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. 

app English International Certificate (PEIC)

app English International Certificate (PEIC) also uses automated assessment technology. With a two-hour test available on-demand to take at home or at school (or at a secure test center). Using a combination of advanced speech recognition and exam grading technology and the expertise of professional ELT exam markers worldwide, our patented software can measure English language ability.

Read more about the use of AI in our learning and testing here, or if you're wondering which English test is right for your students make sure to check out our post 'Which exam is right for my students?'.

Explore our courseware and exams

More blogs from app

  • Students sat ina library studying with laptops in front of them chatting to eachother

    Teaching engaging exam classes for teenagers

    By
    Reading time: 4 minutes

    Teachers all over the world know just how challenging it can be to catch their students’ interest and keep them engaged - and it’s true whether you’re teaching online or in a real-world classroom.

    Students have different learning motivations; some may be working towards their exam because they want to, and some because they have to, and the repetitiveness of going over exam tasks can often lead to boredom and a lack of interest in the lesson.

    So, what can we do to increase students’ motivation and add variation to our classes to maintain interest?

    Engage students by adding differentiation to task types

    We first need to consider the four main skills and consider how to differentiate how we deliver exam tasks and how we have students complete them.

    Speaking - A communicative, freer practice activity to encourage peer feedback.

    Put students into pairs and assign them as A and B. Set up the classroom so pairs of chairs are facing each other - if you’re teaching online, put students in individual breakaway rooms.

    Hand out (or digitally distribute) the first part of a speaking exam, which is often about ‘getting to know you’. Have student A’s act as the examiner and B’s as the candidate.

    Set a visible timer according to the exam timings and have students work their way through the questions, simulating a real-life exam. Have ‘the examiners’ think of something their partner does well and something they think they could improve. You can even distribute the marking scheme and allow them to use this as a basis for their peer feedback. Once time is up, ask student B’s to move to the next ‘examiner’ for the next part of the speaking test. Continue this way, then ask students to switch roles.

    Note: If you teach online and your teaching platforms allow it, you can record the conversations and have students review their own performances. However, for privacy reasons, do not save these videos.

    Listening – A student-centered, online activity to practice listening for detail or summarising.

    Ask pairs of students to set up individual online conference call accounts on a platform like Teams or Zoom.

    Have pairs call each other without the video on and tell each other a story or a description of something that has happened for their partner to listen to. This could be a show they’ve watched, an album they’ve listened to, or a holiday they’ve been on, for example. Ask students to write a summary of what their partner has said, or get them to write specific information (numbers, or correctly spelt words) such as character or song names or stats, for example. Begin the next class by sharing what students heard. Students can also record the conversations without video for further review and reflection afterwards.

    Writing –A story-writing group activity to encourage peer learning.

    Give each student a piece of paper and have them draw a face at the top of the page. Ask them to give a name to the face, then write five adjectives about their appearance and five about their personality. You could also have them write five adjectives to describe where the story is set (place).

    Give the story’s opening sentence to the class, e.g. It was a cold, dark night and… then ask students to write their character’s name + was, and then have them finish the sentence. Pass the stories around the class so that each student can add a sentence each time, using the vocabulary at the top of the page to help them.

    Reading –A timed, keyword-based activity to help students with gist.

    Distribute a copy of a text to students. Ask them to scan the text to find specific words that you give them, related to the topic. For example, if the text is about the world of work, ask students to find as many jobs or workplace words as they can in the set amount of time. Have students raise their hands or stand up when they have their answers, award points, and have a whole class discussion on where the words are and how they relate to the comprehension questions or the understanding of the text as a whole.

    All 4 skills –A dynamic activity to get students moving.

    Set up a circuit-style activity with different ‘stations’ around the classroom, for example:

    • Listening
    • Reading
    • Writing (1 paragraph)
    • Use of English (or grammar/vocabulary).

    Set a timer for students to attempt one part from this exam paper, then have them move round to the next station. This activity can be used to introduce students to certain exam tasks, or a way to challenge students once they’ve built their confidence in certain areas.

  • Students sat at a desk looking at a textbook together, with a teacher pointing to it

    Real-world English: How GSE Job Profiles bridge learning and work

    By Sara Davila
    Reading time: 5 minutes

    Did you know that improving your English proficiency can increase earning potential by up to 50% and safeguard your career against AI? Recent research by app highlights that English is not just a skill but a career-defining advantage in today’s globalized workplace. For millions of adult learners, the journey from the classroom to the workplace requires more than general conversational abilities—it’s about gaining targeted, job-ready skills as quickly as possible. For English language educators, understanding what “jDz-𲹻” English is and how to identify “jDz-𲹻” skills can provide a significant advantage in ensuring learners are prepared to communicate effectively and collaborate with their future coworkers.

    That’s where the Global Scale of English (GSE) Job Profiles comes in. For educators and program developers, it offers a bridge between real-world job skills and the English learners need to perform them. Whether you’re creating programs for nursing assistants, hospitality workers, or IT professionals, this tool ensures that learners build the precise English skills they need to thrive in their roles.

