Particip@tions Vol. 1 (1): Blumler et al. Chapter 4

The Third Survey and its Typological Implications: Television News, The Saint and Callan

Three objectives shaped the design of the investigators’ third main excursion into the field. First, they aimed to extend the range of broadcast programmes concerning which they would have collected and interpreted gratification data. Having perfected an empirical procedure for eliciting the structures of audience gratifications relating to particular programmes, it remained to apply this to two or three still unexplored areas of television content. It was hoped that such an extension would help to build up a typology of the gratifications associated with television viewing – a task which had always been envisaged as a vital element in the project’s preparations for undertaking an eventual nationwide survey of viewers. (The main body of this chapter deals with this part of the outcome of the third exploratory survey.)

Second, the investigators used the third survey as an occasion for trying out a version of a ‘general’ gratification instrument on a fairly large number of respondents. They had been working for some months on various methods of tapping the audience satisfactions that stem from television viewing in general (as distinct from those which derive from a single programme), and the time was now ripe to conduct a field trial of a seemingly promising technique. (This aspect of the work in discussed in Chapter 5 of this report.)

Third, the investigators aimed to collect material that would help them to assess and refine the project’s stock of social indicator variables. Consequently, the questionnaires for the third survey included many items about the social position of respondents – some of which had been incorporated into previous surveys,[1] while others were being administered for the first time.[2] It was intended in the autumn of 1970 to carry out AID analyses of the social determinants of the cluster scores generated by two of the new programme studies, after which the project’s social indicator needs would be critically reviewed.

Procedures

1. Choice of further programmes for investigating. It was initially decided that the third survey should focus on the regular viewers of three quite different types of programmes: one in the news and current affairs field; an adventure series; and one which could be supposed to establish ‘para-social’ relationships between performers and audience members.[3] Television news itself offered the most familiar available focus for the first of these investigations; as a long-running and highly popular series, The Saint was adopted as the basis for the second study; and as a talk programme featuring interaction between a personable compere and outstanding personalities from the entertainment world and other fields of public activity, The Dee Show seemed a suitable choice for the third enquiry. Consequently, a series of tape-recorded group discussions with male and female viewers of each of these programmes was held in April and May, 1970, and the material thus obtained was analysed by the research team before formulating items for inclusion in inventories. In the middle of these preparations, however, a dispute between Simon Dee himself and London Weekend was widely publicised, after which the latter announced that the programme would shortly be terminated. Fearing that this incident might distort viewers’ responses to questions about the programme, the Dee study was abandoned. Meanwhile, it had been noticed that many fans of The Saint had in group discussions spontaneously compared and contrasted its tone and approach with that of Callan, a then popular adventure series, featuring less glamorous settings and a reluctant member of the British secret service as an ‘anti-hero’. It was decided, therefore, to substitute Callan for The Dee Show as the third programme focus. Since shortage of time precluded the holding of group discussions about Callan, it was also decided to use the same scales that had been generated for investigation of The Saint, in the hope that two programmes which differed in style while belong to the same broad content category could be fruitfully compared.

2. Gratification inventories. The format of the gratification inventories combined modified versions of those that had previously been used in the quiz and Coronation Street studies. The news inventory consisted of a list of 25 gratification statements, which were followed by five-point agreement/disagreement scales. Sensitisation to the perils of response set resulted, however, in the preparation of a more spacious and clearly differentiated format, which seemed likely to avoid this type of bias. A specimen of the first two scales of the revised inventory illustrates this approach:

The same format was used for 24-item and 25-item inventories of gratification statements about The Saint and Callan. But in addition, respondents’ images of these programmes were elicited by administering 16 five-point semantic differential-type scales.

3. Sample size and composition. Interviews were held in the week of July 12^th, 1970, with the members of a quota sample (controlling for age, sex and social grade) of viewers who claimed to have watched The Saint or Callan regularly in recent months. The investigators’ target of 200 respondents was greater than in previous surveys in order a) the give the AID analysis larger sub-groups to work with and b) to give the recently developed ‘general instrument’ a broadly based trial. But since fieldwork supervision necessitated the rejection of the work of one of the nine interviewers as unreliable, the achieved sample actually comprised 177 respondents. All but three of these individuals filled in news viewing inventories, and either or both the adventure programme inventories were completed by the following numbers of respondents:[4]

The Results: A Cluster Analysis of News Viewing Gratifications

More has probably been written about the functions of news consumption to the individual citizen than about the gratifications sought from almost any other body of media fare. However, much of this literature has focused on newspaper reading (broadcast news satisfactions have been relatively neglected), and its outstanding themes serve to illustrate some of the confusions that can arise when research effort lacks a firm empirical foundation.

Three principal approaches have dominated contributions to this field. First, there is a tradition of qualitative endeavour which stems from Berelson’s classic study of what people missed when deprived of their newspapers by a strike. The chief outcome of intensive examination of open-ended material obtained from 60 respondents was the following list of five supposed news reading functions:

It is true that Berelson acknowledged that ‘“qualitative” interviews only suggest the proper questions which can then be asked in lesser detail for “quantitative” verification’; nevertheless, he did not personally venture onto the latter path.[5]

Shortly afterwards Wilbur Schramm initiated a second influential approach – one that stemmed essentially from a conceptual analysis. Drawing on Freud’s distinction between the pleasure and reality principles, he postulated two opposed motivational orientations towards newspapers – which he termed immediate reward- and delayed reward-seeking, respectively. The former was defined in terms of ‘either a reduction of tension or discomfort (e.g. curiosity, worry) or an increase of satisfaction (e.g. from a vicarious enjoyment of the achievements of a winning team).’ The latter was said to be sought ‘not because it is pleasant, but because it is realistic. It is not pleasant to be afraid or to anticipate danger; but it is necessary if one is to avert harm and avoid danger’. Thus, delayed reward news reading reflected a concern to acquire information that might help to orient future thought or action in responding to the problems of the real world. Immediate reward news reading, on the other hand, was virtually inspired by ‘escapist’ motives, as Schramm’s own remarks suggest:

When a reader selected delayed reward news, he jerks himself into the world of surrounding reality to which he can adapt himself only by hard work. When he selects news of other kinds, he retreats usually from the world of threatening reality toward the dream world.[6]

Schramm’s theory certainly did not fail to attract criticism. For example, William Stephenson drew attention to its tendency to confuse a distinction of type of reward (the pleasures of fantasy indulgence vs. those of reality manipulation) with one of time span (immediate vs. delayed reward).[7] But a more fundamental objection would have criticised the lack of a concerted effort to find out whether readers themselves were attuned to receive news materials on these two different wavelengths. Indeed, one contributor to this area of audience research has recently recognised that, ‘There is very little direct empirical evidence supporting Schramm’s theory because of the scarcity of motivational… studies in the field of mass communications’.[8]

Nevertheless, this shortcoming did not deter that investigator from adopting a third much-followed approach – that of administering to respondents a check-list of items purporting to reflect Schramm’s conceptual distinction (in his case in order to test a hypothesis about the content preferences that may accompany immediate and delayed reward newsreading dispositions). Perhaps the most glaring example of a pitfall to which this approach is prone can be found in the work of McLeod, Ward and Tancill on the influence of alienation on readers’ habits of newspaper use. These investigators formulated the following three-item scales to represent what they called informational (delayed) and vicarious (immediate) reasons for reading newspapers:

But although McLeod et al based their findings of their study on the assumption that these groupings stood for coherent tendencies, they had to admit in the end that:

Within the vicarious reasons, an assessment of the dimensionality of the various reasons should be made. In our present data, the correlation of ‘escape’ and ‘contact with other people’ was only +.097, perhaps indicating the presence of two or more vicarious sub-dimensions.[9]

All this illustrates the need for evidence which would disclose how citizens themselves structure the satisfactions they derive from following the news. The material presented below originates in the responses of 174 viewers to a 25-item inventory of ‘things that people have said about television news’, the component statements of which were drawn from analysis of group discussion material. Each item was followed by a five-point scale, ranging from ‘Strongly agree’ through ‘Undecided’ to ‘Strongly disagree’. The endorsements of sample members were punched onto paper tape and programmed to produce a 25 x 25 product moment correlation matrix. Application of McQuitty’s technique of elementary linkage analysis initially yielded a set of four clusters. Before accepting this outcome as final, however, it was decided to see whether the disclosed structure could be refined by applying statistical criteria. Using a coefficient of homogeneity (the mean of all intercorrelations of items in a cluster) and alpha, a coefficient of reliability recommended for use in cluster analysis, several items were shifted from one cluster to another, four items were eliminated from the analysis altogether, and one large cluster was split into two smaller ones. These procedures eventually produced five clusters, the individual items and homogeneity and reliability coefficients of which are set out in Table IV.1:

25. They shouldn’t show really unpleasant things on the news because there is nothing can we do about them.

The outcome of this analysis is exceptionally clear. All the assigned items have found homes in clusters, the meanings of which are not difficult to interpret. If the two-item Cluster 4 (which seems mainly to pick out a judgement of what TV is good at rather than a gratification sought from following the news on the medium) is ignored, then four main functions of news reading emerge from the results. Moreover, the cluster pattern seems broadly to confirm the validity of Schramm’s original distinction. At least what he would regard as a delayed reward has been differentiated in the findings from certain other item groupings that are more suggestive of immediate reward-seeking. Nevertheless, it also seems necessary to sub-divide the latter category into two main types of gratifications. And in addition to the orientations that Schramm has in mind, the cluster analysis locates yet another type of satisfaction that following the news can afford.

Cluster 3, which apparently represents a surveillance motive for news-viewing, is closest to Schramm’s notion of delayed reward-seeking. This has combined statements which refer to: following what is going on in the world generally (items 2, 7 and 10); the relationship between such events and the viewer’s own personal statements (20); the relationship between such events and the forming of judgements about the performance of power-holders in society (6); and the process whereby news materials can help viewers to make up their minds on current issues (8 and 11).

Cluster 3 may also be cited as an example of how the calculation of coefficients of homogeneity and reliability helped to strengthen the bonds between items in a group both statistically and substantively. Compared with its revised composition as shown in the table, the new Cluster 3 had included items 12 (‘It sometimes makes me feel sad’) and 22 (‘There is always something different on the news’) and excluded item 2 (‘Watching the news keeps me in touch with the world’). But after item 12 was reallocated to Cluster 5, item 22 was eliminated altogether, and item 2 brought into the surveillance group, Cluster 3’s coefficient of homogeneity had increased from .21 to .25 and alpha coefficient of reliability from .70 to .73.

Cluster 1 is one of the groups that is reminiscent of Schramm’s notion of immediate reward-seeking, for it combines four items that are all removed from the instrumental spirit of comprehending the world of reality cognitively. In this cluster the news is treated partly as a source of companionship (as any other type of broadcast fare can be) and partly as an equivalent to gossip. But the cluster is dominated by an ‘escapist’ meaning which is conveyed through both item 5, with its seemingly incongruous suggestion that following the news can help one to get away from one’s problems, and item 25, with its assertion of a definite desire to avoid the presentation of unpleasant incidents on the news.

The three items of the empathic Cluster 5 express yet another facet of what Schramm had in mind by immediate reward-seeking. In this case news viewing is said at one and the same time to promote an understanding of the problems that other people have to face (item 15), fresh insights into the nature of one’s own situation (13) and a feeling of sympathetic sadness (12). It is interesting to find in the distinction between this group and Cluster 1 a confirmation of McLeod’s suspicion that reasons for news reading worded, ‘To help me get away from my daily worries’ and ‘To feel as though I am taking part in others’ lives without actually being there’, had represented quite different motivational dimensions. It is also noteworthy that the various items of this empathic cluster seem to straddle the so-called clusters of reality exploration and personal reference that had emerged from the Dales study. Nevertheless, the separate existence of Clusters 5 and 3 in the news analysis suggests that a concern to use broadcast materials to reflect upon one’s personal identity must be differentiated from a concern to undertake a surveillance of the wider world of social and political reality. In fact, this distinction has been built into the project’s guiding typology of gratification categories.

Finally, Cluster 2 draws attention to an appeal of news viewing that Schramm’s concepts did not subsume. This refers to the use that can be made of news materials in social interaction with other people, either in the more casual spirit of giving the viewer ‘plenty to talk about’, or in the service of a more definite news-relaying role, as is suggested by item 4.

The Results: A Cluster Analysis of the Appeals of The Saint

The outcome of a cluster analysis of viewers’ endorsements of 40 items about The Saint is presented in Table IV.3.[11] It can be seen that a strongly defined pattern has emerged yet again from the application of this technique of analysis. There are four groups of items, which appear to locate the most important satisfactions to be derived from Saint viewing, followed by a set of three clusters which lack a direct motivational reference and represent little more than descriptions of certain facets of the programme. Perhaps the fact that most of the items in Clusters 5-7 were semantic differential scales points to the unsuitability of that instrument for use in a gratification study. But otherwise, four main functions of an adventure programme in the mould of The Saint appear distinctly in the results.

Cluster 3 – Identification .48 .73

First, the large Cluster 1 suggests that The Saint has a well-defined escapist appeal. This interpretation is specifically suggested by the inclusion of items 21 (‘It helps you to escape from the boredom of everyday life’) and 12 (‘It takes you out of yourself’) in the cluster. The other items are also exceptionally interesting, however, for the glimpses they provide of the ‘mechanics’ by which an adventure programme can serve an escape function. The various statements refer, for example, to the role of excitement (32), suspense (8 and 30), unusual backgrounds (9), conflict (20), and the larger-than-life accomplishments of the hero (10).Yet items 3 and 11 (with their references to the desiderata of family entertainment) also suggest that full enjoyment of these elements depends, in the case of The Saint at least, on an observance by the producers of certain conventions and proprieties.

Perhaps the emergence of Cluster 2 would be more of a surprise to some casual viewers of The Saint. The combination of items suggests that a second function which the programme can serve, for at least some fans (AID analysis would establish what kinds of people they are), is the provision of a stimulus to reflect upon real-life situations and problems. It may be that Cluster 2 derives in part from a tendency (which was previously encountered among the followers of The Dales and Coronation Street) for committed fans uncritically to regard their favourite programmes as ‘realistic’ – and suited, therefore, to ‘reality exploration’. But it could be that the functional potential of even a seemingly restricted range of fictional materials is richer and more diverse (especially when directed at audience members whose imaginations are free to roam) than the critic of popular culture is initially disposed to suspect.

According to Cluster 3, a third appeal of a programme like The Saint is the opportunity it affords to identify oneself with the qualities of a successful hero, either directly (as in item 7) or just possibly as a desired companion and sexual partner (as in item 15). In contrast to the richness of Cluster 1, the component items of Cluster 3 shed little light on the mechanics of this process. There is a hint, however, that viewer identification may be facilitated by the self-confident air of Simon Templar, which, when coupled with his easy mastery of every situation, must represent a marked contrast to the more fumbling approach of many viewers to the resolution of their own everyday problems.

Finally, these elements of identification seem to facilitate yet another function of The Saint, which is represented in Cluster 4. This is the personal reference appeal, or the by now familiar tendency for viewers to enjoy using fictional materials in the course of a dialogue with themselves which helps them to underline or to characterise some facet of their own situation and circumstances.

The Results: A Cluster Analysis of Viewers’ Endorsements of Statements About Callan

As can be seen from Table IV.3, the cluster analysis of viewers’ endorsements of 41 statements about Callan proved less satisfactory than the outcome of the corresponding Saint analysis. Even after they had been revised to enhance their homogeneity and reliability, the Callan clusters embraced more heterogeneous sets of items. And insofar as these could be related to identifiable appeals, they added nothing new to the Saint results. Probably the intention to compare the programmes by requiring respondents to apply statements, drafted originally to refer to The Saint, to both The Saint and Callan had itself been ill-conceived. For what it is worth, however, it may be pointed out, first, that the Callan analysis has yielded a personal reference cluster (Cluster 6), which is identical in composition to its counterpart in the Saint analysis); second, that it has also produced a similar reality exploration cluster (Cluster 4); third, that the escapist appeal of adventure programmes which had been expressed in one large cluster in the Saint analysis has seemingly been dispersed into three clusters in the case of Callan (Clusters 2, 5 and 8); and fourth, that insofar as Cluster 1 conveys a recognisable meaning at all, it seems to represent that identification process which was less ambiguously reflected in one of the Saint clusters.

Developing a Typology of Audience Gratifications

It is appropriate at this point to stand back from the detailed examination of particular results in order to raise a wider question. Can any general conclusions be drawn from the entire sequence of the project’s first-stage activities? Two answers suggest themselves. One is that the project’s original premise, which maintained that distinct sets of different gratifications orient important segments of the audience to popular forms of television programming, has been firmly supported. Of the validity of this assumption the technique of cluster analysis itself provided a stern test. It is true that the technique was bound to produce clusters. What was not determined in advance, however, was the fact that so many of the emergent clusters should have proved so coherent and clear in meaning. Only two of the six programme studies did not disclose a comprehensible gratification structure, and in each case the most likely source of the failure was a technical weakness. Thus, the first-stage research has underlined the place of audience gratifications in the mass communications system, as well as having generated a battery of techniques for detecting and measuring them and relating their incidence to other variables.

A second general conclusion concerns the manner in which the various cluster structures of the individual programme studies have overlapped and reaffirmed each other. This is an impressive outcome of the first-stage research, first, because the results stemmed from four quite different types of programme content (a domestic serial, an adventure serial, quizzes, and news materials), and second, because steps had been taken to ensure that investigator preconceptions about gratification patterns did not unduly influence the initial source material (which was invariably culled afresh from group discussions with fans of the programmes concerned). And of course the discovery of overlapping gratification structures enhanced the feasibility of the investigators’ original expectation that a typology of audience gratifications might be educed from the results of a series of programme studies.

