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

  • Two school girls getting onto. a schoolbus

    4 steps to make sure the transition back to school goes well

    By Donatella Fitzgerald MBE

    As we start thinking about returning to school, the big question for teachers and parents is: How can we help our children get off to a smooth start?

    After the long break, students might be keen to see their friends again. But it's not always easy to get back into the class routine. It's especially difficult when students are moving into a newclass orare facing important exams.

    So what can you do to support children in transitioning back to school at the beginning of the academic year?

    Here are 4 steps to help them get off to a great start.

    1. Be organized and create routines

    Organizing is what you do before you do something, so that when you do it, it's not all mixed up. – A. A. Milne

    Have a "family meeting" to brainstorm the routines for the school year,e.g., tidying room, getting school clothes/uniform/bag ready the evening before, homework routines, familymeal times and exercise. Ensure each member of the family has input into the routine too. Create a chart with the routines so everyone can see what has been agreed upon and how they are being adhered to.

    2. Start bedtime routines early

    Early to bed aԻ early to rise, makes a man healthy, wealthy, and wise. – Benjamin Franklin

    Agreeing on firm bedtimes during the school week is very important for everyone in the family.

    If children and teenagers do not get enough sleep, it can negatively impact their healthand academic achievement. It's important that they are aware of this too.

    In an ideal world, we should ensure that adults and children are not exposed to any form of screen time for at least 30 minutes to one hour before bedtime. This will help everyone sleep better.

    Start the back-to-school sleep transition gradually. To help your child adjust, move bedtime up by 30 to 60 minutes at a time over the course of a few days or a week before the start of school so the transition from a later bedtime to an earlier one is progressive.

    3. Talk about homework and teach organizational skills

    When it's obvious that the goals cannot be reached, don't adjust the goals, adjust the action steps. – Confucius

    Organization needs to be taught and practiced. As children get older, homework usually increases. Take an active interest in your child's homework.

    Parents can be supportive by demonstrating organizational skills and helping children with time management. Assist them with creating a plan for their homework and encourage your child (whatever age) to have a study plan of some sort and to set goals for their homework which are SMART:

    • Specific – Describe in detail what activities they are going to do.
    • Measurable – How will they know when they are progressing/finished?
    • Achievable – Do they have the skills and resources to get to their goal?
    • Relevant – How does this goal connect to short and long-term targets?
    • Time-bound – Set a concrete deadline.

    Teach them how to approach homework with a "strategy". How much homework do they have? How long will it take to do it? Prioritize urgent homework and do more difficult things when they are less tired.

    Make sure there's adequate space in your house and set up a homework-friendly area that is welllit, andhas a table with enough room to put their pens and books. Of course, this is preferably somewhere quiet.

    Also, it's important that parents are motivators and monitors and that they try to make themselves available for advice. Always praise children for their work and efforts. If you spot any problems, try and address them. Keep distractions to a minimum at home while they are doing their homework.

    4. Make mealtimes quality family time: listen and share one good thing about each day

    Kids who grow up having family dinners, when they're on their own, tend to eat more healthily. – Anne Fishel 

    A meal around the table can bring to the entire family and be an important opportunity for daily interaction. Sitting down to eat as a family provides the opportunity to have an influence over both short and long-term familyhealth, andcan help children establish resilience and the ability to cope with the demands of life as we know it now and in the future. It can also be an opportunity to introduce mindful eating too – being more aware of what and how we are eating.

    Additionally, this time together allows for members to talk and share things about their day and also offers an opportunity to establish a strong and powerful bonding experience. How can we make sure family mealtimes are quality time?

    • Be attentive and offer undivided attention during this time.
    • Turn all modern technology off during the meal so everyone is focused on each other.
    • Talk to each other about topics such as: What lessons do you have today (at breakfast)? What did you enjoy about today? What did you have for lunch (while sharing the evening meal). Tell me one thing you learned today. What made you laugh today? What made you happy today?
    • Listen mindfully to your child's thoughts and worries (if any).
    • Encourage each member to talk about one good thing that has happened to them that day. This lets them know you are there for them.
    • Assign mealtime jobs to involve everyone, e.g.: setting and clearing the table and putting away the dishes.

