On whether America’s obsession with STEM is dangerous

A couple of weeks ago I was sent Fareed Zakaria’s article “Why America’s obsession with STEM education is dangerous” (March, Washington Post). The article looked like a counter to the endless commentary on the need to increase the technical skills of students. This movement is largely USA driven but also reaches Europe, Australia and a bunch of what I would generalise as ‘Western education systems’.

The gist of the article is that a hardcore drive towards technical skills (i.e. the learn to code movement), at the expense of liberal arts education, would erode the very elements that made the USA a success – creativity, innovation and problem solving. Zakaria points to the USA’s low global rankings in maths and science but, instead of seeing that as an issue, says that it has always been this way and the USA has been successful regardless.

Personally I believe this is a misguided, clickbait-y, point of view. I have never heard anyone say that liberal arts degrees, and diverse interests, are not valuable. The line is always that technology is a tool to deliver your ideas. These ideas come through big thinking, innovation, creativity etc. It’s not that coding advocates want to do away with liberal arts degrees – its more that our students are coming through the elementary school system without even a basic exposure to tools that will benefit them immensely during their lives (and careers).

The article suggests that revised curriculums have a “narrow” STEM focus. I disagree with this as well. The revised curriculum in Australia proposes a strong focus on problem solving and logical thinking. It does this in the form of computational thinking  – and backs it up by exposing students to the programming and coding skills needed to deliver this creative thinking. There is a large gap between our generations comfort with technology and our ability to make things with it. This gap between confidence and competence is what we are working to reduce – hopefully creating a generation of students that are flexible thinkers, capable of finding jobs that deliver on their passions. In the 21st century its the students with the best skills set that will have the greatest chance of making the impact they want to see.

On the learn-to-code movement in 2015

Last December, Ryan Seashore, CEO and founder of CodeNow, put up his thoughts on the learn-to-code movement – where its at and where it needs to be. I like this article for several reasons. First, it attempts to put a time line around the host of initiatives – making it very clear how recent most of them are. Second, it adds structure to the movement and distinguishes between the different players.

Here’s how Seashore broke them down:

Awareness: the purpose of these orgs is to raise awareness of the need for increased computer science/coding education. The main player here is Code.org, with Made with Code also referenced. Success is measured in terms of publicity, social media and uptake of campaigns such as Hour of Code and Computer Science Education Week. [JW note: I’m pretty sure Code.org would argue that, through their teacher support materials, they moving well down the chain into the exposure and immersion categories, but in general, they play an important role in raising awareness of this issue]

Exposure: main goal of these orgs is to give students a taste of coding as a discipline. The idea is to give students exposure so they can then decide if its an area of interest they might want to pursue in college. Players: CodeNow, Black Girls Code, CoderDojo, Technovation, Rails Girls etc. Success is measured by the number and diversity of students that attend the programs.

Immersion: a subset of the exposure groups – here the aim is to bridge the gap between the first 5-30 hours and full blown curriculum. Programs include SMASH, Girls Who Code, TEALS, ScriptED, UrbanTXT. Once again, success is measured by number and diversity of attendees.

Vocational: generally for-profit. Think General Assembly, Dev Bootcamp, Hack Bright Academy etc. These guys have grown fast and can offer specialisations like UI design, front end coding, app creation, data science etc.

Online: these are online courses, often free, from CodeAcademy, Khan Academy, CodeSchool and MOOCs facilitated through Coursera and Udacity. These can range from hour long tutorials through to 8-12 week courses. Here Seashore comments on the frequently quoted 5% completion rate and the huge amount of self-discipline needed to actually finish the courses.

The ‘where to from here’ section of Seashore’s article touched on a few things: a) the creation of a national association to coordinate the linkages between all the above organisations – to help them move up the education ladder, b) setting public goals for education, c) pushing the public to challenge the government to make coding mandatory in schools, and d) pushing tech companies to do more than just donate money – they can play an active role in educating students and taking interns.

My thoughts:

• Not-for-profits: Seashore starts his article saying that not-for-profits are stepping into this space because the tech industry has diversity issues and sectors of society (women, African American, Hispanic) are not being included in this incredibly progressive, well paying industry sector. This is true for Australia as well – the tech sector has low representation from women, Indigenous Australians and other groups. However our base uptake is so low that, in general, Australians as a whole are missing out on this opportunity. Not-for-profits are stepping in, not for diversity reasons, but because they can move faster (and arguably have less responsibility) than official government education. Very few of the organisations above have a physical presence in Australia – which relegates them all to the ‘self driven’ online category.
• Educational progression: Seashore’s article warns of offering false hopes. Imagine for example that enthusiastic volunteers (from organisations or companies) come into a school, run an amazing workshop on coding, generate a healthy amount of interest in technology but then leave an under supported teacher to figure out the next steps. It would make sense for each outreach activity to have a ‘next steps’ component to their activities.
• Localisation: in Australia we are starting to see a few organisations emerge in for-profit category of tech education – General Assembly, CoderFactory and Code Rangers (relatively new) are some that come to mind. Code Rangers is interesting because it is playing in the traditionally not-for-profit space of youth education but is tapping the structured, and paid, after school networks. Essentially providing a quality alternative to after school care for young students.

