The sample report seems to confirm that score reports will include the new 1-6 score scale as well as a legacy 1-120 overall score… but no legacy 1-30 scores for each individual skill. I suppose those could be included elsewhere, but it doesn’t really LOOK like that is the case.
The challenge here is that some key stakeholders (including a majority of Russell Group schools) haven’t yet updated their cut scores for individual skills. American and Canadian schools don’t generally institute section score requirements, so they have the luxury of dilly dallying for two years thanks to the inclusion of an overall legacy score. Schools in the UK, which usually require specific section scores, do not.
But it isn’t just schools that need to get moving. Indeed, quite a few foreign trained pharmacists I’ve talked to are a bit anxious about the fact that the National Association of Boards of Pharmacy hasn’t yet issued new score requirements.
The University of British Columbia now fully accepts the Duolingo English Test for undergraduate admissions. Previously, the school’s website still mentioned that it was acceptable only in cases where no other test could be taken.
I think Duolingo has now achieved near-universal acceptance among undergraduate programs in Canada.
There is an excellent new article on the IELTS website which notes how “[r]esearch shows that the average minimum entry score is an IELTS band 6.6 across countries, below our expert’s recommended levels.” It says “[w]hen minimum English language scores are set too low, the consequences can be far-reaching.”
We sometimes hear university staff complaining about how new international students don’t have the linguistic competency needed to excel in their new environment. But what do they expect when their institution requires only an IELTS score of 6.5? Or even a score of 7.5?
The people setting cut scores at universities aren’t wholly to blame when those scores are poorly set. It’s a tough thing to do, and they may not have enough information to work with. The aforementioned article links to the IELTS writing band descriptors. But while those are detailed, their usefulness is somewhat limited. For instance, here’s what they say about the grammar in a band seven essay:
“A variety of complex structures is used with some flexibility and accuracy. Grammar and punctuation are generally well controlled, and error-free sentences are frequent. A few errors in grammar may persist, but these do not impede communication.”
It’s a start, I guess.
If readers of the article dig two links deeper, they’ll find some sample essays with band scores. But there are only a dozen of them. And, curiously, they are all from the paper edition of the test, and barely legible.
As scrutiny of tests increases, it is more imperative than ever that test makers provide access to whatever it is that test takers produce during the test. Technology has caught up with the needs of decision makers, and it isn’t terribly hard to let university staff read all of the IELTS essays produced by all of their applicants. Obviously they probably won’t use this information to make admissions decisions, but it will give them crucial information about how to set the best possible cut scores. They’ll finally know, perhaps, what an applicant who has submitted a score of 6.5 is capable of.
A handful of ancient samples from the test maker can only go so far. Scores can sometimes drift, inter-rater agreement isn’t always fantastic, the scoring criterions interact in curious ways that only become apparent when you’ve got a bunch of responses to look at… and there is always the niggling fear that the samples have been graded by item writers or test developers rather than actual raters. So… the more the better. And instead of talking about providing “more” essays, we should probably be talking about providing all the essays.
And we haven’t even talked about the speaking section.
It is worth mentioning that the TOEFL and Duolingo tests have provided sample responses along with every score report for ages. Other tests have as well.
“Thousands of migrants may have been given visas despite failing mandatory English language tests following a blunder over marking, The Telegraph has learnt.
Up to 80,000 people sitting a language test run by the British Council were given the wrong results, meaning many of them were given pass marks even though they had failed.
Separately, evidence of cheating has been discovered in China, Bangladesh and Vietnam, where criminals sell leaked test papers to migrants so that they know the answers in advance.
It means students, NHS workers and other migrants with a poor grasp of English have been given study or work visas to which they were not entitled.”
And also:
“Because it took so long to discover the problem, many people who were wrongly told they had passed would have been able to obtain visas and come to Britain legally.”
That’s all sort of true, though unmentioned is the fact that a majority of the people who got incorrect scores likely used them to head to Canada, Australia and other places. Or that many IELTS tests are used for domestic purposes. And the fact that the British Council runs the test in partnership with a few other parties.
In any case, it isn’t a great situation. And the IELTS partners could do better in terms of transparency. For the record, here (once again) are the questions I would ask the partners if I were a real journalist:
What exactly was the problem?
Exactly how many tests were affected by this problem? “Less than 1%” could mean anywhere from 1 to 70,000 tests.
Were any administrations from before August 2023 impacted by the problem? IELTS has contacted test takers who must be given new score reports. But what of test takers whose results have expired since they took the test? Did any of them receive incorrect scores?
Are there any instances where a test was incorrectly scored, but a band score change was not necessary (and, by extension, the test taker was not contacted)? If so, how many? And if so, were these included in the “less than 1%” figure given to score users?
