tag:theconversation.com,2011:/global/topics/benchmark-21349/articlesbenchmark – The Conversation2019-07-30T20:03:37Ztag:theconversation.com,2011:article/1188302019-07-30T20:03:37Z2019-07-30T20:03:37ZAlgorithms are everywhere but what will it take for us to trust them?<figure><img src="https://images.theconversation.com/files/285800/original/file-20190726-43104-1ythktr.jpg?ixlib=rb-1.1.0&rect=433%2C274%2C3636%2C2343&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">An algorithm is just following rules designed either directly or indirectly by a human.</span> <span class="attribution"><span class="source">Shutterstock/Billion Photos </span></span></figcaption></figure><p>The <a href="https://www.vox.com/technology/2018/10/1/17882340/how-algorithms-control-your-life-hannah-fry">role of algorithms</a> in our lives is growing rapidly, from simply suggesting <a href="https://www.google.com/search/howsearchworks/">online search results</a> or <a href="https://newsroom.fb.com/news/2019/05/more-personalized-experiences/">content in our social media feed</a>, to more critical matters like helping doctors determine our <a href="https://www.bloomberg.com/opinion/articles/2018-08-24/ibm-s-watson-failed-against-cancer-but-a-i-still-has-promise">cancer risk</a>. </p>
<p>But how do we know we can trust an algorithm’s decision? In June, <a href="https://edition.cnn.com/2019/06/26/us/google-maps-detour-colorado-trnd">nearly 100 drivers in the United States</a> learned the hard way that sometimes algorithms can get it very wrong.</p>
<p>Google Maps got them all stuck on a muddy private road in a failed detour to escape a traffic jam heading to Denver International Airport, in Colorado.</p>
<figure>
<iframe width="440" height="260" src="https://www.youtube.com/embed/lHd793hZVU8?wmode=transparent&start=0" frameborder="0" allowfullscreen=""></iframe>
<figcaption><span class="caption">Google Maps glitch sends Colorado drivers to muddy backroad.</span></figcaption>
</figure>
<p>As our society becomes increasingly dependent on algorithms for advice and decision-making, it’s becoming urgent to tackle the thorny issue of how we can <a href="http://theconversation.com/finding-trust-and-understanding-in-autonomous-technologies-70245">trust</a> them.</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/whats-not-to-like-instagrams-trial-to-hide-the-number-of-likes-could-save-users-self-esteem-120596">What's not to like? Instagram's trial to hide the number of 'likes' could save users' self-esteem</a>
</strong>
</em>
</p>
<hr>
<p>Algorithms are regularly accused of <a href="https://theconversation.com/biased-algorithms-heres-a-more-radical-approach-to-creating-fairness-109748">bias</a> and <a href="https://theconversation.com/big-data-algorithms-can-discriminate-and-its-not-clear-what-to-do-about-it-45849">discrimination</a>. They have attracted concern from <a href="https://www.huffingtonpost.com.au/entry/ocasio-cortez-facial-recognition-technology_n_5ce5d3a3e4b0db9c29929f86">US politicians</a>, amid claims we have white men developing facial recognition algorithms trained to work well only for white men.</p>
<figure>
<iframe width="440" height="260" src="https://www.youtube.com/embed/pxZk6IQxaLY?wmode=transparent&start=0" frameborder="0" allowfullscreen=""></iframe>
<figcaption><span class="caption">US committee investigates racial bias in facial recognition software.</span></figcaption>
</figure>
<p>But algorithms are nothing more than computer programs making decisions based on rules: either rules that we gave them, or rules they figured out themselves based on examples we gave them.</p>
<p>In both cases, humans are in control of these algorithms and how they behave. If an algorithm is <a href="https://theconversation.com/algorithms-might-be-everywhere-but-like-us-theyre-deeply-flawed-66838">flawed</a>, it’s our doing. </p>
<p>So before we all end up in a metaphorical (or literal!) muddy traffic jam, there is an urgent need to revisit how we humans choose to stress-test those rules and gain trust in algorithms.</p>
<h2>Algorithms put to the test, kind of</h2>
<p>Humans are naturally suspicious creatures, but most of us can be convinced by evidence. </p>
<p>Given enough test examples – with known correct answers - we develop trust if an algorithm consistently gives the correct answer, and not just for easy obvious examples but for the challenging, realistic and diverse examples. Then we can be convinced the algorithm is unbiased and reliable.</p>
<p>Sounds easy enough, right? But is this how algorithms are usually tested? It’s harder than it sounds to make sure that test examples are unbiased and representative of all possible scenarios that could be encountered. </p>
<p>More commonly, well studied benchmark examples are used because they are easily available from <a href="https://skymind.ai/wiki/open-datasets">websites</a>. (Microsoft had a database of celebrity faces for testing facial recognition algorithms but it was <a href="https://www.bbc.com/news/technology-48555149">recently deleted</a> due to privacy concerns.)</p>
<p>Comparison of algorithms is also easier when tested on shared benchmarks, but these test examples are rarely scrutinised for their biases. Even worse, the performance of algorithms is typically reported on average across the test examples. </p>
<p>Unfortunately, knowing an algorithm performs well on average doesn’t tell us anything about whether we can trust it in specific cases. </p>
<p>It’s not surprising to read that doctors are <a href="https://theconversation.com/people-dont-trust-ai-heres-how-we-can-change-that-87129">sceptical</a> of <a href="https://edgylabs.com/google-ai-cancer-diagnosis">Google’s algorithm for cancer diagnosis</a>, which offers 89% accuracy on average. How does a doctor know if their patient is one of the unlucky 11% with an incorrect diagnosis?</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/treat-or-trick-we-asked-people-how-they-feel-about-sharing-fitness-data-with-insurance-companies-119806">Treat or trick: we asked people how they feel about sharing fitness data with insurance companies</a>
</strong>
</em>
</p>
<hr>
<p>With increasing demand for personalised medicine tailored to the individual (not just Mr/Ms Average), and with averages known to hide all sorts of sins, the average results won’t win human trust. </p>
<h2>The need for new testing protocols</h2>
<p>It’s clearly not rigorous enough to test a bunch of examples - well-studied benchmarks or not - without proving they are unbiased, and then draw conclusions about reliability of an algorithm on average. </p>
<p>And yet paradoxically this is the approach on which research labs around the world depend to flex their algorithmic muscles. The academic peer-review process reinforces these inherited and <a href="https://www.researchgate.net/publication/225782608_Testing_Heuristics_We_Have_It_All_Wrong_Journal_of_Heuristics_11_33-42" title="Testing heuristics: We have it all wrong">rarely questioned</a> testing procedures.</p>