On the Digital Technologies Curriculum review process

I thought I’d put up a piece on further developments in the ongoing discussion on the role of the Digital Technologies (K-10) curriculum in Australia. This curriculum would deliver on calls for mandatory technology education, coding in particular, from foundation to Year 10. Last October I put up a piece on the then recent review of the Australian curriculum. This review suggested, amongst other things, that technology should be taught as an elective from Year 9 and older (roughly the same as we have now). The tech industry, and tech education advocates, saw this as a worrying sign.

However, things took a positive turn in December, when the Commonwealth Government (Department of Education) announced it would refer the Review’s recommendations to the Australian Curriculum, Assessment and Reporting Authority (ACARA). ACARA is the joint Commonwealth/State Government education group that led the original consultations and designed the Digital Technologies curriculum. ACARA has been asked to report to the Education Council at its first meeting in 2015.

There are still a few more areas to address. Fran Foo’s December article on this matter specifically asked for comment from NSW and WA governments as to whether they would commit to supporting this industry call for mandatory tech education (and the Digital Technologies curriculum). The NSW response was “NSW is committed to its current practice where technologies learning commences in early stage 1 (kindergarten) with the Science and Technology syllabus and continues into years 7 & 8“. A quick glance at the NSW Board of Studies website shows that NSW has just launched an updated Science K-10 (incorporating Science and Technologies K-6) syllabus in 2015. The article then quotes the NSW Board of Studies as saying “Should the Technologies Curriculum be endorsed by education ministers and BOSTES (Board of Studies, Teaching and Education Standards) decide to  adopt it, consultation with stakeholders including the advocates of coding and algorithmic thinking will ensue”.

My thoughts:

I guess we wait and see. I can absolutely see how a review could decide that the Digital Technologies would be a challenge to implement – by default it features content (coding) that is not particularly strong in Australia. This is all the more reason that it should be taught. The curriculum review focused on whether the content of the Digital Technologies curriculum was achievable from an educators point of view – was it written in a style that people could deliver. This one I’ll leave to educators. Another point was whether teachers would be supported to deliver the content – this was/is a concern for those delivering the technology subject in the UK, so it will be interesting to see how that plays out. Finally I can see why State Government education groups would be cautious about radical overhauls of education systems. Education is vital and not something that is easy to switch up – so naturally they would want to follow due process.

On the plus side, at least the conversation is continuing.

Please Don’t Learn to Code (2012)

CEO of StackOverflow and blogger Jeff Atwood’s 2012 post “Please Don’t Learn to Code” looks to douse the enthusiasm of those that think code will solve the world.

The STEM/coding movement is strong right now – code.org, codeclub, code academy, year of code etc are in full swing and countries around the world (such as Australia and UK) are pushing to increase students’ exposure to computer science through revised curricula. In his post Jeff disagrees with the increasingly popular idea that coding is as essential as readying, writing and math.

My thoughts on this are a) I don’t believe people are saying that ‘coding’ is essential – but rather, computational thinking and problem solving, developed through coding/programming skills and b) while it may not be essential to everyone – if you increase the volume of skilled students coming through then you have at least given all students an opportunity to participate in the digital economy. Much of this push is a preemptive guess that there will only be more technology in our lives from this point on.

I’ve pulled up this two year old post because I like how Jeff outlines his counter argument (summarised below):

• This movements celebrates writing code: in a perfect world there would be as little code as possible.
• It assumes writing code is a goal: the act of coding itself is not that exciting – what is exciting is the thrill of the outcome – making things and seeing how they impact society is fantastic.
• It puts method before the problem: learning how to define a problem is more important than coding. [nb I would argue that understanding your tools and limitations helps you define your problem – I’d also argue that much of this movement is also about computational thinking and problem solving]
• More coders are good: It assumes adding uncertain (bad) programmers to workforce is a good thing
• Is it a career?: It assumes there is a link between learning how to program and then getting paid for it professionally.

Similar to the other “anti-STEM” articles I looked at a few weeks ago – Jeff closes with:

I suppose I can support learning a tiny bit about programming just so you can recognize what code is, and when code might be an appropriate way to approach a problem you have.

That sounds like a happy middle ground.

On Computational Thinking

Article: Computational Thinking – What and Why? (Jeannette M. Wing, Carnegie Mellon University, 2011)

Last week I reviewed three articles on the ‘myth of the STEM crisis’. The articles suggested that there are now more STEM workers than ever before and that the current focus on increasing STEM graduates was unwarranted. However, one thing they did agree on was that increased computational thinking among school students is a good thing that would aid the students in whatever career path they chose.

