Explaining computerized English testing in plain English

ɫèAV Languages
a pair of hands typing at a laptop

Research has shown that automated scoring can give more reliable and objective results than human examiners when evaluating a person’s mastery of English. This is because an automated scoring system is impartial, unlike humans, who can be influenced by irrelevant factors such as a test taker’s appearance or body language. Additionally, automated scoring treats regional accents equally, unlike human examiners who may favor accents they are more familiar with. Automated scoring also allows individual features of a spoken or written test question response to be analyzed independent of one another, so that a weakness in one area of language does not affect the scoring of other areas.

was created in response to the demand for a more accurate, objective, secure and relevant test of English. Our automated scoring system is a central feature of the test, and vital to ensuring the delivery of accurate, objective and relevant results – no matter who the test-taker is or where the test is taken.

Development and validation of the scoring system to ensure accuracy

PTE Academic’s automated scoring system was developed after extensive research and field testing. A prototype test was developed and administered to a sample of more than 10,000 test takers from 158 different countries, speaking 126 different native languages. This data was collected and used to train the automated scoring engines for both the written and spoken PTE Academic items.

To do this, multiple trained human markers assess each answer. Those results are used as the training material for machine learning algorithms, similar to those used by systems like Google Search or Apple’s Siri. The model makes initial guesses as to the scores each response should get, then consults the actual scores to see well how it did, adjusts itself in a few directions, then goes through the training set over and over again, adjusting and improving until it arrives at a maximally correct solution – a solution that ideally gets very close to predicting the set of human ratings.

Once trained up and performing at a high level, this model is used as a marking algorithm, able to score new responses just like human markers would. Correlations between scores given by this system and trained human markers are quite high. The standard error of measurement between ɫèAV’s system and a human rater is less than that between one human rater and another – in other words, the machine scores are more accurate than those given by a pair of human raters, because much of the bias and unreliability has been squeezed out of them. In general, you can think of a machine scoring system as one that takes the best stuff out of human ratings, then acts like an idealized human marker.

ɫèAV conducts scoring validation studies to ensure that the machine scores are consistently comparable to ratings given by skilled human raters. Here, a new set of test-taker responses (never seen by the machine) are scored by both human raters and by the automated scoring system. Research has demonstrated that the automated scoring technology underlying PTE Academic produces scores comparable to those obtained from careful human experts. This means that the automated system “acts” like a human rater when assessing test takers’ language skills, but does so with a machine's precision, consistency and objectivity.

Scoring speaking responses with ɫèAV’s Ordinate technology

The spoken portion of PTE Academic is automatically scored using ɫèAV’s Ordinate technology. Ordinate technology results from years of research in speech recognition, statistical modeling, linguistics and testing theory. The technology uses a proprietary speech processing system that is specifically designed to analyze and automatically score speech from fluent and second-language English speakers. The Ordinate scoring system collects hundreds of pieces of information from the test takers’ spoken responses in addition to just the words, such as pace, timing and rhythm, as well as the power of their voice, emphasis, intonation and accuracy of pronunciation. It is trained to recognize even somewhat mispronounced words, and quickly evaluates the content, relevance and coherence of the response. In particular, the meaning of the spoken response is evaluated, making it possible for these models to assess whether or not what was said deserves a high score.

Scoring writing responses with Intelligent Essay Assessor™ (IEA)

The written portion of PTE Academic is scored using the Intelligent Essay Assessor™ (IEA), an automated scoring tool powered by ɫèAV’s state-of-the-art Knowledge Analysis Technologies™ (KAT) engine. Based on more than 20 years of research and development, the KAT engine automatically evaluates the meaning of text, such as an essay written by a student in response to a particular prompt. The KAT engine evaluates writing as accurately as skilled human raters using a proprietary application of the mathematical approach known as Latent Semantic Analysis (LSA). LSA evaluates the meaning of language by analyzing large bodies of relevant text and their meanings. Therefore, using LSA, the KAT engine can understand the meaning of text much like a human.

What aspects of English does PTE Academic assess?