<p>A new algorithm is publishable if it’s better on average than existing algorithms on well-studied benchmark examples. If it’s not competitive in this way, it’s either hidden away from further peer-review scrutiny, or new examples are presented for which the algorithm looks useful.</p>
<p>In this way, a warm, flattering light is shone on each newly published algorithm, with little attempt to stress-test its strengths and weaknesses, and present it warts and all. It’s the computer science version of <a href="https://theconversation.com/register-all-trials-report-all-results-its-long-overdue-11603">medical researchers failing to publish the full results of clinical trials</a>.</p>
<p>As algorithmic trust becomes more crucial, we urgently need to update this methodology to scrutinise whether the chosen test examples are fit for purpose. So far, researchers have been held back from more rigorous analysis by the lack of suitable tools.</p>
<h2>We’ve built a better stress-test</h2>
<p>After more than a decade of research, my team has launched a new online algorithm analysis tool called <a href="https://matilda.unimelb.edu.au">MATILDA</a>: Melbourne Algorithm Test Instance Library with Data Analytics.</p>
<p>It helps stress-test algorithms more rigorously by creating powerful visualisations of a problem, showing all scenarios or examples an algorithm should consider for comprehensive testing.</p>
<p>MATILDA identifies each algorithm’s unique strengths and weaknesses, recommending which of the available algorithms to use under different scenarios and why. </p>
<p>For example, if recent rain has turned unsealed roads into mud, some “shortest-path” algorithms may be unreliable unless they can anticipate the likely impact of weather on travel times when advising the quickest route. Unless developers test such scenarios they’ll never know about such weaknesses until it is too late and we are stuck in the mud.</p>
<p>MATILDA helps us see the diversity and comprehensiveness of benchmarks, and where new test examples should be designed to fill every nook and cranny of the possible space in which the algorithm could be asked to operate.</p>
<p>The image below shows a diverse set of scenarios (dots) for a Google Maps type of problem. Each scenario varies conditions - like the origin and destination locations, the available road network, weather conditions, travel times on various roads - and all this information is mathematically captured and summarised by each scenario’s two-dimensional coordinates in the space.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=524&fit=crop&dpr=1 600w, https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=524&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=524&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=658&fit=crop&dpr=1 754w, https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=658&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/283069/original/file-20190708-51253-duqja2.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=658&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">A Google-maps-type problem with diverse test scenarios as dots: Algorithm B (red) is best on average, but Algorithm A (green) is better in many cases.</span>
<span class="attribution"><a class="source" href="https://matilda.unimelb.edu.au/matilda/">MATILDA</a>, <span class="license">Author provided</span></span>
</figcaption>
</figure>
<p>Two algorithms are compared (red and green) to see which can find the shortest route. Each algorithm is proven to be best (or shown to be unreliable) in different regions depending on how it performs on these tested scenarios.</p>
<p>We can also take a good guess at which algorithm is likely to be best for the missing scenarios (gaps) we haven’t yet tested. </p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/consumer-watchdog-calls-for-new-measures-to-combat-facebook-and-googles-digital-dominance-120077">Consumer watchdog calls for new measures to combat Facebook and Google's digital dominance</a>
</strong>
</em>
</p>
<hr>
<p>The mathematics behind MATILDA helps to create this visualisation, by analysing algorithm reliability data from test scenarios, and finding a way to see the patterns easily. </p>
<p>The insights and explanations mean we can choose the best algorithm for the problem at hand, rather than crossing our fingers and hoping we can trust the algorithm that performs best on average.</p>
<p>By rigorously stress-testing algorithms in this way – warts and all – we should reduce the risk of rogue algorithm decisions, securing the trust of Mr/Ms Average, and perhaps even the most sceptical humans.</p><img src="https://counter.theconversation.com/content/118830/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Kate Smith-Miles receives funding from the Australian Research Council as a Georgina Sweet Australian Laureate Fellow, and as a Chief Investigator in the Australian Centre of Excellence in Mathematical and Statistical Frontiers (ACEMS). She is Immediate Past President of the Australian Mathematical Society, and a member of the Australian Research Council's College of Experts.</span></em></p>Algorithms are only human (well, designed by humans) but we need to trust they’ll do what they’re supposed to do. And that means we need a better way to test them.Kate Smith-Miles, Professor of Applied Mathematics, ARC Laureate Fellow, Chief Investigator in the Australian Centre of Excellence in Mathematical and Statistical Frontiers, The University of MelbourneLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/1063862018-11-12T22:06:27Z2018-11-12T22:06:27ZEducation does not always equal social mobility<figure><img src="https://images.theconversation.com/files/244012/original/file-20181105-74783-1h7gtbu.jpg?ixlib=rb-1.1.0&rect=0%2C455%2C5070%2C2261&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">Some countries seem to provide more equitable opportunities in schools and society in general. Others have work to do if they want to advance the adage that hard work and education afford success regardless of one's existing social status. </span> <span class="attribution"><a class="source" href="https://www.shutterstock.com/image-photo/income-inequality-distribution-276410669?src=ZuHwG833Ir3iGSxCfuZcTg-1-56">www.shutterstock.com </a></span></figcaption></figure><p>Educators around the world, particularly those in secondary schools, often default to a compelling story when they are trying to motivate their students: Work hard, achieve well and you will secure a successful future with attractive job prospects. </p>
<p>This is currently the conventional wisdom across much of the Western world, with strong links drawn between education, meritocracy and upward social mobility. </p>
<p>But what does the research suggest about intergenerational mobility? Do children from poorer backgrounds have the same potential to realize their dreams if they achieve high standards in their education systems? </p>
