Film style and narration in Rashomon

And so after a long (and much enjoyed break) I return to the blogosphere with the first draft of paper on film style and narration in Rashomon. This paper is different to other statistical analyses of film style I have published on this site and to all other studies of film style and narration because it uses multivariate analysis to look at several different aspects of film style together. The method used is multiple correspondence analysis, and you can find a good introductory chapter on MCA here. The software I used is FactoMineR for R, and the website explaining how to do the analysis can be found here.

Multivariate analysis has been used in the quantitative study of literature for some time (see the links below the abstract), but this is the first time multivariate analysis has been applied to film style and it appears to work very well. I am currently looking at some other applications, particularly in distinguishing between the different parts of portmanteau horror films (which is a proper scholarly endeavour and not simply an excuse to watch lots of portmanteau horror films).

The pdf file can be accessed here: Nick Redfern – Film style and narration in Rashomon

An Excel file contain the data used in the analysis can be accessed here: Nick Redfern – Rashomon. This file contains two worksheets: the first is the shot length data for the film, and the second is that data used in the multiple correspondence analysis.

Abstract

This article analyses the use of film style in Rashomon (1950) to determine if the different accounts of the rape and murder provided by the bandit, the wife, the husband, and the woodcutter are formally distinct by comparing shot length data and using multiple correspondence analysis to look for relationships between shot scale, camera movement, camera angle, and the use of point-of-view shots, reverse-angle cuts, and axial cuts. The results show that the four accounts of the rape and the murder in Rashomon differ not only in their content but also in the way they are narrated. The editing pace varies so that although the action of the film is repeated the presentation of events to the viewer is different each time. There is a distinction between presentational (shot scale and camera movement) and perspectival (shot types) aspects of style depending on their function within the film, while other elements (camera angle) fulfil both these functions. Different types of shot are used to create the narrative perspectives of the bandit, the wife, and the husband that marks them out as either active or passive narrators reflecting their level of narrative agency within the film, while the woodcutter’s account exhibits both active and passive aspects to create an ambiguous mode of narration. Rashomon is a deliberately and precisely constructed artwork in which form and content work together to create an epistemological puzzle for the viewer.

On the multivariate analysis of literature see the following:

Hoover DL 2003 Multivariate analysis and the study of style variation, Literary and Linguistic Computing 18 (4): 341-360.

Stewart LL 2003 Charles Brockden Brown: quantitative analysis and literary style, Literary and Linguistic Computing 18 (2): 129-138.

Tabata T 1995 Narrative style and the frequencies of very common words: a corpus-based approach to Dickens’s first person and third person narratives, English Corpus Studies 2: 91-109.

The Veritiphone system

I have previously written three posts on the efforts of the Leeds inventor Claude Hamilton Verity to develop a synchronisation system for motion pictures using a sound-on-disc system. In 1923 he sailed to America to work with the Vitagraph Film Company, though the result of this collaboration remains unknown. His efforts were reported worldwide but he has disappeared from the history of British cinema. You can read my earlier posts here, here, and here.

I had not thought about Verity for many months until Luke McKernan asked me a question yesterday, and I took the opportunity to have a quick search to see if anything new was available.

Rather wonderfully I have just found a discussion at Gramophone Collecting which has images of two articles. One is by Verity himself written for The Sound Wave 1922 describing his ‘Veritiphone’ system complete with a picture of this unusual machine.There is even a picture of the man with his machine. The other is a description of his efforts.

The original discussion can be found here.

The introduction to the article reads:

We have had an opportunity of testing the acclaimed merits of the Veritiphone. This is the invention of Mr. Claude H. Verity, of Leeds, who has made a deep study of the synchronisation of moving pictures, and who has admittedly accomplished what at one time appeared to be an impossible feat, that of timing the movement of the lips of the speaker  with the recorded speech given coincidentally. The Veritiphone is, indeed, the outcome pure and simple of Mr. Verity’s pursuit of the science of synchronisation.

