The BFI and ‘Opening Our Eyes’
Last Friday the BFI launched its report on the cultural contribution of the cinema to the UK. I was unable to attend the launch, but the report is no available online here.
The motivation for this report is partly to do what is says and explore the contribution to British cinema, but it is also necessary to justify the UK’s economic film policy. Under EU law, financial support in the form of tax relief or lottery funding can only be provided for the representation of regional and national cultures and NOT for competitive economic advantage. Therefore, by establishing the importance of the cinema to British cultural life everyone gets to pretend that economic film policy in the UK is cultural and not economic at all.
Leaving that collective self-delusion aside, let’s focus on the cultural aspects of the report. The problem here is that we have what is largely a marketing report that does not really address the question of ‘cultural contribution.’ Films, we are told, are one of the ‘most popular entertainment and leisure interests,’ but popularity is not a measure of cultural contribution. One of overriding problems I have with this report is the lack of detail. The list of films that are often mentioned as having a ‘significant effect on society or attitudes in the UK’ is frequently used as a basis for drawing conclusions about the relevance of the cinema in British cultural life but this effect isn’t qualified beyond the personal anecdotes of some of the interviewees. This isn’t necessarily a bad thing, but the number of interviewees is small: ten paired interviews were conducted giving a sample size of 20 participants from major cities in Scotland (Glasgow) or England (Manchester and London). Perhaps no-one in the UK lives east of the Pennines (thereby plunging me into a severe existential crisis); or perhaps no-one in Wales or Northern Ireland goes to the cinema. I recognise that measuring how a film affects society is difficult, but then this at the core of what the report is trying to do and it never really achieves it. We can certainly see how Cathy Come Home affected the UK, leading to the founding of Shelter; but what was the ‘significant effect’ of Trainspotting? This is never explained. I also recognise that it is a lengthy and costly process, but if you’re going to justify national film policy based on research it should be national research.
One aspect that has been commented upon in a number of news articles based on this report (and which I therefore assume was part of a press release) is that people are more interested in the cinema than they are in religion. Specifically, 32% percent of respondents indicated they were interested in religion compared to the 84% who were interested in the cinema. The question we really want to ask here is why religion was included in the survey at all. Does the BFI think that religion is a form of popular entertainment and leisure interest comparable to the cinema? In what ways is an interest in films comparable to an interest in religion? Surely we cannot be expected to interpret this as meaning that the people of the UK are 2.5 times more interested in the cinema than religion, because this is how it has been presented in reporting typical of the standard you can expect from the British press. This is really a measure of popularity and not a measure of contribution at all: the contribution of the Church of England to the cultural life of the UK is far greater than that of the cinema, even if no one appears to be interested in it any more. I may be an atheist but, dear God, this was a stupid question to ask.
A further confusing statement is that a ‘strong interest in film correlates with a stronger than average interest in many other forms of leisure and cultural pursuit,’ because no correlation coefficient, correspondence analysis, correlation matrix, or any other statistic is presented. Correlation is a statistical term, and in the context of this report that presents such a large amount of data in terms of graphs, percentages, and other measures, it has no meaning. Nor is there any evidence presented to link strong interest in the cinema, strong interest in other arts, and increased levels of participation. It is simply asserted that
higher interest in a range of other social and cultural activities tends to follow through into higher levels of activity (Figure 5), with more people very interested in film also going more often to the theatre, concerts, pubs and clubs than those with no interest in film.
The reference to Figure 5 leads us only to a series of bar charts that present frequencies and not correlations. This is just poor research, and a better approach would have been to perform a correspondence analysis on this data. (This is exactly what Pierre Bourdieu does in Distinction to answer exactly the same sort of questions). If nothing else, this would have presented us with only a single graph to deal with rather than thirteen separate bar charts.
The same is true of the words associated with different forms of entertainment (Figure 11). The endless bar charts have by now become somewhat tiring and the plots of the first and second principal and first and third principal axes of the correspondence analysis of this data would have made life so much easier (Figure A). To interpret these plots, points in the same data set that are near each other have similar profiles in terms of the attributes assigned to them while points in the same data set that are distant have different attribute profiles; and points in different sets that lie away from the origin and in the same direction are related. In both plots we see that film (FIL) lies along the same direction as escapism (ESC), and so the cinema is strongly correlated with this attribute.
A1: Dimensions 1 and 2
A2: Dimensions 1 and 3
Figure A Biplots of the correspondence between entertainment/leisure activities and attributes associated with them, based on percentages in the BFI ‘Opening Our Eyes’ report. For the detailed numerical summary of the analysis see here: BFI CA Summary.
As we would expect, pubs/clubs (PUB) and dining out are similar and news (NEW) is strongly correlated with the attribute ‘informative’ (INF). Interestingly, art galleries and museums (GAL) lies along the same axis as exciting (EXC) in both plots, and playing sport (SPP) produces similar results to going to the theatre (THE) but watching sport (SPW) does not. Also, pubs/clubs seem to correspond with thought-provoking (THO).
The attributes do appear to spread out into a number of subsets. Emotion (EMO) and Inspiration (INS) appear to be related, as do exciting and informative. Relaxing (RLX), entertainment (ENT), escapism, and thought-provoking appear to have profiles more similar with another than with the other four attributes. This relationship is apparent in both plots. Television (TEL) appears to be associated with no particular attribute, contributing nothing to the variation in dimension 3.
Note that religion (REL) in the first plot is unrelated to the profiles of the other forms of entertainment and leisure activity and contributes nothing to the variation of the second plot (CTR = 0.00), and it would be better if this category had not been included in the survey at all.
A further problem arises when we look at the section on what films people are watching. Blockbusters emerge as the most frequently watched type of films, but as these types of films dominate the UK exhibition market what else did you expect?
The information on the factors that people use to decide on what to go and see is much more useful. There is a fairly large body of literature on this topic for Hollywood cinema (and you can find some information on this topic here), but there is a lack of research in this area for the UK (but see here). This is also the case for sources of information on films on release. Hopefully, this information will lead researchers to explore these areas in more depth, because we really know very little about film audiences in the UK compared to what we know about the US.
The section on opinions about the proportion of films distributed in the UK on specific subjects (Figure 17) can be ignored. I don’t quite see the point of the question about whether there are too many films about homeless people – this seems a needlessly specific question that doesn’t really have anything to with the cinema. I’ve never seen a British film about a quantity surveyor, but they didn’t ask about that. How would making films about the homeless contribute to British culture? If this were to prompt filmmakers to start making films about the homeless, asking the question again would increase the number of responses for too many as the public becomes bored by a glut of such films. This part of the survey does not appear to take into account the films currently on release that would influence the respondents. It is a silly question that tells us nothing about anything.
More interesting is the revelation about how audiences perceive stories as being related to national identity rather than the source of the production:
Respondents saw films with British stories set in the UK as close to ‘entirely British’, even if they were partly produced or financed from the USA. Films about non-British people set outside the UK were seen as being mainly of another nationality. This is illustrated at the borderline by Slumdog Millionaire which, though an independent British film, was seen as being slightly more of ‘another nationality’ (Figure 19). Some interviewees saw it as a British film set in a non-UK location.
Now this is interesting information that can make a substantive contribution to the study of British national cinema and will, I’m sure, feature in many books and articles on the subject. The analysis based on the seven-point rating scale (Figures 19 and 20) is interesting but I have problems with the statistic used: the average position on the scale appears to be the mean and the use of a continuous measure for ordinal data is problematic. Either the mode or the median could be more appropriate because we cannot be sure that respondents interpreted the phrases used (‘entirely British’) in the same way or that the differences between the categories are equal. (There is no reason to assume that responses to this scale would be normally distributed so some additional justification for relying on the central limit theorem should have been provided, particularly given the possibility of central tendency bias).That said, the differences in the score between Slumdog Millionaire on the one hand and St. Trinian’s and The Full Monty are the real gem in this report. As scholars of British cinema we should be exploring this in great depth.
The section on the regions of the UK brings us back to the endless bar charts, and as before a different approach might have reduced the cognitive burden and eye strain on the reader. Generally, these results reflect the responses given to a question asked as part of the 19th British Social Attitudes Report published in 2002, which found that regional identity in England increased with distance from the capital and a high level of identification with the metropolis in London. This would explain why people in the North of England, the North West, and Yorkshire and Humberside want to see more films based in their regions relative to the regions in the south and the east. It is surprising to see that the percentages for ‘Too many’ and ‘About the right amount’ for the South West are consistent with East Anglia and the South East, as this region tends to have a relatively higher sense of regional identity. This is an interesting result that could be followed up.
Overall, this report is just like every other report produced by an official body that you’ve read: relentlessly upbeat, and riddled with oddities of design and analysis. There are some really interesting points to be found, and if this report is going to make a real contribution to our understanding of British film culture it will be as a starting point for researchers to try to get behind the data.
Finally, the best news of all: the detailed responses and the full data sets are being made available by the BFI, and can be accessed here. This means hours of endless fun of doing your own analysis on the data.
Heath A, Rothon C, and Jarvis L 2002 ‘English to the Core?’ in A Park, J Curtice, K Thomson, L Jarvis, and C Bromley (eds.) British Social Attitudes: The 19th Report. London: Sage: 169-184.
Posted on September 22, 2011, in British Cinema, Film Industry, Film Policy, Film Studies, Motion Picture Distribution, Motion Picture Exhibition, Motion Picture Production and tagged British Cinema, Film Industry, Film Policy, Film Studies, Motion Picture Distribution, Motion Picture Exhibition, Motion Picture Production. Bookmark the permalink. 2 Comments.