KS2 SATs 2025: reading paper analysis

After thousands of young people took the KS2 SATs 2025 reading paper, we had to ask SATs expert analyser extraordinaire Sophie Bartlett for her take. So here it is, the post-SATs reading paper analysis from Sophie B - did the paper hit the mark?

If you would like a slides version of this blog as well as GPS and maths papers analysis to show to colleagues and friends, sign up to our mailing list and get them sent to your inbox.

Sophie Bartlett’s SATs 2025 reading paper analysis

Yes, I know. No poem. Again.

This is future-Sophie, telling past-Sophie that this introduction that I wrote pre-embargo is going to make me laugh because - guess what - I’ve since read the mark scheme. Read the naive “I-haven’t-seen-the-mark-scheme-yet” Sophie’s views here:

“Anyway, let’s talk about the elephant in the room… I liked that reading paper. Never in a million years (or 10, really, as that’s how many it’s been since the first SATs paper since the curriculum change) did I think I’d say that. You only have to skim last year’s analysis to see me describe the 2024 reading paper as ‘confusing’, ‘mean’ and ‘challenging’ (an opinion shared by many other Year 6 teachers at the time). 2023’s paper was so bad that this BBC article was written days after children sat the test, describing how ‘distraught’ the children were, and the papers were released earlier than they ever have been due to ‘public interest’ in them.

This year, however, I thought they were reasonable, and not even that boring! Watch me bite my tongue when the pass mark comes out as something ridiculous like 70%...”

To be fair, I stand by the fact that the texts and questions themselves were okay. But the rigid, preposterous mark scheme? A true celebration of form over substance - because what’s education without a bit of pedantry masquerading as rigour?

Reading paper 2025 facts and figures

In the reading paper, the children are tested on eight content domains.

Retrieval and inference have been the most represented content domains in every SATs paper. The proportion of retrieval questions dropped this year to 30%, whilst inference questions were at the highest they’ve ever been - nearly half the paper!

Related content:

Get this blog and the GPS & maths analysis in slide format - sign up to our mailing list, today

KS2 SATs 2025: GPS papers analysis

KS2 SATs 2025: maths papers analysis

Point 3.2 in the KS2 English reading test framework states that the paper will consist of ‘a selection of texts, [between] 1500–2300 words’. The lowest word count we’ve seen was in 2018 (although this also came with one of the highest pass marks); this year, the total word count was around 1,900 (rounded to the nearest 100).

Point 6.2.1 in the framework says ‘a range of texts will be included in the tests, including fiction, non-fiction and poetry’. I think I’ve been predicting that a poem will appear for about three years now… only one poem has appeared in a SATs paper under the new curriculum! This year’s structure followed the same as last year’s: one non-fiction, followed by two fiction.

The test framework also states that there could be up to four 3-mark questions in the paper, but every year so far, we have only seen two.

If we combine the number of words in the question booklet (usually around 1,000), that takes our word count total up to 2,900, which sits nicely in the middle of previous word counts we’ve seen before.

The question is - what will the pass mark do this year? It’s been on a turbulent ride from 2017, increasing to its highest in 2022 at 58%, only to drop by a huge 10% the following year. It has since been on the rise… only time will tell where it’ll end up in 2025!

The SATs reading paper questions

Point 6.2 of the test framework says that ‘the questions are placed in order of difficulty, where possible, while maintaining chronology with the text’. I personally found all the questions to be very fair, and would even go as far to say I enjoyed answering some of them! (Does that make me a bit sad? Probably.)

Analysis of extract 1: A Life-changing Game (non-fiction)

The ‘easiest’ text in the test was a non-chronological report about a Ugandan girl who learned to play chess. It was just under 600 words (similar to other first extracts in past papers), and had 14 marks allocated to it (the lowest amount we’ve seen, in line with 2019 and 2023). I believe there is actually a film about Phiona’s story, called Queen of Katwe.