From this we can infer the Veritphone system worked, performing exactly as Verity claimed and as reported around the world. And yet he is utterly unknown to historians of British cinema.

Here are the images from the forum.

The mAR index of Hollywood films

UPDATE: reviewing the methodology of the mAR index in general, Mike Baxter noted an error in the data whereby I had reported the exponent of the negative exponential function instead of the mAR index for films from the 1960s. I have now corrected this and redone the analysis and the graphs (which are still cool). This mainly effects the conclusions regarding differences between genres. Overall, it turns out that, as a result of this error, I had actually underestimated the difference between the classical and rank mAR indices. If anyone finds any other errors then feel free to add a comment to this post and I’ll try to correct it as soon as possible.

And so to finish the month as we started, looking at robust estimates of the mAR index of film style. Below is the first draft of a paper comparing the mAR index  based on the methods used by James Cutting, Jordan De Long and Christine Nothelfer to describe the clustering of shots in motion picture with a rank-based alternative that is resistant to outliers. Naturally, it features some pretty cool graphs.

The pdf file is here: Nick Redfern – The mAR index for Hollywood films

Robust estimation of the modified autoregressive index for high grossing films at the US box office, 1935 to 2005

The modified autoregressive (mAR) index describes the clustering of shots of similar duration in a motion picture. In this paper we derive robust estimates of the mAR index for high grossing films at the US box office using a rank-based autocorrelation function resistant to the influence of outliers and compare this to estimates obtained using the classical, moment-based autocorrelation function. The results show that (1) The classical mAR function underestimates both the level of shot clustering and the variation in style among the films in the sample.; (2) there is a decline in shot clustering from 1935 to the 1950s followed by an increase from the 1960s to the 1980s and a levelling off thereafter rather than the monotonic trend indicated by the classical index, and this is mirrored in the trend of the median shot lengths and interquartile range; and (3) the rank mAR index indentifies differences between genres missed by the classical index.

 

Robust estimation of the modified autoregressive index of film style

Earlier this I looked at the time series structure ITV news bulletins using robust methods of autocorrelation. This post follows on from that earlier study, this time looking at BBC news bulletins. This paper was written with three goals in mind. First, I wanted to improve on the method used before. Second, I wanted to try the rank based method of estimating the mAR index. Third, I wanted to apply these methods to a different cluster of data sets to see if I would come up with similar results.

The paper can be accessed as a pdf file here: Nick Redfern – Robust estimation of the modified autoregressive index of film style

Abstract

The modified autoregressive index (mAR) describes the tendency of shots of similar length to cluster together in a motion picture but is not resistant to the influence of outliers if derived from the classical moment-based partial autocorrelation function. In this paper we calculate robust estimates of the modified autoregressive index based on outlier-resistant partial autocorrelation function based on the ranks of the shot length data and robust measure of scale. The classical, rank, and robust methods of determining mAR are compared for a sample of BBC news bulletins.

Genre trends in five European countries, 2006 to 2010

This post is an updated and extended piece I wrote last year on genre trends at the box office in five Eurpoean countries with the data cleaned up and new variables considered. Although the numbers have changed slightly from lasty year’s version the orignal conclusions remain valid.

The pdf can be accessed here: Nick Redfern – Genre trends in five European countries

Abstract

This paper analyses box office trends of the genres for the top 50 grossing films in each year from 2006 to 2010, inclusive, in five European countries – France, Germany, Italy, Spain, and the United Kingdom. We find that, generally, the frequency of genres is homogeneous and that the same types of films dominate the highest reaches of the box office charts; while the number of films unique to a country and the variation among production sources within a country is strongly associated with the distinction between international ‘technology-friendly’ films (action/adventure, fantasy/science fiction, and animated family films) and domestically produced ‘technology-unamenable’ genres (comedy, drama, crime/thriller, romance, and non-animated family films). The results suggest the concepts of national cinema and genre are closely interrelated, and that for audiences in these five European countries the decision about which films to see presents itself as a choice between genres that is often also a choice between Hollywood films and domestic films.

