The Changing Nature of Party Election Broadcasts: The Growing Influence of Political Marketing

This paper reports findings from a study of the changing nature of the narrative contents and production formats of Party Election Broadcasts (PEBs) produced by the Labour, Conservative, and Liberal Democratic parties for UK general elections from 1979 to 2010. This analysis tracked production changes that might signal a movement on the part of the political parties toward using marketing-oriented techniques of the kind found in televised advertising. Although PEBs are not technically classified as advertisements by the broadcasting industry, but rather as programs, they nevertheless present an opportunity to political parties to promote themselves and their policies. Using content analysis, it was found that PEBs have grown progressively shorter from 1979 to 2010 and become faster paced. They have become more sophisticated as productions with wider use of dramatized documentary formats rather than talking heads, popular music, and professional performers.

it represents part of a wider malaise in political communication that places image ahead of substance and ultimately undermines democratic principles (Mazzoleni and Schulz 1999;Mayhew 1997).
The application of marketing principles to the promotion of political parties is understandable in a world in which the mass media play an increasingly crucial role in politics (Palmer 2002). Television, in particular, has become a key medium through which politicians reach their electorates. Politicians are placed under almost continuous media scrutiny and they must learn to use the media-especially television-effectively to ensure that the exposure they receive presents the messages they wish to get across in a manner that ordinary citizens will not only understand but also pay attention to. This last point is crucial in a world in which people are growing increasingly alienated from politics and distrustful of politicians (Hay 2007).

POLITICAL MARKETING
In early political marketing research, special attention was paid to political communication in American election campaigns. These campaigns were seen as providing the best examples of the application of marketing techniques in a political context (Scammell 1995). Manifestations of a marketing orientation include the way political candidates conduct campaigns, the forms of communications they use, and the ways in which their parties manage their affairs as organisations (Scammell 1999).
The application of marketing methods in party politics has also emerged in Europe (Palmer 2002). Here parties are the ''product'' being marketed as much as candidates and this is particularly true in the UK (Norris et al. 1999). Product sales or marketing-oriented techniques have been distinguished in relation to party approaches to campaigns (Lees-Marshment 2001). Parties increasingly try to ''advertise'' their credentials in a fashion similar to the promotion of commercial products. Electorates are regarded as ''markets'' to which political messages must be ''sold.'' Ultimately the parties hope that their ''markets'' will ''buy'' their messages (policies or personal credentials) and that this purchase will be manifest in terms of their voting choices.
While the use of advertising is a prominent feature of political campaigns in the U.S., across all major media in the UK political advertising is restricted to the non-broadcast media. Broadcast political advertising is outlawed. In its place, the political parties are permitted to take airtime that is freely allocated to them during election campaigns to fill with their own party election broadcasts (PEBs). According to the Electoral Commission (2003a), party broadcasts are designed to allow the parties to freely publicize their platforms and policies to voters. PEBs are technically classified as programs. Over the years, however, they have become shorter and more professionally 230 B. Gunter et al. and glossily produced and as such have come more closely to resemble advertisements. Its proponents believe that a marketing approach in politics can enhance party or candidate credibility. By sticking to tried and tested marketing principles, advertising can provide an effective mechanism through which to convey political messages to voters (Newman 1994;Maarek 1995;Lees-Marshment 2001). Critics have countered that marketing approaches tend to favor image over substance. Communications styles focus on candidate personality and qualities rather than on issues and policies. This means that political marketing can undermine democratic principles founded on the notion of an informed electorate making decisions about political candidate support based on the policies that candidates stand for (Mazzoleni and Schulz 1999;Mayhew 1997).
The effectiveness of PEBs is an important matter given that they have consumed increasing shares of campaign funds over successive general election campaigns (Norris et al. 1999). The use of glossier formats absorbs significant costs and may also indicate some degree of movement toward a marketing approach in British politics. To be clearer about whether PEBs are really just another form of political advertising, however, we need to examine what changes have occurred internally to these broadcasts.
Political advertising on television in the United States resembles other types of advertising in physical terms. Televised political advertisements are typically around 30 seconds in length and adapt a style and tone that are persuasive in nature. The advertising of politics has been further differentiated in terms of whether it is positive in nature (saying something distinctive about the advertised candidate) versus negative (saying something critical about an opponent) and image oriented (focusing on the personal qualities of the candidate) versus issue oriented (emphasizing the policies the candidate stands for; Trent and Friedenberg 2008).
This study investigated whether PEBs can legitimately be regarded as programs anymore or whether they have taken on the form of advertisements. This shift, if it has occurred, will be manifest not simply in the length of these broadcasts but also in the kinds of production techniques that characterize them.

