Can computers really mark exams? Benefits of ELT automated assessments

ɫèAV Languages
Hands typing at a laptop with symbols

Automated assessment, including the use of Artificial Intelligence (AI), is one of the latest education tech solutions. It speeds up exam marking times, removes human biases, and is as accurate and at least as reliable as human examiners. As innovations go, this one is a real game-changer for teachers and students. 

However, it has understandably been met with many questions and sometimes skepticism in the ELT community – can computers really mark speaking and writing exams accurately? 

The answer is a resounding yes. Students from all parts of the world already take AI-graded tests.  and Versanttests – for example – provide unbiased, fair and fast automated scoring for speaking and writing exams – irrespective of where the test takers live, or what their accent or gender is. 

This article will explain the main processes involved in AI automated scoring and make the point that AI technologies are built on the foundations of consistent expert human judgments. So, let’s clear up the confusion around automated scoring and AI and look into how it can help teachers and students alike. 

AI versus traditional automated scoring

First of all, let’s distinguish between traditional automated scoring and AI. When we talk about automated scoring, generally, we mean scoring items that are either multiple-choice or cloze items. You may have to reorder sentences, choose from a drop-down list, insert a missing word- that sort of thing. These question types are designed to test particular skills and automated scoring ensures that they can be marked quickly and accurately every time.

While automatically scored items like these can be used to assess receptive skills such as listening and reading comprehension, they cannot mark the productive skills of writing and speaking. Every student's response in writing and speaking items will be different, so how can computers mark them?

This is where AI comes in. 

We hear a lot about how AI is increasingly being used in areas where there is a need to deal with large amounts of unstructured data, effectively and 100% accurately – like in medical diagnostics, for example. In language testing, AI uses specialized computer software to grade written and oral tests. 

How AI is used to score speaking exams

The first step is to build an acoustic model for each language that can recognize speech and convert it into waveforms and text. While this technology used to be very unusual, most of our smartphones can do this now. 

These acoustic models are then trained to score every single prompt or item on a test. We do this by using human expert raters to score the items first, using double marking. They score hundreds of oral responses for each item, and these ‘Standards’ are then used to train the engine. 

Next, we validate the trained engine by feeding in many more human-marked items, and check that the machine scores are very highly correlated to the human scores. If this doesn’t happen for any item, we remove it, as it must match the standard set by human markers. We expect a correlation of between .95-.99. That means that tests will be marked between 95-99% exactly the same as human-marked samples. 

This is incredibly high compared to the reliability of human-marked speaking tests. In essence, we use a group of highly expert human raters to train the AI engine, and then their standard is replicated time after time.  

How AI is used to score writing exams

Our AI writing scoring uses a technology called . LSA is a natural language processing technique that can analyze and score writing, based on the meaning behind words – and not just their superficial characteristics. 

Similarly to our speech recognition acoustic models, we first establish a language-specific text recognition model. We feed a large amount of text into the system, and LSA uses artificial intelligence to learn the patterns of how words relate to each other and are used in, for example, the English language. 

Once the language model has been established, we train the engine to score every written item on a test. As in speaking items, we do this by using human expert raters to score the items first, using double marking. They score many hundreds of written responses for each item, and these ‘Standards’ are then used to train the engine. We then validate the trained engine by feeding in many more human-marked items, and check that the machine scores are very highly correlated to the human scores. 

The benchmark is always the expert human scores. If our AI system doesn’t closely match the scores given by human markers, we remove the item, as it is essential to match the standard set by human markers.

AI’s ability to mark multiple traits 

One of the challenges human markers face in scoring speaking and written items is assessing many traits on a single item. For example, when assessing and scoring speaking, they may need to give separate scores for content, fluency and pronunciation. 

In written responses, markers may need to score a piece of writing for vocabulary, style and grammar. Effectively, they may need to mark every single item at least three times, maybe more. However, once we have trained the AI systems on every trait score in speaking and writing, they can then mark items on any number of traits instantaneously – and without error. 

AI’s lack of bias

A fundamental premise for any test is that no advantage or disadvantage should be given to any candidate. In other words, there should be no positive or negative bias. This can be very difficult to achieve in human-marked speaking and written assessments. In fact, candidates often feel they may have received a different score if someone else had heard them or read their work.

Our AI systems eradicate the issue of bias. This is done by ensuring our speaking and writing AI systems are trained on an extensive range of human accents and writing types. 

We don’t want perfect native-speaking accents or writing styles to train our engines. We use representative non-native samples from across the world. When we initially set up our AI systems for speaking and writing scoring, we trialed our items and trained our engines using millions of student responses. We continue to do this now as new items are developed.

The benefits of AI automated assessment

There is nothing wrong with hand-marking homework tests and exams. In fact, it is essential for teachers to get to know their students and provide personal feedback and advice. However, manually correcting hundreds of tests, daily or weekly, can be repetitive, time-consuming, not always reliable and takes time away from working alongside students in the classroom. The use of AI in formative and summative assessments can increase assessed practice time for students and reduce the marking load for teachers.

Language learning takes time, lots of time to progress to high levels of proficiency. The blended use of AI can:

  • address the increasing importance of formative assessmentto drive personalized learning and diagnostic assessment feedback 

  • allow students to practice and get instant feedback inside and outside of allocated teaching time

  • address the issue of teacher workload

  • create a virtuous combination between humans and machines, taking advantage of what humans do best and what machines do best. 

  • provide fair, fast and unbiased summative assessment scores in high-stakes testing.

We hope this article has answered a few burning questions about how AI is used to assess speaking and writing in our language tests. An interesting quote from Fei-Fei Li, Chief scientist at Google and Stanford Professor describes AI like this:

“I often tell my students not to be misled by the name ‘artificial intelligence’ — there is nothing artificial about it; A.I. is made by humans, intended to behave [like] humans and, ultimately, to impact human lives and human society.”

AI in formative and summative assessments will never replace the role of teachers. AI will support teachers, provide endless opportunities for students to improve, and provide a solution to slow, unreliable and often unfair high-stakes assessments.

Examples of AI assessments in ELT

At ɫèAV, we have developed a range of assessments using AI technology, including  is aimed at those who need to prove their level of English for a university place, a job or a visa. It uses AI to score tests and results are available within five days. 

Read more about the use of AI in our learning and testing here.

Explore our courseware and exams

More blogs from ɫèAV

  • Students sitting outside a building chatting and smiling together

    Understanding resolutions: Setting your language goals for success

    By Charlotte Guest
    Reading time: 3 minutes

    Millions of people make resolutions every New Year (or any significant milestone). Many learners set inspiring goals to acquire new skills, explore new academic fields, or develop better study habits. But how often do these resolutions stick? And how can you turn your learning-related goals into lasting habits? This post dives into the science of resolutions and practical strategies to make your learning commitments stick.

    Why do learning resolutions fail?

    that the majority of New Year’s resolutions fail by February. When it comes to learning, the reasons often include overly ambitious goals, lack of clear plans and failure to sustain motivation over time.

    One main reason resolutions fail is that they are not specific enough. Goals like "get fit" or "learn a language" are too vague and don’t include steps to take. Without a clear plan, it’s easy to feel lost and lose motivation.

    For example, it’s one thing to decide, “I’m going to learn French this year,” but without a structured plan, consistent practice and measurable benchmarks, the resolve often fizzles out.

    Another factor is having unrealistic expectations. While it's good to be ambitious, setting too high goals can cause frustration and burnout. For example, trying to master a skill in just a few weeks ignores the time and effort needed to improve.

    Distractions and other priorities can slow down progress. Life demands, like work or family, can easily push resolutions aside if they aren't included in everyday routines.

    How long does it take to form a habit?

    One key to sticking to a learning resolution is habit formation. Popular belief often quotes the “21-day rule,” but research says differently. A 2009 study published in the found that it takes an average of 66 days to form a new habit.

    However, this number varies based on personal factors, the difficulty of the habit and the surroundings. For instance, if your goal is to study for 20 minutes each day, it may feel more comfortable after a few weeks of practice. In contrast, picking up a harder learning habit, like spending two hours a day studying Mandarin grammar, could take much longer to become a regular part of your routine.

    Strategies for making resolutions stick

    1. Start small and specific

    Instead of setting a big goal like “become fluent in Spanish," set smaller, easy-to-reach goals. For example, try "learn 10 new Spanish words each day" or “listen to one Spanish podcast each week.” This makes your goal seem less overwhelming and helps you feel accomplished over time.

    that mixing big-picture goals with small, specific ones can be a useful way to stay on track when working toward long-term objectives.

    2. Use the power of routine

    Connect your new learning habit with one you already do. This is called “habit stacking.” For example, if you have coffee every morning, decide to study for 15 minutes as you finish your drink. Linking the new habit to a routine you already have makes it more likely to stick.

    3. Track your progress

    Keeping track of your learning helps motivate you and shows where you can improve. Whether you check off lessons in a language learning app like Mondly by ɫèAV or write notes in a journal, seeing what you’ve accomplished keeps you interested.

    4. Build accountability

    Tell your friends, family, or study groups about your learning goals. When someone else knows your goal, it can help you stay on track. It’s even better to join a community of others who want to learn, like on social media or in online classes.

    People who do well often set clear, achievable goals and share them with friends for support.

    5. Reward yourself

    Small rewards can go a long way toward maintaining motivation. Celebrate milestones with meaningful perks, such as a favorite treat, an afternoon off, or buying yourself a book on the subject you're learning.

    6. Anticipate and plan for setbacks

    Life happens—sometimes work is stressful or things go wrong. Remember that missing a day or getting behind doesn’t mean you've failed. Keep going by recognizing challenges and getting back to your study plan with fresh motivation.

    7. Don’t fear failure

    Mistakes are part of progress. Every mispronounced word or awkward conversation is a step closer to fluency.

    8. Reassess and adjust goals

    If your language learning plan isn’t working, change it. If weekly goals seem too much, try monthly goals instead. The key is flexibility.

    The reward of resolutions

    Keeping resolutions can be tough, but the benefits are great. Picture yourself reading a book in your target language, traveling more easily, or making stronger friendships with people from different cultures. With determination and these tips, you can achieve your language learning goals.

    Start today -Whether it's January or June, it's never too late to set new goals. Draft a language learning plan, break it into small steps, and watch as each day brings you closer to your destination.

  • A teacher leaning over a desk in a classroom talking to her students, some who have their hands up in the air

    Educator wrapped 2024: A year in review for educators

    By Charlotte Guest
    Reading time: 3 minutes

    As the year comes to a close, it’s time to reflect on the highlights, achievements and innovations that shaped education in 2024. For educators everywhere, this has been a year of growth, adaptation and pushing boundaries to empower both students and teachers. Whether you’ve been focused on refining your teaching practices or adopting new educational technologies, there’s plenty to celebrate and learn from this year.

    Here’s a look back at some of the major ɫèAV Languages educator highlights in 2024.

  • A woman dressed in a halloween costume smiling holding a cat shaped pumpkin outdoors

    The intriguing etymology of spooky words

    By Charlotte Guest
    Reading time: 4 minutes

    There's a certain allure that surrounds spooky words. Their very sound can send shivers down your spine and their meanings often carry chilling tales of the past. For those who revel in the peculiarities of language, exploring the origins of these eerie expressions offers a hauntingly delightful experience.

    Language is full of mystery and the etymology of words related to the supernatural is no exception. Let's take a closer look at some of the most spine-tingling words in the English language and unearth their origins.

    1.Ghoul

    The word "ghoul" has its roots in Arabic folklore. Derived from the Arabic word "ghūl," it refers to an evil spirit that robs graves and feeds on the dead. This sinister entity first appeared in English texts around the 18th century, becoming synonymous with creatures that haunt our nightmares.

    2.Witch

    "Witch" is a word steeped in history and lore. Its origins can be traced back to the Old English word "wicce" (for a female witch) and "wicca" (for a male witch). These terms are believed to be linked to the Proto-Germanic root "wikkjaz," meaning "one who wakes the dead." Over the centuries, the image of witches transformed, influenced by cultural narratives and historical events such as the infamous witch trials.

    3.Vampire

    The word "vampire" conjures images of blood-sucking fiends that prowl the night but its linguistic origins are equally fascinating. It likely comes from the Serbian word "vampire," which gained popularity in the 18th century in Western Europe. This term was used to describe beings that rise from the grave to feast on the living, a concept that has since been romanticized in literature and film.

    4.Specter

    Derived from the Latin "spectrum," meaning "appearance" or "vision," the term "specter" is often used to describe a ghostly apparition. In the 17th century, it came to be associated with the haunting phantoms that drift through abandoned halls and eerie landscapes. Its spectral connotations are timeless, evoking images of translucent figures and the eerie rustle of bygone whispers.

    5.Zombie

    While the concept of reanimated corpses exists in various cultures, the word "zombie" has its origins in West African folklore. It is derived from the Kikongo word "nzambi," meaning "spirit of a dead person." The term was introduced to the Western world through Haitian Vodou practices and gained prominence in popular culture during the 20th century.

    6.Poltergeist

    The term "poltergeist" originates from the German words "poltern," meaning "to make noise," and "Geist," meaning "spirit" or "ghost." This eerie word describes a type of supernatural entity that is known for its mischievous and sometimes malevolent behavior, often manifested through unexplained noises or objects moving without apparent cause. Poltergeist occurrences have long featured in folklore and horror stories, capturing the imagination with tales of restless spirits causing chaotic disturbances in the world of the living.

    7.Banshee

    The word "banshee" is rooted in Irish mythology, deriving from the Old Irish term "bean sídhe," meaning "woman of the fairy mound." Banshees are believed to be heralds of death, their mournful wails seen as an omen that someone is soon to pass away. These spectral figures often appear as women shrouded in gray or white garments, their cries echoing the sorrow and mystery that enshroud their presence. The legend of the banshee has endured in popular culture, continuing to haunt the imaginations of those who hear her tales.

    8.Doppelgänger

    The term "doppelgänger" originates from the German language, combining "doppel," meaning "double," with "Gänger," meaning "goer" or "walker." It refers to the unsettling phenomenon of encountering one's double, often considered an omen of bad luck or death. In folklore, a doppelgänger is thought to be a spirit or supernatural entity that takes on the appearance of a living person. This eerie concept has been a source of fascination in literature and art, exploring themes of identity and the dual nature of the self.

    9.Wraith

    The word "wraith" has Scottish origins and is commonly used to describe a ghost or apparition, particularly one that portends death. Its etymology is somewhat obscure, though it shares a kinship with words indicating spectral or eerie appearances. Wraiths are often portrayed as shadowy, ethereal figures that linger between the realm of the living and the dead, haunting desolate landscapes with their sorrowful presence.

    10.Mummy

    While the practice of mummification is most famously associated with ancient Egypt, the word "mummy" itself has an intriguing history. Derived from the Persian word "mūmiya," meaning "bitumen" or "asphalt," it referred to the embalming substance used in the preservation process. This term was absorbed into medieval Latin and later English, coming to define the preserved bodies themselves. Mummies have captivated imaginations and spurred countless myths and stories, bridging the gap between ancient rituals and modern horror tales.

    11.Werewolf

    The word "werewolf" has deep linguistic roots, stemming from the Old English "were," meaning "man," combined with "wulf," meaning "wolf." This term describes the mythical entity that transforms from human to wolf, often during a full moon. Such legends have been present in numerous cultures, with various explanations and lore surrounding the transformation process. The enduring allure of werewolves in fiction and folklore highlights humanity's fascination with the primal, untamed aspects of nature and identity.

    The power of spooky language

    Spooky words hold a unique power over us. Understanding their origins not only enriches our linguistic knowledge but also deepens our appreciation for the stories and cultures that have shaped these words over time.

    For linguaphiles, unraveling the mysteries behind spooky words is a thrilling adventure. Each term carries a legacy, a tapestry woven with tales of terror and wonder. Whether you're penning a chilling tale or simply enjoy the art of language, these eerie expressions continue to captivate and inspire.