What is a 21st Century Internship Program?

This post is inspired by a discussion with the great team examining and shaping the future of education in emerging markets at Pearson.

In the last post, I discussed how overcoming the waste and exploitation that is currently endemic in the unpaid internship system in the US will require educational providers to more clearly define their overall learning objectives and structure internships to achieve them. Digging deeper into this approach leads to two critical broader questions that we at Kepler – and hopefully others – will be exploring as develop the work-based learning elements of the model: what is the purpose of internships? And what should optimal 21st century work-based learning experiences look like?

The obvious answers to the first question are to learn and facilitate desired employment as I have assumed in my arguments in the last post. But it is useful to unpack these objectives further before designing an internship program. A thoughtful student will pursue an internship for some or all of the following benefits:

  1. Development of relevant skills through practical experience;
  2. Build their professional network
  3. Associate themselves with a professional brand
  4. Establish a direct pathway to a future job
  5. Explore a career to determine if they want to invest themselves further in it
  6. Receive credit towards a degree

 For undergraduates, the first objective is arguably the most important and is what most educational providers will likely say is their sole interest. Delving further into that objective, that learning will ideally apply to three critical categories of skills: 1) specific technical skills (e.g., financial modeling), 2) generic cognitive skills (e.g., critical thinking and writing), and, perhaps most importantly, 3) non-cognitive skills such as professional demeanor and communication.

The internship experience can help the student develop those skills through three channels that they are unlikely to benefit from in a normal campus environment:

  1. Application of their skills in real workplace environments;
  2. Receipt of feedback on their products from experienced professionals;
  3. Observation of experienced professionals;

 As Carey’s article suggests, many traditional unpaid undergraduate internships offer little or no value for many of these elements. Students end up doing grunt work that provides them with few opportunities to practice meaningful skills and given their bottom-caste status, only a few and fortunate make valuable connections. In my personal experience, the greatest value often comes from the observation of the workplace and the corresponding information to help the student decide whether to pursue work in the field (in many cases the brand on the CV is of limited value as employers know the truth of intern experience).

There were traditionally few alternatives for students interested in work-based learning in a professional environment. However, work is becoming increasingly democratized. Increasingly, anyone with a computer, an internet connection, and relevant skills can seek gainful work experience? Sites like Odesk enable workers to sell their skills around the globe. Imagine if students spent the time they would have otherwise devoted to making photocopies at an unpaid internship contributing to projects through those sites. How would the experiences compare?

Such remote work can never enable students to observe and learn from full professional environments and they would have less opportunity to make connections. On the other hand, they would be guaranteed to have opportunities to hone their skills and receive valuable feedback (employers will only post projects if they value the work and will therefore have quality standards). If they are good, they will develop online credentials and connections that could be equally or more valuable than those they would get from a traditional office internship. And they will be paid.  

Traditional office environments, like degrees, will not disappear soon. But the global office-of-one is clearly a major trend in the future of employment so experiences that help students develop this approach to work may be an attractive investment despite the drawbacks. For a new institution like Kepler, the answer may to support both internship approaches. Remote projects could complement well-designed traditional internships for those students who are able to secure them and could be an attractive replacement for the low quality internships or lack of work experience that would otherwise be available to the other students. And clever, strategic institutions could work with partners and students to craft remote projects that could maximize both learning and future employment opportunities for students. We look forward to exploring these and similar opportunities.

Better, Not Fewer, Interns(hips) Please

Kevin Carey, a consistently articulate and insightful observer of the forces buffeting traditional higher education at the moment, has a trenchant piece in the NYT today on the often exploitative world of college internships. He chronicles how universities and businesses often both receive free money (in the form of tuition for absent students and unpaid labor respectively), while a growing number of students struggle to unbury themselves from mountains of debt and to find relevant work. As the lively responses in the comment thread of the article show, there of course a wide spectrum of internship experiences, with some universities and businesses providing significant value to the students in the return for their labor. For those of us who are focused on designing better ways to prepare students for life-long success to close the education-employment gap, it is useful to explore the roots of the problems Carey reveals.