PARTY ELECTION BROADCASTING
Political parties and candidates in the United Kingdom are forbidden to buy advertising time on television and radio. Instead, a committee of the political parties and broadcasters allocates airtime for parties to produce their own promotional messages (PEBs) based upon previous electoral support and the number of candidates standing at the election. Over time, conventions have been established such as the maximum number of broadcasts should be five per party and that the governing party and main opposition party receive the same number of broadcasts (for more on the historical development of the PEB system see Electoral Commission 2003a andFranklin 2004). These broadcasts have the status of programs and are bound by broadcasters' codes of practice, for instance, in terms of taste and decency, although they are exempt from the Advertising Standards Authority code of practice that applies to commercial advertising, and content is otherwise entirely the responsibility of the political parties. In the early 1950s, when television news was in its infancy and bound by significant limitations on what could be reported, PEBs were the only television material produced during general election campaigns themselves. Over the years, however, PEBs have gone from being the focal point of elections on television to an ever more marginal feature as news coverage became ever more dominant (Franklin 2004). New media too have begun to feature in party election communication activities, though nowhere near as prominently or influentially as in the U.S. (Electoral Commission 2003b), and in the 2010 general election Britain held its first series of televised debates between the main party leaders which had a major impact on the 2010 campaign (Harrison 2010) including the PEBs (see below).
Over time PEBs' value has been increasingly brought into question, particularly in terms of whether they would continue to be sufficiently engaging for viewers (Scammell 1990), and certainly evidence over time has shown them to be, while widely viewed, a consistently unappreciated mode of campaign communication by viewers (Scammell and Langer 2006). An apparent response to this competition for attention with other kinds of campaign communication content, on the one hand, and electorate dissatisfaction on the other, is that PEBs have increasingly taken on the form of advertisements since the early 1980s, for instance, having reduced significantly in length and adopting advertising-like production techniques. For some observers this pattern of change represents part of a wider shift in approaches taken by political parties in promoting themselves to the electorate (Norris et al. 1999). The use of marketing strategies is an increasingly prominent and central feature of political communications, particularly during election campaigns.
In addition to this observation, however, distinct patterns were identified in respect of the types of productions made by the major political parties. The Conservatives' PEBs in 1992, for example, were judged to be more personality oriented than those of the Labour Party. The Liberal Democrats were similarly found to rely on political personalities to convey their message. The latter used simpler production techniques than the two larger parties, most likely because of a more limited budget (Scammell and Semetko 1995). Some thematic differences were found in relation to the issues that received greatest emphasis on the part of each political party. The Conservatives focused on the economy, Labour on social welfare issues, and the Liberal Democrats on education. Perhaps the most significant observation made by this study was an apparent ''Americanization'' of PEBs that effectively embraced the utilisation of marketing-oriented techniques such as focus on personalities, use of emotional appeals, and negative campaigning (Scammell and Semetko 1995).
A subsequent study analyzed PEBs broadcast during the 1992 and 1997 general election campaigns. This research again noted an increased use of personalities and negative campaign attacks in these broadcasts. Issues and policies were not discussed in depth. In many ways, PEBs were becoming more like American televised political advertising (Hodess, Tedesco, and Kaid 2000).
By the 2001 general election campaign, a further study observed that while the PEBs in that campaign covered key issues, they adopted a more ''personal'' style of delivery. Just a few privileged issues were covered at all, however, and the three major parties often imitated each other in relation to the issues they debated in their PEBs (Dermody and Scullion 2002).
To pin down whether PEBs have become more like advertisements, though, we need to look at more than how long they are. There are specific production techniques that typically characterize commercial advertising, and evidence that PEBs have become more like advertisements is likely to be found in relation to the extent to which such techniques have been deployed by the major parties in their own televised election broadcasts.

