Architectural preferences in the UK

Words by Robert Adam
29th March 2021

When it comes to architecture, there is a divide between the preferences of experts and the general public. What do we know about public opinion on architecture?


This article was first written in 2005.

Architecture comes in many forms and according to various ideologies. There is no one style or type of architecture both historically and currently. There is, however, one predominant type, style or ideology in contemporary practice and, consequently, this is generally presented as modern architecture. I have called this mainstream modern architecture.

The public, similarly, has different attitudes and interests in architecture. If architecture is a public art, relevant not just to the commissioning client, then a responsible profession needs to understand and respond to the public view of architecture.

Recent attitude research

Research by Cheltenham and Gloucester Building Society in October 2004, based on a self-selected sample of 2,000, produced a list of the ten worst and ten best buildings in Britain spontaneously given by the sample (there was no list to choose from). There were no recent buildings on the “Best Buildings” list. The Dome, the Gherkin, Canary Wharf, the Scottish Parliament, St George’s Wharf in Vauxhall, BBC White City, Tate Modern, the Bullring in Birmingham, the Post Office Tower and Centre Point all appeared in the ten worst buildings list. This is a curious survey with surprisingly specific results and we have no further indication of the nature of the sample but it is reasonable to assume that this is more likely to be completed by those who feel strongly on the subject and so have a wider cultural interest. Other than a general dislike for mainstream modern architecture, however, there is little further interpretation that can be put on these results.

Chetwood Associates commissioned a very simple piece of omnibus research in August 2004. This included a question whether the respondent “liked modern architecture.” 33% responded positively and this was fairly evenly distributed across the age and social range. This omnibus research was based on only a few elementary questions and there is little else to be deduced from this and there should be some caution with results using the unqualified adjective “modern” (see below).

In May 2002 CABE (The Commission for Architecture and the Built Environment – a government agency) commissioned a much higher quality piece of research based on 1,000 face-to-face interviews. This included two statements: “Architects should concentrate on designing buildings which appeal to as many people as possible” and 80% agreed; and “New buildings should be adventurous and different, even if they shock or offend some people” and 30% agreed. These questions had no visual prompts and so we do not know what the interviewees thought was meant by “adventurous and different” or what might “appeal to as many people as possible.” BMRB Research in 1998 (see below) produced similar results. When asked in separate parts of a quantitative survey whether “Old styles are right for new houses” and “New houses should not imitate old houses,” 63.5% thought old styles were right for new houses and 15.5% did not and 54% thought new houses should imitate old houses and 25% did not. This survey included both prospective buyers for new houses and those who would not buy a new house and there was no significant difference. As the questions were essentially the same and were separated and given opposite positives as a check, they can be averaged as approximately 60% positively favouring traditional design and 20% positively favouring more unusual design.

These last two research findings are extracts from research primarily concerned with housing preferences. Because of the sheer volume of housing in comparison with other forms of new building and because houses are purchased freely by individuals, thereby giving taste commercial significance, most architectural attitude research is for housing.

While people may make choices on preferred house design that differs from that of other buildings, it is most likely that personal choice will correspond to general architectural taste. At the same time, most buyers are concerned with re-sale and probably have an understanding of general preferences for traditional designs. This can lead some who might prefer a mainstream modern design to choose a more traditional design for their own house.

RIBA Journal research from 1994 produced results that indicated that where 67% would “prefer an older looking property or copy of an older design” for their house, by implication, 33% would prefer something less traditional. This does not fit very comfortably with similar research commissioned by the Halifax Building Society from Mulholland Research in 1997 where from a sample of 302 selected intending and recent buyers only 12% would like a house “more innovative and up-to-date in appearance”. The difference can be explained by the clear bias in the RIBA Journal Research which used omnibus research designed by an architect and used architectural terminology (see below) and, on the other hand, the Mulholland sample which was undertaken more intensively by a professional research firm and concentrated on a smaller group of those contemplating or completing a house purchase.

At this point it is worth noting that differences between professional and normal vocabulary can be misleading. For example, preliminary discussion group research by BMRB in 1998 (see below) indicated that in common usage the word “modern” meant merely “that which is being done now” and had no necessary stylistic meaning with reference to buildings. This simple fact moderates many of the conclusions drawn from the 1994 RIBA Journal research and the Chetwood Associates research (see above) and explains some of the apparent anomalies in the RIBA Journal research. (This may also apply more to housing than other use types as there is a preponderance of traditional designs in housing and a virtual monopoly of overtly modern buildings in all other functional types.)

The only way of overcoming uncertainties in use of vocabulary is to use illustrative material. While this, through choice of material presented, can create a bias, it is more likely to give a result free from bias in the interpretation of the results. Three pieces of recent research have used carefully selected visual material for housing preferences: the 1998 BMRB Research, the 2002 CABE Research and the 2005 CABE Research.

Research undertaken for the Popular Housing Group in 1998 by BMRB had a wide range of visual information. This indicated that 4.5% of the sample (both prospective buyers and others – that is the whole population) would choose a house of a mainstream modern design.  Interviewees were shown 6 cards with different types of houses grouped according to categories drawn from six discussion groups. Two of these were specifically non-traditional and attracted 1.5% (predominantly glass) and 3% (mainstream modern with natural materials) of the sample. This was not the purpose of this research but a by-product. Nonetheless, the illustrative choices are particularly clear and it was based on face-to-face quantitative surveys on a sample of 829 nationally selected by the Acorn method.