 

Genre and European box office, 2006-2010

UPDATE: This post has now been superseded by a revised version that cleans up the data and extends the analysis and should be referred to in place of this. See here for the new version.

To round off a series of posts on genre and box office this August, I look at the frequency of different genres in five European countries – France, Germany, Italy, Spain, and the UK – to see what we can learn about different national markets.

For each of the five countries, I accessed the data from Box Office Mojo for the top 50 grossing films in each year from 2006 to 2010, inclusive. (For some reason, Box Office Mojo lists some films twice in the same year if they have slightly different titles; and I removed these duplicates to replace hem with the next film in the box office rankings). This gives a total sample size of 250 films for each country, and a total of 1250 data points overall. Obviously this does not mean we have data on 1250 films because many of the films reached the top 50 in more than one country. Overall, this sample has data on 596 different films.

Usually I use a system of nine categories for sorting films according to genre; but due to the fact that the number of horror films reached double figures for Spain and the UK only (with 11 and 10 films, respectively) and were very small in number for the other countries (France = 2, Germany = 7, and Italy =6), I have put this films into the category of ‘Other.’ Obviously the fact that horror infrequently reaches the top of the box office charts is interesting in itself, as is the French aversion to horror.

The eight categories used are, therefore, Action/Adventure, Comedy, Crime/Thriller, Drama, Family, Fantasy/Science Fiction, Romance, and Other. Alongside Horror films, Other also includes Westerns, War films, Musicals (including concert films), and Documentaries.

First, we look at the frequency of films occurring in each country in each genre (Table 1).

Table 1 Genre frequency in the top 50 grossing films in five countries, 2006-2010 (NB: the Total column to the right is the number of data points for each genre and NOT the number of different films)

Overall, the number of films from each genre to make it into the top 50 films in the five years covered is similar in each country. To test if the proportion of films from each genre was the same in the five countries, I performed a chi-square test of homogeneity (corrected α = 0.0131, based on 8 tests and an experiment-wise error rate of α = 0.10). These results are presented in Table 2, and show that the only statistically significant difference occurs for the comedy genre. Post-hoc analysis of the adjusted standardized residuals (based on a two-tailed critical z-value of 2.5596) revealed that this is due to Spain having fewer comedy films than expected (z = -3.6880), but the effect size for omnibus test is small (V = 0.1122).

Table 2 Chi-square test of homogeneity for the proportion of films in each genre in five countries

With the exception of the missing comedy films in Spain, these five different markets appear to otherwise very similar for each genre. However, this does not mean that audiences in these five countries are necessarily watching the same films.

To find out if the same films were making it into the top 50, I counted the number of times a film featured in the list of films for each genre. For example, if a film only made it into the top 50 in Germany (e.g. Elementarteilchen (Atomised)) then it would appear only in the list of drama films only once, while a film that made it into the top fifty in all five countries (such as one of the Harry Potter films) would appear in the list of Fantasy/Science Fiction films five times. This is a somewhat crude measure, but it does allow us to see some basic commonalities and differences. This information is presented in Table 3.

Table 3 Frequency with which individual films make the top 50 highest grossing films in five countries from 2006 to 2010 (NB: the Total column to the right is the number of different films in each genre in the overall sample)

Table 3 reveals three distinct patterns:

• Action/Adventure films tend to feature in the lists for four or five different countries (59%). This is the only genre for which this is the case.

Generally, these films a big-budget Hollywood franchise films such as The Fast and the Furious: Tokyo Drift and Fast and Furious, Iron Man and Iron Man 2, Pirates of the Caribbean: At World’s End and Pirates of the Caribbean: Dead Man’s Chest, and the like. Just less than a quarter of these films feature in only one list, but even these tend to be Hollywood films (e.g. Watchmen or Resident Evil: Extinction*).

* Resident Evil: Afterlife did much better though, ranking everywhere except the UK.

• The genres of Comedy, Crime/Thriller, Drama, and Romance and dominated by films that appear in one list only.

If Hollywood is able to dominate the global market with its action movies, then it is much less successful when it comes to these four genres. Comedy, in particular, seems to be very different with 78% of films appearing in the list for only one country. Some of these are individual Hollywood films that have performed well in one country not the others; but many are films that only feature in the list of the country in which they were produced. For example, the series of Christmas comedy films from Italy directed by Neri Parenti has performed exceptionally well in that country: one film has made the top 5 grossing films in each year in the sample, with Natale in crociera (2007) and Natale a Rio (2008) both taking the number 1 ranking. However, these films have not made any impact at the box office in any of the other European countries included here. Four comedy films made it into list of each country (Burn After Reading, Mr. Bean’s Holiday, The Devil Wears Prada, and The Hangover).

The Crime/Thriller genre features several big-budget Hollywood films that were successful in all five countries (The Da Vinci Code, Angels and Demons, No Country for Old Men, The Bourne Ultimatum, etc), but again these five markets are more different than they are similar. Some films that appear only once are Hollywood films (e.g. The Taking of Pelham 123, State of Play – neither of which are as good as the originals); but most are successful only in the country in which they originate. So Un prophète and Ne le dis à personne feature in the French box office charts only; and Gomorra and Milano-Palermo: il ritorno only in the Italian charts.

Only a few drama films appear in the top 50s of all countries (Australia, Blood Diamond, Brokeback Mountain, Shutter Island, and The Pursuit of Happyness), while 73% feature in one list only. Romance films show the same pattern, with only seven (13%) films featuring five times (and three of these are from the Twilight franchise), and 65% of films featuring once only. The drama and romance films that appear once tend to feature only in the country from which they originate, but when drama films do cross borders they go between the continental countries and not tot the UK. For example, Das Leben der Anderen (The Lives of Others) features in every country except the UK. There does not appear to be the same level of cross-over for the romance films, and when a film from this category appears more than once it tends to be a Hollywood film.

Laughter and love do not apparently travel well – in the cinema at least. And nor do crime and drama. The five markets are much less homogenized in these categories, unlike the Action/Adventure films where they are much more consistent in terms of the films in circulation. This clearly raises question about the extent to which we can speak of the Americanization or globalization of European cinema, as it appears to affect some categories of films more than others.

Finally, the third set of genres:

• The genres of Family and Fantasy/Science Fiction are split between films that feature in one list only and films that feature in the box office charts of all five countries.

For the family genre, 41% of films feature once and 39% of films feature five times. For the Fantasy/Science Fiction films, the equivalent statistics are 42% and 29%. This suggests that there is a divide in the market for these films. The majority of the films in these two genres are Hollywood blockbusters no matter how many time they occur. But we do see a clear split between films that are broadly successful against films that do not travel across borders so well; especially when it comes to animated family films that perform well in all markets (e.g. Cars, Flushed Away, Ice Age: The Meltdown) alongside several European animated films that appear – yet again – only in the country of their production (e.g. Konferenz der Tiere in Germany, El ratón Pérez in Spain, or Azur et Asmar in France). Separating out the UK is much harder as many of the Hollywood films are produced here anyway.

As Other is a category comprising films from several other genres it makes little sense to speak of trends, but it is interesting to note that the three films that feature in all five lists are High School Musical 3: Senior Year, Inglorious Basterds, and Mamma Mia!

As I said before, this is a crude way of measuring differences in audience taste, and I won’t have a much richer picture until I start to compare the box office gross of films in each country directly. But what the information in the above tables provides is a means of describing the national specificity of film a markets based on the types of in circulation and which achieve the highest box office rankings. There are many similarities between these five countries, but we should want to know why the Spanish do not go and see as many comedy films as the British, Germans, French, and Italians? Why do we all seem to watch the same Action/Adventure films but not the same Drama films? Perhaps the specificity of a national cinema is only evident in some categories of films and not others; or Hollywood has cornered the market on such blockbusters to the exclusion of all other producers. Why, if the audiences in these five countries are watching mostly different Romance films, is the proportion of films from this genre in the 250 films for each country so similar? Is there a common underlying structure to European film a markets? Why did the British not pay to see Resident Evil: Afterlife unlike the rest of Europe? And where are the French horror films?