The NABP (National Association of Boards of Pharmacy) held a taskforce to discuss their continued use of the TOEFL iBT as a requirement for foreign-trained pharmacists seeking to practice in the USA. I’ve attached a photo from their social media here just to show how this was a pretty big deal. This was clearly not a formality or a rubber stamp sort of thing.
Notably, the task force was called for before ETS announced changes to the test. One hopes that the changes were added to the agenda at some point. Indeed, the NABP’s current TOEFL cut scores (introduced only in October) will soon be obsolete. They’ll need to issue new ones… and pretty soon.
I’ll be keeping an eye on this, of course. I asked when we can expect to see the task force’s report, and was told it would be made public “in several months.”
According to friends in Tehran, IDP Education Ltd will stop offering the IELTS in Iran at the end of January. This means that it will no longer be possible to take the IELTS in Iran at all, as the British Council pulled out of the country in 2009.
As a result of a special license issued to ETS from the Office of Foreign Assets Control, the TOEFL remains an option for Iranian test takers.
While chatting with John Healy about the new test on Facebook (yes, there is still some chatter about English tests on Facebook), we got to talking about templates. I wrote:
“Here’s some interesting food for thought: After years of resisting, Pearson finally introduced human raters in order to kill off template usage in the PTE Test. Conversely, the new TOEFL will not make use of human raters for most responses. Some might argue that this will make templates even more effective than before.”
It is worth thinking about. Without human raters checking every single response, there is the chance that some test takers seek to game the test with heavy duty templates. This was the story of the PTE test for many years… until humans started checking almost every task.
If anyone here has the ear of the IELTS partners, it is worth asking if they still plan to bid on the HOELT as a group. It’s likely that they will do so – a bid that brings a sort of “white label IELTS” to the table would definitely have a lot of gravitas and psychometric heft. That said, given that test center delivery doesn’t seem to be a factor anymore, the partners suddenly have less need for each other. Any one of them could easily fulfill both the delivery and development requirements on their own. In the early days of the HOELT we talked a lot about how the massive test center networks of IDP Education and the British Council were huge advantages, but obviously they don’t matter any more.
If you are interested in the possibilities, do take a moment to explore some of the tests that the partners currently develop and deliver on their own. Cambridge University Press & Assessment, for instance, has a very nice test called LinguaSkill. I’d say that it leans “traditional” (if you will forgive the term). It is delivered digitally at test centers but could probably be adapted to a remote format. The British Council, meanwhile, runs the Aptis test which is also pretty traditional. This product is mostly taken at test centers, but a remote version has been available for some time. IDP Education Ltd has a test called Envoy. It is mostly unknown at the moment but has a lot of modern features (adaptive content, 90 minute duration, wholly AI scoring, scores in two hours, etc) that could help it stand out from the crowd. I think proctoring of the Envoy test is async, but I couldn’t confirm that by looking at the website today.
Now, regular readers know that the IELTS partnership has endured for 36 years without any hint of disagreement or conflict between the partners. So I’m pretty sure that a joint bid will take advantage of that history of positive cooperation. But, as I said, it is worth asking for clarification.
It should also be mentioned that the winning test is unlikely to be (strictly speaking) the IELTS test, the Duolingo English Test, the PTE Test, or whatever. It will build on the research and designs of one of those tests, but will be adapted to meet the requirements of the Home Office.
I took this photo at the Youngpoong Bookstore at Mario Outlet in Seoul! As you can see, Hackers Education Group is first out the door with updated TOEFL books in Korea, and they even scored a nice little table display at the front of the store’s test prep section. Basic editions of the books are also available, round back.
Despite its somewhat silly name, Hackers is the biggest publisher of prep books for English tests in Korea. They also publish some of the very best books in this category. They hire some of the very best test prep minds in the country as staff writers.
When ETS held a gathering to promote the revised TOEFL in Seoul a few weeks ago, representatives from Hackers were on the guest list. That’s noteworthy because according to prosecutors here in Korea, between 2007 and 2012 Hackers (allegedly) sent staff to (allegedly) copy questions from administrations of ETS tests using hidden cameras and microphones. The chairman of Hackers, allegedly surnamed Cho, received an 18 month prison sentence for his part in the alleged scheme. The sentence was later suspended.
Press coverage at the time indicated that ETS would be seeking compensatory payments to cover the cost of “Korea only” versions of tests it created in response to these alleged activities, but I don’t know if they ever went ahead with that.
Anyway. I don’t know if the Hackers invite is a sign that the chilly relationship between these two parties has thawed. Or if it is just a reflection of the alleged fact that institutional memory at these organizations isn’t what it used to be.
One more Google Trends chart. This one shows interest in the search terms “IELTS” and “TOEFL” since 2004. As you can see, back in 2004 there was a lot of interest in TOEFL. Back then, IELTS was something of an afterthought. Today, though, IELTS is utterly dominant.
