Can computers really mark exams? Benefits of ELT automated assessments

app 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.  aԻ 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 app, we have developed a range of assessments using AI technology.

Versant

The Versant tests are a great tool to help establish language proficiency benchmarks in any school, organization or business. They are specifically designed for placement tests to determine the appropriate level for the learner.

PTE Academic

The  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. 

More blogs from app

  • A group of friends sat outside smiling and talking

    Lesser-known differences between British and American English

    By Heath Pulliam
    Reading time: 5 minutes

    Heath Pulliam is an independent education writer with a focus on the language learning space. He’s taught English in South Korea and various subjects in the United States to a variety of ages. He’s also a language learning enthusiast and studies Spanish in his free time.

    British and American English are two well-known varieties of the English language. While the accent is often the first difference people notice, there are also subtle distinctions in vocabulary, grammar and even style. Many know about how Brits say boot and lift, while Americans would say trunk and elevator, but what about a few lesser-known differences?

    Here, we take a look at a few of the more obscure differences between British English (BrE) and American English (AmE).

    Note: British English is underlined and American English isitalicized.

    1.Footballerandfootball player

    Along with the well-known difference of how in the U.S., football refers to American football, while football in Britain is what Americans like me call soccer, Americans also use player after the sport to denote someone who plays the sport. In British English, the sport with an added -er suffix is more common, like footballer and cricketer, not football player or cricket player.

    This is not universal, though. For some sports, the -er suffix is used in both dialects. Both Brits and Americans use the term golfer, not golf player. There are also sports where the -er suffix is never used, like for tennis, cycling and gymnastics. Nobody says tenniser, tennis player is used instead.

    People who cycle are cyclists and people who do gymnastics are gymnasts. Sometimes, badminton players are even called badmintonists. Overall, there aren’t really any concrete rules for what to call each player of a sport. Each sport has its own way of calling someone who participates in that sport.

    2.I dzܱ’t care lessandI could care less

    The American version (I could care less) means the same thing. Although technically incorrect, it is still widely used in North America as an idiom and will be interpreted as not caring at all about something. Although popular, both variations can be heard in North America. Regardless, miscommunications do happen surrounding this phrase.

    “I could care less about who Harry Styles is dating right now.”

    “Oh, I didn’t know you were interested in tabloid news.”

    “I’m not! I just said I didn’t care about it.”

    “No, you said that you could care less, meaning that it is possible for you to care less about who he’s dating.”

    “Ugh! What I mean is that I dzܱ’t care less. Happy?”

    3. American simplification

    Both British and American dialects are filled with many minuscule differences in spelling and phrasing. For example, the words plough (BrE) and plow (AmE) mean the same thing, but are spelled differently.

    When two words differ, American English generally favors the simpler, more phonetic spelling. Hey, there’s another one! Favour (BrE) and favor (AmE). It’s apparent in pairs like analyse (BrE) and analyze (AmE), and neighbour (BrE) and neighbor (AmE).

    Many of these small spelling differences can be attributed to Noah Webster, author of Webster’s Dictionary, who sought to distinguish American from British English by simplifying many of the words.

    Some of his simplifications to American English are swapping the s for z, (specialised to specialized), dropping the u in words ending in our, (colour to color), and changing words ending in -tre to -ter (theatre to theater).

    4. Courgette and zucchini

    The history of this vegetable, whatever you may call it, tells us why zucchini is used in American English and courgette is used in British English. If you’ve studied languages, you can probably guess what country each name originated from. England was introduced to this cylinder-shaped vegetable in the 19th century by its French neighbors, while Americans were introduced to it in the early 20th century by the large influx of Italian immigrants.

    The word zucchini is something of a mistranslation from Italian, however. What Americans use (zucchini) is the plural masculine form of the proper Italian word, (zucchino).

    5.Anticlockwiseand counterclockwise

    These terms mean the same thing, the rotation against the way a clock runs. In British English, this movement would be called anticlockwise, and in the U.S., they use counterclockwise.The prefixesanti- andcounter- mean similar things.Anti- means against, andcounter- means contrary or opposite to.

    You should use antibacterial soap in order to stop the spread of germs.Buying cheap clothes that only last you a few months is counterproductive in the long term.

    Can you guess how they described this movement before the invention of clocks with hands and circular faces? English speakers this long ago used sunwise. This direction at the time was considered auspicious and the opposite of the other direction.

    6.Haveand take

    Have and take are used often before nouns like shower, break, bath, rest and nap.In the U.S., peopletakeshowers andtakenaps, while in the U.K., peoplehaveshowers andhavenaps. Another example of this is how Americanstakea swim and Britshavea swim. These are called delexical verbs and we use them all the time in English, both British and American.

    Although often different, both groups of English speakers have arguments, make decisions and take breaks.

    7.Quite

    This word is spelled the same in both American and British English, but means something different. In the U.S.,quiteis typically used as an intensifier, like the wordvery.In the U.K., it’s normally used as a mitigator, like the wordsomewhat.

    It can also mean completely if it modifies certain adjectives. (e.g., It’s quite impossible to learn a language in one month.)

    American English: That Mexican food we had yesterday was quite spicy.

    Translation: That Mexican food we had yesterday was very spicy.

    In British English, quite means something more on the lines of kind of, or a bit.

    British English: Thank you for the meal, it was quite good.

    Translation: Thank you for the meal, it was somewhat good.

    8. Clothing differences

    The category of clothes is one of the richest, with differences between the two English variants.How about those pants that people used to only wear at the gym and around the house, but now wear them everywhere?

    Brits call themtracksuit bottomsand Americans call themsweatpants. What about a lightweight jacket that protects from wind and rain?Brits might call this ananorak(derived from the Greenlandic word), but Americans would call it awindbreaker. Both variants also useraincoatfor this article of clothing.

    9.Torchandflashlight

    As an American, I’ve been confused before when coming across the word torch while reading the work of an English author.

    To Americans, a torch is a piece of wood with the end lit on fire for light.What Brits are referring to when they use the wordtorchis aflashlight (AmE), a small, battery-run electric lamp.

    10.’t and don’t need to

    Ah, the English contraction. Many English learners don’t particularly love learning these, but they are an essential and everyday part of the language. ’t, however, is one that I don’t think I’ve ever heard another American say.

    In the U.K., this contraction is fairly common. ’t, when separated, becomes need not.

    British English: “You needn’t come until Tuesday night.”

    Americans would say the relatively simpler don’t need to.

    American English: “You don’t need to come until Tuesday night.”

    Don’t be fooled into thinking British English has necessarily more difficult contractions than the U.S., though. Just come to the American South and prepare to hear famous (or infamous) contractions like y’all (you all) and ain’t (am not, is not, are not)!

    Conclusion

    There are hundreds of differences between British and American dialects, we’re only scratching the surface here.Some of these make more sense than others, but luckily, both Brits and Americans can usually understand the meaning of any English word through context.

    Some people would even say that Brits speak English while Americans speak American.Although each dialect from across the pond seems very different, they have far more similarities than differences.

  • A group of students sat together in a library around a laptop smiling

    Forgetting to remember – What active recall is all about

    By Heath Pulliam
    Reading time: 5 minutes

    Heath Pulliam is an independent education writer with a focus on the language learning space. He’s taught English in South Korea and various subjects in the United States to a variety of ages. He’s also a language learning enthusiast and studies Spanish in his free time.

    Almost everyone has studied for a test at some point in their life. Have you ever studied a lot for a test and still not received a great grade? Probably so. This is because not all study methods are created equal. Just because you’ve ‘studied’ doesn’t necessarily mean that the material has stuck in your brain.

    Some common study methods don’t do as much as you might think. One technique, however, stands out for its efficiency and effectiveness. It’s called active recall, and it’s regarded by many as the best way to study.

    What is active recall?

    In short, active recall is the process of pulling information from your brain without prompts. The idea is that when you force yourself to remember something, you memorize it much faster than you would by only reading material or passively studying.

    When you’re taking a test and you’re not sure about an answer, you search your brain for the answer; that’s active recall. The action of trying to remember is what essentially crystallizes the information in your head for the long term. Multiple study methods use active recall or the action of searching your brain for an answer.

    Active recall forces you to pull information from your mind. It’s not just for doing well on tests but also for long-term retention.

    Active recall is praised for its effectiveness, but it is also mentally taxing. Attempting to recall things that are not in the front of your mind is tiring, which is part of why some opt out of using it.

    Dr. Cal Newport, author of Deep Work, says:

    “It’s almost like you have a pseudo-photographic memory when you study this way.”

    Passive and active study

    Most people are taught to study, but less about how to study. People end up using passive study techniques, like rereading material or rewatching lectures. Although not particularly effective, studies show that reading and rereading notes remains university students’ most common study method. (Dunlosky, 2013).

    These methods might feel like effective studying but aren’t really helping much. Yes, passive study methods such as reading, highlighting, or watching can help, but can also give you a false sense of understanding.

    With these methods, when the time comes to retrieve the information, you struggle to recall it because it is not in your long-term memory.

    The first step to becoming excellent at studying is understanding how important active study methods are, such as flashcards, practice testing, or practice teaching – all methods that use active recall.

    Spaced repetition

    If you’re researching study methods, you’ve probably encountered the term spaced repetition. This technique can be incorporated into active recall study methods, primarily flashcards.

    Spaced repetition involves reviewing information at increasingly longer intervals in order to improve long-term memory. You are using this technique when you learn something, so review it a day later, then a week later and then a month later. It helps fight your brain’s natural process of forgetting things.

    Combining spaced repetition with active recall is always a recipe for successful study.

    So, what methods can you use to employ active recall in your study sessions?

    Summarize what you know

    The simplest way to practice active recall is to try to remember the material actively. To do this, read or reread some material, then take some time away—take a walk, snack break, or whatever you like to do to take a break.

    During this break, try to replicate in your mind as much about the topic from scratch as if you had to teach about it. Don’t be discouraged if you have trouble remembering; part of the learning process is forgetting material and then remembering it.

    When using this for language acquisition, make up relevant sentences and translate them into your target language. What type of vocabulary were you learning? What grammatical structures were new to you?

    Take a mental note of which elements you were able to remember and which ones gave you trouble.

    When finished, go back to the material and review. Were you able to remember things? Whether you could or not, you’ve helped yourself. Your attempt to remember is you practicing active recall.

    When you do successfully remember something, you’ll realize that it is locked in your long-term memory. This method is as simple as it is productive.

    Teach a friend

    Similar to mental summarizing, after studying, find a friend and do your best to explain the topic to them. Language learning could involve explaining how a new grammar pattern works in your target language or taking sentences and phrases and translating them to a friend.

    As a teacher, your best lessons are the ones where you know the material inside and out. If you can effectively explain and teach someone about your target language, you are that much closer to mastering the information yourself.

    Flashcards

    Flashcards, especially when combined with spaced repetition, are the king of study methods for language acquisition.

    Between all that they can be used for, they are especially good for language learning, no matter what part of the language you’re studying. This is one of the best ways to learn sentence structure that does not mirror your native tongue, conjugations for speedy use and general vocabulary.

    I have found that using a spaced repetition-based flashcard program is the best way to study a language. This way, you don’t have to schedule a bunch of paper flashcards manually.

    Among the most popular programs for language learners is Anki. Anki is a spaced repetition flashcard program. With it, you create your cards and the program schedules them at increasingly longer intervals. It is also used widely among students studying for large tests, like medical and law exams.

    To make the best flashcards, make them yourself. Downloading a big set of flashcards for your target language is tempting but not particularly helpful. Instead, after each study session, make a few flashcards yourself covering the new vocabulary/grammar you’ve just learned about.

    One more thing: do not use multiple-choice answers for your cards. Active recall happens when you produce information without prompts or assistance.

    Personally, I use flashcards every day to learn Spanish through . This program has a learning curve, but the payoff makes it worth it.

    Test yourself

    The goal of studying for a test is to be able to answer questions about a topic or, even better, know the material for real-world application. As stated earlier, active recall happens during testing, so why not use this form of active recall before testing?

    Use your materials to make a mock exam in this method, then take it. Through actively remembering to answer each question, the ones you get right will stick in your brain. Review the questions that you dzܱ’t remember and retake the test focusing on them.

    While not particularly complicated, these tips and methods’ effectiveness comes from their use of active recall. As you forget information and pull it from the depths of your memory, you solidify it for yourself.

    Forgetting plays an important role in learning something for long-term memory.

    Now, get to studying; how are you going to use the power of active recall in your routine?

    References

    Dunlosky, J. "Improving Students’ Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology." Association for Psychological Science, 2013, 1-6

  • A teacher sat in a classroom with a child, sharing crayons with eachother and smiling

    Four ways to keep kindergarten ESL students focused all day

    By Heath Pulliam
    Reading time: 5 minutes

    Heath Pulliam is an independent education writer with a focus on the language learning space. He’s taught English in South Korea and various subjects in the United States to a variety of ages. He’s also a language learning enthusiast and studies Spanish in his free time.

    Those who have taught children anywhere between the ages of 4 and 8 know that one of the biggest challenges of getting through to them is keeping your presentation style interesting. As someone who taught ESL in South Korea to kindergarteners, there are a few factors that make keeping students engaged a challenge. In countries where students learn English, students often have a heavy courseload and high expectations. As a first-year teacher, I learned a lot about what worked and what didn’t through trial and error. These are four methods that I consistently used to keep my students interested and engaged all day.

    Students are quick to lose focus at such a young age. You’re not speaking their mother tongue and some parts of an ESL curriculum are less than exciting. With young students, you can’t lecture your way through the material all day. Kindergarteners have a small window of focus and it must be capitalized on. The following methods are ones that worked for me and can be modified to cover any topic you’ll run into in an ESL curriculum.