KS2 English SATs paper analysis 2023

The embargo has been lifted and the dust is beginning to settle on what can only be described as a controversial SATs week. We asked year 5/6 teacher, Sophie Bartlett, to give us her unfettered opinion on the reading and GPS papers this year, and she did not disappoint.

Grab a brew and a biscuit, and settle in. It’s quite the ride.

2023 KS2 Reading SATs analysis

Well, hasn’t this year’s reading paper caused quite the stir? So much so that this BBC article was written days after children sat the test and the papers were released earlier than they ever have been “due to public interest in the tests”. I think a hotspot of Texan bats is going to go down in the SATs hall of fame, alongside the bewilderingly milling warthogs of 2016…

Predictions vs the real thing

Firstly, let’s have a look to see how our predictions for this paper earlier this year held up…

Prediction: The majority of the questions in the reading papers have always been retrieval (2b) and inference (2d) – this was still the case in 2022 and so it seems sensible to expect it to be the same in 2023 as well.

As you can see below, this indeed continues to hold true. In fact, this year’s test contained the highest proportion of inference questions that we’ve ever seen – nearly half the paper!

The amount of vocabulary-focused questions (content domain 2a) was back up to levels similar to those seen in 2016, 2017 and 2018. Like last year, there were no questions based on authorial choice (2g) or comparisons (2h). Prediction questions (2e) also had no representation this year, just like years 2017, 2018 and 2019.

Prediction: A poetry extract hasn’t appeared since 2018, so it may be sensible to assume that a poem might be included in the 2023 reading papers.

I know, I know – I’m sorry! SURELY poetry has to appear next year, right?

Wordiest ever Reading SATs paper

Since the new style SATs began in 2016, in this year’s reading paper (I’m using ‘wordy’ here to mean the amount of words!):

• Extract 1 was the second wordiest extract 1 ever (closely beaten by 2019’s Extract 1)

• Extract 2 was the wordiest extract 2 ever

• Extract 2 was the wordiest non-fiction extract ever

• Extract 2 was the second wordiest extract ever

• Extract 3 was one of the least wordy extract 3s (last year’s was less wordy)

• The entire paper was the second wordiest paper ever (closely beaten by 2019’s paper)

Other points to note:

• No poetry… again!

• Extracts comprised of two fiction and one non-fiction – the same as every other year but 2018

• Order of extracts: fiction, non-fiction, fiction – a pattern that has appeared four times now

• 40% of the marks allocated to one extract (Extract 2: non-fiction) is the highest since 2016 (where 44% of the marks were allocated to Extract 2: fiction)

Gráinne Hallahan (@heymrshallahan) wrote this article for TES, highlighting why this year’s paper was perceived to be so difficult. Her analysis states that there are “just 21 and a half minutes to answer 38 questions - meaning you have to give an answer every 34 seconds (this includes the time it will take to go back and re-read a passage to get their answer).”

Totalling the word counts for all previous SATs papers (both reading and answer booklets) reveals that this year saw the highest number of words so far, closely followed by 2019’s paper. (Please note that you may see slightly different word counts on different people’s analyses – there is obviously a margin for both human error and understanding of what is considered a ‘word’, e.g. hyphenated words, numbers, etc.)

The difficulty, or ‘reading level’ (e.g. Lexile) of each text is obviously also not taken into account here.

Reading question types

As in every paper since 2016, there were two 3-mark questions. A personal favourite of mine, the ‘Find and copy’ questions are still appearing, albeit but only twice in this paper (it appeared three times last year, four times in both 2019 and 2016, five times in 2017 and six times in 2018).

Interestingly, the structure of the 3-mark questions have changed (for the good, in my opinion) from 2018 onwards. 3-mark questions in 2016 and 2017 were more vague and perhaps seen as quite overwhelming with so many lines for the children to write on.

Since 2018, these more valuable questions have been more scaffolded and – arguably – easier for the children to achieve the full three marks.

Word on the street pre-embargo

So what did everyone else think? Firstly, some “pre-embargo” thoughts from both Facebook and Twitter…

Post-embargo thoughts

And some thoughts now that the papers are in the public domain!

Conclusion

This year's paper will certainly leave its mark. Up until now, 2016 was the notorious 'avoid-that-past-paper-like-the-plague-unless-you-want-to-feel-bad-about-yourself' paper, but I think we may now have a new contender!