“Computational thinking is used in the design and analysis of problems and their solutions. It is not just or all about computer science. The education benefits of being able to think computationally  – starting with abstraction, the process of defining problems – enhance and reinforce intellectual skills, and thus can be transferred to any domain.”

“Computational thinking overlaps with logical thinking and systems thinking. It includes algorithmic thinking and parallel thinking which in turn engage other kinds of thought processes, such as compositional reasoning, pattern matching, procedural thinking and recursive thinking.”

Educators like computational thinking because it is a foundation logic skill. It sits well with students of all ages and those that have a talent in this area can easy proceed to other forms of computer science. Those that don’t can enjoy the problem solving and logic aspects of computational thinking challenges. I consider it as pre-programming – the type of activity you can introduce to any student, regardless of their competency, and have them succeed.

The reason all three ‘anti STEM crisis’ articles focused on computational thinking, as opposed to coding, is that the basic logic it teaches can help people in any profession. Lawyers, doctors, accountants, marketing, sales, operations managers are all careers that respond positively to increased computational thinking. This is because it is a life skill, not a coding skill, that helps people break down complex problems into manageable pieces.

The value of a STEM based education

Here is an interview with Australia’s Chief Scientist Professor Ian Chubb (after the 2013 Maths and Science Symposium):

Speaking of changing our culture, there’s an opinion abroad that if someone studies science but doesn’t work in that discipline, something’s failed. That is ridiculous. A broad science education means a person has career options in multiple sectors of the economy and we should be really pleased if they go off and work in a non-science field. They will add value…

The background of a science education means that they think about evidence, that they don’t take statements at face value, that they are constructively sceptical, they analyse evidence carefully, validate and replicate and then use it.

These are characteristics inherent in a science education. If people want to work in science that’s great, but if they decide to move to something else they have the skills and talents they have acquired and honed as a consequence of an education in science.

They will be valuable members of the workforce wherever they choose to use their skills.

I believe these same words can apply to computational thinking. I also like the idea that we are not expecting every child that studies computing to go on to become a software developer. I would love a scenario where social policy makers had combined skills in anthropology and technology. Positioning STEM as a foundation skill set, not an end goal in itself, should help people calm down a touch (see previous post on the STEM crisis).

A summary of “Is Coding the New Literacy?”

This post is a summery of the reasonably detailed article by Tasneem Raja, ‘Why Computer Literacy is Key to Winning the 21st Century‘ (Motherjones, June 2014). This 6,000 word piece touches on the role technology can play in driving productivity, the decline of both student and teacher numbers, the difference between using tech and being fluent in it as well as gender and race divides. Its long but bang on.

Computational Thinking vs Coding:

What I like about this article is that pulls back from the argument that everyone should learn to code and refines the challenge as everyone should have capacity to think logically and engage computational thinking methodologies to solve challenges. The article suggests that the biggest issue we are facing is that “Unless you can think about the ways computers can solve problems, you can’t even know how to ask the questions that need to be answered” (Annette Vees, University of Pittsburgh). Computational thinking involves solving problems, designing systems, and understanding human behavior,” she writes in a publication of the Association for Computing Machinery. Those are handy skills for everybody, not just computer scientists.

Why you need computational literacy for future jobs:

The article proposes that computational fluency is fast becoming a divide in the same way that numeracy/literacy used to keep a stable of scribes employed to write on behalf of citizens. While not everyone will go on to become a programmer, at some point, everyone will need to work with programmers to solve problems. Knowing how to talk in technical terms will make this process easier. The article also draws a clear line between knowing how to use technology and knowing how to make it. It raises the idea that knowing how to make a powerpoint or edit in iMovie is not the same as being technologically literate.

US Computer Science Study Trends:

Between 1989 and 2009, while almost all other STEM subjects grew in the US, computer science dropped from 25% of high school students enrolled to 19% (USA, National Centre for Education Statistics). In 2014, only 20 states allowed computer science to count as part of a core graduation requirement. Part of this is the lack of qualified teachers. In the US, states manage teacher accreditation and there is no clear path for tech teachers as they move from university into the education system.

Gender, messaging and tech studies:

The New Image for Computing survey (2009, link) tested various messages about computer science with college-bound teens. It found that explaining how programming skills can be used to “do good”—connect with one’s community, make a difference on big social problems like pollution and health care—reverberated strongly with girls. Far less successful were messages about getting a good job or being “in the driver’s seat” of technological innovation—i.e., the dominant cultural narratives about why anyone would learn to code.

In the US, women currently make up only 20% of the tech workforce. This is a drop from 37% in 1987. Excluding huge swaths of the population also means prematurely killing off untold ideas and innovations that could make everyone’s lives better. There is evidence that girls exposed to very basic programming concepts early in life are more likely to major in computer science in college (Increasing Student Interest and Attitudes in STEM: Professional Development and Activities to Engage and Inspire Learners, 2011, link).