In fact the four successful programme studies yielded a total of 19 clusters to which substantive labels could be attached (six from the Dales study, five for news viewing, and four each for quizzes and The Saint), and when these were compared with each other, a relatively small number of recurrent categories promptly emerged. For example, escape clusters appeared in the Saint, news and quiz analyses. Reality exploration clusters were found in the Saint and Dales results. Personal reference clusters were remarkably ubiquitous, having emerged from the Dales, quiz and Saint studies (as well as having a ‘near relation’ in the empathic cluster of the news viewing analysis). A use of programme materials to reinforce viewers’ value preferences figured in both the Dales study (in two different clusters) and the quiz analysis. Their use as a ‘coin of exchange’ produced clusters in both the quiz and news studies. In addition, there were two clusters which, though each surfaced only once in the project’s four analyses, are probably strong enough to key groups of viewers to deserve as much attention as the previously mentioned satisfactions – e.g. the substitute companion function (found definitely among Dales fans and possibly in a minor quiz cluster) and the surveillance function (an important component of the news analysis).

The insistent repetition of this small number of themes was such a dominant feature of the first-stage research that the investigators were virtually obliged to use them as the building blocks of their typology. Once this inference was accepted, it remained only to consider how the various categories should be grouped in relation to each other. For this purpose four sorts of ordering principles were applied. One of these involved an acceptance of something like Wilbur Schramm’s distinction between immediate and delayed reward-seeking. It was noticeable, that is, that while the escapist clusters reflected a disposition to use television to get away from or forget certain restricting or unpleasant features of the viewer’s environment, others (especially the surveillance function) represented a concern to acquire information about the wider environment. A second set of considerations affected treatment of the escapist category itself. It was decided to refer to it, as far as possible, through the terminology of ‘diversion’ instead of that of ‘escape’, in order to avoid the possibly misleading pejorative implications of the latter expression. It was also decided that the orientation of diversion could be usefully divided into three sub-types, reflecting, respectively, a wish to get away from a) the constraints of excessive involvement in routine (as in the Saint escape cluster’s emphasis on boredom in everyday life), b) the burden of problems (as in the news and quiz clusters) and c) inhibitions on self-expression (as suggested by the Dales cluster of emotional release). A third distinction revolved around the fact that some clusters primarily refer broadcast materials in some way to the viewer’s own self (including most of the personal reference, reality exploration and value reinforcement clusters), while other clusters were more concerned with his relations with others (other real-life people, as in the coin of exchange function). Finally, because of the evident depth of viewer involvement in programme materials that many of the clusters evinced, it was decided to characterise the categories of the investigator’s typology as diverse forms of ‘media-person interaction’.

From all these considerations, then, the simple typology that is set out below was evolved. Its categories manage to cover all the substantive clusters that have emerged from four programme studies, omitting only two relatively programme-specific clusters – i.e. the identification appeal of The Saint and the reality-piercing function of television news.

[1] Numbers of close friends, numbers of acquaintances in the vicinity, household size, size of accessible extended family, sense of attachment to the community, place of birth, size and position in family of origin, geographical mobility, school-leaving age, occupation, intergenerational occupational mobility, and subjective estimates of strain of work and fatigue after work.

[2] Involvement in certain leisure-time activities (sports and games, betting, attendance at spectator sports), index of opportunities to travel outside Leeds, and self-estimated opinion leadership role.

[3] Horton, D and Wohl, R., ‘Mass Communication as Para-social Interaction’, Psychiatry, Vol. XIX, 1956, pp. 215-29.

[4] Three respondents claimed never to watch television news, and a late filter in the interview indicated that seven respondents had not really seen either The Saint or Callan sufficiently recently to be given the adventure series inventories.

[5] Berelson, Bernard, ‘What “Missing the Newspaper” Means’, in Lazarsfeld, Paul F. and Stanton, Frank N. (eds), Communications Research, 1948-1949, Harpers, New York, 1949.

[6] Schramm, Wilbur, ‘The Nature of News’, Journalism Quarterly, Vol. XXVI, 1949, pp. 259-69.

[7] Stephenson, William, The Play Theory of Mass Communication, Chicago University Press, Chicago, 1967.

[8] Pietila, Veikko, ‘Immediate versus Delayed Reward in Newspaper Reading’, Acta Sociologica, Vol. XII, 1969, pp. 199-208.

[9] MeLeod, Jack, Ward, Scott and Tancill, Karen, ‘Alienation and Uses of the Mass Media’, Public Opinion Quarterly, Vol. XXIX, 1965-1966, pp. 583-94.

[10] As in the quiz study, two-item clusters have been ignored in further calculations and analyses.

[11] As in the case of the news viewing analysis, the results were revised in the light of calculated coefficients of homogeneity and reliability.