Written scoring

Spoken scoring
  • Word choice
  • Grammar and mechanics
  • Progression of ideas
  • Organization
  • Style, tone
  • Paragraph structure
  • Development, coherence
  • Point of view
  • Task completion
  • Sentence mastery
  • Content
  • Vocabulary
  • Accuracy
  • Pronunciation
  • Intonation
  • Fluency
  • Expressiveness
  • Pragmatics

More blogs from ɫèAV

  • What’s it like to teach English in Nepal?

    By
    Reading time: 3 minutes

    Anandi Vara was trained in teaching English in Kathmandu, Nepal before teaching at a monastery in Pokhara. There she taught students ranging from six to 10 years of age, both in groups and individually. Here she reveals the lessons she learned during her time there– including how to avoid being perturbed by a cockroach attack.

    Whatever you do: don’t freak out

    It just makes everything worse. It’s easy for things to get overwhelming – a sense that can be made worse by the feeling of homesickness, especially if it’s your first time living abroad – but thinking rationally, and getting to the source of what’s causing the worry, usually helps. It’s important, as you don't want to share your fear in lessons because you’re the teacher and need to show confidence.

    This was, however, tested to the limit when I had a cockroach dangled in front of my face. It took all my strength to stay calm. I gave an unimpressed look, thereby establishing myself as the figure of authority, which seemed to work.

    Be aware of cultural traditions

    It is important to remember that every country has its own traditions. I was teaching in a monastery, so I made sure to wear respectful clothing, even in the face of soaring temperatures. The more I learned about the Tibetan culture, the more fascinated I became by it. The students taught me how to write my name in Tibetan and the meaning behind it. I learned about Tibetan history and Tibetan culture.

    I also found that the more I showed willingness to learn about the Tibetan culture, the more I bonded with the students, so that when it was time to teach, the students were more cooperative in lessons, engaging and participating more.

    Teaching is two-way learning

    There is so much I learned teaching abroad, both in the classroom and out. Making mistakes as you begin is only natural, but it’s whether you can learn from these mistakes that counts. No two students are ever the same so it’s a constant process of learning as you go. As a result, I learned about the environment I was in – from traditional prayer ceremonies to the Tibetan alphabet– and about myself, notably organizational skills and a renewed curiosity about the English language.

    Be Flexible

    Sometimes it doesn’t matter if you’ve planned your lesson down to the smallest detail – if it doesn’t take, then it doesn’t take. I was only 10 minutes into a lesson once and I could tell that I was beginning to lose the students’ attention. Not only did it show that they were uninterested, it also distracted me from what I was doing. It was at this point that I threw out my existing plan and tried a whole new lesson: I had the students up on their feet and engaging with each other and, although completely improvised, it was very successful.

    Patience is a must

    During my one-on-one mentoring session, my student seemed to have no motivation. He wasn’t learning as well as the other students and had therefore given up. No matter what I tried, he refused to cooperate, but I didn’t let it put me off. I kept trying different methods until finally finding one that he responded to. I made sentence structuring into a game. It wasn’t anything fancy and consisted of scraps of paper with words written on them.

    Although it took a lot of time to find the right angle, it was worth it because he soon realised that although it might take longer for him to pick things up, he would eventually get there and have a greater sense of accomplishment.

  • Children engaged in a classroom activity, with colorful educational posters and a banner in the background.

    How to use flipped learning to support your learners

    By
    Reading time: 6 minutes

    What is flipped learning?

    To understand better what flipped learning is, first let’s see how it differs from blended learning, a term with which it is often confused.

    Blended learning is a way of teaching that combines face-to-face classroom teaching with online resources. We freely use online resources to create a more personalized learning experience.

    Flipped learning is a little different. As the name suggests, it "flips" a traditional lesson. It tells us exactly which stages of the lesson should go online. In a flipped learning class, all of the more traditional aspects (also called "study stages"), are completed online and the homework (also called "application stages") comes into the classroom.

    Let’s look at an example.

    A typical receptive skills lesson normally has six stages:

    1. Lead in
    2. Set context
    3. Pre-teach vocabulary
    4. Gist task
    5. Detailed task
    6. Follow up

    With a traditional teaching model, we do the first five stages in class and set the last one for homework. With flipped learning there are a few ways to tackle these stages, but a basic model would look like this:

  • Precision teaching with AI: Aligning GSE objectives with generative AI for targeted materials

    By
    Reading time: 4 minutes

    English teachers today face increasing demands: create engaging content, differentiate instruction and address diverse learner needs – all within a limited time. The rise of Generative AI, like ChatGPT, offers a promising solution. But without proper guidance, AI-generated content can lack educational value. This blog post introduces a practical, research-informed approach to using AI tools aligned with the Global Scale of English (GSE). You will learn how this framework helps educators design accurate, personalized and level-appropriate English teaching materials quickly and confidently.