The romance genre at the box office in five European countries, 2006 to 2010

Assuming I have not been defeated by the rivers Wharfe, Aire, and Ouse I shall today be presenting a paper at the International Association for the Study of Popular Romance conference in York (though it is entirely possible that I’m stuck in York railway station). Below is the basic text of my presentation from which I will have inevitably digressed enormously. The pdf file is below. This is based on the same data I used in earlier post on genre and European box office although it has been cleaned up a little so the results are slightly different, though this does not have any impact on the conclusions.

Nick Redfern – The romance film at the box office in five European countries

We analyze the box office performance of romance films in five European countries – France, Germany, Italy, Spain, and the United Kingdom – from 2006 to 2010, inclusive, based on the top 50 grossing films in each country in each year. The results show that romance films account for only a small proportion of the films to reach the top 50 highest grossing films, and that there is no statistically significant variation in the proportion of romance films among the highest grossing films in each country. However, few romance films achieve a high box office ranking in more than one of these countries, indicating a lack of commonality across different markets with different audiences watching different romance films. Romance films achieving top 50 rankings in Germany, Spain, and the UK originate almost exclusively from outside these countries, whereas domestically produced films account for a larger proportion of romance films in France and Italy. Romance films perform consistently at the box office in three of the five countries, albeit lacking the very high grosses achieved by action/adventure, family, and fantasy/science fictions films; while this genre performs particularly poorly in Italy and Spain. Romance films emerge as a fixed part of the exhibition market in all five countries, but the variation in the films viewed, source of productions, and box office grosses indicates some important national differences.

Take a deep breath …

For the past three and a half years I have added a post to this blog every Thursday covering a range of topics and hopefully introducing you to some new ideas that aren’t just the same old film studies.

At present it is not feasible for me to carry on producing one post a week as I have just too much data to work through on slasher films, a whole host of RKO musicals, Scandinavian crime films (they really do get everywhere), and Asian horror films. As time passes it gets harder to work with data because you start to forget all the little details you observe during the collection phase and so I want to make some progress in these areas without too many distractions. (It turns out you can play Championship Manager 99-00 on a Windows 7 machine, and that is all the distraction anyone needs).

Consequently, I won’t add any  more posts until I’ve managed to get on top of the mass of data I have accumulated (assuming I don;t add more to the pile) or unless something particularly annoys me or I find something worth commenting on (so it’s entirely possible normal service will be resumed next week).

To keep you going in the meantime, here is an interesting article published 9 days ago in Frontiers in Human Neuroscience:

Dudai Y 2012 The cinema-cognition dialogue: a match made in brain, Frontiers in Human Neuroscience 6: 248.

That human evolution amalgamates biological and cultural change is taken as a given, and that the interaction of brain, body, and culture is more reciprocal then initially thought becomes apparent as the science of evolution evolves (Jablonka and Lamb, 2005). The contribution of science and technology to this evolutionary process is probably the first to come to mind. The biology of Homo sapiens permits and promotes the development of technologies and artefacts that enable us to sense and reach physical niches previously inaccessible. This extends our biological capabilities, but is also expected to create selective pressures on these capabilities. The jury is yet out on the pace at which critical biological changes take place in evolution. There is no question, however, that the kinetics of technological and cultural change is much faster, rendering the latter particularly important in the biography of the individual and the species alike. The capacity of art to enrich human capabilities is recurrently discussed by philosophers and critics (e.g., Arsitotle/Poetics, Richards, 1925; Smith and Parks, 1951; Gibbs, 1994). Yet less attention is commonly allotted to the role of the arts in the aforementioned ongoing evolutional tango. My position is that the art of cinema is particularly suited to explore the intriguing dialogue between art and the brain. Further, in the following set of brief notes, intended mainly to trigger further thinking on the subject, I posit that cinema provides an unparalleled and highly rewarding experimentation space for the mind of the individual consumer of that art. In parallel, it also provides a useful and promising device for investigating brain and cognition.