There is a lesson to be taught here.
Which is that it didn’t necessarily have to be this way. You could say that IELTS is at the top of the heap because of the increased importance of new receiving markets over the past twenty years. But I don’t think it was a given that a test from the UK would become the de facto English test for people headed to Australia. And, needless to say, it wasn’t a given that a test from the UK would be the thing taken by people headed to Canada. TOEFL probably shoulda cornered that market.
So what happened? Well, the TOEFL originally launched at some point before the beginning of recorded history and for several decades it was fairly dominant in its category. Meanwhile, the IELTS was handed down on stone tablets in 1989 and quickly grew in popularity. The data suggests that its popularity surpassed that of TOEFL sometime around 2009.
Note that the peak of TOEFL’s popularity, according to the data, was around 2006. This was the year that the old paper-based TOEFL was largely replaced with the TOEFL iBT (note that a few markets got the iBT in 2005). A key aspect of this transition is that the TOEFL jumped from being a roughly 2-hour test to a roughly 4 hour and five minute test. Four hours! Old timers will remember that the iBT included unscored content in either the reading or listening section, but might not know that the reading section could contain TWO unscored passages.
(Why so long? Well, remember that the original TOEFL tested neither writing nor speaking.)
To me, this is the key reason why the test declined in relative popularity. I’d say that the 2:44 length of the IELTS suddenly became a lot more attractive. I think this was a major reason why test takers started piling into that particular test.
The TOEFL was quietly shortened to just under four hours sometime in the early 2010s with the removal of the second unscored reading passage. In 2019 it hit about 3:20 with the removal of certain reading and speaking questions. In 2023 it was reduced to just over two hours with the removal of all unscored questions and the longer writing task. In each case the numbers were massaged to make the test look even shorter than it actually was. In 2026 the test will clock in at just under 90 minutes, according to information that is currently available. And that will include plenty of unscored material, so further reductions are entirely possible.
Test duration remains a big deal. People like short tests. Go figure. The PTE has eaten away at the IELTS in part because it has always been a roughly two-hour test. And the Duolingo English Test’s one-hour duration has helped it capture a lot of market share in recent years.
I wonder, though, if we have reached a test-duration floor. The length of the PTE was quietly increased a few months back when certain revisions were introduced. Even the Duolingo Test seems to be a sliver longer than it used to be due to the introduction of new integrated tasks a few months ago (not to mention the time it takes for test takers to carry out new security features). And the IELTS team seems to be in no rush to shorten their test. Meanwhile, to my jaundiced eyes it almost seems like the TOEFL team is actually a little embarrassed about the potential length of the new version of their test. And the so-called “long tail” of minor tests that trail the big four in popularity seem to be distinguishing themselves in ways that don’t relate to test length.
It will be interesting to watch how this goes in the years ahead. Maybe the big tests will all become longer than they currently are. Stranger things have happened.
The HOELT Tender has been published. And I mean the actual tender, not another “request for information.”
Here’s a link. The key takeaway is that the Home Office seeks “a fully remote language testing service.” Says the tender:
“HOELT will move to a digital-by-default approach, while upholding the rigorous standards necessary to support our immigration decisions. The successful provider will deliver a service that combines technological innovation with security assurance, enabling customers to demonstrate their English language skills with confidence and convenience throughout their UK immigration journey.”
And:
“The Supplier shall provide a fully managed Remote Testing service. HOELTs shall be delivered securely in any location where the Customer chooses to take the test, provided secure test conditions are met.”
Test centers aren’t even mentioned in the publicly available version of the tender, as far as I can tell. How do you like that? The final RFI mentioned the possibility of 268 physical test centers. Note that the bidders will get a longer form of the tender, which could reference the possibility of test centers being included.
I’m not terribly familiar with the tender process, so it is absolutely fascinating to me how the document breaks down the weighting that will be given to certain criteria that make up the “Quality” of each bid. Do check that out to learn how the winner will be selected. Just note that price is pretty important too.
A winner will be selected in November of 2026.
A few other key details:
The contract will run from ’26 to ’31. That’s the same as in the final RFI… but the start date has been pushed back about four months to December of 2026.
The tender will not be divided into lots, though the possibility of collaborative bids is mentioned.
As expected, the HOELT will be available in 2-skill and 4-skill versions.
The estimated value remains £680,000,000 (excluding VAT) and £816,000,000 (including VAT).
Needless to say, this is good news for the team from the Duolingo English Test. They likely have the most experience with and enthusiasm for this sort of remote testing. But it is also good news for a variety of smaller providers who also have ample experience with remote testing. It probably isn’t great news for the IELTS partners, who have been pretty vocal in recent months regarding perceived shortcomings of certain online-only tests.