The flaws in many current internship systems grow from the same deep root of the other major problems plaguing higher education: the lack of clarity of, focus on, and accountability for the outcomes of college. Carey suggests that a principal challenge with internships is that they are unlike the oversight and assessment of academic classes that is the core business of colleges. In fact, the challenge is that they suffer from the same malaise. The authors of the scathing Academically Adrift attribute the lack of learning improvement they observed in their study partly to a “compact of disengagement” in which professors do not design challenging classes and inflate grades in order to ensure positive student feedback – and therefore job security – and more time to focus on research, while students reward this cakewalk so they can coast to the piece of paper they see as the primary purpose of their college investment. As long as the focus is on inputs (e.g., credit hours or a brand on a resume) rather than outputs (true learning), academics and internships will both be riddled with waste and abuse.

The answer to the problems Carey highlights is not reduce the focus on internships: practical work experience is the right prescription for the diagnosis of the epidemic of unemployed graduates unprepared for the modern workforce. Nor is it to tighten regulation. Rather it is to integrate internships into a broader system of definition of and support for student learning outcomes. In his recent book, Sal Khan profiles the internship program at Waterloo University in Canada. Students spend significant time interning throughout their four-year course of study at institutions like Google and Facebook and are paid appropriately for their time. As a result, they graduate with bright employment prospects and little or no debt. What would it take to expand these types of internships?

First, educational institutions will need to employ a “build it and they will come” approach. In this example, the employers clearly see value in the students. They have the skills to add at least some value during their paid time at the company and they have a high potential to grow into badly needed effective full time employees. This requires some vision on the part of the company. But the responsibility rests primarily on the educator to help their students achieve the quality during and after their studies that will make their value unavoidably clear to employers. This will undoubtedly be much more difficult or impossible in some industries where supply of potentially qualified students vastly exceeds demand (respondents to the article frequently reference the entertainment industry). But with many employers struggling to find the talent they need, it should be possible to apply this approach well beyond Silicon Valley.

Second, colleges will need to see internships as an integral part of the educational path that is their mission and not an ugly but wealthy stepchild. That begins with being clear and specific about what students are expected to learn during their time at the institution. With such an objective in place, staff, students, and employers can all be held accountable for the extent to which an internship experience achieves the necessary progress towards it. Employers can be vetted based on the learning their internships generate and delisted if they are not adding sufficient value. And students can see the outcomes they need from their internship time, apply themselves accordingly, and challenge both the college and the employer if they are not receiving the support they need.  There will always be those who will pursue an internship for the perceived value of the brand on their resume and the potential connections. But as this approach expands, more students and employers will see the benefits it generates compared to the often-rotten fruit of the current system.

Lastly, colleges will need to structure their staffing and other systems to support this approach. While some professors clearly take internship oversight seriously as the comment thread shows, many professors do not have the incentives to invest adequately in academic learning let alone work-based learning. Institutions will need to design the right mix of staff between academic and work-based learning and ensure that there are sufficient individuals with the skills, interest, and incentives to robustly support the latter.

These changes will not be easy, particularly for entrenched existing institutions. But as Waterloo U and other leaders like Northeastern U have shown, it is possible to overcome the challenges and realize the potential win-win-win for employers (better hires and skilled interns), colleges (higher demand), and students (better learning and employment). This approach will be core to Kepler and we have begun to develop how it will be best applied in the Rwandan context.

But the most important question may not be how to design an optimal workplace internship, but what an internship that prepares students for successful work over the next three decades should be. I will explore this question in a subsequent post.

The Best University in the World

You probably haven’t heard of the best university in the world.

The best university in the world has a perfect, 100% employment rate, with all graduates going into stable careers in a growing company. And almost every single student graduates. It is free and most students actually get paid to attend.

It leverages technology to deliver a cutting-edge, blended experience. And all content is designed explicitly to prepare students to start working the day after graduation. They have a best-in-class non-cognitive curriculum.

It’s amazingly accessible as well – like the UN they have a staff of translators on hand to enable content delivery in 28 languages, and the university has campuses throughout the developed and developing world.  It is accredited by the American Council on Education – one of the toughest accreditation bodies in the country.

And it’s not easy to get in. When a new campus opened in Shanghai, the acceptance rate was under 1%, making Harvard admissions seem easy.