Upon initial reading, I noticed a few words that might trip some pupils up: maize, intriguing, recklessly and anticipation (I was clearly onto something here, as the children were tested on the word ‘intriguing’ in Q2): this just consolidates what we already know - we must continue to expose children to a wide range of vocabulary.

Some of the retrieval questions weren’t quite as ‘retrieval-y’ as you may expect - for example, the first question required the children to relate the phrase ‘make money’ (in the question) to ‘no longer afford’ and ‘support’ (in the text), as opposed to just searching for the word ‘money’.

1) How did Phiona make money for her family? [1 mark]

‘...her family could no longer afford to send her. She started selling… to support her family.’

Having a fluent understanding of vocabulary is a running theme: question 3 expects the children to understand that being ‘not familiar’ (in the question) with something is synonymous to it being ‘unusual’ (in the text).

3) How can you tell that people in Uganda were not familiar with chess? [1 mark]

‘It was so unusual in Uganda at that time, there was no word for it in Phiona’s language.’

The ability to fluently switch between semantically similar language is a key aspect of competent reading. Here are some more examples of how children are required to use this knowledge in order to answer the question in the first text.

5) Look at page 5. According to the text, what is it important to have when learning something new?

And this is where madness ensues. You’ll notice another running theme throughout the rest of this analysis: the absolute ludicrousness of the mark scheme.

• An accepted answer to this question? → ‘someone to help you learn’

• An unacceptable answer? → ‘someone to help you’

Why on earth would the second answer not be accepted? The word ‘learn’ is literally in the question, so surely it doesn’t need repeating? I don’t want to get too enraged for this one, because there is plenty more of this stupidity to come.

8) Look at the section: Coaching a champion. Why did Phiona play carelessly when she started playing chess?

• An accepted answer to this question? → Reference to Phiona being too eager to win, e.g. ‘she’s so eager’

• An unacceptable answer? → Reference only to Phiona’s desire to win, e.g. ‘she’s eager to win’

There are only three words that are different between these answers: ‘so’ and ‘to win’. You could argue (if you were being pedantic / annoying) that the ‘correct’ answer didn’t include ‘to win’ - but the reference acknowledges that this point is correct. So that leaves the ‘so’ - now you tell me the difference between ‘she’s so eager’ (‘correct’) and ‘she’s eager’ (‘incorrect’) - is there something wrong in diminishing her apparent eagerness?

10) Look at the section: A roaring success. Phiona’s skills in chess took her abroad. Why?

An acceptable answer to this was any reference to Phiona becoming an international chess champion - but that just doesn’t make sense. She didn’t travel abroad because she became an international chess champion. To me, this is far more unacceptable than some of the actual unacceptable answers (‘she was really good at it’, etc.)

Analysis of extract 2: In the Cave (fiction)

The middle text was an extract about two boys exploring a cave (which, after a little searching, I found to be taken from the story Aquila by Andrew Norriss). I actually really enjoyed reading this one - it felt perfectly readable and actually mildly interesting! It was 550 words long (amongst one of the shortest second extracts we’ve seen), and had 18 marks allocated to it. We also had our first 3-marker at Q26 - a lovely one, might I add, as any Year 6 teacher will know that teaching ‘impressions + evidence’ is a pretty standard practice in a Year 6 classroom!

Some great vocabulary again in this text: beckoning, intently, inaudible and reluctantly (the first and last of which children were tested on in Q17 and Q27 respectively).

Analysis of extract 3: Longbow Girl (fiction)

The final, and most complex, text was an extract from a story about a young archer (from the story Longbow Girl by Linda Davies). It was just over 700 words long and had 18 marks allocated to it. It also contained the second, and final, 3-mark question - which I did feel was quite tricky (but not mean!). As expected from a final text, there were some more complex words and turns of phrase: dimly aware, jeers, marshal, keep at bay, spectacle, the air of ridicule and flush of euphoria to name a few (mind you, all lovely bits of vocab).