<p>In fact, education is important but not enough to change inequities around the world. Intergenerational mobility, referring to changes in social status for different generations in the same family, is far from normal. </p>
<h2>The American dream in Denmark</h2>
<p>Public health researchers <a href="https://www.equalitytrust.org.uk/resources/the-spirit-level">Richard Wilkinson and Kate Pickett argued</a> outcomes in social mobility and education are significantly worse in rich countries with more inequality, that is, with populations that show larger gaps between the wealthy and the poor. For example, the United States and United Kingdom have close associations between fathers’ and sons’ incomes, compared to countries such as Denmark, Finland, Sweden and Norway. </p>
<p>Wilkson went so far as to jokingly comment in a TED talk “if Americans want to live the American dream, they should go to Denmark.” </p>
<figure>
<iframe width="440" height="260" src="https://www.youtube.com/embed/Ndh58GGCTQo?wmode=transparent&start=42" frameborder="0" allowfullscreen=""></iframe>
<figcaption><span class="caption">Richard Wilkinson says income means something very important within our societies.</span></figcaption>
</figure>
<h2>Great mobility?</h2>
<p>The relationship between national levels of income inequality and lower levels of intergenerational mobility is known as the Great Gatsby Curve. The Great Gatsby is the hero of the same-titled F. Scott Fitzgerald novel, who first appears as the enigmatic host of roaring parties in his waterfront mansion. Later, he is revealed as the son of poor farmers. The curve thus seeks to measure how much a person can move up in social class in a given society. </p>
<p>A <a href="https://doi.org/10.1093/sf/sov075">2015 study</a> used cross-national comparable data from <a href="http://www.oecd.org/skills/piaac/">the Programme for International Assessment of Adult Competencies (PIAAC)</a> to shed new light on the role of education in relation to this curve: the study examined the relationships between a person’s education, their parents’ education and labour-market outcomes such as income.</p>
<p>In countries such as Denmark, Finland, Norway, Sweden, Austria, Germany, Belgium and the Netherlands, the results suggested that parental education had little additional impact on a child’s income; it was the child’s level of education that mattered.</p>
<p>But in France, Japan, South Korea and the United Kingdom, the impact of parents’ education on their offspring was substantial. In these countries, the children whose parents came from a low education group earned 20 per cent less than children whose parents had higher levels of education, even though these individuals held the same level of qualification in the same subject area.</p>
<p>Collectively, this research suggests that a range of social mobility exists across different countries in relation to how much education a person gets. Equal education does not always mean equal opportunity. </p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/hard-evidence-do-grammar-schools-boost-social-mobility-28121">Hard Evidence: do grammar schools boost social mobility?</a>
</strong>
</em>
</p>
<hr>
<h2>Benchmark measures</h2>
<p>In a globalized economy, reliance on patronage and nepotism has little use. Rather, the global economy requires countries to maximize their human resources, regardless of the social status of particular individuals or groups, to remain competitive.</p>
<p>Not surprisingly, governments are increasingly concerned with addressing socioeconomic disadvantages within school systems so that they are able to maximize their nations’ human capital and promote intergenerational mobility. </p>
<p>Indeed, policymakers around the world have shown an affinity for the results of international benchmark measures such as PIAAC and <a href="http://www.oecd.org/pisa/">the Programme in International Student Assessment (PISA)</a>. They often rely on such measures to <a href="https://www.routledge.com/The-PISA-Effect-on-Global-Educational-Governance/Volante/p/book/9781138217416">assess the performance gaps</a> that exist among students of different socioeconomic backgrounds. </p>
<p>Ideally, countries strive for high performance and small achievement gaps, since the latter is a sign of an effective education system. Not surprisingly, some countries seem to be doing a better job at promoting better educational outcomes for students coming from lower socioeconomic groups. </p>
<p>For example, PISA 2015 results indicated that more than 30 per cent of economically disadvantaged students in Canada, Denmark, Estonia, Finland, Germany, Hong Kong, Ireland, Japan, Korea, the Netherlands, Norway, Singapore and Slovenia were considered “academically resilient.” This means <a href="https://www.oecd-ilibrary.org/docserver/66e037e8-en.pdf?expires=1541085901&id=id&accname=guest&checksum=10424661411DE01E7C99F68CAFF21A63">they performed at high levels despite coming from the bottom quarter of the socioeconomic status classification system</a>.</p>
<p>While the apparently better-performing countries may take pride in their outcomes, it is worth noting that a high global ranking does not necessarily capture how inequities manifest nationally. For example, Canada has a <a href="https://theconversation.com/why-there-are-so-few-indigenous-graduates-at-convocation-96782">noticeable gap between Indigenous and non-Indigenous education outcomes</a>.</p>
<h2>Policy for equality</h2>
<p>When one considers the capacity of education to influence social mobility around the world the results appear to be mixed. We need more research to understand exactly how some countries seem to provide more equitable opportunities in schools and society, and for whom. </p>
<p>Where there are disparities, governments need to consider more policy options across multiple sectors — to create a situation where equal abilities and qualifications translate to equal prospects and outcomes. Failure to do so casts doubt on our cherished notion of meritocracy. </p>
<p>In other words, in many countries education will only equal social mobility with further government intervention.</p><img src="https://counter.theconversation.com/content/106386/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Louis Volante receives funding from the Social Sciences and Humanities Research Council of Canada (SSHRC).</span></em></p><p class="fine-print"><em><span>John Jerrim does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>Conventional wisdom across much of the Western world says there’s a strong link between education and upward social mobility. Really?Louis Volante, Professor of Education, Brock UniversityJohn Jerrim, Lecturer in Economics and Social Statistics, UCLLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/869272017-11-10T05:50:14Z2017-11-10T05:50:14ZExplainer: what exactly is a living wage?<p>Australia’s national minimum wage should become a “living wage”, according to a <a href="https://www.theguardian.com/australia-news/2017/nov/02/unions-seek-dramatic-pay-increases-to-ensure-minimum-living-wage">new campaign</a> from the Australian Council of Trade Unions (ACTU). But what exactly is a living wage?</p>