However, even the best university in the world does have a couple of weaknesses. It doesn’t have the brand or prestige of Harvard, and it offers a fairly narrow set of courses, although in the UK it’s starting to offer a broader business degree.

In the US right now it only offers one degree: a Bachelors of Hamburgerology.


A BA in Ham doesn’t have the same ring to it as a PhD from Yale, but the rise of corporate training programs as a real alternative to traditional degrees, particularly for higher ed consumers who don’t have access to elite universities is genuinely exciting. As David Fairhurst, “chief people officer” at McDonald’s, told The Telegraph, “Tuition fees are rising and university places are increasingly hard to come by. There is a role for employers to fill that gap and provide a new model of learning.” Infosys’s 335 acre campus in Mysore now trains 10,000 students per year according to the Economist, and Wallmart has partnered with an online degree program to give its employees on-the-job education.

Whether the future of education is just linked more tightly to jobs, or whether this kind of employer-sponsored degree could actually replace traditional higher ed for some students, any trend toward better outcomes for the students who need them most is exciting.  And HU’s outcomes – 100% graduation, 100% employment, $0 student loan debt – should make every college in the country pay attention.

Assorted Links

Many of the best articles I find come from the Twitter feed of the prolific and insightful Matt Rascoff. If you’re interested in innovation in education, follow him as fast as you can. (@mzrascoff)


MOOCs get the Thomas Friedman Treatment

Thomas Friedman’s wide eyed Times op-ed on MOOCs is everything you would expect it to be, given that MOOCs represent the education world’s apotheosis of what he loves most: the democratizing, world-flattening, poverty-killing power of technology and communication. He’s certainly not cautious in his vision:

I can see a day soon where you’ll create your own college degree by taking the best online courses from the best professors from around the world — some computing from Stanford, some entrepreneurship from Wharton, some ethics from Brandeis, some literature from Edinburgh — paying only the nominal fee for the certificates of completion. It will change teaching, learning and the pathway to employment.

First off, at the broadest level this op ed is just about perfect. MOOCs deserve Friedman’s brand of technological determinism. They’re part of a set of technologies that are about to turn global higher education on its head, and it’s great to have Friedman out there banging the drum for them.

A small quibble: we believe, and I suspect if pushed that Friedman would agree, that there’s more to a university education than a composite of subject-specific courses. But it’s not so clear at first glance what the essential elements actually are. Part of Kepler’s mission is to figure out what level of real-world facilitation and coaching could turn the vision Friedman describes above into a world class education for a broad base of students in the developing world. Until blended-learning MOOC projects like ours and others begin to take off, we’ll have to make due with the anecdotal Mongolian or Ivorian geniuses that currently populate articles like Friedman’s.

Nudge U

David Brooks wrote a column this week urging greater use of behavioral science to shape policy decisions, applying relevant evidence to correct the inefficiencies that often result from our common sense assumptions of what will influence people to adopt desired behaviors (he cites, for example, the fact that immediate grief counseling as been shown to entrench rather than relieve trauma of police officers whose partner is killed). He is essentially advocating for the doctrine of “libertarian paternalism” that was compellingly articulated by Cass Sunstein and Richard Thaler in the book Nudge and which has been tentatively embraced by the Obama (by bringing Sunstein in to scour new policies for opportunities to nudge rather than smother) and Cameron (through the creation of a recently renewed “nudge unit”) administrations.

The doctrine holds considerable appeal. Through steps as simple and cheap as having a box to contribute to a 401k start as checked rather than unchecked when someone starts a job, we can significantly improve social outcomes without obstructing individuals’ freedom to choose for themselves. As Brooks suggests, there are likely a host of no-brainer such switches across our current public policies that are begging for a combination of relevant research and public leadership.

Brooks concludes by calling for greater application of the approach to tackle the largest problems facing the country.  Perhaps some well-designed nudges could make modest contributions towards resolving the US government’s medium-term fiscal train wreck. But it is hard to imagine such a massive and entrenched issue being resolved without more sweeping change and disruptive leadership.

