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.

Robust time series analysis of ITV news bulletins

I have mentioned numerous times on this blog the importance of using robust statistics to describe film style. This week I continue in this vein, albeit in a different context – time series analysis. In a much publicised piece of work James Cutting, Jordan De Long, and Christine Nothelfer (2010) calculated partial autocorrelation functions and a modified autoregressive index for a sample of Hollywood films. While I have no problems with the basis of this research, I do think the results are dubious due to the use of non-robust methods to determine the autocovariance between shot lengths in these films. The paper attached below analyses the editing structure of the set of ITV news bulletins I discussed in a paper last year, comparing the results produced using classical and robust autocovariance functions.

Robust time series analysis of ITV news bulletins

In this paper we analyse the editing of ITV news bulletins using robust statistics to describe the distribution of shot lengths and its editing structure. Commonly cited statistics of film style such as the mean and variance do not accurately describe the style of a motion picture and reflect the influence of a small number of extreme values. Analysis based on such statistics will inevitably lead to flawed conclusions. The median and  are superior measures of location and dispersion for shot lengths since they are resistant to outliers and unaffected by the asymmetry of the data. The classical autocovariance and its related functions based on the mean and the variance is also non-robust in the presence of outliers, and leads to a substantially different interpretation of editing patterns when compared to robust time statistics that are outlier resistant. In general, the classical methods underestimate the persistence in the time series of these bulletins indicating a random editing process whereas the robust time series statistics suggest an AR(1) or AR(2) model may be appropriate.

The pdf file is here: Nick Redfern – Robust Time Series Analysis of ITV News Bulletins

My original post on the time series analysis of ITV news bulletins can be accessed here, along with the datasets for each of the fifteen bulletins.

My new results indicate the conclusions of Cutting, De Long, and Nothelfer are flawed, and that it is very likely they have underestimated the autocovariance present in the editing of Hollywood films. The discrete and modified autoregressive indexes they present are likely to be too low, though there may be some instances when they are actually too high. This is not enough to reject their conclusion that Hollywood films have become increasingly clustered in packets of shots of similar length, and I have not yet applied this method to their sample of films. It is, however, enough to recognise there are some problems with the methodology and the results of this research.

References

Cutting JE, Delong JE, and Nothelfer CE 2010 Attention and the evolution of Hollywood film, Psychological Science 21 (3): 432-439.

Time series analysis of ITV news bulletins

Back in the summer I wrote a post looking at the relationship between the discourse structure and the formal structure of BBC news bulletins (see here). This week I have the first draft of a similar paper looking at news bulletins from ITV.

The pdf file can be accessed here: Nick Redfern – Time series analysis of ITV news bulletins

Abstract

We analyze shot length data from the three main daily news bulletins broadcast on ITV 1 from 8 August 2011 to 12 August 2011, inclusive. In particular, we are interested to compare the distribution of shot lengths of bulletins broadcast on different days and at different times across this time period, and to examine the time series structure by identifying clusters of shots of shorter and longer duration in order to understand the relationship between this aspect of the formal structure to the discourse structure of these broadcasts. The discourse structure of the bulletins in this sample is fixed, and remains constant irrespective of the subject of news items themselves suggesting that content is adapted to meet the needs of this structure. The statistical results show that neither the day nor the time of broadcast has any impact on the distribution of shot lengths, and the editing style is consistent across the whole sample. There is no common pattern to the time series of these bulletins, but there are some consistent features in the time series for these bulletins: clusters of longer takes are associated with static shots of people talking on-screen, while clusters of shorter takes occur with montage sequences, sports reports, series of news items, and footage from non-ITN sources. Consequently, the presence and order of discourse elements in a bulletin shapes its formal structure.

The data for the bulletins used in this study can be accessed as an Excel 2007 file here: Nick Redfern – ITV News Bulletins

I’m a little wary of making direct comparisons between this data and that of the BBC news bulletins as they are separated by three months and deal with news presentation in very different circumstances. The data used in the ITV study covers the week of the riots in the UK this August, and this presents a very different news cycle to that seen in the BBC data from April. However, some general points can be made:

• In both samples clusters of longer shots are associated with people speaking at length on camera, and these shots are framed in the same way.
• In both samples clusters of shorter shots are often associated with montage sequences accompanied by a description from an off-screen reporter or with footage that is derived from other sources (e.g. library footage, other broadcasters).
• In both samples, there is no evidence of any trends or cycles in the time series.
• There is no significant difference in the median shot lengths and dispersion of shot lengths in the two samples of bulletins (BUT remember these are from different times of the year, so this information is only of limited use).
• Day and time of broadcast have no impact on news bulletins for either broadcaster (but again the comparison is not as direct as I would like).