And here is the National Media Museums report on the first ever colour motion picture:

The report form the Guardian is here.

Yet more visual illusions

We haven’t had any visual illusions on this blog for a while, and since the poster recently released for Ram Gopal Varma’s Bhoot Returns depends on a visual illusion this seems as good as time as any.

It’s surprising that more films do not choose to use visual illusions in their marketing materials, but some  nice examples based on Disney films by Rowan Stocks Moore can be found here. The Peter Pan and Snow White posters in particular stand out.

Archimedes Lab has many different illusions and oddities from Gianni Sarcone and Marie Waeber, which  you can access here. There is also a great selection of vintage illusions dating back 2500 years.

The finalists for this year’s Illusion of the Year contest can be found here, with attractive celebrities that turn ugly and a great interactive demonstration of the wagon wheel illusion. There is also an illusion inspired by the infamous twisting neck scene from The Exorcist which you can see below if you’re brave enough. The effect is much more eerie than anything you could do with CGI.

io9 has a dedicated illusions channel, which has lots of different examples of visual illusions and articles covering a range of issues including the art of anamorphic illusions and why our pupils contract when looking at illusions that are not bright lights.

This last example comes from the pages of Akiyoshi Kitaoka, and you can find details of his latest work here.

Finally, a good collection of illusions, including some of those listed above and in my other posts on this topic (here and here), can be found in New Scientist’s ‘Friday Illusion’ column.

Some notes on cinemetrics V

This post addresses some issues raised by Mike Baxter as part of the ‘cinemetrics conversation’ at the Cinemetrics website (and is the post I would have produced last week had I been able to remember which bit of software had the right command to create the necessary graph). You can find an introduction to the conversation here and my first response to some of the issues raised here.

I want to address two issues: first, the nature of outliers in shot length distributions and better methods of representing such distributions than I have used up to now; and, second, the straw-man the median shot length has become in Baxter’s comments.

Baxter’s comments in response to the earlier can be found in the second tab under his name here. In section 2 Baxter questions my use of the term ‘outlier’ and the definition used to identify such shots.  This is fair enough – we wouldn’t get very far if such definitions weren’t questioned. In the examples of Lights of New York and Scarlett Empress, Baxter argues there is no evidence of outliers since

it’s difficult to identify any point at which ‘extremes’ begin, or discontinuities in the distribution of the kind I think are needed to assert, with any confidence, that you are dealing with ‘outliers.’

Baxter never defines what such a discontinuity would look like and so his argument is vague. (Arguably this is the semantic version of a slippery slope).

Figure 1 shows the kernel density and box plot of Lights of New York. There is a 12.2 second gap between the five shots of longest duration and the sixth longest, presumably the sort of discontinuity Baxter refers to and he does concede he might be prepared to accept five shot lengths as extreme values (though he does not say on what basis). From Figure 1 we can see there are in fact several such discontinuities, and that the kernel density is zero at several points in the upper tail (indicating the kernels do not overlap), particularly above 30 seconds (which corresponds to the 22 extreme outliers identified using this type of box plot). However, a limitation of this boxplot is that it does not take into account the skew of the distribution and so over identification of outliers is a problem.

Figure 1 Kernel density and boxplot of shot lengths in Lights of New York (1928)

Figure 2 presents the same data using an adjusted boxplot that takes into account the skewed nature of the data. This method uses the med-couple, a robust measure of skewness, to identify outliers. The adjusted boxplot can be generated using the adjbox() command in the R package robustbase.

The number of outliers in Figure 2 is much less than in the original boxplot: in the upper tail 10 shots greater than 55 seconds are identified as outliers (or 3% of the total).  Nonetheless, there are still some values which are sufficiently removed from the rest of the data to be classed as outliers even when accounting for the asymmetry of the distribution. Whether or not Baxter would accept this definition would depend on the interpretation of his use of the term ‘discontinuity,’ which he does not define.

