Are bigger sixth forms better?

6th December 2021 by Timo Hannay [link]

Sixth forms come in a wide variety of shapes and sizes, from Year 12/13 cohorts at small secondary schools to huge standalone sixth-form colleges with thousands of students. But what, if any, difference does this make to their effectiveness? Conventional wisdom and anecdote suggest that sixth forms with fewer than 200 students can suffer from insufficient scale, making it harder to attract good teachers and provide a suitably wide range of subjects. This post looks at the quantitative evidence, mainly using Department for Education (DfE) data from 2019 (ie, the most recent full year before COVID-19 struck).

We conclude that:

• There is a huge range of sixth form sizes, from almost zero up to many thousands, but among the mid-sized providers there appears to be a mild preference to have more than 250 students.
• Larger sixth forms tend to receive less money per student, but enter them for a wider range of A-level subjects and show better academic performance. They also tend to be more affluent, so this, rather than their size, could be the source of their relative success.
• However, the same effects persist even after controlling for disadvantage, suggesting that larger sixth forms do indeed tend to perform better, independently of the effects of disadvantage. This trend plateaus at around 200-300 students. Even the lower of these thresholds suggests that as many as 54% of sixth forms, accounting for 28% of students, might be below their optimum size.
• These results also suggest that disadvantaged sixth-form students face a double whammy: not only are they poorer, but their sixth forms also tend to be smaller, both of which correlate with lower academic performance.

Sizing up sixth forms

Figure 1 shows the distribution of all 1,927 mainstream state-funded sixth forms in England, based on the total number of students in Years 12 and 13. (Independent and special schools are excluded, as are further education colleges.) Whether viewed in terms of school/college numbers or percentages, there is a wide range of sizes. The most common student number is around 120, but note also the second small peak at 260, which suggests a mild preference to be above this size where possible.

There is also a long tail of very large sixth forms, extending well into the thousands, represented here by the '500+' column. Since the aim of this study is mainly to understand the performance of small and medium-sized sixth forms, Figures 2 and 4 below will omit those with 600 or more students (49 institutions, accounting for about 2.5% of the total and 14% of students).

(Use the menu below to switch between numbers and proportions of schools/colleges. Hover over the columns to see corresponding values.)

So much for the overall distribution – does the size of sixth forms vary much by school type or location? It does, as shown in Figure 2.

Among the regions, London has the largest average sixth form size, while the North East and the Midlands have the smallest. Interestingly, Ofsted 'Outstanding' sixth forms tend to be much larger than the those with lower ratings. This suggests that bigger institutions might indeed be better – but it could be for reasons other than their size. In particular, sixth forms with lower levels of disadvantage and those with lower levels of local poverty also tend to be bigger, so student affluence rather than educational effectiveness might be the main driver of the differences by Ofsted rating.

(Use the menu below to explore these and other categorisations. Hover over the columns to see underlying values and sample sizes.)

For richer or poorer?

Allowing for the effects of poverty is especially important in any analysis of educational effectiveness. To that end, Figure 3 shows the same distribution as Figure 1, but this time with low-, medium- and high-deprivation sixth forms split out separately. Whether viewed by numbers of schools/colleges or their proportions, it's clear that sixth forms with low levels of disadvantage tend to be considerably larger (ie, shifted to the right) compared to those with medium or high levels. (Show all groups.) It's essential to take this into account when comparing sixth forms of different sizes, as we will attempt to do in the next section.

(Use the menu below to switch between numbers and proportions of schools/colleges. Click on the figure legend to turn individual groups on or off. Hover over the columns to see corresponding values.)

Sliding scale

Figure 4 compares the characteristics of sixth forms across different size categories.

To begin with, let's focus on the results for 'All schools/colleges' (blue columns). As expected, we see that small sixth forms tend to have more disadvantaged students and poorer local populations, but get more funding per student. However, they also enter students for a narrower range of A-level subjects, achieve worse A-level grades and display lower A-level value-added (a measure that takes into account each student's prior attainment at GCSE).

Now look at the results for 'Selected schools/colleges' (red columns). These only include sixth forms with disadvantage rates in the range 10-17% (405 schools or colleges in total), which has the effect of evening out the disadvantage gap between smaller and larger institutions (and does the same for local poverty too). Even here, smaller sixth forms tend to get more funding and perhaps unsurprisingly, they also enter students for a narrower range of subjects. Even more interestingly, they continue to show lower attainment and value-added even after controlling for disadvantage.

(Use the menu below to explore these different measures. Click on figure legend to turn each group on or off. Hover over the columns to see corresponding values.)

None of this proves that smaller sixth forms are doomed to underperform. There's plenty of variation in the performance of establishments of all sizes. It's also possible that the differences are related to the variations in subject ranges (though these seems unlikely to be responsible for the whole effect). Furthermore, the causal link, if there is one, might go the other way: perhaps sixth forms that are already low-performing simply attract fewer students.

Nevertheless, it does support the idea that sixth form size can influence educational outcomes. Based on the analyses presented here, these effects seem to plateau at around 200-300 students. Even the lower of these corresponds to fully 54% of sixth forms and 28% of sixth-form students in England.

Furthermore, if sixth forms catering for poorer families tend to be smaller, and if they also tend to be lower-performing even after allowing for the effects of poverty, then those students may find themselves doubly disadvantaged.

As ever, we welcome your thoughts: [email protected]