Monday

How well are the people of the world speaking English?

I recently attended the launch of the EF English Proficiency Index 2013 (full PDF report here) in Montevideo (left), where participants were told that Uruguay is still in the 'low proficiency' bracket and has dropped in points (from 53.42 to 51.49) since the EPI was last published in 2012. It was not included in the first EF EPI published in 2011.

The 2013 edition of the EFP ranks 60 countries and territories by adult English proficiency, was produced with data collected from 750,000 test-takers and is a report that attempts to rank average level of English skills amongst adults (Wikipedia).  Most of the data is collected by EF from the free placement test taken by people on their website.  As such, the report has been criticised for a lack of representative sampling in each country. As such, this makes the EPI an online survey rather than a statistically valid evaluation (Wikipedia). When we asked about the data at the launch in Montevideo, EF were open about this and details about data collection is also clearly mentioned in the report on page 42.

Despite the above, it has been very interesting to see just how much attention is being paid to the results. In Uruguay, the 'decline in English' in the country stated in the report made the major newspapers and the same seems to be true in many other places - it's been featured in the New York Times, the World Bulletin, the Bangkok Post, the Swedish Wire, and the Wall Street Journal, and I'm sure many other periodicals, newspapers and websites. It's all fabulous marketing for EF, who have found a way of delivering information that the world wants to know, and which no other organisation is trying to provide. 

The AI learning paradox

On his substack, Jason Gulya outlines a paradox: "Learning with AI tools suffers from a paradox. To use AI as an effective tool, learn...