METHODOLOGY
The methodology was originally developed by Juarez (2004), who examined PEBs for the three major UK political parties across all general elections from 1979 to 2001. This paper focuses on the 2010 and 2005 general election campaigns, but comparisons are made with findings from the earlier campaigns. The use of a content analysis based on concrete, objectively specifiable and quantitatively measured variables facilitated a concise and systematic investigation of prominent and emerging content and format trends within these broadcasts. The research was underpinned by a need to explore whether marketing-oriented perspectives were present in these broadcasts. The analysis over time enabled that investigation to determine whether political marketing techniques had become more prevalent and more prominent across successive election campaigns. Furthermore, it determined whether any such trends were common to all three political parties or more prominent for one than others (Juarez 2004).
Content analysis focuses on the surface features of media outputs. It does not explore deeper-level meanings that may reveal cultural and social forces at play. Nor does it demonstrate anything about the effects of content or production formats on audiences.
The Changing Nature of Party Election Broadcasts

Broadcast Sample
The research comprised a content analysis of PEBs for the three leading political parties (Conservatives, Labour, Liberal Democrats) broadcast on television during the last two general election campaigns in 2005 and 2010. In 2005, five televised PEB slots were allocated in each case to the Conservatives and Labour Party, and four were allocated to the Liberal Democrats. However, these broadcast slots were not always filled by different PEBs. While Labour produced five different broadcasts and the Liberal Democrats produced four distinct broadcasts, the Conservative Party produced just three different broadcasts. Two of these broadcasts were televised twice. In 2010, the Conservative and Labour parties were again allocated five PEBs each, while the Liberal Democrats were allocated four broadcasts.
All the major TV channels (BBC1, BBC2, ITV1, Channel 4, Five) carried these broadcasts, although not always at the same times in the evening on days when PEBs were transmitted. Editorial control of these broadcasts was controlled by the parties themselves, but they were required to observe the terms of Ofcom's Broadcasting Code for television programs. In 2010, the major parties' PEBs were also sometimes transmitted on different days in England, Scotland, and Wales.

Coding and Analysis
Three levels of content analysis are presented: party PEB level, party leader level, and proposition level.
The PEB level took the broadcast as unit of analysis. Categories of content were informed by previous research (Kaid and Johnston 2002;Hodess et al. 2000). Each PEB was coded in terms of audio, visual, and verbal structural complexity incorporating measures of camera shot changes, format style (documentary style, video clip, talking head), category of dominant speaker, gender of speaker, use of music, and genre of music used.
The party leader analysis reflected growing interest in the rise of the political personality (Stewart and Clarke 1992;Swanson and Mancini 1996). Party leaders were coded in terms of whether they were seen, heard, or seen and heard; whether they addressed the audience directly or were in an on-screen interview or other scenario; formality of dress; nature of camera shot (close-up, long shot); nature of appeal (emotional versus rational); valence of appeal (positive versus negative); and whether music was present or absent during the leader's appearance.
The proposition level of analysis represented a response to rising interest in negative versus positive comparative advertising in politics (Basil, Schooler, and Reeves 1991;Kaid and Johnston 2002). Negative attacks in political advertising have been found to mediate the way it affects voters' attitudes toward candidates and the advertising itself (Johnson-Cartee and Copeland 1991). This analysis level therefore examined specific propositions contained in PEBs that were self-promotional, attacks against opponents, or neutral. The nature of each proposition is also classified in terms of whether it is emotional or rational; the type of emotional strategy used (fear, humor, enthusiasm); the type of rational appeal used (factual, opinion); and the presenter of the proposition (professional actor, party member, party leader, member of an interest group, other).
Each PEB was viewed by a trained coder using the coding frame and code book developed by Juarez (2004). Separate viewings were required to complete each level of analysis. The coding frame's reliability had been established by Juarez (2004). Reliability scores were 80 percent at the PEB level, 88 percent at the proposition level, and 93 percent at the party leader level.
At each level of analysis a different base unit of analysis was used. At the PEB level, attributes were coded for each PEB with averages being reported for and percentages based on the number of televised broadcasts transmitted in the 2010 general election campaign. At the party leader level, appearances within made by party leaders were counted within PEBs and these appearances (whether on camera of off camera=voice only) represented the base units on which data were computed. At the proposition level, propositional statements that occurred within PEBs were used as the base units of analysis and all averages and percentages were derived from these measures.
Data were analyzed within SPSS and comparisons made between the findings for the 2010 and 2005 PEBs and with those obtained in earlier research with PEBs from general elections held from 1979 to 2001.