    Let’s walk through how to create a GSE Job Profile and explore its practical use for building programs that align with today’s professional realities.

  • A older techer sat with students, in particular looking at a laptop with one of the students

    Teacher technology adoption: What’s age got to do with it?

    By Jennifer Williams
    Reading time: 4 minutes

    Educators, by design, are innovators. Living and working in a constant state of beta, teachers bring to the classroom a natural desire to explore and better understand new practices and methodologies to support students in their search for knowledge.

    Guided by an internal drive to make the world a better place and powered by the rewards of working daily with the youth that can create that positive change, teachers collectively are part of a profession different from any other. Within one classroom on any single day, a teacher can become a scientist, a historian, an inventor, or a scholar. Over time, whether in a one-room schoolhouse a century ago or in a technology enhanced classroom of the digital age, these defining characteristics of “teacher” have seemed to stay constant.

    Bridging the generational and digital divide

    Today, as citizens of our digital and global world, we are in an extraordinary time where divides are closing and collaboration becomes standard – where differences are truly celebrated and associations are surpassing separations in society.As educators, this change reflects what we have consistently asserted in our classrooms.

    Yet, terms like 'generational gap” and “digital divide” suggest a significant separation between us as teachers and our students. In truth, teachers and students are increasingly using asharedlanguage of innovation and exploration. Rather than sticking to notions of a technology-based generational divide, many are transcending the labels of “digital immigrants” and “digital natives”.

    This is further reinforced by current research demonstrating that technology adoption in education has less to do with age and instead is based more on exposure and experience (Bennett & Maton, 2010; Bullen, Morgan, & Qayyum, 2011; Guo, Dobson, & Petrina, 2008; Helsper & Eynon, 2010). As opposed to a focus on divides, this new evidence offers supportive pathways of possibilities to bring together older and younger generations in the process of learning in our world through exploration and wonder.

    Embracing skill development

    Moving attention away from a fixed age factor to emphasizing a continuum of proficiency based on exposure and experience enables teachers of all ages to identify as part of this digital age of education. Though digital technologies are fixtures in the daily lives of many (if not most) of our students, the ways students use technologies are not always consistent.

    Students of today – like students of previous generations – continue to need the guidance of teachers in finding ways to take and apply their understanding of technology to effectively convey perspectives with the world, influence the opinions of others, and contextualize and synthesize information in meaningful ways.

    Timeless lessons from teachers

    Teachers as innovators – young and old, tech-savvy and tech-novice – have extraordinary gifts to bring to their students, who are also navigating through digital environments. Though there are countless lessons teachers bring to students, here are several timeless ones that transcend any generation of teaching. Teachers guide students to use advanced technologies so that they can be...

    Critical consumers

    Today, students in our classrooms are bombarded with information at nearly every moment in time. With basic searches for information often leading students in multiple directions, research at every level has now become a highly complex and sophisticated process. The result is that we now see learning practices shifting from finding answers to discovering more questions.

    Teachers can guide students to evaluate and consume information critically. By promoting transparency and sharing, they help learners consider context and sources, reason with evidence and build knowledge.

    Creators of rich content

    As classrooms advance as blended learning environments, seamless integration of technologies and innovative teaching practices offers students opportunities to transform from passive learners to active content creators. With countless digital tools available for content creation, students can seek guidance from teachers on matching the best tool to a particular communicative purpose.

    Teachers who are skillful in creating synthesized messages can show students ways to comprehend and compose with combinations of digital technologies. Less in the role of directing lessons, the teacher becomes a catalyst for learning, inspiring students to work creatively and collaboratively to solve problems and respond to instructional lessons as producers of information.

    Centered on relationships

    A defining quality of a master teacher is the ability to create a positive learning community built on empathy and understanding. The focus on relationships in these classrooms models the continued importance of valuing alliances and friendships.

    Soft skills, such as joy, honesty, trust and respect, become even more essential as our students continue to interact in a highly networked and diverse world. Balance, too, becomes a critical area for discussion as students look to divide attention, time and social-emotional reserves between onscreen and in-person experiences.

    Within responsive classrooms and digital learning spaces that keep emphasis on the formation of relationships, teachers can offer opportunities for students to develop through collaboration, teamwork, solidarity and conflict resolution.

    Resilient in endeavors

    Our students are looking out onto a world where anything is possible. Digital age technologies are empowering students to believe: “if you can dream it, you can do it”. The impossible is within reach, and a mindset of grit and determination moves sparked ideas to a place of accomplishment and excellence.

    Through the creation of learner-driven classrooms centered on the student, teachers can help guide students to narrow their interests and ultimately find their passions. With relevance and authentic purpose, teachers can connect students in the process of learning through design thinking and trial and error – shifting the outlook from fear of failure to resilience and 'stick-to-itiveness' to advance learning to a place of discovery and invention.

    And it is this mindset – a position of inquiry and exploration – held by students and teachers, young and old, that can show that great will come from risk. We all just need to be ready to try.