But what about applying nudges to the ills of the world’s higher education institutions. Imagine if universities had small nudge units like the UK government striving every day to identify simple, cost-efficient behavioral influences to improve student outcomes. In the recent NYT profile, Angelica was buried crippling student debt because she did not meet the deadline for financial aid. Could Emory have used other simple nudges (text messages?) to increase her chances of meeting the deadline? Could carefully designed use of social pressure (like the voting example mentioned by Brooks) or steers to access support services increase the college completion rates?

As with public policy, it is likely that there is an initial crop of no-brainer higher education nudges (whether within individual institutions or by local or national governments) waiting to be reaped. Taking advantage of those opportunities will require a number of key steps.

  1. The institution needs to be willing to focus on improving the target outcome. While seemingly obvious, we have seen that many institutions do not see major challenges such as graduate employment or measurable improvements in generic skills as their problem or role. This will be at least half the battle in many cases.
  2. Relevant, rigorous, timely, and efficient research needs to be conducted. This research will be most efficient if it is coordinated across multiple institutions, particularly given the focus on cost consciousness. But it will be worth even individual institutions investing in this given the potentially high ROI that successful nudges could generate. We certainly plan to experiment in this area as much as possible in Kepler.
  3. Strong leadership and good management will be needed to execute the nudges identified and be willing to make mistakes and change course based on ongoing learning.

Sunstein recently left the administration and returned home to Harvard. Perhaps he could look inward and launch the first such university nudge unit. Now if only we could identify a nudge to get universities to develop more nudges…  

University Ventures view of the universe

University Ventures is out with another insightful post. We haven’t yet come across an organization whose outlook on higher education is more aligned with ours.

The latest post takes a crack at explaining the public perception gap between private for-profit (scuzz/slime) and private nonprofit colleges (ivory/teflon) in the US. It argues that the gap is due to the path- and time-dependent evolution of our current higher ed landscape: a ‘firmament’ of mission-driven private nonprofits that evolved 100+ years ago, joined only recently by a wave of financially-motivated for-profits after the private nonprofit category’s reputation had been cemented. Takeaways as I understand them: 1) US-based for-profits should therefore look internationally for their growth opportunities where they won’t face the same uphill perception battle; 2) innovation in the ‘exceptional’ US market will require partnerships where clearly nonprofit/public colleges are in the visible student-facing role and for-profit companies are behind the emerald curtain providing capital, technology and services.

The retrospective description of how the perception gap emerged is compelling. And no doubt that US-based for-profits should be looking for growth abroad, particularly in the near term given the current climate. Whether the second takeaway will hold true in the longer term is less clear.

UV’s earlier holiday letter — which I found to be the best piece of writing in 2012 on higher ed’s future — argued cogently that perceptions of online degree quality/value will improve quickly (especially 5-15 years out). Quality will improve intrinsically, and the movement of teflon nonprofits into the online space will enhance online learning’s reputation. At the moment, it seems that perceptions of private for-profits and perceptions of online learning are closely intertwined (private for-profit is to online as PB is to jelly). Likewise, perceptions of motivation and and perceptions of quality are hard to separate. If UV is right that perceptions of online learning quality will improve dramatically — and if for-profit online learning quality improves (partly through new entrants) at a rate comparable to nonprofit online learning quality — then mightn’t that also dramatically improve the reputation of for-profit providers? And if that happens, then for-profit players could step out from behind the emerald curtain.

All that said, what matters most is that the innovation happen. Ideally, it’d be innovation driven by all three sectors — for-profit, private NP, and public. It’d be a shame for one of those sectors to be mostly on the sidelines or hidden from view for hollow reasons.

Also wanted to share a few small thoughts building on Oliver S’s earlier response to the UV holiday letter. Main points of the letter were all spot on. My few reactions other than ‘wow, great piece’ were:

1) Agree with Oliver S. that the ‘onground’ vs ‘online’ is too simplistic; very much believe there will be a third category of ‘blended/hybrid’ institutions. Every ‘onground’ college will start to incorporate online learning pieces into their curriculum, and many of today’s ‘online’ providers already have at least a small piece of ‘onground’ in their program. But we aren’t thinking of around-the-edges moves toward the middle of the spectrum. We believe a new category will emerge with colleges that have deliberately-designed hybrid model in their DNA from the get-go.