<p>In theory, a living wage is no different to a minimum wage. Both set a binding “floor” on wages, below which no employee can (legally) be paid. But in practice there are several differences between minimum and living wages, in their value, purpose, and adjustment. </p>
<p>A living wage is set higher than a minimum wage and may be “pegged” to (fixed as a percentage of) some other measure of living standards, such as average weekly earnings. This ensures that the living wage holds its relative value over time.</p>
<p>Essentially, while the minimum wage sets a <em>bare minimum</em>, the living wage aspires to be a <em>socially acceptable minimum</em>. Typically, this is seen as a level that keeps workers out of poverty.</p>
<p>But the point at which workers fall into poverty varies widely, due to differences in family responsibilities, and complex interactions between low wages and welfare payments. These factors necessarily affect how the level of the living wage would be set and adjusted.</p>
<p>The idea to shift to a living wage follows a string of bad news about pay. Many vulnerable workers <a href="http://fbe.unimelb.edu.au/home/newsroom/the-gig-economy-and-migrant-workers-changing-the-face-of-work-2017-foenander-lecture">have been denied</a> their minimum entitlements by employers. Wage growth is so slow that even the Reserve Bank Governor has <a href="https://www.theguardian.com/australia-news/2017/jun/19/rise-up-and-demand-pay-increases-reserve-bank-chief-urges-workers">encouraged</a> workers to demand pay increases. And workers are <a href="https://blogs.imf.org/2017/04/12/drivers-of-declining-labor-share-of-income/">getting less</a> of the national income, as capital owners increase their share.</p>
<h2>Living vs. minimum wages</h2>
<p>Australia’s national minimum wage is set each year by an expert panel of the Fair Work Commission (FWC). The panel <a href="https://www.fwc.gov.au/awards-and-agreements/minimum-wages-conditions/annual-wage-reviews">receives submissions</a> from a wide range of organisations and <a href="https://www.fwc.gov.au/resources/research/annual-wage-review-research/previous-research">conducts research</a> to inform its decisions. </p>
<p>Increases to the minimum wage are based on <a href="http://www8.austlii.edu.au/cgi-bin/viewdoc/au/legis/cth/consol_act/fwa2009114/s284.html">objectives</a> enshrined in law. These refer to different factors, including business competitiveness, employment growth, and the needs of the low paid. There is no specific mention of poverty in the current objectives. Nor is there a fixed relationship with any other measure of living standards. </p>
<p>In other countries, minimum wages and living wages co-exist. In the United States, <a href="http://www.epi.org/minimum-wage-tracker/">long periods can pass</a> without increases in the federal minimum wage, as there is no mechanism for its regular adjustment. This has led many local governments to set their own mandatory <a href="http://laborcenter.berkeley.edu/minimum-wage-living-wage-resources/inventory-of-us-city-and-county-minimum-wage-ordinances/">living wage ordinances</a>, above the federal (and state-level) minimum wages. </p>
<p>The situation is different in the United Kingdom, where the <a href="https://www.gov.uk/government/organisations/low-pay-commission">Low Pay Commission</a> recommends a national minimum wage increase each year. Even there, the movement for a voluntary “<a href="https://www.livingwage.org.uk/what-real-living-wage">real living wage</a>” has strong support from employers.</p>
<p>If the ACTU plan became law, Australia’s living wage would differ from the US and UK models. It would <em>replace</em>, rather than complement, our national minimum wage, substantially raising the wage floor. This would require the FWC’s expert panel to have different wage-setting objectives, with its primary goal being to eliminate working poverty.</p>
<h2>Would a living wage help the poor?</h2>
<p>Regrettably, poverty is the reality for many of Australia’s lowest-paid workers. Some struggle to make ends meet and go without basic necessities, such as meals and heating - particularly those <a href="http://journals.sagepub.com/doi/abs/10.1177/0022185610397138">in single-income families</a>.</p>
<p>Neither our current minimum wage, nor the proposed living wage, is a pure “anti-poverty” tool. This is because the poorest people <a href="http://www.natsem.canberra.edu.au/storage/Poverty-Social-Exclusion-and-Disadvantage.pdf">do not have paid jobs</a> - often due to serious socioeconomic disadvantage. A living wage only helps those who rely on paid work (their own or someone else’s) for an income.</p>
<p>The intention of a living wage is therefore not to eradicate all poverty, but to end poverty among those who work - “the working poor”.</p>
<p>This laudable ambition is complicated by differences in personal and family circumstances. A living wage cannot vary from person to person, yet low-paid workers <a href="https://www.fwc.gov.au/sites/afpc2006wagereview/research/AFPCResearchSeries0406AnUpdatedProfileontheMWWorkforceinAus.pdf">are not all alike</a>: some live alone, some have children, and <a href="http://onlinelibrary.wiley.com/doi/10.1111/j.1475-4932.2007.00432.x/abstract">many are in</a> dual-income families.</p>
<p>Who should a living wage be set for? The income needed to prevent poverty is inevitably much higher for workers with families than for those who live alone. </p>
<p>The Social Policy Research Centre (SPRC) produces “<a href="http://apo.org.au/node/103781">budget standards</a>” that show the minimum income required by different types of families to reach a healthy living standard. Their evidence has been widely used by <a href="https://www.fwc.gov.au/documents/sites/wagereview2010/downloads/healy_phd_thesis.pdf">the ACTU</a> and <a href="http://www.acoss.org.au/wp-content/uploads/2017/03/ACOSS-minimum-wage-submission-2017_FINAL.pdf">other advocacy groups</a> in submissions to the Fair Work Commission.</p>
<p>According to their analysis, an employed single adult <a href="http://unsworks.unsw.edu.au/fapi/datastream/unsworks:46141/binaaacbcf3-915f-40bc-a70f-2052746ab643?view=true">currently needs</a> A$597 per week (before tax, and including housing costs) to live healthily. A couple with two young children needs almost twice as much: A$1,173. </p>
<p>The national minimum wage <a href="https://www.fwc.gov.au/documents/decisionssigned/html/2017fwcfb3501.htm">is currently</a> A$695 for a full-time worker. So, according to the SPRC’s research, that worker already earns enough for a healthy life if they live alone, but not nearly enough if they have a family. This highlights the difficulty of setting a single living wage that would universally prevent working poverty.</p>