FINDINGS PEB Level of Analysis
At the PEB level the aim was to provide a general impression of the content and format of these televised political messages in terms of their physical dimensions, production treatments, and broad political references. Changes in duration; use of camera shots; nature of speakers (professionals versus party members); references to minorities, social class, and partisanship; use of different production styles; and use of music were coded. More detailed analyses of PEB ingredients were provided at the other levels. The key measures reported here were derived from earlier research (see Juarez 2004) and comparisons are made with the findings from that study which covered general election campaigns from 1979 to 2001 inclusive.

Length of PEBs
Although the number of PEBs for the main parties have remained very stable, the length of PEBs has declined significantly from around 15 minutes per PEB The Changing Nature of Party Election Broadcasts in the 1950s to under 10 minutes by the early 1970s, with the standard official allocation per today being 5 minutes per PEB (see Franklin 2004). A significant shortening of average PEB durations was observed over the 1979 to 2001 period (as shown in Table 1), with 1992 (524 seconds) proving an exception to this prevailing trend. Since 2001, PEBs have grown longer across the three major parties, although they still remain much shorter than they were up to the early 1990s and shorter than the time available to them to use. The Labour Party produced the longest broadcasts in the 2005 general election, at 219 seconds (range: 160 seconds to 273 seconds), with the Conservatives producing the second longest on average at 196 seconds (range: 162 seconds to 219 seconds) and the Liberal Democrats producing the shortest average broadcasts among the major parties, at 176 seconds (range: 157 seconds to 221 seconds). In the 2010 general election, the Conservatives had the longest PEBs (241 seconds), followed by the Liberal Democrats (203 seconds) and finally Labour (163 seconds). For the Conservatives, their 2010 PEBs attained the longest duration seen since 1979, while for Labour their PEBs were about the shortest they have been since 1979. The Liberal Democrats did not break any duration records for their PEBs in 2010, but they were on average much longer than in 2005.

Pace of PEBs
The pace of PEBs can be determined by the average number of shot changes that occur within these broadcasts and by the average length in seconds of camera shots. A larger number of shot changes and shorter duration shots signal a faster pace. As Table 2 below shows, post-1990 PEBs contained more camera shot changes on average than did pre-1990 PEBs and average shot lengths (in seconds) has progressively reduced since 1979. The average shot length in 2005 remained stable compared with the previous election in 2001. At an average shot duration of three seconds, however, there is little scope for shot lengths to get much shorter. What these findings do signify, though, is that PEBs have not only gotten shorter in total length over time but they  The type of spokesperson used by the three major parties exhibited marked shifts in 2010 from 2005. While the Conservative Party (80 percent) and Liberal Democrats (100 percent) moved away from using professional actors to showing politicians as spokespersons, Labour moved in the reverse direction and did not use politicians at all, only actors or celebrities. What is particularly significant here is that Labour decided against using the prime minister as a front for the party. This stands out given that since the first leader-centered PEB (on Neil Kinnock in the 1987 election), in every election since then the leaders of all three major parties have been the focus of at least one PEB and present in some form or other in many others. The absence of Gordon Brown from Labour's 2010 PEBs may say a lot about his perceived contribution to the Labour Party brand at that time. Note. Bold=Italics used to distinguish new original data produced by the study from contextual earlier data.
The Changing Nature of Party Election Broadcasts 237