2) On a related note, the discussion of student outcomes was unsatisfyingly thin. It treated outcomes as an ‘engagement’ or ‘focus’ problem. In other words, if the learning isn’t captivating you, it’s easier to walk away from an online program because you’re just walking away from a machine. Seeing this as the problem, UV turns to solutions that are largely related to engagement & learning quality — particularly synchronous learning and ‘flow.’

While there’s a big kernel of truth in this diagnosis, I don’t think it explains the outcomes gap at least as regards persistence rates — and particularly not for lower-income students. I think the bigger explanatory factors are overall program quality (not just learning quality) and relationships. When a student drops out, it’s not generally because they’re bored. More often, it’s the kind of shit-happens, life-throws-curveballs, uphill-battle stuff that was described in the great ‘Getting off the Island’ NYT piece over the holidays. This is especially true for low-income students but also for others; and for older students (= many online learners) there’s a lot more going on in life than is true for 18-year-olds. Colleges with an onground presence don’t all do a great job with this — witness the NYT piece — but to the extent they’re doing at least somewhat better my hunch is it’s due to a) the power of in-person relationships with both peers and adults at the college who can pull you through a challenging issue or moment; and b) the fact that ‘onground’ colleges as a category have invested more in building non-academic support services that help with retention.

So the solution to poor outcomes for online providers will have to involve more than learning approaches like synchronous technology and ‘flow.’ And therein lies, in our mind, one of the big structural advantages for hybrid models over purely online models. The academic side of high-quality hybrid models will progressively go more and more online as the technology and content improves; but the peer interaction and student support elements (not to mention others like the local-employer-relationship element) of college will always be more powerfully delivered in person.

3) On the concern about the ‘Dear Colleague’ letter reifying the credit-hour as the basis for Title IV eligibility — let’s hope SNHU is about to blow this out of the water! Fingers crossed. College for America degree went live today — Title IV eligibility to be added soon if the gods aren’t crazy.

Laying out a vision of the next generation

The always-insightful Tom Vander Ark has just posted a trenchant overview of an optimal “next gen” university. Two of his points – and one omission – are worth exploring in more depth.

First, he rightly emphasizes that better preparing students for employment should be a central pillar of a new higher education institution. The more I read about developments in higher education over the past several decades, the clearer it becomes that most current institutions are lousy at developing the employability of their students not because they fail, but because they do not try. Directly linking students to employment is at best a minor priority for most colleges and universities. This is vividly depicted in McKinsey’s recent report on Education to Employment, with more than half of surveyed institutions not knowing or over-reporting their graduates employment rates and most citing partnering with other universities and companies (and, of course, research) as equally or more important that helping students find employment.  Given this playing field, Vander Ark is right that any new institution that actually prioritizes employability will find it has quickly exceeded the status quo.

That is not to say that increasing employment will be simple or easy. Vander Ark helpfully breaks the task of increasing employability down into three core functions that an institution will have to execute:

      • Job relevant knowledge and skills: The result of back-mapping from job requirements definitions;
      • Market signaling: Degree or certificate from a reputable institution, badges that reflect recognized achievements, portfolio of quality work, and/or references based on value delivered; and
      • Connections: Building and/or plugging in to a network of people with relevant connections.

Each of these will require intense and consistent legwork to understand the needs and interests of the employers in the surrounding market and, equally importantly, ensure they understand the value of the school and its graduates. This will take time, creativity, and, in the form of dedicated staff, money. And we would add one more function to this list: providing students with opportunities for practical work experience prior to graduation as has been effectively demonstrated by Waterloo and Northeastern Universities. The primary challenge therefore is not whether a new institution can significantly increase employment rates, but rather how to do so will minimizing costs and maintaining affordability. Overcoming this challenge will require careful experimentation and iteration, which we look forward to contributing to through Kepler.