We all like a bit of challenge, and the children deserve a bit of challenge, but I feel this year's paper went a little too far. Perhaps I could've forgiven it more if the extracts were even a tad more engaging for the children... and if we don't have some poetry next year, people, I'll eat my hat!

2023 KS2 grammar, punctuation and spelling SATs analysis

Less controversial but still slightly thorny, let’s take a look at the GPS papers.

The spelling test

The spelling tests contain 20 words – these are split fairly evenly between the Year 3/4 spelling rules and the Year 5/6 spelling rules, although there has always been a few more from Lower Key Stage 2.

See the analysis of all the SATs spelling tests below.

• An italic word means it appears under more than one content domain

• A blue word is one of the first three of the test – generally considered slightly easier to spell

• A red word is one of the final three of the test – generally considered slightly harder to spell

The most commonly used content domain in SATs spelling tests is S38: adding suffixes beginning with vowel letters to words of more than one syllable (a Year 3/4 spelling rule). This is followed by S47: endings that sound like /ʃən/, spelt –tion, –sion, –ssion, –cian (also Year 3/4) and S60: words with ‘silent’ letters (a Year 5/6 spelling rule).

The least used (only once each across all six tests so far) content domains are S42: the suffix -ation; S49: words with the /ʃ/ sound spelt ch; and S53: endings which sound like /ʃəs/ spelt –cious or –tious.

Those spellings generally considered to be the most difficult (at the end of the test) have often come from S58: words with the /i:/ sound spelt ei after c – words from this content domain have appeared four times in the final three spellings across all the SATs tests.

Question 16 on the spelling test

The spelling test seemed to pass by without much fuss this year. Social media was fairly quiet on this front, apart from some light discussion around Spelling 16.

We are so used to drilling the children that, in the grammar paper, capitals must be used if a word is at the start of the sentence! However, for the spelling test, this isn’t the case. In point 6.2 of the mark scheme (General marking guidance for Paper 2: spelling), it states that “the pupil can answer in upper or lower case, or a mixture of the two. The exception to this is for days of the week, which must be written in lower-case letters with an initial capital letter for the award of the mark.”

The grammar and punctuation paper

Then onto our favourite – the grammar and punctuation paper. Children do love a good old tick box question! This paper makes up 50 of the 70 marks allocated for the overall GPS score. This year, the proportion of content domains covered in the paper has been similar to previous years.

G1 (grammatical terms or word classes) and G5 (punctuation) remain the most commonly used content domains, comprising over half of the test questions.

Vexing verbs and tricksy adverbs

The most common topics assessed throughout the grammar paper were conjunctions, commas and questions/commands/statements/exclamations, with three or four questions based on each topic.

A few teachers commented on the awkwardness of Q15.

The correct answers are both fed and ate for 1 mark, but this turns the sentence into ‘I usually fed my cat before I ate breakfast’, which is – to put it kindly – grammatically clumsy, and it threw a lot of children. Why not just remove the word ‘usually’? Or, if it’s just a case of conjugating verbs, take the words out of context completely? Perhaps it should be ‘I usually write my test questions before I check if they make any sense.’

Q40 also caused quite a fuss. One tweeter felt particularly strongly about this question, but I would’ve had to redact most of the words if we’d shared their view here verbatim! So here’s a more PG opinion:

Now, it may not be the most grammatically correct way of thinking about adverbs, but we all know that most children recognise many adverbs to end in -ly. If not for that, then they may recognise an adverb to be doing just as its namesake suggests – adding to a verb. So knowing this, you would, of course, when writing a test for 10- and 11-year-olds, choose an adverb that does neither of these things.

Conclusion

Those two grammar questions were a bit like stepping on a piece of Lego: annoying, but won’t ruin your day. All in all, I feel the grammar paper was fair - and so did most other teachers on social media it seems (it’s always nice to fit in with a crowd). The reading paper certainly caused much more uproar, and quite rightly.

It will be fascinating to see where the pass mark lies… it has risen each year since 2016, but something tells me it’ll be lower again this year. Either that, or we’re not far off reading variations of “primary schools are sending too many children to secondary school unable to read” tabloid headlines. Ah, the joys of teaching!

From everyone at Learning by Questions, congratulations to all the teachers and pupils who took on the week of national testing - you are heroes!

Have you read Sophie's maths SATs papers 2023 analysis?