Party Identification References
The extent to which the three main political parties included references within their PEBs designed to reinforce party identity varied. Such references are designed to appeal to members and supporters and can therefore also be labeled ''partisanship'' references. In 2005, only the Labour Party included any such references in its PEBs (in one PEB out of five). This was on a par with the 1979-2001 election campaigns average for the party. The Conservatives and Liberal Democrats were found to have included no such references in their PEBs in 2005, which was consistent with 2001 but a departure from PEBs in earlier campaigns where such references had been detected, usually in one or two PEBs per campaign (Juarez 2004).
Overall, partisanship references occurred at a very low level in the major parties' PEBs in 2005, reinforcing the finding of 2001 where no such references were found. These findings confirm a post-2000 election phenomenon with the major parties whereby addressing the party membership exclusively is apparently regarded as less of a priority by parties.
Party identity references increased dramatically in 2010 for Labour PEBs occurring in all five of their PEBs. Conservative PEBs also exhibited partisan references to the party in two out of five PEB, whereas they had been completely absent in 2005. These references were once again completely absent in Liberal Democrat PEBs in 2010. This is a result that might indicate a tendency on the part of Labour to focus on the party's ''achievements'' in government. The opposition parties instead were more focused on challenging Labour's track record.

Format Features of PEBs
An important aspect of this analysis has been to establish whether PEBs have changed in terms of the production techniques they use. Earlier research, covering PEBs from 1979 to 2001, had indicated that a documentary style of production was dominant for all three political parties (Juarez 2004). Trends over time indicate that 2005 witnessed a fall in the use of documentary style format across all parties and a further growth in the use of issue dramatizations and video clips. By 2010, the use of a documentary style diminished still further. Instead, a montage of video clips was preferred. Thus, the pattern emerging is that PEBs have assumed more complex production formats overall with a greater mix of different production techniques, even though the use of the ''talking head'' format where a presenter is addressing the audience by directly looking at the camera in the style of news bulletin broadcasts has been completely abandoned post-2000 (see Table 3).
This above pattern held true in 2005 for each of the three political parties, but with the use of documentary style disappearing altogether in 2010 for the Liberal Democrats. In 2010, with the Conservatives and Labour the use of video clips occurred in three out of five PEBs in each case.

Music in PEBs
One aspect of the production treatment of PEBs is whether these broadcasts contain any music. If so, what types of music do they contain? Music was relatively rarely used in 1979 and 1983 (in under one in five PEBs). In 1987 and 1992, music occurred in nearly nine out of ten PEBs. In 1997, the use of music slipped back to around seven in ten PEBs, before climbing again to nine in ten by 2001. In 2005, music appeared in all PEBs for the three major parties and in nine out of ten in 2010. The analysis continued by examining the genre of music used in PEBs.

Party Leader Level of Analysis
The party leader level of analysis focused on the physical presence of party leaders in party election broadcasts. The unit of analysis in this context was any appearance by a party leader, whether shown on camera (speaking or silent) or heard as a voice-over off camera. These appearances were timed and classified in terms of a range of presentational and narrative attributes. These attributes included the degree of visual versus verbal presence of party leader in PEBs, the target of the party leader's message, camera shots used, use of emotional versus rational and positive versus negative appeals, and past and future reference to party's performance. Earlier published research has indicated that party members have a diminishing presence in party broadcasts as presenters of party statements about performance, policy, or intentions (Franklin 2004;McNair 2007). Another feature identified by other researchers in the field has been the deployment by political parties of techniques gleaned from advertising (Scammell 1999;Kaid and Holtz-Bacha 1995).

PARTY LEADER PRESENCE
For many years, the dominance of party leaders in party general electioneering on television was underlined by their visible and verbal presence in PEBs. By the late 1970s, party leaders' voices were always present, and from the 1980s, leaders also became visibly more prominent in these broadcasts (see Table 4). This leader prominence has disappeared in the last ten years, as leaders have been placed more in the background and by 2010 were heard rather than seen, at least in the front of the screen.
For Labour, 2005 PEBs displayed a greater emphasis on visually presenting the party leader (57 percent of appearances compared with 36 percent between 1979 and 2001). As mentioned, in 2010, however, Gordon Brown was never used as the visible lead spokesperson. If he was seen, it was always in the background. For the Liberal Democrat PEBs in 2005, the party leader was visible more often (in 50 percent of his appearances) than between 1979 and 2001 (38 percent). By 2010, for the Liberal Democrats, however, more than nine out of ten appearances by the party leader were voice-over only (91 percent). For the Conservatives, the party leader was used much more in voice-over mode in 2005 (43 percent of appearances) than was usual over the previous six election campaigns (26 percent between 1979 and 2001). For the Conservatives, though, the party leader was less often visible in 2005 PEBs than over 1979 to 2001 (43 percent versus 74 percent of appearances). By 2010, the Conservative Party leader was used as a voice-over in one-third of his appearances (33 percent), otherwise he was shown in the background (66 percent).