Second, Vander Ark highlights the implications of these alternative priorities for the staffing structure of new institutions. Instead of content creators and researchers (i.e., the typical professor), these institutions will rely on staff who are purely focused on – and, ideally, well incented to achieve – the target student outcomes:

“These new broad access learner-centered institutions will have very different staffing patterns than traditional universities.  Next-gen institutions will have fewer staff involved in content delivery and more involved in student success.  Teams will support academic success. A advisory [sic] will track progress, guide course taking, and facilitate work experiences.  Next-gen institutions won’t have a research institution (i.e., R1) but they will be all about continuous improvement and supportive of iterative development.”

The enabling of this new staffing structure is one of the most important effects of the MOOC mania of the past year as institutions will be able to increasingly source world-class content at low – or no – cost. But this aspect of the model raises another priority that Vander Ark implies but does not adequately discuss: quality. He may have omitted this since it goes without saying that a new institution will have to ensure adequate quality of student learning. However, for an institution that is trying to ensure prices as low as $1,000 per year, this too will require innovation and change. As such, he might have included at least one other paragraph that read something like this.

A next-gen institution will set clear standards for what students will learn during their tenure and consistently, rigorously, and efficiently measure progress against those standards. It will then use that data to develop and test additional methods for improving those outcomes while maintaining affordability, including different approaches to tutoring and peer learning and/or adaptive learning software. The institution will use competency-based learning not just to increase efficiency and flexibility, but also to better tailor instructional support and improve many students’ learning outcomes (and motivation). A next-gen institution will hold its management and staff accountable for core outcomes such as learning and employment and transparently share those outcomes to, over time, clearly differentiate itself from traditional players and establish a new standard in the field.

But above all Vander Ark’s post captures the essence of what is energizing us at the moment: building such a next-gen institution will be a fascinating and exciting challenge.

Home cooking

Over at Slate, Matthew Yglesias wonders whether online education could be useful but not very profitable:

For a skeptical comparison, I would look to cooking. There’s no doubt at all in my mind that the Web has changed how people cook. It is much cheaper and easier than ever before to look up a recipe for just about anything. Digital technology is also very helpful with weight and measure conversions. And I’m sure I’m not the only one who’s learning cooking techniques in part by watching videos. So online learning with regards to cooking is a huge success. But it’s not a huge business success.

Sure, but then let’s also assume that 99.9% of people want formal training as chefs, but few can afford culinary school. How’s business development at Cook’s Illustrated looking then?

University system, how ya like them apples?

Both the Boston Globe and Boston Magazine have great articles this week on the disruption of higher education (h/t Wired Academic). But one of the best Bostonian commentaries on our university system comes from none other than Matt Damon.  My favorite scene in Good Will Hunting is the incredibly satisfying dressing down of a cocky intellectual.  Take a minute and watch this, it’s just as good as you remember:

The riff on higher ed (starting at 3:17) is as eloquent as it is entertaining:

Will: See the sad thing about a guy like you, is in about 50 years you’re gonna start doin’ some thinkin’ on your own and you’re gonna come up with the fact that there are two certainties in life. One, don’t do that. And two, you dropped a hundred and fifty grand on a fuckin’ education you coulda’ got for a dollar fifty in late charges at the Public Library.

Clark: Yeah, but I will have a degree, and you’ll be serving my kids fries at a drive-thru on our way to a skiing trip.

Watching the movie you’re rooting for Will but you can’t help cringing a bit as you realize that in fact Clark is probably right.  In our legacy model of higher education, no matter how smart you are, if you aren’t born into enough wealth to foot the bill for an expensive education, and you don’t have the non-cognitive skills you need to make it through a challenging education, your employment prospects are pretty grim.  Will might not know it, but what he needs is a new model of higher education.  He needs a competency-based degree that gives him credit for the skills he’s already mastered, delivered at least partially online so he has the flexibility to continue working to pay the bills, all capped with strong character education to help him stick with it and succeed.  Sound familiar?

For more Good Will Hunting lessons on reinventing higher ed, check out Stellan Skarsgard’s attempt at competency-based learning - offering college credit for completion of a single assessment, as well as Robin Williams’ intense approach to character education – instilling Will with enough grit and optimism to stick with the job search and get hired.

And don’t worry, even though I can’t quite find a connection to the disruptive higher ed ideas in this next clip, I won’t leave you hanging.  The punchline you’ve been waiting for: