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. 

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

Explore our courseware and exams

More blogs from app

  • A teacher sat with students at a table, the students are using tablets.

    Benefits of using tablets in the primary classroom

    By Jacqueline Martin

    Reading time: 5 minutes

    Interactive whiteboards, PCs and laptops are common in many schools worldwide, but have you ever considered using tablets in your young learners' classes?

    Tablets can be used for many things. Online research, watching and creating videos, playing games, and digital storytelling are just a few examples. Of course, there's also the added environmental benefit of going paper-free.

    In this post, we're going to explore some of the reasons why using tablets can be beneficial in the young learner's classroom and what to consider before you do so.

    What are the benefits of using tablets in class?

    1. Facilitating engagement

    With good direction from the teacher, tablets can emulate natural social interaction and interactivity. They can also offer problem-solving activities, set achievable goals and provide instant feedback.

    Moreover, when young learners are truly engaged in an activity, it may be perceived as effortless - and they learn and use their second language (L2) without even realizing it.

    2. Introducing authenticity and autonomy

    In terms of content, tablets allow us to bring the real world into the classroom at the tap of a screen. We can provide learners with authentic materials via level-and-age-appropriate videos and real-life communication. As well as interaction with other teachers and learners through teams or by using a secure app such as .

    Tablets also promote learner autonomy. They are easy to use, allowing us to take a step back and let our students work at their own pace, being on stand-by as a facilitator when students require help or a little push in the right direction.

    3. Promoting creativity, communication and inclusion

    Nearly all tablets have a webcam and voice recorder, which means that learner-generated content can be created easily - even without dedicated software.

    You can have your students make their own vlogs (video diaries), ebooks, comics, cartoons and movie trailers. All you need to do is to install apps such as or from Duck Duck Moose. While these apps have been created for 'fluent-speaker' classrooms, they can easily be adapted to an ELT context.

    Tablets also promote communication. This can help improve students' L2 oral skills at any level, when the teacher is there to support and guide them.

    One of the greatest advantages of a tablet as opposed to a computer is that anyone can use one and they are much more portable.

    For students with special educational needs, tablets can be an essential learning tool and they can also be used by students with low-level motor skills, such as very young learners. Similarly, tablets can work really well with multi-level classes, as they allow you to offer differentiated materials, activities and support where necessary.

    4. Enabling online assessment

    Tablets can also facilitate interactive online exams or help measure progress. Tests such as 'English Benchmark - Young Learners' are designed with primary learners in mind, to be taken anytime, anywhere.Its game-like format engages students and takes the fear out of being assessed. It also provides instant feedback to the teacher with informative reports and advice for future study.

    5. Building relationships with caregivers

    Finally, as with any online content, tablets allow you to connect with our learners outside the classroom. You can quickly send links to classwork and feedback to the children's caregivers, fostering a positive relationship and a greater interest in their child's progress and learning.

    Tips for using tablets in class

    Before implementing the use of tablets in your classroom, there are some things you should consider. Here are some useful tips that will help you gain the maximum benefit from tablets.

    Usability:

    • Decide what you are going to use the tablets for and when. Are you going to allow students to use the tablets for all parts of the lesson or only for specific activities? This may depend on the number of tablets you have available.
    • Use technology to improve an activity or design new activities that would not be possible without the tech, rather than using it to carry on as normal. Think about when a tablet will help learners do something they wouldn't be able to do without one, e.g., make a video or create and share a piece of writing with the whole class.
    • Think about using tablets for creation rather than consumption. Your students can (and probably do) spend a fair amount of time consuming videos in their free time. Whether they do this in English or not is another story, but in the classroom, students should use the language as much as possible (see the next point).
    • Use the tablets for collaborative tasks that require social interaction and communication. It's unlikely that you will have one tablet per student. Make the most of this limitation by having students work in pairs or small groups. Students can use their own devices individually outside the classroom.
    • Try to incorporate tablets into regular classroom activities and interactions. Avoid making them a "reward" or just for "games". Even if games are part of your planned tablet usage, make it clear that students are playing them in order to learn English. Encourage students to think of the tablet as a tool to help them on their learning journey.

    General tips

    • Try out any apps or widgets before asking students to use them. If necessary, make or find a step-by-step tutorial to help students use an app. There's nothing worse than having a class of twenty-five students all raising their hands at the same time because they don't know where to start.
    • Have clear rules and guidelines for tablet use. Educate students about using the equipment responsibly. Do this before you hand out tablets the first time.
    • Provide students and parents with a list of recommended apps to continue their home learning. Whether you have a class set of tablets or are using BYOD (Bring Your Own Device), many students will have access to a tablet or mobile phone at home, which they can use for further practice. Students will likely be motivated to continue playing games at home and may wish to show their parents and friends any content they've created in class.

    Practicalities

    • Consider the hardware and technical requirements. Do you need a Wi-Fi connection? How many devices will you have? Which apps and programs do you want to use?
    • Ensure the features and apps you plan to use suit the age group you're teaching. Do some research, and if possible, choose apps designed for educators, avoiding freebie apps that may contain advertising. Block any websites you think unsuitable and install a search engine with child-friendly filters.
    • Set the language of the devices to English. Even if your students are very young, they'll pick up useful language and will be more inclined to use English as they are using the tablet.
    • Decide where you will keep the tablets and how they will be maintained. How often and where will they be charged?
    • Think about how you can flexibly set up your classroom to incorporate collaborative tablet use. Move tables together to make group work easier. Create workstations or even have cushions or bean bags in a corner of the classroom.

    Using tablets to assess student progress with Benchmark

    With the right software, tablets can allow us to conduct formative assessments through immediate feedback and learning analytics.

    We have developed our own English-language test for children aged 6 to 13 in an app designed specifically for tablet use. This fun, game-like test is highly motivating and assesses all four skills in a relaxed environment, removing the stress of traditional exams. It also allows you to see where each learner needs more improvement, providing recommendations on what to teach next and suggested activities in selected app courseware.

    Find outmore information about the English Benchmarktest.

  • A child running with a rugby ball outside, with children behind them

    How can gaming support language learning?

    By Jacqueline Martin

    Reading time: 5 minutes

    Academics and teachers have been writing about the benefits of using games in the language classroom for many years. Wright et al (1984), Lee Su Kim (1995), Ubermann (1998), Ersoz (2000), Yong Mei and Yu-Jin (2000) and Thi Thanh Huyen and Khuat Thi Thu Nga (2003) all pretty much agreed that games provide a useful and meaningful context for language use; encourage students to interact and communicate; can both challenge and reduce anxiety (as the emphasis is on the message, not the form); provide practice in all four skills; and help students to make and sustain the significant effort involved in learning a language.

    Kim and others have also noted that games can offer a welcome break from the usual routine of the language class. Playing a game after an intensive test or with over-excited students after break time can help re-engage learners instantly in your lesson, and you'll maximize your time with them.

    Lengeling and Malarcher (1997) took the list of potential benefits of games in the classroom even further.

    Affective

    • Games lower the affective filter
    • They encourage the creative and spontaneous use of language
    • They promote communicative competence
    • Games are both motivating and fun

    Cognitive

    • Games reinforce learning
    • They both review and extend learning
    • Games focus on grammar in a communicative manner

    Class dynamics

    • Games are extremely student-centered
    • The teacher acts only as a facilitator
    • Games build class cohesion
    • They can foster whole-class participation
    • Games promote healthy competition

    Adaptability

    • Games can be easily adjusted for age, level and interests
    • They utilize all four skills
    • Games require minimum preparation after the initial development stage

    It is important to bear in mind that when the above was written over 20 years ago, it was with reference mostly to more traditional games. But more recent evidence seems to indicate that the same principles apply. Some additional benefits cited by teachers I've spoken to are that:

    • Games could make language lessons less threatening for less confident pupils as their concern about getting sentence form wrong was reduced, and so their production greater.
    • Students learn more than just the language of the lesson when playing a game; they may learn instructional language through discussion or rules and sometimes negotiation skills and a lesson in cultural differences too.
    • Students can form a greater variety of emotional connections with language through playing games, for example acting out a word or seeing another student do so, or remembering a clue for a word.

    So, playing games can help students learn a language – but is just playing them enough? Some teachers like using games with less motivated classes who won't engage with straight practice activities and will willingly use key vocabulary and structures in a game, gaining much-needed practice without even realizing it. In today's language-learning context, though, is that a good thing?

    Motivating the unmotivated

    In recent years, much research has shown that students learn better when the intention or objective of the lesson is clear to them. In short, they understand what they're supposed to be learning and why and, when taking it to the next level, can assess their own learning and be actively involved in planning their next steps.

    Would knowing that the games they play are actually a way of doing some additional language practice make these students engage less? Opinion differs, and some discussion seems to center around the actual activity involved. Some games are thinly veiled group-work tasks, but other games that are at the right proficiency level (or slightly above) and take into account factors like cultural context, available time, learning topic and the classroom setting are generally considered to have a positive impact.

    Another major influence on improving motivation is the feedback a student receives, and this is something games can also support. Online games can provide richer simulated learning experiences and immediate feedback to students in a variety of ways.

    Above all, the main issue for the less motivated students is usually that they can't see why they need to learn English. Playing games not only simulates 'real' contexts but also helps them understand that they can accomplish a variety of tasks using English as a medium, which is motivational in itself.

    As teachers, there is a responsibility to explain how or why games will help students learn. This can equally motivate learners (or parents) who fear that playing games is just frivolous time-wasting. For example, informing even adult students that a simple hangman or hot seat game helps them improve spelling skills, gets their brains focused on recognizing the shape and structure of new words, and facilitates their learning of new vocabulary soon helps them see the value (Simpson 2011).

    Can games help learners acquire 21st-century skills?

    Maybe we can draw the conclusion that games can positively impact learning – but is that even enough? Today's teachers have to ensure not just that their students learn but that they acquire the skills they need for life and jobs in the 21st century. Can games help here too? This is a newer area of research, but evidence seems to indicate that games can help students learn a variety of important skills such as critical thinking skills, creativity, teamwork and good sportsmanship.

    These ideas were taken seriously by Robert Morris University Illinois, who offered an e-sports scholarship for the first time in 2014. They studied two groups of students – football players and gamers – and found that levels of competitiveness, perseverance, focus and determination were very similar. Both groups showed a similar desire to excel as part of a team. Both 'sports' required the team members to be detail-orientated, have good hand-eye coordination and have a strategic mind. The only difference was in the level of cardiovascular activity. Both groups received performance analysis and tactical advice from coaches and both subsequently made improvements.

    How many universities will start to offer these types of programs remains to be seen. Still, the idea that online competitive gaming can improve performance is being brought to the workplace too. Think about what virtual teams could learn from playing role-based collaborative games. Team members have set roles and clear and shared goals and have to work closely together to formulate an action plan to achieve them. Teamwork, skill, strategic thinking and communication are essential.

    All these are important skills for today's workplace, so maybe gaming can provide an opportunity to hone these in a lower-risk environment and improve business performance.

    These examples are clearly far from the norm, but they do seem to indicate that using gaming to support learning in the classroom is not a waste of time. When you get the right mix of gaming and learning, it develops a student's autonomous learning skills and encourages them to spend more time on task – both of which greatly impact learner outcomes.