A particular favourite question of mine was question 31 - I just found it so satisfying! Read the text, understand the text, picture it in your head and then identify the correct image. Chef’s kiss.

35b) Why were the rules changed after the first round? Write one reason.

The difference between the accepted and unacceptable answers here is that the answer must reference something being increased, or ‘more’ (referencing either the audience’s increased appreciation of the archers’ skills, the archers’ increased level of challenge, or the crowd’s increased enjoyment).

• An accepted answer to this question? → ‘to make it more interesting to watch’

• An unacceptable answer? → ‘to make it interesting to watch’

You can fight me on this, but I think it’s clear that the child who wrote that second answer does understand the question. Aren’t markers supposed to be assessing ‘positively’? To me, reading comprehension is just too subjective for this kind of rigid marking. It’s like asking different people what a particular element in a piece of artwork represents - there isn’t always a clear ‘right’ or ‘wrong’ answer.

Sure, there are definitely incorrect answers, but this wasn’t one of them. Losing 2% over something like this doesn’t seem fair, especially when we all know how crucial every mark can be.

36) Read the paragraph beginning: She waited till... What made Merry worry that her bow might be close to breaking?

• An accepted answer to this question? → ‘she pulled it back as far as it could go’

• An unacceptable answer? → ‘she pulled it back hard’

Do I need to say anything else? Can I just remind you that the little humans taking this test have barely been on this planet more than a decade?

I’ve saved this bit until the end, because perhaps I’m being a bit of a pedant (or just generally have poor judgement) - but I really wasn’t a fan of the syntactic choices made in this text. I gather this was a stylistic choice by the author, as it happened a few times, but the lack of conjunctions in certain sentences really bothered me:

‘She turned, walked back from the line as the crowd, which had been stunned into silence, started to clap.’

‘She just looked at the mountains rising behind the castle, tried to keep at bay the noise and the attention.’

‘She nocked it, drew back her bow to its fullest extension.’

‘She felt the first flush of euphoria, pushed it down, selected her third and final arrow.’

I think it frustrates me because, grammatically, these are comma splices, and if a Year 6 child wrote these, we’d be picking them up on it. However, in this text, it is obviously a technique employed for stylistic effect (the author is quite clearly an extremely competent writer). We all understand it’s highly unlikely that an 11-year-old would deliberately, and consistently, use the same grammatical ‘mistake’ as a literary device, but still - it just seems a bit like a double standard! Luckily, it doesn’t affect how the questions would be answered, but I’ve barely moaned at all about this paper, so I had to find some sort of gripe, didn’t I?!

The reading paper results

To achieve the expected standard in reading, the pass mark has previously been around 50% (between 48-58%). The mark allocation for extract 1 (the ‘easiest’ text) has always been between 14-17 marks; for extract 2, it’s 15-20 marks. ‘Passing’ the reading paper (achieving around half of the answers correct) is roughly equivalent to getting most (~80%) of the answers correct in the first two extracts.

To achieve what is generally accepted as GDS, children have had to score between 76-82%: this is equivalent to only losing 9-12 marks overall (out of 50)​.

On Tuesday 8th July, we shall find out what this year’s mark boundaries are!

So, did the 2025 reading paper hit the mark? While the paper showed promise, the mark scheme remains a stubborn relic of over-engineering, where semantics eclipse sense and ‘unacceptable’ answers are sometimes just actual answers minus a conveniently placed adverb.

If there’s one thing this year’s paper reminded us, it’s that reading is a nuanced, interpretive process, and perhaps it’s time the mark scheme caught up with that fact. Until then, we’ll keep training children not just to understand texts, but to try and become mind-readers…

Roll on the pass mark reveal; I’ll be over here, adjusting my expectations and muttering “so eager” under my breath.

If you would like a personalised revision roadmap for your year 6s, you can try out LbQ’s SATs Springboard for free. 98% of teachers we asked agreed that they saw greater progress using it!