    Why GSE and AI are a game-changing combination for ELT

    The Global Scale of English (GSE) is a CEFR-aligned framework developed by ɫèAV, offering detailed "can-do" learning objectives. It includes nearly 4,000 descriptors across speaking, listening, reading and writing skills, offering more precision than traditional level labels like A2 or B1. At the same time, Generative AI tools such as ChatGPT can generate entire lessons, tasks and assessments in seconds. The challenge lies in ensuring this content is aligned with clear pedagogical outcomes.

    Pairing AI’s creative speed with the GSE’s structured outcomes offers a scalable way to meet learner needs without compromising instructional quality.

    Unlocking measurable, differentiated and efficient teaching with GSE and AI

    The GSE makes objectives measurable

    Unlike generic teaching goals, GSE objectives are specific and measurable. For example, a B1-level learner objective might state:

    “Can identify a simple chronological sequence in a recorded narrative or dialogue.” (GSE 43)
    This clarity helps teachers define outcomes and ensure each AI-generated task targets an actual language skill, not just generic content.

    Generative AI enhances productivity

    Teachers using Generative AI can create draft lesson materials in minutes. By inputing a structured prompt such as:

    “Create a B1 reading activity that helps learners summarize the main points of a short article.”
    ChatGPT can instantly generate content that meets the learning goal. When guided by the GSE, AI becomes a collaborative assistant as well as a time-saver.

    The GSE + AI combination supports differentiation

    Because the GSE includes descriptors across a wide proficiency range (from pre-A1 to C2), teachers can tailor AI-generated content to meet the exact needs of their students. Mixed-level classrooms or tutoring contexts benefit especially from this, as teachers can create multiple versions of a task with consistent scaffolding.

    Practical tips

    • Use the GSE Teacher Toolkit to select objectives based on skill, level or function.
    • When prompting ChatGPT, include the GSE descriptor in your input for more precise results.
    • Always review and adapt the AI output to match your learners’ context, culture and curriculum.
    • Create a prompt library mapped to GSE codes to save time in future planning.

    A step-by-step example of the GSE and AI in action

    Here is a typical application of the workflow:

    1. A teacher selects a GSE objective, such as:
      “Can write a basic formal email/letter requesting information.” (GSE 46).
    2. Within seconds, a sample formal email, accompanied by a short reading comprehension task and a vocabulary activity, is generated.
    3. The reading task serves as a model to help learners analyze the structure, tone, and key language features of a well-written email before attempting their own.
    4. The teacher then reviews and refines the output for clarity, appropriateness, and context relevance.

    This process supports targeted teaching while significantly reducing preparation time.

    Overcoming challenges: Ensuring quality and relevance

    Challenge: AI outputs may lack cultural context, level appropriateness or instructional clarity.
    Solution: Always pair AI with professional judgment. Use the GSE to check that skills match the intended outcome, and adjust the complexity of the language as needed.

    Challenge: Teachers may be unfamiliar with how to write effective AI prompts.
    Solution: Start simple with templates like:

    “Create a [skill] activity at [level] that supports this GSE objective: [insert objective].”

    Challenge: Risk of over-relying on AI for instruction.
    Solution: Use AI as a starting point, not the final product. Combine AI-generated content with classroom interaction, feedback and your own creativity.

    Teaching tools that make this easier

    • : for exploring and selecting level-appropriate learning objectives
    • : for generating customizable teaching content
    • GSE Smart Lesson Generator: an AI-powered lesson creation tool developed by ɫèAV that uses the GSE framework to automatically generate high-quality activities and lesson plans
    • Google Docs or Word: for editing and organizing your materials before class

    Confidently transforming English teaching

    Combining Generative AI with the Global Scale of English allows teachers to design materials that are both fast and focused. The GSE provides the structure; AI provides the speed. Together, they offer a sustainable solution for personalized English instruction that respects both learner needs and instructional quality.