Overall, there is some evidence that news bulletins are stylistically homogenous across these broadcasters. I will do another study looking at the comparing the bulletins from the both the BBC and ITV from a single week, but this will have to wait for another day.

Empirical research on television style

The various selections of empirical research on film style that I have posted on this blog have been dominated by feature films, and television has received rather less attention. In part, this is because there is a lack of research on television style: Schaefer and Martinez (2009) (see below) write:

that formal study of the craft of editing television news appears to have suffered from a lack of a conventional vocabulary for describing and analyzing structural techniques used in what is primarily an audio-visual phenomenon, maintaining that television journalists have traditionally learned the evolving art of news shooting and editing through an immersion process that does not readily lend itself to conscious articulation of forms. Hence, it should not be too surprising that discussions of the evolution of trends in journalistic editing are often based on scant anecdotal evidence.

The ‘dearth of formal analysis’ is something that can be easily remedied: the materials are easy to access and the methods are well known. A question worth exploring is why has no one done this research?

A few weeks ago I posted a draft version of a paper on the statistical analysis of style in BBC news bulletins (here), and I am currently part way through a similar paper on news broadcasts on ITV. But until then, here are some papers worth reading on television style that address various issues that have been raised elsewhere on this blog.

Bolls PD, Muehling DD, and Yoon K 2003 The effects of television commercial pacing on
viewers’ attention and memory, Journal of Marketing Communications 9 (1): 17–28.

This study investigated the effects of advertising pacing (i.e. the number of visual cuts in an advertisement) on viewers’ voluntary and involuntary attention to an advertisement, as well as its effects on the recall of claim-related and non-claim-related components of the advertisement. Using a limited capacity model of information processing/retrieval as its theoretical base and physiologically oriented measures of attention, this study provided some evidence that fast-paced advertisements (as compared to slower paced ones) may have a positive effect on viewers’ involuntary (automatic) attention towards an advertisement, but have little differential effect on their voluntary attention. Furthermore, it appeared that the enhanced involuntary attention gained through the use of fast-paced advertisements comes in the form of attention directed towards the non-claim (advertisement executional) elements of an advertisement as opposed to the message-based (copy) elements of the advertisement. The practical and theoretical implications of these findings are discussed.

Choi YJ and Lee JH 2006 The role of a scene in framing a story: an analysis of a scene’s
position, length, and proportion, Journal of Broadcasting and Electronic Media 50 (4):
703-722.

A scene is proposed as the unit of analysis in broadcast news studies as a way to measure a more accurate representation of perspectives and arguments of a story. Based on film studies, a scene is defined as a unit that represents continuity in time, place, character, ideas, or themes in a news story. The role of a scene in a news story is analyzed by examining how the position, length, and proportion of a scene frame and valence are related to story frame and valence.

McCollum JF and Bryant J 1999 Pacing in children’s television programming, Annual Meeting of the Association for Education in Journalism and Mass Communication, 4-7 August 1999, New Orleans, LA.

Following a content analysis, 85 children’s programs were assigned a pacing index derived from the following criteria: (1) frequency of camera cuts; (2) frequency of related scene changes; (3) frequency of unrelated scene changes; (4) frequency of auditory changes; (5) percentage of active motion; (6) percentage of active talking; and (7) percentage of active music. Results indicated significant differences in networks’ pacing overall and in the individual criteria: the commercial networks present the bulk of the very rapidly paced programming (much of it in the form of cartoons), and those networks devoted primarily to educational programming–PBS and The Learning Channel–present very slow-paced programs. (Contains 26 references, and 12 tables and a figure of data.)

Pak H 2007 The Effects of Incongruity, Production Pacing, and Sensation Seeking on TV Advertisements, unpublished Ph.D. Thesis, Cornell University.