<p>Families with children also receive other government assistance through targeted welfare payments. This further complicates the task of setting a living wage.</p>
<h2>What are the alternatives?</h2>
<p>There are other ways to tackle working poverty. In the US, an “earned income tax credit” reduces the taxes of low-paid workers, so their wages stretch further. Such a scheme <a href="https://openresearch-repository.anu.edu.au/bitstream/1885/40576/3/DP450.pdf">has been recommended</a> for Australia.</p>
<p>Another very different approach to welfare is a universal basic income (UBI). This would provide a guaranteed minimum income, regardless of whether someone works, and without eligibility tests like those behind Centrelink’s recent <a href="https://www.theguardian.com/australia-news/2017/apr/10/centrelink-debt-scandal-report-reveals-multiple-failures-in-welfare-system">“robo-debt” debacle</a>. </p>
<p>Supporters of UBI also <a href="https://www.weforum.org/agenda/2017/01/why-we-should-all-have-a-basic-income/">see it as</a> a solution to job losses caused by rapid automation.</p>
<p>Living wages and UBI are radically different ways of tackling poverty. Work remains vital for a living wage, but is optional for a UBI. A living wage would raise the value of paid work, but might make life harder for some jobseekers whose labour becomes more expensive. A UBI would provide income <em>without</em> work, which might encourage more people to drop out of the labour force altogether. </p>
<p>In pushing to “make work pay”, the ACTU is hoping to capture both the public imagination and, for workers, a larger slice of the economic pie.</p><img src="https://counter.theconversation.com/content/86927/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>The authors do not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.</span></em></p>The ACTU has proposed Australia adopt a “living wage”. This might improve the incomes of some people, but it wouldn’t solve “working poverty”.Josh Healy, Senior Research Fellow, Centre for Workplace Leadership, The University of MelbourneAndreas Pekarek, Lecturer in Management, The University of MelbourneLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/833662017-09-05T19:39:27Z2017-09-05T19:39:27Z‘Indicator frenzy’ : the ‘economicist’ tendency of public policy and alternative indicators<p>The craze for quantitative data has become a hallmark of public policy on both national and local levels. Exploring the limits of quantification allows us to better understand the importance of developing alternative indicators.</p>
<p>Economic indicators currently play a key role in the elaboration of public policy and undoubtedly play an important role in <a href="https://www.canal-u.tv/video/meshs/statistiques_societe_et_decision.13203">government decision-making</a>. This is easily understandable because the quantification of social phenomena has many advantages. In addition to facilitating comparison and counteracting doubt, indicators are tools of comprehension and the objectification of “realities”. They are also used to compare and rank territories; subsequently, we could consider them as <a href="http://onlinelibrary.wiley.com/doi/10.1111/j.1540-5893.2012.00473.x/abstract">“technologies of governance”</a>. </p>
<p>The scientific value of indicators is also undeniable, but it is important to remember that they are social constructions. As Eva Mineur points out in her PhD thesis, <a href="http://www.diva-portal.org/smash/record.jsf?pid=diva2%3A141019&dswid=9015">“Towards sustainable development: Indicators as a tool of local governance”</a> (2007):</p>
<blockquote>
<p>“Indicators are loaded with values in that they include and exclude certain characteristics of what is to be indicated and measured”. </p>
</blockquote>
<p>Caution is therefore necessary when using indicators. We can observe that the use of biased, large-scale indicators within the <a href="http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780198289029.001.0001/acprof-9780198289029">New Public Management framework</a> and <a href="https://vimeo.com/87556137">public-policy benchmarking practices</a> can induce authorities and local actors to pay more attention to the indicators than the actual goals. </p>
<p>While indicators can enlighten stakeholders, the lack of balance in the measurement process may contribute to rampant <a href="https://www.jstor.org/stable/3708368">“quantophrenia”</a>, in the words of Pitirim Sorokin. In other words, <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2645155/">“indicator madness”</a> can be observed, where the increase or decrease of the indicator’s values becomes an end in itself.</p>
<h2>And what if means become ends?</h2>
<p>Benchmarking practices highlight the dominance of economic indicators through the emergence of “economicist” responses to public policy making, to the detriment of other issues requiring more qualitative assessment methods, such as <a href="http://ec.europa.eu/eurostat/documents/118025/118123/Fitoussi+Commission+report">well-being, quality of life or social and environmental sustainability</a>.</p>
<p>This trend also substantially reinforces the <a href="http://www.uv.es/ecoinst/ipe/toboso-static94.pdf">“objective-technocratic stance”</a>, according to which “reality” can be directly measured, and ultimately results in denying the socially and technically designed aspect of the indicators. Cloaked in the language of objectivity, indicators are associated with the reification and naturalization of social phenomena. The performative power of indicators thus <a href="https://www.sociology.northwestern.edu/documents/faculty-docs/rankings-and-reactivity-2007.pdf">tends to increase</a> without researchers and professionals being fully aware of their influence, which in turn occults the <a href="https://www.researchgate.net/publication/253831647_Surveys_versus_administrative_records_reflections_on_the_duality_of_statistical_sources">quantification conventions underlying their design</a>. In this context, indicators not only reflect certain “realities”, but they also tend to establish them.</p>
<p>Indicators also appear as “instruments of domination”: They reflect the balance of power, and a certain way of viewing the rationalisation of public policy. In fact, indicators currently tend to be the instruments of a narrow rationalisation method, and are characterised by a strong “economicist” dimension. The surge of performance indicators also has tendency to distort actors’ logic.</p>
<h2>Regional indicators in France</h2>
<p>To a lesser extent, this “indicator frenzy”, prevalent at the national level, is also found at the regional level in France.</p>
<p>The emergence of new demands for statistical information on the subnational scale can be explained by a number of factors: the increased power of local authorities following the decentralization process, the development of contract-based links between the various levels of action, the evaluation of public policies, the new spatial dimension of public action and the transformation of the public statistical system (big data, open data, decentralization of data production, etc.).</p>