Surprisingly, this method identifies three outliers in the lower tail of the distribution (which I wasn’t expecting and will have to think about more).

Figure 2 Kernel density and adjusted boxplot of shot lengths in Lights of New York (1928)

The following article describes the adjusted boxplot and its calculation:

Vandervieren E and Hubert M 2008 An adjusted boxplot for skewed distributions, Computational Statistics and Data Analysis 52 (12): 5186-5201. An ungated, earlier version of this paper can be accessed here.

Even if we accept Baxter’s argument that there are no outliers in Lights of New York it remains necessary to be aware of the problems caused by outliers in data sets and to check the distribution of shot lengths so that we  are not be fooled by non-robust statistics. Certainly more effort will have to be devoted to defining what is or is not an outlier (in either statistical or filmic terms) in research if this type. (But it is much easier when you remember which bit of software to use).

Finally, I wish to address a misrepresentation that has taken a hold at this early stage in the ‘cinemetrics conversation.’

Baxter writes

the use of either the ASL or median as the statistic for attempting to summarise ‘style’ doesn’t make much sense (as Salt observes) [original emphasis].

This argument is a straw-man.

I have never stated that the median shot length is the statistic for describing film style. I have argued that the median shot length is better than the mean shot length for describing film style, and should therefore be preferred for the following reasons:

• the median is conceptually simple and easy to calculate, and is certainly no more difficult than the mean.
• the median shot length has a clearly defined meaning and the difference between two median shot lengths is also meaningful, whereas the meaning of the mean the difference between two mean shot lengths is not clear in either case (and seem to change every time I raise an objection against them).
• the median shot length is not affected by a monotone transformation (the median of a data set is the same as the median of the logarithmic transformation of a data set), while the possibilities for confusing the arithmetic and geometric means are endless.
• the median locates the centre of a distribution irrespective of its shape, whereas this is not true of the mean.
• the median is not affected by outliers or extreme values (however you choose to define them), whereas this is not true of the mean.
• interpretations of film style based on the median shot length are consistent with graphical methods and (it turns out) with dominance statistics (Cliff’s d, HLΔ), while those based on the mean shot length are not.

But I have always argued that it is important use a range of statistical methods to get a full understanding of the nature of film style.

As far as I am aware I am the only person writing about film style to even consider the dispersion of shot lengths in a motion picture and the appropriate methods to use this. I am also the only person to use a range of graphical methods (probability plots, boxplots, empirical cumulative distribution functions, kernel densities, order structure matrices, running Mann-Whitney Z statistics, rank-frequency plots) to describe film style. I am the only person in film studies to employ confidence intervals, statistical hypothesis tests, effect sizes, or even to describe the methodologies I use in studying film style. (Others working outside films studies in disciplines where quantitative methods are commonplace also use such tools as a matter of routine, and those within film studies would do well learn by their example).

I am also the only person who has attempted to describe these methods so that others may try to analyse film style for themselves. I am the only person who has brought to the attention of researchers in film studies the availability of free learning resources and software for statistics. I am the only person to look outside film studies for empirical research on film style and to bring it to the attention of film scholars. I am the only person to address the issue of statistical literacy in film studies (here and here).

Baxter writes that

the accessibilty of computational power, and essential simplicity of important statistical ideas (however mathematically complex) is a hobby-horse of sorts.

I am glad to hear this, because it means that if someone else is prepared to devote some time and effort to explaining statistical concepts and methods to film scholars then I won’t have to do it on my own.

However, as Baxter presents the argument I am interested in the median shot length only while Barry Salt apparently does not have a narrow attachment to a particular statistic of film style and embraces a pluralistic approach. However, I am not aware of any forum in which Salt has made any concession to his view that the mean shot length is the only appropriate statistic of film style. In fact, I am unaware of any other statistics of film style used by Salt besides the average shot length and the histogram (while his odd comments on the calculation of kernel density estimates indicates he may not properly understand other methods).

Baxter has his argument back to front here: you won’t find methodological ecumenism in the statistical analysis of film style in the work of Barry Salt.