School in Houston. Since that time, he has developed a strong acumen for QHSE (Quality, Health, Safety and Environment) related analytics and data science. Marek also runs his own consulting company in Austria, specialising in QHSE analytics and how to decrease risk and increase business outcomes.
- 3. The obvious: we are in a period of rapid technological change ● We see new technologies every month ● “Would you like fries an app with for that?” But why is technological change so rapid and what are the consequences? “The only way to win is to learn faster than anyone else.” “The only way to win is to learn faster than anyone else.”
- 4. With technological change, comes social change Social change follows technological change (though at a more non-uniform pace) as we adapt, e.g: ● We adapted to literacy and numeracy as reading became the norm ● We became computer literate as computers became the norm How will we adapt to the widespread use of data & analytics? “The only way to win is to learn faster than anyone else.” “The only way to win is to learn faster than anyone else.”
- 6. Two key questions addressed in this talk: Why is technological change so rapid and what are the consequences? How will we adapt to the widespread use of data & analytics? NEXT
- 7. We’re getting better at running projects ‘Eric Ries - The Lean Startup, London Edition’ (cropped, B&W re-colour) by Betsy Webber available at https://www.flickr.com/photos/betsyweber/6730126217 under a CC by 2.0. “The only way to win is to learn faster than anyone else.” Why is technological change so rapid?
- 8. We’re getting better at running projects We’ve seen improvements in: ● Project methodology ● How we manage large groups of people ● What technology can enable: ○ New applications ○ Leverage pre-existing infrastructure ○ Reduction in project costs ○ Reduction in project risks ‘Eric Ries - The Lean Startup, London Edition’ (cropped, B&W re-colour) by Betsy Webber available at https://www.flickr.com/photos/betsyweber/6730126217 under a CC by 2.0. “The only way to win is to learn faster than anyone else.” Why is technological change so rapid?
- 9. Projects accessible to large companies / nation states are more accessible to you and me Traditional projects / start up Today you can also... Set up infrastructure: ● Hire technical expertise ● Buy fixed hardware ● Build / buy software ● Maintain software / hardware Rent infrastructure / use the cloud: ● Variable pricing ● Scalable Hire permanent staff, hire contractors through agencies etc. Access short-term experienced labour onsite or remote (e.g. AlphaZetta analytics consultancy) Require manual labour to interact with customers (stores / branches, calls etc.) Automate some interactions (i.e. via mobile apps) Why is technological change so rapid?
- 10. But even if we can build anything... “The big question of our time is not Can it be built? but Should it be built?” ― Eric Ries, The Lean Startup The key challenge is in deciding what to do in order to achieve key outcomes under uncertain conditions, competition and change. ‘Sipping Bird’ (cropped, B&W re-colour) by RobinLeicester available at https://commons.wikimedia.org/wiki/File:Sipping_Bird.jpg under a CC BY-SA 3.0. Consequence of technological change
- 11. The nature of work is changing: less about doing, more about deciding ● Doing things is becoming easier and faster ● Data literacy is playing a central role in business ● Many tasks are being automated across many sectors, meaning: With automation people will have to do what machines can’t do: DECIDE! Today ‘Tesla Autobots’ (cropped & BW re-colour) by Steve Jurvetson available at https://www.flickr.com/photos/44124348109@N01/6219463656 under a CC by 2.0. Consequence of technological change
- 12. Good decisions are more important than ever before ‘wocintech (microsoft) - 229’ (cropped & BW re-colour) by WOCinTech Chat available at https://www.flickr.com/photos/wocintechchat/25392653883 under a CC by 2.0. In particular good decisions are important when: ● You're in a in a highly uncertain & complex environment ● You have competitors ● You have to make a large volume of decisions ● Not making a decision is not an option, and making a bad decision has has an immediate and very bad outcome. Today
- 13. Using data can enable better decision making A key purpose of data and data analytics is to facilitate better decision-making in particular it enables us to: ● Process and use all information available to us ● Apply reason to all possible options and outcomes ● Decide in a timely manner Today
- 14. But… we’re not using data for good decision making Currently we are in an anomalous time: data-related technology and skills is changing but our professional environment is not necessarily changing with it: ● Social change is slow (e.g. similar to the computer revolution where some executives couldn’t use their web browser/email 10+ years ago) ● We are in comfortable economic times with little competition There are exceptions: where outcomes are ambiguous and they matter cultures tend to be data literate (e.g. hedge funds, sports betting, political parties, military intelligence etc.) ● Management class (and corporate cultures) are not necessarily data literate even if they hire data scientists. If data scientists are expensive tropical fish, then water is a data literate environment ‘Tesla Autobots’ (cropped & BW re-colour) by Steve Jurvetson available at https://www.flickr.com/photos/44124348109@N01/6219463656 under a CC by 2.0. Today
- 15. and… we erroneously think “data” = good decisions Socially we haven’t all caught up with what it means to use our data in professional environments: ● Buzzwords still abound: “data” + “analytics” with no mention of their role in decision making ● Many data scientists are miserable in lucrative careers ● Many data scientists are hired for roles other than that of a data scientist ● Many data scientists work in data illiterate environments Today ‘Confused’ (cropped & BW re-colour) by CollegeDegrees360 available at https://www.flickr.com/photos/83633410@N07/7658298768 under a CC by 2.0.