TARGET OF PARTY LEADER'S NARRATIVE
At the party leader level, PEBs were assessed in terms of who the leader was speaking to when he or she appeared in a broadcast. A broad distinction was made between speaking directly at viewers in the television audience and someone else (possibly within the broadcast itself). Across the 1979 to 2001 general election campaigns, the party leader was judged to be speaking to the viewer in the audience much less often than someone else in the case of Labour PEBs (38 percent versus 63 percent) and Conservative PEBs (28 percent versus 72 percent), but with direction of speech more evenly divided for Liberal Democrat broadcasts (44 percent versus 41 percent, with other instances being difficult to judge). On occasions when the party leader was speaking in PEBs in the 2005 campaign, the Labour and Conservative PEBs displayed mirror opposite patterns, while the Liberal Democrats addressed viewers at home and someone else fairly evenly (47 percent versus 53 percent). With Labour PEBs, the great majority of cases when the party leader spoke (89 percent), he or she addressed someone else rather than the viewers, while in Conservative PEBs, the party leader addressed viewers directly in most instances (71 percent).
There was a greater tendency for party leaders to speak directly to viewers in 2010 than had been the case in 2005. This represented a return to a preferred style in 2001, which at the time represented a dramatic shift of perspective from the three previous election campaigns. The Liberal Democrats' leader was much the most likely to speak directly to viewers (91 percent of appearances), with the Conservatives leader (33 percent) much less likely to do so. Labour's leader did not speak at all in any of his party's PEBs.
One interesting feature about 2010 is how the original plans of the Conservative Party for their PEBs changed at the last minute due to the impact of the televised debates. Initially the plan for their PEBs had been to use at least some of them to attack their opponents, particularly Labour leader Gordon Brown. The success of Liberal Democrat leader Nick Clegg in the first TV debate, and the ensuing significant increase in news media attention paid to his party, forced a last minute rethink, with one of the later Conservative PEBs given over to a hastily edited together piece all about Cameron, with lots of footage and audio from campaign speeches made by him, designed to try to counter the perceived threat of Nick Clegg (Harrison 2010, p. 276). This late change also indicates how PEBs, far from being a highly formulaic, preplanned, and rather archaic form of parties' electoral communication output, are seen as a significant part of parties' dynamic proactive marketing strategies during election campaigns.

NATURE AND VALENCE OF APPEALS MADE BY PARTY LEADER
Political party leaders' appeals to the electorate were differentiated in terms of whether they were ''emotional'' or ''rational'' and positive or negative in nature.