<p>The regional administration of France was reformed in 1992 and 2015, contributing to the development of institutions for intermunicipal cooperation, pooling the expertise of existing municipal bodies and giving rise to more localized observation processes. The interdependence of various government authorities is linked to increased contractual operations in public policy, and, more broadly to the surge of neoliberal ideology and thus budget cuts.</p>
<p>The neoliberal approach is implemented at the local level via the imperative of the <a href="http://onlinelibrary.wiley.com/doi/10.1111/j.0435-3684.2000.00074.x/abstract">territory’s attractiveness and competitiveness</a>, to the detriment of human and environmental factors. Accordingly, attractiveness, as the concept is <a href="http://journals.sagepub.com/doi/abs/10.1177/0969776411430289">currently understood</a>, does not reflect a territory’s ability to provide everyone with the opportunity to lead a “good life”. Most of the time, attractiveness is viewed through the prism of economics. The French Regional Action Service Centres, which provide measures of attractiveness, focus on the supply and not on the needs of population. The indicators of attractiveness are aligned with a certain idea of economic growth, namely the setting up of companies in the territory, or traditional employment indicators (unemployment rate, job creation in the region, etc.). It’s a concept that is closely related to the <a href="http://onlinelibrary.wiley.com/doi/10.1111/j.0435-3684.2000.00074.x/abstract">competitiveness imperatives</a>:</p>
<blockquote>
<p>“To attract and retain capital within its territory is a means to maintain and strengthen competitiveness in the global economy”.</p>
</blockquote>
<p>Furthermore, the goal is demographic growth, but not just any growth. The territories’ residents are only considered as human resources from the point when they are active and educated, and subsequently become a potential source of wealth creation. The neoliberal city, based on neoliberal planning, leads the definition of <a href="https://www.researchgate.net/publication/226661039_Neoliberal_Planning_Does_It_Really_Exist">“everything and everybody as either economic gain or loss”</a>. Consequently, many social and environmental resources are missing from current observations and planning.</p>
<p>Based on this conception of urban governance, economic issues have a priority over human ones. Competition prevails over cooperation, economic violence over <a href="https://theconversation.com/le-corps-et-lame-de-la-paix-economique-72936">economic peace</a>. And there are invisibles costs we pay beyond those represented by money.</p>
<h2>New measures of neglected phenomena</h2>
<p>A change in approach is possible, however: indicators can be used to uncover neglected phenomena and thus play the role of protest instruments. Examples include <a href="http://footprint.wwf.org.uk/">environmental footprint</a> and the French <a href="http://www.bip40.org/">Inequality and Poverty Survey</a> (BIP40). In this context, indicators are not fated to become instruments of an “economicist” rationalisation method. This approach is close to that of <a href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2466882">“stat-activism”</a>, a term used by <a href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2466882">Isabelle Bruno, Emmanuel Didier and Tommaso Vitale</a> that must be </p>
<blockquote>
<p>“understood both as a slogan to brandish in struggles and as a descriptive concept, which is used to quantify experiences aimed at re-appropriating the emancipating power of statistics”.</p>
</blockquote>
<p><a href="https://www.cairn.info/load_pdf.php?ID_ARTICLE=RERU_153_0533">Several experiments in France</a> have employed this approach, including indicators for <a href="https://link.springer.com/chapter/10.1007%2F978-94-007-6501-6_2">social health</a> in the Nord-Pas-de-Calais region, <a href="http://www.theses.fr/2015GREAE003">sustainability</a> in the Grenoble area and <a href="http://www.oecd.org/site/progresskorea/44109891.pdf">well-being</a> in Brittany. Similar initiatives have been conducted in other countries, including an early one in <a href="http://jcciweb.wixsite.com/jcci/indicators">Jacksonville, Florida</a> in the United States, and another in <a href="https://www.sfu.ca/%7Emholden/documents/Urbanindicatorsandtheintegrativeideals.pdf">Vancouver</a>, Canada. All these projects show the importance of the democratic re-appropriation of indicators.</p><img src="https://counter.theconversation.com/content/83366/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Les auteurs ne travaillent pas, ne conseillent pas, ne possèdent pas de parts, ne reçoivent pas de fonds d'une organisation qui pourrait tirer profit de cet article, et n'ont déclaré aucune autre affiliation que leur organisme de recherche.</span></em></p>The craze for measurement has become a hallmark of local and national public policy. Exploring the limits of quantification allows us to understand the advantages of developing alternative indicators.Fiona Ottaviani, Enseignante-chercheuse en économie - Chaire Mindfulness, Bien-être au travail et Paix économique - Political Economy and Sustainable Competitiveness Initiative, Grenoble École de Management (GEM)Anne Le Roy, Enseignante chercheuse en Economie au CREG à l'UGA, Université Grenoble Alpes (UGA)Licensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/784812017-06-05T16:37:50Z2017-06-05T16:37:50ZUniversity tests should be part and parcel of teaching - not stand-alone events<figure><img src="https://images.theconversation.com/files/171790/original/file-20170601-25658-1hvw188.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">
</span> <span class="attribution"><span class="source">Shutterstock</span></span></figcaption></figure><p>Standardised tests exist in education systems <a href="https://www.theatlantic.com/education/archive/2015/06/exams-around-the-world/395540/">all over the world</a>. South Africa is no exception – and its test results often make for depressing reading.</p>
<p>In 2014 education accounted for 6.2% of South Africa’s Gross Domestic Product and 19.1% of total <a href="http://www.treasury.gov.za/publications/igfr/2015/prov/03.%20Chapter%203%20-%20Education.pdf">government expenditure</a>. International benchmarking studies like the <a href="http://www.timss-sa.org.za/">Trends in International Mathematics and Science Study</a> and regional assessments such as the <a href="http://www.nbt.ac.za/">National Benchmark Tests</a> suggest this money is not being well spent. The level and quality of South African schools’ learning outcomes tend to be lower than those of countries that invest significantly less in their schooling sectors. For example, Singapore spends 3% of its GDP on education, yet <a href="http://www.channelnewsasia.com/news/singapore/singapore-tops-oecd-s-global-education-ranking-report-8278406">scores top</a> in Mathematics and Science among 76 countries.</p>
<p>This shouldn’t be surprising. South Africa’s National Benchmark Tests are little more than a screening system for university applicants. They only assess students’ ability to combine aspects of prior learning in a few competency areas. They don’t address the <a href="http://www.universityworldnews.com/article.php?story=20090816082047397">systemic problem</a> of poor learner performance. </p>
<p>This points to a much larger problem: namely, the deficiencies in South African schools’ and universities’ assessment practices. In our book <a href="https://www.peterlang.com/view/product/78618">Education, assessment and the desire for dissonance</a> we argue that assessment should not be considered as a one-directional activity that’s detached from teaching and learning. Rather, assessment needs to involve doing things with others, rather than for others. It has to be embedded with teaching and learning.</p>
<h2>Exam stress</h2>
<p>Assessment should be a part of teaching and learning at universities. It’s important because it will subvert exclusion and allow all students to take responsibility for their work. </p>
<p>The problem is with how this assessment is currently structured. As university lecturers we have experienced it in quite a negative way.</p>
<p>Our students – we teach educational theory and practice – are enthusiastic about their education, and committed to it. But most worry a great deal about exams, as if their learning can be appropriately ascertained just through their performance in these tests. This concern is compounded by the overwhelming emphasis many lecturers place on examination results. Exams <a href="http://www.nber.org/papers/w12199">are seen as</a> the ultimate corroboration of students’ understandings and insights.</p>
<p>This only happens because assessment is considered as that which “measures” learning: that is, assessment of learning. On top of that many of the formal opportunities students receive to consolidate and expand their insights are often erroneously connected to short pieces of work. These are mostly tests and assignments, according to which they are assessed. </p>
<p>At university, the repercussions of focusing on short bursts of measurable tasks often only present themselves when students opt to pursue postgraduate degrees. It’s not unusual, at the institution where we teach, to encounter students who access the master’s programme with remarkable results – since this is the criterion our institution uses to attract students in the first place. </p>
<p>These students come to us with particular understandings of their capacity to study and learn based on the results that we have given them.</p>
<p>Yet many of these “excellent” students battle at master’s level. Any attempt to extend the basics on which they had been assessed in the previous academic year, such as short pieces of work, is often met with serious ineptitude. They struggle to proffer their own perspectives or to articulate their own opinions. This is seemingly because they were not required to do so in previous programmes.</p>
<p>So neither the student nor the lecturer can exercise his or her governmentality. This is <a href="http://www.palgrave.com/us/book/9781403986528">a term</a> coined by French philosopher and social theorist Michel Foucault, which refers to the way in which the state exercises control. </p>
<p>The student doesn’t have the autonomy to bring what’s required to fruition, such as a research proposal. This affects the student’s responsibility and self-esteem. The teacher does not have the autonomy to undo what’s been put in place in the first instance: institutional sanctioning via results.</p>
<p>There seems to be a disconnect between teaching and learning and assessment.</p>
<h2>A new approach</h2>
<p>Assessment in university remains at odds with a plausible notion of assessment because it does not lend itself to responsible and esteemed pedagogical action: that is, teaching and learning. For this shift to happen a different understanding of assessment is needed; one that transcends the notions of assessment <em>of</em> learning and assessment <em>for</em> learning.</p>
<p>We consider assessment as embedded in teaching and learning. It needs to unfold <em>within</em> teaching and learning. This implies that lecturers and students give an account of their actions to one another by making assertions, expressing doubts through questioning and indicating their desires by making particular requests. </p>
<p>As it stands now, you teach a particular concept and then assess whether a student has understood and learnt by posing particular questions. </p>
<p>What we’re suggesting, quite differently, is that assessment for learning occurs while the teaching unfolds. A student is able to give their own initial thoughts on a particular concept while it’s being taught. Students collaborate with their teacher to reach particular understandings. Teaching and learning takes on a deliberative rather than an instructional form.</p>
<p>In this space both lecturers and students are in a position to ask what is “good” and “best” in the enactment of their learning desires. They can do things differently so that assessment enhances rather than regulating teaching and learning. That is, without entirely doing away with assessment, start using assessment for learning as well.</p><img src="https://counter.theconversation.com/content/78481/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Nuraan Davids receives funding from NRF. </span></em></p><p class="fine-print"><em><span>Yusef Waghid does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>Assessment should be a part of teaching and learning at universities. It’s important because it will subvert exclusion and allow all students to take responsibility for their work.Nuraan Davids, Senior Lecturer in Philosophy of Education, Stellenbosch UniversityYusef Waghid, Distinguished Professor of Philosophy of Education, Stellenbosch UniversityLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/528342016-01-06T15:15:44Z2016-01-06T15:15:44ZSouth Africa’s universities may not be ready for the latest crop of school leavers<figure><img src="https://images.theconversation.com/files/107378/original/image-20160106-29944-1ujzfdy.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">Prospective students rush the gates of the University of Johannesburg during a deadly 2012 stampede. Are South Africa's universities ready for the latest crop of matriculants?</span> <span class="attribution"><span class="source">Adrian De Kock/EPA </span></span></figcaption></figure><p>When South Africa’s annual matric results are announced, it is important to listen to what is said - and, more importantly, to what is not. </p>
