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

  • A group of students stood around a teacher on a laptop

    The ethical challenges of AI in education

    By Billie Jago
    Reading time: 5 minutes

    AI is revolutionising every industry, and language learning is no exception. AI tools can provide students with unprecedented access to things like real-time feedback, instant translation and AI-generated texts, to name but a few.

    AI can be highly beneficial to language education by enhancing our students’ process of learning, rather than simply being used by students to ‘demonstrate’ a product of learning. However, this is easier said than done, and given that AI is an innovative tool in the classroom, it is crucial that educators help students to maintain authenticity in their work and prevent AI-assisted ‘cheating’. With this in mind, striking a balance between AI integration and academic integrity is critical.

    How AI impacts language learning

    Generative AI tools such as ChatGPT and Gemini have made it easier than ever for students to refine and develop their writing. However, these tools also raise concerns about whether submitted texts are student-produced, and if so, to what extent. If students rely on text generation tools instead of their own skills, our understanding of our students’ abilities may not reflect their true proficiency.

    Another issue is that if students continue to use AI for a skill they are capable of doing on their own, they’re likely to eventually lose that skill or become significantly worse at it.

    These points create a significant ethical dilemma:

    • How does AI support learning, or does it (have the potential to) replace the learning process?
    • How can educators differentiate between genuine student ability and AI-assisted responses?

    AI-integration strategies

    There are many ways in which educators can integrate AI responsibly, while encouraging our learners to do so too.

    1.Redesign tasks to make them more ‘AI-resistant’

    No task can be completely ‘AI-resistant’, but there are ways in which teachers can adapt coursebook tasks or take inspiration from activities in order to make them less susceptible to being completed using AI.

    For example:

    • Adapt writing tasks to be hyperlocal or context-specific. Generative AI is less likely to be able to generate texts that are context-bound. Focus on local issues and developments, as well as school or classroom-related topics. A great example is having students write a report on current facilities in their classroom and suggestions for improving the learning environment.
    • Focus on the process of writing rather than the final product. Have students use mind maps to make plans for their writing, have them highlight notes from this that they use in their text and then reflect on the steps they took once they’ve written their piece.
    • Use multimodal learning. Begin a writing task with a class survey, debate or discussion, then have students write up their findings into a report, essay, article or other task type.
    • Design tasks with skill-building at the core. Have students use their critical thinking skills to analyse what AI produces, creatively adapt its output and problem solve by fact-checking AI-generated text.

    2.Use AI so that students understand you know how to use it

    Depending on the policies in your institution, if you can use AI in the classroom with your students, they will see that you know about different AI tools and their output. A useful idea is to generate a text as a class, and have students critically analyse the AI-generated text. What do they think was done well? What could be improved? What would they have done differently?

    You can also discuss the ethical implications of AI in education (and other industries) with your students, to understand their view on it and better see in what situations they might see AI as a help or a hindrance.

    3.Use the GSE Learning Objectives to build confidence in language abilities

    Sometimes, students might turn to AI if they don’t know where to start with a task or lack confidence in their language abilities. With this in mind, it’s important to help your students understand where their language abilities are and what they’re working towards, with tangible evidence of learning. This is where the GSE Learning Objectives can help.

    The Global Scale of English (GSE) provides detailed, skill-specific objectives at every proficiency level, from 10 to 90. These can be used to break down complex skills into achievable steps, allowing students to see exactly what they need to do to improve their language abilities at a granular level.

    • Start by sharing the GSE Learning Objectives with students at the start of class to ensure they know what the expectations and language goals are for the lesson. At the end of the lesson, you can then have students reflect on their learning and find evidence of their achievement through their in-class work and what they’ve produced or demonstrated.
    • Set short-term GSE Learning Objectives for the four key skills – speaking, listening, reading and writing. That way, students will know what they’re working towards and have a clear idea of their language progression.
  • A teachet stood in front of a class in front of a board, smiling at his students.