This study addresses an important area of research that has fascinated advertising professionals who are eager to make more attractive ads: understanding how the viewing audience perceives and processes television advertisements. Ad incongruity, the introduction of unexpected elements that are atypical of a given ad category, and production pacing were tested to explore the roles of these stimuli in capturing higher levels of arousal, which can produce both better evaluations and clearer memories of ads. Sixty subjects, who were recruited from among undergraduate students at Cornell University and patrons of a local shopping mall, participated in an experiment in which a set of TV ads was shown. Participants then answered questions immediately following exposure to the ads, providing data pertaining to sensation seeking, ad evaluation, arousal, and memory. The ads themselves represented six different conditions: incongruent and slow paced, incongruent and medium paced, incongruent and fast paced, congruent and slow paced, congruent and medium paced, and congruent and fast paced. The main findings involved Lang?s limited capacity model. It was found that the mental capacity or cognitive load required to process incongruent fast-paced ads exceeded study participants? cognitive capacity to process the information in such ads. When ads with both fast paced and incongruent elements were shown, participant?s memory for that particular kind of ads declined. The study provided confirmation of Lang?s (2000) limited capacity model. The study?s contributions include a key finding pertaining to incongruity effects that should help to resolve discrepancies in the literature on incongruity. As expected, incongruent ads were evaluated more positively, and were more arousing and better remembered than congruent ads. Production pacing also had some effect on participants. As pacing increased, participants remembered better and ad evaluations tended to be more positive. However, ad type had a significant influence on the processing of ads. Car ads were evaluated more positively, were more arousing, and were better remembered than over-the-counter drug ads. There were no significant relationships between sensation seeking and incongruity or sensation seeking and production pacing.

Schaefer RJ and Martinez TJ 2009 Trends in network news editing strategies from 1969 through 2005, Journal of Broadcasting and Electronic Media 53 (3): 347-364.

Four editing variables were tracked through a content analysis of U.S. commercial network editing that spanned a 36-year period. The analysis revealed that synthetic-montage increased and continuity-realism decreased from 1969 through 1997. Network news editors also embraced faster pacing, shorter soundbites, and more special effects between 1969 and 2005. When taken together, the results suggest that U.S. network television journalism has evolved from more “camera of record” and realistic news techniques in favor of a variety of synthetic editing strategies that convey complex audio-visual arguments.

Finally, this paper has appeared in the latest issue of Pediatrics though I am somewhat dubious about its methodology due to the way they have defined the pace of the programmes used, which appears to be based on scene transitions rather than shot transitions:

Lillard AS and Petersen J 2011 The immediate impact of different types of television on young children’s executive function, Pediatrics 128 (4): doi: 10.1542/peds.2010-1919.

Objective: The goal of this research was to study whether a fast-paced television show immediately influences preschool-aged children’s executive function (eg, self-regulation, working memory).

Methods: Sixty 4-year-olds were randomly assigned to watch a fast-paced television cartoon or an educational cartoon or draw for 9 minutes. They were then given 4 tasks tapping executive function, including the classic delay-of-gratification and Tower of Hanoi tasks. Parents completed surveys regarding television viewing and child’s attention.

Results: Children who watched the fast-paced television cartoon performed significantly worse on the executive function tasks than children in the other 2 groups when controlling for child attention, age, and television exposure.

Conclusions: Just 9 minutes of viewing a fast-paced television cartoon had immediate negative effects on 4-year-olds’ executive function. Parents should be aware that fast-paced television shows could at least temporarily impair young children’s executive function.

Time series analysis of BBC news bulletins

Following on from my use of running Mann-Whitney Z statistics to look at the time series structure of Top Hat (here), this week I have the first draft of an analysis of 15 BBC news bulletins using the same method.

The pdf file can be accessed here: Nick Redfern – Time series analysis of BBC News bulletins

Abstract

Shot length data from 15 news bulletins broadcast at 1300, 1800, and 2200 on BBC 1 between 11 April 2011 and 15 April 2011, inclusive, is used to compare the editing style between different bulletins broadcast at different times on different days and to examine the time series structure by identifying clusters of shots short and long duration. The results show there is no evidence that shot length distributions of BBC news bulletins vary with the time or day of broadcast, and the style of editing is consistent across the sample. There is also no evidence the highly structured format of television news is related to the time series of shot lengths beyond the opening title sequence, which is associated with a cluster of short shots in every bulletin. The number, order, and location of clusters of longer and shorter shots is different for each bulletin; and there are several examples of abrupt transitions between different editing regimes, but no evidence of any cycles present in the time series. Although there is no overall common pattern to the editing, there are some consistent features in the time series for these bulletins: clusters of shorter shots are associated with footage derived from non-BBC sources (library footage, other broadcasters, public information films) and montage sequences; while clusters of shots of longer duration are associated with shots in which the viewer is addressed directly by the presenter or reporter (including graphics), live-two-way interviews, and speeches or interviews with key actors in a news item.

The data is described in the above paper and can be accessed as an Excel 2007 file here: Nick Redfern – BBC News Data