<p>Basic Education Minister Angie Motshekga offered a sober account of the 2015 results, particularly about the <a href="http://mg.co.za/article/2016-01-05-matric-pass-rate-drops-to-707">significant drop</a> in pass rate from 75% in 2014 to 70% in 2015. There were some reasons to celebrate: the ministry’s interventions to support “progressed learners” are to be commended. These are pupils who were pushed through from Grade 11 to matric even though their results didn’t qualify them to progress.</p>
<p>Nearly 40% of these learners passed and a not insignificant number obtained Bachelor’s passes, which means they can now apply for university study, and many were awarded distinctions, or “A” symbols. This suggests that the investment has been worthwhile. Many of these learners were given the opportunity to show their true potential.</p>
<p>But what about the things Motshekga didn’t say? In the higher education sector, we are interested in what the results say about the extent to which the 2015 matriculants are prepared for university study. To answer this question, one needs to look a little deeper.</p>
<h2>Methods of measurement</h2>
<p>There are two ways to measure admissions into and preparedness for university in South Africa. One is the National Senior Certificate, the other the National <a href="http://www.nbt.ac.za/node/89">Benchmark Tests</a>. The benchmark tests are taken by nearly 80,000 university applicants across the country each year. Together these measures offer complementary data for universities’ admissions decisions. The senior certificate is an indicator of learners’ past school achievement in a range of subjects. The National Benchmark Tests are an indicator of likely future performance against more general competencies, such as academic literacy, quantitative literacy and mathematics. </p>
<p>Not all university programmes require Maths and Science, but these subjects are good indicators of university preparedness. They are also essential for entry into programmes such as Science, Commerce, Engineering and Health Science and some specialisations in the Social Sciences arena. Motshekga said during the results announcement that the number of passes in Maths and Science increased between 2014 and 2015. She qualified this by noting, though, that the percentage pass had dropped. </p>
<p>A technical report is always released alongside the official results announcement. The detail about Maths and Sciences passes in <a href="http://www.education.gov.za/LinkClick.aspx?fileticket=JgmMMi8JtFI%3d&tabid=358&mid=1325">this report</a> is both interesting and important:</p>
<p>While the number of students enrolled for each of these subjects increased, the overall percentage of students enrolled from these subjects remained the same across the two years: 33% of learners in both 2014 and 2015 wrote Mathematics and 24% learners wrote Physical Science. In comparison the percentage of learners writing Mathematics Literacy increased from 46% to 49% between the two years. Mathematics Literacy is a subject introduced in 2006 and typically perceived to be <a href="http://www.bdlive.co.za/national/education/2014/07/24/department-prefers-pupils-to-take-maths-rather-than-maths-literacy">less academically challenging</a> than Mathematics.</p>
<p>In terms of performance, while <a href="http://www.education.gov.za/LinkClick.aspx?fileticket=JgmMMi8JtFI%3d&tabid=358&mid=1325">more learners</a> passed in each of these subjects the percentage of passes for Mathematics is down: in 2015 49% of those writing scored 30% and above, compared to 53% who scored the same in 2014. This means that of the 33% of learners who wrote Mathematics in 2015, only half scored above 30%, which constitutes a pass. Of the approximately 25% who wrote Physical Science as a subject, 59% scored 30% and above, down from 61% in 2014. </p>
<p>Overall these results are of concern for learners who want to go to university and study a science-related subject.</p>
<h2>A sobering picture</h2>
<p>The National Benchmark Test results for the 2015 writers provide an equally sobering picture of performance in these key competency areas. Pupils are scored against three benchmarks: basic, intermediate and proficient. These benchmarks are set by disciplinary experts around the country who represent most of South Africa’s universities.</p>
<p>‘Basic’ predicts that students will face serious challenges and require extensive and long term support. ‘Intermediate’ predicts that students will face some challenges and require some support. ‘Proficient’ predicts that students will be able to cope with regular programme of study. Here’s what the 2015 test results tell us:</p>
<ul>
<li><p>45% of those writing the Maths benchmark test scored ‘basic’. This means that they will have serious challenges with university level Mathematics. </p></li>
<li><p>Only 10% scored ‘proficient’, meaning that they would be expected to cope with regular mainstream provision. This provides strong evidence that, despite all the achievements of the Department of Basic Education, the overwhelming majority of matriculants are not prepared for university level study in science-based fields. </p></li>
</ul>
<h2>Are universities ready?</h2>
<p>There is a flip side to the issue of student preparedness: are universities ready for these incoming matriculants? The data I’ve outlined above provides a compelling body of evidence for universities to revisit their assumptions about what constitutes entry level readiness.</p>
<p>This requires a hard and critical look at the undergraduate curriculum. For instance, the University of Johannesburg is <a href="http://www.enca.com/south-africa/uj-ready-enrol-10500-new-students">preparing</a> to enrol 10,500 new students in 2016. To what extent does it have the necessary Maths and Science curriculum in place at entry level to cater for the diversity of educational preparedness among those new students? </p>
<p>Some will need no support. But the majority will need extensive academic guidance and help. </p>
<p>The reality is that until there is significant reform of South Africa’s undergraduate curriculum, with a particular focus on Mathematics and Science, universities will remain unprepared - even for those who made it through matric, obtained Bachelor’s passes and are, in Motshekga’s own words, “the best of the best” that the schooling system has to offer.</p><img src="https://counter.theconversation.com/content/52834/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Suellen Shay receives funding from the National Research Foundation and from Department of Higher Education and Training (both are research funding). </span></em></p>South Africa’s matric results and data from national benchmarking tests suggest that many school leavers aren’t ready for university. It’s also worth asking: are universities ready for them?Suellen Shay, Dean and Associate Professor, University of Cape TownLicensed as Creative Commons – attribution, no derivatives.