    However, sometimes it's not possible to share meal times during the week so plan at least one on the weekend if possible.

    The benefits of any small moment of time can have long-lasting positive influences on your child's mental and physical health. Children model adultbehaviorand if they see you eating and engaging positively with them and others, they will carry this into their own lives.

    With a bit of preparation, the leadup and transition back to school can be smooth and enjoyable not only for children but also for the rest of the family.

  • friends chatting around a laptop

    Realistic practice scenarios to boost your English skills

    By app Languages

    It's essential to engage in real-life conversations to truly become proficient in a language. By immersing yourself in realistic situations, you can really boost your English skills. Of course, it can be difficult to find people to speak to if you aren’t in an area with many English speakers. By practicing different possible scenarios, when the time or situation arises, you know you’ll be ready to take on the challenge.

    You might be preparing to work or study abroad, trying to connect with a new friend or improving your workplace communication. Let’s explore some realistic practice scenarios that can help you feel more confident and fluent when speaking English in your everyday life.

  • A server standing and smiling at a group of ladies smiling sat at a table

    Language and employability skills: Critical thinking, creativity, and communication

    By

    Why learn a language?

    For most, it's part of academic studies. For some, it's a fun social opportunity. But for many, language learning is aimed at getting a job.

    Language teachers didn't always consider the reasons students were motivated to learn a language. Instead, they focused solely on the central parts of language learning: phonology (sounds of letters and words), morphology (the meaning of parts of words), lexicon (vocabulary), grammar (word order) and to a lesser extent, discourse (the intent of language).

    But today, beyond the mechanical aspects of language teaching and learning, language teachers and their teaching and learning materials try to align with students' motivations. This includes exploring a wide variety of social issues from global warming to racism to homelessness. Reasons for teaching these issues are based on the notion that language is culture, and students want to learn broad topics and be able to contribute to conversations about the issues of the day.

    Employability skills

    A related challenge facing students is employability skills. In the past, students were largely taught the types of language expected of factory workers: giving and responding to simple instructions. Most students learning via the audio-lingual method would consider the question "How are you?" to always be answered with the response, "I'm fine, thank you." The reality, of course, is that you might just as well say, "I'm okay." "Can't complain!" "Not too bad." or even the little-used but truthful, "I feel terrible!"

    The Communicative Approach challenged this pre-programmed speech and reflected changes in the workplace. As robots and artificial intelligence agents take over more and more factory work, today's language students are graduating into jobs that require critical thinking, creativity, and broad communication skills. What are these skills and how do they relate to employability?

    Critical thinking is about examining problems to better understand them. Sometimes critical thinking helps students make choices between one or more alternatives. Like creativity and communication, critical thinking is vital in both academic and employment situations where, for example, staff might try to decide between two locations to build a new factory.

    Creative thinking is about looking for new solutions. In the factory example, a solution might be to build a factory on a boat so it travels between where the raw materials are collected to the market where they're to be sold.

    Communication is about explaining ideas, listening to others' views, and using persuasive speaking and writing to structure arguments. Is the factory boat the best idea? It might be, but without clear communication and debate, it will be tossed aside.

    In terms of employability, the app series Step Up outlines the varied needs faced by adult learners: "to improve their employability skills to get their first job, secure a promotion, find a different job, re-enter the workforce after an absence or change fields."

    Meeting these needs requires new teaching and assessment approaches.

    Be collaborative

    Teaching has to become more collaborative. This reflects the nature of modern work, where most people work in teams, rather than in the factory model where workers were interchangeable parts of a machine. Workers today need to identify problems, share ideas about how to solve them and negotiate, using critical and creative thinking.

    Assess positively

    Similarly, assessment needs to change to a model that allows students opportunities to show what they know in open-ended ways with multiple opportunities to achieve success. Tests with closed-ended questions aimed at tricking students are a thing of the past. Assessment today needs to present students with chances to learn and try again and again until they and their teachers are confident of their abilities.

    Learning a language and related abilities, like employability skills, is not a narrow classroom-bound experience. Students continue to learn and improve throughout their lives. More than anything else, the role of today's teachers is to set their students on a path of lifelong learning.