- 16. Data literacy is required for good decision support Everyone will be a data professional, doing what computers can’t do: searching for insights to support good decision making. This requires data literacy. A data literate professional is: ● comfortable with uncertainty ● ready and willing to conduct experiments ● Comfortable with basic mathematics and statistics ● Versed in some common data visualisation techniques Future ‘Tesla Autobots’ (cropped & BW re-colour) by Steve Jurvetson available at https://www.flickr.com/photos/44124348109@N01/6219463656 under a CC by 2.0. ‘Three people sitting beanbag chairs working’ (cropped & BW re-coloured) by katemangostar https://www.freepik.com/free-photo/three-people-sitting-beanbag-chairs-working_993093.htm How we adapt
- 17. Examples of data literacy Future Traditional white collar literacies New white collar literacies Literacy and simple numeracy Financial literacy Computer literacy Process and project literacy Basic visualisations (bar chart, pie chart etc.) Logic Narratives and storytelling See left + Probabilistic reasoning Experimentation / causality Scientific method Basic statistics Common visualisations (histograms etc.) Basic understanding of data science Abstraction Future Future Today How we adapt
- 18. The dangers of ignoring data literacy Ignoring the development of data literacy in your organisation means running the risk of: ● Being unable to process all the information available to you / your organisation ● Being unable to understand and question the output of data professionals ● Contributing to a floundering data analytics environment ● Missing opportunities, inability to pivot, no competitive edge FutureHow we adapt
- 19. Next steps Improve your data literacy: ● Statistical training ● data science courses & MOOCs Network: see data science meetups in your city Find a mentor Aim to be hired in a decision support role. Participate ‘Man suit thinking’ (cropped and BW re-colour) by asierromero available at https://www.freepik.com/free-photo/man-suit-thinking_927831.htm
- 20. Next steps: improve your data literacy Statistics: ● Bedtime reading: OpenIntro Statistics (www.openintro.org) ● If interested consider a masters in statistics / econometrics / biostatistics Data Literacy courses: ● Courses on data visualizations (see MOOCs) ● AlphaZetta Academy Relevant tools, e.g. SQL, Python and/or R courses ● Massive Open Online Courses (MOOCs) ● Study groups (see meetup.com) ● AlphaZetta courses Participate
- 21. Next steps: network & There are meetups in multiple capital cities for: ● Data science ● Tools (R, Python) ● Statistics
- 22. Next steps: find a mentor Look for a good mentor (see networking), expect them to: ● Have relevant experience & willingness to share their knowledge ● Be your champion ● Help you understand realistically your strengths and weaknesses ● Guide you in developing your data literacy (or in building your data science skills) Participate ‘wocintech (microsoft) - 229’ (cropped) by WOCinTech Chat available at https://www.flickr.com/photos/wocintechchat/25677265212under a CC by 2.0.
- 23. Next steps: working in decision support role ● You can work in data driven decision making as a data professional (skills permitting) or as a savvy data consumer (data literacy permitting). ● Very competitive hiring environment. Job-seekers have varied backgrounds / qualifications ● No fixed qualification / no centralised qualifying body / unprotected environment ● Hiring managers are often unqualified to hire data professionals or data literate professionals ● Fast paced technological environment ● High churn in the industry Participate Untitled (cropped and BW re-colour) by pxhere available at https://pxhere.com/en/photo/410489 under a CC BY 2.0.
- 24. The Independent Analytics Experts
Editor's Notes
- -------------- Image source: https://pxhere.com/en/photo/1376465 No attribution required
- -------------- Image source: https://pxhere.com/en/photo/1458907 No attribution required
- The rapid technological change in data analytics is best summarized by Eric Ries Eric is the who is the creator of a very useful concept called the Lean Startup Eric: motivation for lean start up is we’re getting better at doing stuff NEXT: Examples of how well we can run projects -------------- Image source: https://www.flickr.com/photos/betsyweber/6730126217/ Attribution 2.0 Generic (CC BY 2.0)
- Hammer home point: we’re going at doing stuff. Useful analogies: Can create a bank infrastructure at home using cloud based technologies Doing stuff = there’s an app for that. Apps are the “can you have fries with that” of the tech world NEXT: Examples of how doing stuff has gotten easier -------------- Image source: https://www.flickr.com/photos/betsyweber/6730126217/ Attribution 2.0 Generic (CC BY 2.0)
- NEXT: Doing stuff is less important than deciding what to do (i.e. making decisions)
- NEXT: nature of work is changing -------------- Image source: https://commons.wikimedia.org/wiki/File:Sipping_Bird.jpg Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0)
- Doing stuff is less important than deciding what to do (i.e. making decisions) Tell building bridge analogy/story doing is becoming a trivial task even though it used to be the higher status task because you could be a town on this side of the river and there's a market on that side of the river it's like everybody knows that if only you could build a bridge things would be great the trick was building that bridge it's hard it's expensive & it's risky Doing stuff is still important (it won’t go away) but increasingly as things become easier to do, we need to look at how we make decisions NExt: good decisions are important
- Tell decision-maker / executive story: In the future important people are going to be called executives they'll be called something that has to do with their ability to decide rather than their ability to make things happen I see a decoupling between the decision the decision function (which is all about wisdom) and the executive function which is all about getting things done. Getting things done may not even about getting people to do things it's about getting machines to do things very different skill Therefore decisions are much more important than ever NEXT: Key point: lots of talk about data, little talk about decisions
- Data scientists exist to facilitate good decision making Good strategic decision making is important Caveat: realistically, data scientists aren’t often hired for decision support Little focus on decision making and more focus on buzzwords (e.g. “big data” etc.) In any case: strategic decision making IS becoming more important than “doing things” NEXT: but we’re not using data to enable better decision making -------- Image source: https://pixabay.com/illustrations/man-notebook-continents-binary-1071773/ Attribution not required
- -------- Image source: https://pixabay.com/illustrations/man-notebook-continents-binary-1071773/ Attribution not required
- Discuss what you’ve seen: Trends: government can’t fire people, people’s roles are changing because I see many people’s jobs changing I run training businesses I train a lot of government people to become data professionals In the future everyone will be data literate NEXT: illustrate current data revolution with computer literacy revolution ------
- If data analytics was like IT - analogy: IT professional in 1985 has at least the same order of magnitude of knowledge as your average 12 year old We all here today have a non-trivial knowledge about computers - in the early 1990s we’d be considered IT professionals We all need this basic knowledge to be functional members of the middle class. Today you can’t get by without a non-trivial knowledge of computers. Therefore there is a stack of knowledge the average manager and member of the middle class should have. NEXT: discuss data revolution ------ Image source: https://www.flickr.com/photos/cityofbostonarchives/10086196513/ Attribution 2.0 Generic (CC BY 2.0)
- Discuss data literacy professional No one will be completely removed from: Data Logical reasoning Scientific experimentation CAVEAT: we are currently an anomalous time - there is no drive to use data scientists & data effectively - no drive for the management class to be data literate. Some fields are very data literate Hedge funds Sports betting Politics Most places are not data literate, because: We’re economically “lucky” When do good decisions matter: when times are tough. When times get tougher data literacy will matter Analogy: expensive fish - data scientists + water - data literacy of the company Analogy: printing revolution without reading. ------ Image source: https://www.flickr.com/photos/cityofbostonarchives/10086196513/ Attribution 2.0 Generic (CC BY 2.0)
- Also discuss, data driven roles which are: statutory roles Not about managing business risk This may be difficult to go from a statutory role into a decision support role Decision support roles Easier: about supporting decisions in the business ----- Image: https://www.pexels.com/photo/photographed-grayscale-of-child-in-dress-1096932/ No attribution required
- Also discuss, data driven roles which are: statutory roles Not about managing business risk This may be difficult to go from a statutory role into a decision support role Decision support roles Easier: about supporting decisions in the business ----- Image: https://www.pexels.com/photo/photographed-grayscale-of-child-in-dress-1096932/ No attribution required
- Also discuss, data driven roles which are: statutory roles Not about managing business risk This may be difficult to go from a statutory role into a decision support role Decision support roles Easier: about supporting decisions in the business ----- Image: https://www.pexels.com/photo/photographed-grayscale-of-child-in-dress-1096932/ No attribution required
- Also discuss, data driven roles which are: statutory roles Not about managing business risk This may be difficult to go from a statutory role into a decision support role Decision support roles Easier: about supporting decisions in the business ------
- Notes: Executives of today will not be like executives of tomorrow ------ Image source: https://pxhere.com/en/photo/410489 Attribution 2.0 Generic (CC BY 2.0)