The 2002 research by CABE indicated that only 3% of the sample would prefer to live in houses or flats that were examples of recent mainstream modern architecture (for example Murray Grove, Broadwell Housing at Coin Street and Gainsborough Studios in Islington) . These same buildings counted in the 93% of those where the sample would least like to live (although this was skewed significantly by an 84% least choice for a 1970s block of flats).

The figures differ from those in the latest research findings “What Home Buyers Want” published in March 2005 by CABE. Based on a selection of three new house types, neo-vernacular, traditional and mainstream modern, 20% of the sample in this survey favoured explicitly modern design. The sample was self-selected from existing home-buyer’s lists, recruited and questioned on-line. The method of selection and interrogation would tend to favour the computer literate and the more wealthy buyers that used home-buyer’s lists and there was, indeed, a strong bias towards upper income groups. Based on further analysis of Chetwood Associates’ research (see below) this would be likely to show a slight bias in favour of mainstream modern. There was also a bias towards London and evidence from the same research shows a bias towards mainstream modern in London. These factors could explain the variation in the figures. It is unlikely that there would be a 300% swing in preferences over a three year period.

Attitude research conclusions 

A survey of this research material produces variable results but the broad conclusions are the same and inescapable – the majority of the public prefer more traditional design to the architecture produced by the architectural mainstream (called professionally “modern” or “contemporary”). The results vary between a range from a zero to 4.5% preference for ‘modern’ architecture (see above) up to 33%.  The higher figures have a bias of some kind, terminological or by sample, while the lower figures include both high quality and less stringent research. If we assume that the true figure simply lies between the maximum and minimum, somewhere between 15% and 20% of the population are likely to have some sympathy for mainstream modern architecture.

Interest in architecture as a subject

Research was commissioned from Surrey University and the LSE by the DCMS (Department of Culture, Media and Sport) in 2003 to assess the level of sporting and cultural activity in the UK. This drew on a sample of 8,584. This demonstrated that only 6% of the public could be classified as “high culture vultures” (the term used in the research). The threshold for judging who might be a “cultural vulture” was quite low: visiting a stately home, visiting museums and going to the cinema relatively frequently would qualify.  There is a second category named as “heritage seekers” that make up a further 8% and here the threshold is much lower. This research also indicated that a maximum of 15% of the population participated in an arts activity. Again, the category is quite wide and includes singing or writing poetry.

It is reasonable to assume that a number somewhat less than this 14-15% would be interested in architecture as a cultural phenomenon. That is not to say that a larger number than the culturally active groups identified in this survey are not interested in architecture as something that affects their everyday lives. If we are, however, to try and establish what part of the population is interested in or sympathetic to architecture as a cultural phenomenon and practiced by the majority of the profession today they are likely to lie within this group. It is unlikely that all of those that participate in the arts or some cultural activity will have an interest in architecture. Music, in all its forms, or watching films has a much wider participatory and interested constituency.

There seems to be no other indication to discover which culturally active people would be interested in architecture as a cultural phenomenon. We can, however, try and assess how many of this group is likely to be sympathetic to mainstream modern architecture. A simple formula would be to apply the 20% figure concluded above to the 15% culturally active group, or 3% of the population. This, however, ignores a demographic bias in these numbers. Based on the Chetwood Associates Research, those who liked “modern architecture” were approximately twice as likely to read a broadsheet newspaper. Similarly, the culturally active group was twice as likely to have achieved an educational level of two A levels or above. Assuming these broadsheet newspaper readers and those with higher educational achievement are roughly coextensive (a sweeping assumption it must be admitted), this will produce a higher figure than 3%. Based on the 2001 census figures which show that 22% of the population has qualifications of 2 A levels or above, the corrected figure indicates that the likely proportion of the population interested in architecture as a cultural phenomenon and sympathetic to mainstream modern architecture is a maximum of 6.5%.


While it is impossible to be more precise with the information available, it is at least clear that a small proportion of the population is sympathetic to architecture currently practiced by the majority of the profession. Over all the public, this would seem to be a maximum of 20%, or one in five.

Amongst the culturally active group, most likely to have an interest in architecture as a cultural phenomenon and as currently practiced by the majority of the profession, a generous estimation would put this at approximately 45%. This represents about 6.5% of the total population. Taking account of a probable lack of interest in architecture amongst the culturally active, this is most likely rather less.

This should come as no surprise. There has been a sharp division between the architectural taste of the profession and the public for some 50 years. The recent date of some of this research indicates that the long hope of the profession that public taste would sooner or later catch them up does not seem to have materialized. Indeed, the figures for housing preferences in the recent research are very similar to those of research undertaken by the RIBA Journal in 1994. Speculative housing development for individual houses for sale, sensitive only to the market and private choice (and to some degree to planning) has been determinedly traditional and reinforces this view.


Prof Robert Adam is the founder of ADAM Architecture, now the largest architecture practice specialising in traditional design in Europe. He has written six books, including the principal textbook on classical architecture, and has won numerous prizes including the world’s highest value architectural award.