The Changing Nature of Party Election Broadcasts 241
Nature of appeals. Party comparisons for 2005 revealed that Labour relied entirely on rational appeals by the party leader in its PEBs, the Conservatives depended more on emotional appeals (76 percent of appeals in PEBs), and the Liberal Democrats used both types of appeal fairly evenly (emotional: 47 percent; rational: 53 percent). This pattern revealed a dramatic shift of emphasis from emotional to rational appeals for Labour PEBs (1979PEBs ( -2001 and slight shifts toward greater use of rational appeals by the Conservatives 24 percent in 2005, compared with 1979-2001 average of 7 percent) and Liberal Democrats (1979Democrats ( -2001 average for emotional appeals À60 percent).
In 2010, the Conservatives (93 percent) and Liberal Democrats (88 percent) relied predominantly on rational appeals to voters. In both cases this marked a significant increase in the use of such appeals in 2010 compared with 2005. Labour failed to register any emotional or rational appeals in their 2010 PEBs.
Valence of appeals. The valence distinguished positive from negative appeals by party leaders. The distinction made here was between whether party leaders spoke positively about their party's own policies or achievements or negatively about the shortcomings of their opponents.
Negative campaigning was far outweighed by positive campaigning in PEBs in 2005. This was most clearly the case for the Labour Party, which was judged to have used only positive statements in its PEBs. The other two parties also relied two-thirds of the time on positive statements (Conservatives: 65 percent; Liberal Democrats: 67 percent), though did also include some negative snipes at the Labour government in one-third of their 2005 PEB appeals. The level of negativity in the PEB campaigns of the latter parties exceeded the average observed in earlier campaigns, especially for the Conservatives (35 percent in 2005;1 percent between 1979) and Liberal Democrats (33 percent in 200512 percent between 1979 and2001).
In 2010, the valence of appeals profile did not change much from 2005. For Labour no negative statements were recorded again nor were there any positive statements. The Conservatives' positive statements (67 percent) outnumbered negative statements (33 percent) by two to one. For the Liberal Democrats, positive statements (74 percent) outnumbered negatives (26 percent) by three to one.
Averaging across parties, positive campaigning has declined and negative campaigning has increased over time. Although negative campaigning is still significantly outweighed by positive campaigning, the growth in its prevalence is quite marked and more than doubled in prevalence in 2005 PEBs compared with 2001 (see Table 5). Given the relatively small number of PEBs, unlike systems like the U.S. where broadcast political advertising is allowed generating lots of ads, making comparisons concerning trends such as increasing negativity is not straightforward, but it is worth noting that an increasing presence of negativity in UK PEBs identified here does parallel 242 B. Gunter et al.
research evidence of increasing negativity in political advertising elsewhere in the world (Kaid and Holtz-Bacha 2006).

Proposition Level of Analysis
A further level of analysis examined propositional content of PEBs. In the original research the proposition unit of analysis was defined as ''communication content structured in constituent parts comprising of an orientation, target, focus, and argument . . . . A proposition item therefore includes at least two of the following characteristics: (1) a particular presenter, (2) an identifiable orientation (i.e., attack, compare, promote), and (3) an explicit target (i.e., party leader, party)'' (Juarez 2004, p. 214).

NATURE OF PROPOSITION
Proposition nature was differentiated in terms of whether it was emotional or rational. Emotional propositions satisfied empirical criteria derived from audience research literature that indicated the potential of certain types of content to elicit emotional responses in the audience (see Juarez 2004). Trends over time by party indicated that the distribution of emotional and rational propositions within PEBs fluctuated from campaign to campaign (see Table 6). In 2005, rational propositions were far more prevalent than emotional propositions for all three parties' PEBs. In every case, there was a dramatic  Note. Bold=Italics used to distinguish new original data produced by the study from contextual earlier data.
appeals. In contrast, for Labour emotional appeals outweighed rational appeals, which showed a reversal of 2005 and a return to the types of appeals used during the early Blair era.

TYPES OF EMOTIONAL PROPOSITIONS
Breaking down emotional propositions further, the coding frame examined specific strategies adopted by political parties when addressing viewers via PEBs. Propositions deemed to be emotional on the grounds that they satisfied certain empirical conditions, established through audience research, for evoking emotional reactions in viewers, were differentiated in terms of whether they might invoke fear, humor, or some other emotional response. Data over time indicated that the use of statements designed to play on fear or enthusiasm responses were generally more prevalent within PEBs than statements of a humorous nature (see Table 7). The use of fear exhibited a steady and unrelenting reduction over the 1979 to 2001 campaigns but recovered sharply in 2005. Propositions designed to invoke another type of emotional audience response (other than fear or humor) exhibited growth from 1987 to 2001 but fell away dramatically in 2005. Humor has never been a major feature of PEBs, but it achieved its highest level over the period monitored in 2005. By 2010, the use of both fear and humor fell away again.
On examining the distribution of emotional strategies by party, it emerged from 2005 PEBs that fear was used more by Labour (86 percent of propositions) than had been the case in elections from 1979 to 2001 (49 percent). By 2010, however, the use of fear by Labour fell dramatically (35 percent). The Conservatives, in contrast, used fear a great deal in 2010

