Features

Museum engagement: Insights hidden in public data

Hidden within public data about museum engagement are some unexpected insights. Guy Turton of Morris Hargreaves McIntyre has been digging beneath the data to see what’s changed – and what hasn’t – since Covid.  

Guy Turton
7 min read

The government’s Participation survey is the largest survey of cultural engagement in the country. It reveals the stark disparity in cultural engagement across different demographic groups. But these patterns aren’t new. And the good news the survey presented this summer – of increased cultural engagement – may be hiding some concerning patterns of widening societal disparity.

A trends analysis of the survey indicates that the pandemic acted as a leveller in museum engagement. It caused demographic disparities in museum visiting to narrow. Post-pandemic, nearly all demographics reported increased museum engagement in 2023/24; though crucially, not all have increased at the same rate.

As a result, we’re now seeing a return to wide gaps in demographic engagement – especially by socioeconomic and disability factors. Indeed, there are signs the societal gap may be getting wider.

You can also see the ripple effect of the cost-of-living crisis across the data. In 2023/24 at least, it likely impacted on the cultural activity of some more than others: notably those younger, those out of work, those with disabilities and those living in higher IMD areas; all profile factors that are intersectional.

Even free museums suffer from this in their pursuit of audience diversity, with high travel costs, along with secondary spend, viewed as major engagement barriers. This trend poses challenges for museums and the support they need to address audience diversity ambitions.

What we’ve learnt

This survey has been running for three years now with the data available to all. It can even be accessed at a ‘case level’, so it can  be analysed by anyone from their bedroom.

The annual data revealed some headlines around increased cultural participation on 2022/23, including a 6% increase in library use to 30%, a 10% increase in museum engagement to 46%, but a plateauing of heritage engagement at 69% (though this may be impacted by an error in data collection). 

What we’ve NOT learnt, yet

The main report goes quite deep in describing variation in all manner of cultural engagement by different demographic groups. This reinforces well established patterns in the disparity of cultural engagement by factors such as socioeconomic status, origin, employment, disability, etc. All fascinating, worrying, and always worthy of deeper exploration to help us address this disparity at a sector level.

But what the analysis doesn’t yet explain is which cohorts have had a disproportionate impact on the change in cultural engagement observed over the past three years.

In other words, if engagement with museums has increased, who is driving this increase most and least? Are there some groups that have significantly rebounded from Covid declines whereas others are lagging behind in their recovery? 

Exploring the increase in more depth

I decided to experiment with one question: physical engagement – as opposed to digital engagement – with museums and galleries in the last 12 months. 

I took a deep dive into the data from the past three years and, where possible, added in data from the earlier Taking Part survey from 2018/19 for good measure. While such comparison should be done with some caution  – change in method and supplier – the 46% of public engagement with museums in 2023/24 is below the 50% reported in Taking Part in 2018/19 (pre-Covid). 

Overall, there aren’t many outlier groups

Increases from 2021/22 and 2022/23 aren’t caused by new patterns in demographics generally. Across the board, demographic groups saw consistent levels of decline and are now broadly increasing in ratio with previous patterns, reinforcing those same societal disparities described earlier.

But there are some notable outliers which speak to patterns we’ve seen across the sector in terms of pace and profile of visit recovery. While a deeper statistical analysis from the source confirms some of these patterns, here are the biggest takeaways:

Young adults have been slower to recover

People at both ends of the age spectrum are less likely to go to museums compared with those between ages 25 and 74. And while the elderly remain the least likely visitors, we especially note the relatively modest upturn in museum engagement from those under the age of 25 – in particular those aged 20 to 24.

This aligns with other research we’ve conducted, where we’ve seen young people less likely to be experimental with leisure activity as they feel the financial pinch.

Ethnicity patterns don’t align to patterns observed in most museums

People identifying as Black/African/Caribbean/Black British remain well behind the national average for museum engagement. While those identifying as Asian/Asian British are almost exactly at the average point, with the same true of those identifying as an ethnic group classified as White.

It’s then ‘other ethnic minority’ groups which are above the national average – for example mixed ethnicity groups such as those of White and Asian ethnicity at 56%.

When we go back to the Taking Part Survey data from 2018/19, this suggests that those identifying as White have shown the slowest levels of recovery in museum-going since Covid. 

For many museums, this data may seem at odds with their experience. Many see the proportion of their visitors from ethnic minority groups underrepresented compared with national and local population proportions.

This suggests a range of possible considerations including visit frequency, overseas museum attendance and even perceptions on what is considered a museum or art gallery. 

Education level sees an increasing engagement gap

In 2021/22, the museum engagement gap between those educated to degree level and those not was 17%. This widened to 20% in 2022/23 and is wider now still at 23% in 2023/24. We don’t have this data point from 2018/19 to compare.

​​The unemployment gap is also increasing

The same pattern for education level is seen when we analyse employment: an 8% gap between those working and those not, compared to 6% in 2022/23 and 5% in 2021/22. But this gap remains behind that reported in 2018/19, which was 11%.

The socio-economic gap is widening 

While margin of error means these differences aren’t consistently statistically significant, the triangulation of data points builds a story of a widening engagement gap based on socio-economic disparity.

For example, the gap between those in the highest IMD category and those in the lowest was 13% in 2021/22, then 17% in 2022/23 and now 20%. While marginal, this is higher than the 19% in 2018/19. 

The disability gap is as wide as ever

The gap in museum attendance between those identifying as disabled and those not is 9%. Again marginal, but this is higher than the gap in other years suggesting the museum access gap isn’t improving and may even be going in the wrong direction. 

The big picture: an exaggeration of disproportionate museum engagement, or just reverting to type?

This exercise was designed to demonstrate what is possible to everyone in public data analysis. Other big data sets are also available for this level of analysis, including the Active Lives survey.

While deeper statistical analysis would sharpen up analysis, my initial big takeaways from this exercise are:

  • Covid acted as something of a leveller in museum engagement.
  • It closed the gap on demographic disparities.
  • Those disparities were still very present, but the gap narrowed.
  • We’re now seeing a reverting to those wide gaps.
  • This means more disparity by socioeconomic factors and disability especially. 
  • Indeed there are signs that the gap may be getting wider.

Looking ahead not in the rear-view mirror

These trends all reflect on the past 12-18 months. The market is moving fast, impacting all sorts of organisations in different ways (e.g. how consumers are being increasingly cost savvy with membership).