Multilevel Language tests: Walking into the land of the unexplored

Duolingo English Test (DET) 
Duolingo English Test (DET), developed by Duolingo Inc., is a computer-adaptive English proficiency test 
that can be taken anywhere without the need to do it from a testing center. One of the main characteristics 
of this test is that there is no need to schedule or make any appointments, which makes it very convenient 
for test takers. Results, which range from 10-160 points, are usually provided within two days, and they are 
accepted in over 2000 institutions worldwide, especially for higher education admission in English speaking 
countries. The registration fee is $49 USD.
As opposed to other tests, the Duolingo English Test is not organized by the four skills of listening, reading, 
speaking and writing. Instead, tasks are integrative, and they focus on Literacy (reading & writing), 
Comprehension (reading & listening), Conversation (listening & speaking) and Production (writing & 
speaking). The first part of the test is adaptive and includes the following item types: c-test, audio yes/no 
vocabulary, visual yes/no vocabulary, dictation, and elicited imitation. The second part of the test is a 10-
minute video interview, where test takers respond to four writing prompts and four speaking prompts. Test 
results include an overall score as well as subscores for Literacy, Conversation, Comprehension, and 
Production. 
Duolingo English Test is an exam that has gained popularity in the last couple of years. Back in 2018, 
Plough et al. stated that tools such as Duolingo were starting to become more used, and in 2019 it was 
considered to be a low-stakes commercial language learning system (Zechner & Evanini, 2019). However, 
since the COVID-19 pandemic, the exam has gained much more popularity due to its “at home nature”, and 
more universities have started accepting it as a measure of applicants’ English language proficiency 
(Wagner, 2020). The exam has previously been criticized for presenting discrete items that are 
decontextualized in many cases (Wagner & Kunnan, 2015). Most of the research found about the test are 
technical reports that are listed in the 
Duolingo website
, which serves as evidence that more research 
published in scientific journals is needed.