    How to assess your learners using the GSE Assessment Frameworks

    By Billie Jago
    Reading time: 4 minutes

    With language learning, assessing both the quality and the quantity of language use is crucial for accurate proficiency evaluation. While evaluating quantity (for example the number of words written or the duration of spoken production) can provide insights into a learner's fluency and engagement in a task, it doesn’t show a full picture of a learner’s language competence. For this, they would also need to be evaluated on the quality of what they produce (such as the appropriateness, accuracy and complexity of language use). The quality also considers factors such as grammatical accuracy, lexical choice, coherence and the ability to convey meaning effectively.

    In order to measure the quality of different language skills, you can use the Global Scale of English (GSE) assessment frameworks.

    Developed in collaboration with assessment experts, the GSE Assessment Frameworks are intended to be used alongside the GSE Learning Objectives to help you assess the proficiency of your learners.

    There are two GSE Assessment Frameworks: one for adults and one for young learners.

    What are the GSE Assessment Frameworks?

    • The GSE Assessment Frameworks are intended to be used alongside the GSE Learning Objectives to help teachers assess their learners’ proficiency of all four skills (speaking, listening, reading and writing).
    • The GSE Learning Objectives focus on the things a learner can do, while the GSE Assessment Frameworks focus on how well a learner can do these things.
    • It can help provide you with examples of what proficiencies your learners should be demonstrating.
    • It can help teachers pinpoint students' specific areas of strength and weakness more accurately, facilitating targeted instruction and personalized learning plans.
    • It can also help to motivate your learners, as their progress is evidenced and they can see a clear path for improvement.

    An example of the GSE Assessment Frameworks

    This example is from the Adult Assessment Framework for speaking.

    As you can see, there are sub-skills within speaking (andfor the other three main overarching skills – writing, listening and reading). Within speaking, these areproductionandfluency, spoken interaction, language range andaccuracy.

    The GSE range (and corresponding CEFR level) is shown at the top of each column, and there are descriptors that students should ideally demonstrate at that level.

    However, it is important to note that students may sit across different ranges, depending on the sub-skill. For example, your student may show evidence of GSE 43-50 production and fluency and spoken interaction, but they may need to improve their language range and accuracy, and therefore sit in a range of GSE 36-42 for these sub-skills.

  • A child sat with a teacher with a tablet

    Writing your own English language materials with the GSE

    By Billie Jago

    Being an English language teacher means you’re also probably (definitely) a materials writer. You likely tailor or create language materials for your students that are suited to their needs and interests, either as supplements to your course materials or for communicative lessons. Alternatively, you might be a teacher who creates paid, published materials available for students worldwide to enjoy.

    With this in mind, think of the materials you’ve developed and ask yourself the following:

    • How do you level your grammar or vocabulary for the content you write?
    • How do you find topic-related vocabulary to extend your students’ knowledge of language?
    • How do you contextualize new grammar or vocabulary?

    You can use many different resources, from online dictionaries to course workbooks to a Google search. Still, the Global Scale of English is a reference that provides everything you need to write great learning materials, all in one place. It can help save you valuable time as a teacher and materials writer.

    For me, the GSE was a game changer as an English teacher, and it continues to be as I write materials. The GSE is not just a tool; it’s a companion in the complex journey of material development, offering clarity and direction at every step. It can guide you in creating effective, engaging learning resources.

    How to use the GSE toolkit to create your own materials

    1. Establishing clear Learning Objectives

    helps you start with a clear roadmap. It provides detailed descriptors for language proficiency at every level, ensuring your materials align with specific learning objectives. For instance, if you’re creating a beginner-level reading comprehension activity, the GSE descriptors will guide you on the appropriate complexity of vocabulary and sentence structures.

    Take a look at the Learning Objectives tab in the GSE Toolkit to learn more.

    2. Designing level-appropriate content

    Once objectives are set, the GSE assists in tailoring the content difficulty to the targeted proficiency level. Its numerical scale, ranging from 10 to 90, allows you to pinpoint the exact level of language skills required and design your materials accordingly. This precision ensures that learners are neither overwhelmed nor under-challenged.

    You can set the level you are looking for by sliding the bar along the scale, so it corresponds to the appropriate CEFR level or GSE range.