ORIENTATION OF PROPOSITIONS
This measure examined whether propositions were used to promote party performance, positions, and policies; to attack other parties on the same terms; or to make comparisons between own party and other parties in these respects. Over time the use of promotional propositions has grown, while use of attack propositions and comparison propositions has declined (see Table 8). Only minor fluctuations from these prevailing patterns were observed over the 1979 to 2005 period. Clearly, the three political parties have placed more emphasis on self-promotion in their political broadcasts in recent general election campaigns and have devoted less attention to attacking other parties or making comparisons between themselves and others.
It is important to go beyond these general time trends and examine any differences that may exist between parties. As with a number of other content and format variables examined in this research, inter-party differences emerged yet again in relation to use of propositions of different orientation.
In 2010, the Conservative Party used mainly self-promotional statements much less often than in 2005 (52 percent versus 81 percent) and more than the 1979 to 2001 average (34 percent). Use of attack statements in 2010 (41 percent) increased dramatically from 2005 (13 percent), which had seen an exceptionally low level of usage compared with 1979 to 2001 (51 percent). In 2010, Labour used ''self-promotional'' statements (55 percent) more than it did in 2005 (48 percent

DISCUSSION
This research has examined the contents and format features of televised PEBs transmitted for the Labour, Conservative, and Liberal Democrat parties across general election campaigns from 1979 to 2010. A content analysis methodology was used throughout this work. The research was conceptually driven by the notion that these televised political communications have evolved over time and have increasingly adopted marketing techniques. In this vein, PEBs have taken on the appearance of political advertisements, even though they are technically classified as programs and broadcast political advertising is banned in the UK.
The empirical evidence accrued by the original research reported by Juarez (2004) indicated that the PEBs of the major political parties did change radically between 1979 and 2001. Most notably they became much shorter and much faster paced. Camera shot durations reduced in length and the broadcasts became increasingly glossy and sophisticated productions. Professional actors were used to provide voice-overs and to present critical messages about party policies and performance, and reliance on party leaders or other party members in these roles declined. While always present to some extent in PEBs, the use of music evolved and the range of different types of music increased. Popular music scores were used as often as classical music scores during the Labour government era.
Glossier documentary formats came to dominate these productions, replacing the more simplistic earlier formats with strong reliance on talking heads. All these trends were found to have continued in the 2005 general election campaign, though the pace of PEBs slowed a little in 2010. Furthermore, the documentary style was displaced to some extent by the use of video montages.
These changes in production treatments have represented radical shifts over time in their nature and reflect the adoption of techniques that are found 246 B. Gunter et al. in televised commercial advertising. PEBs have reduced in length to the point where many barely qualify as ''programs.'' The shortest PEBs in 2010 and 2005 were only slightly longer than the longest commercial advertisements. Of equal significance though is the average length of camera shots (three seconds), which again reflects the style of commercial advertisements and represents a much faster-paced form of presentation. Such rapid shot changes are designed to maintain audience attention and presumes that the average viewer has a short attention span while watching television that continually needs to be reoriented toward the screen by rapid on-screen changes of scene. The increased use of professional actors or celebrities in recent election campaigns also signals a need to match the professionalism of television programming and advertising in this respect. Politicians do not always make to best television presenters. Although leading politicians are increasingly skilled in television interview formats, addressing the audience direct to camera requires a different skill set. This task is often handled better by trained actors.
The more prevalent use of popular music in PEBs also reflects a need to engage with the general audience by using techniques that will attract the attention and interest of the ordinary viewer. It may also represent a tactic designed to appeal to younger people who have generally been perceived to lack any significant interest in politics.
PEBs in 2010 continued a trend observed since 2000 for the major political parties to address their messages beyond the party faithful. Instead the parties focus on highlighting their achievements to all or (mostly in the case of opposition parties) challenging the success of the incumbent government.
Another notable change in PEBs has occurred in the way party leaders are used, or it would be more accurate to say these days, not used-or at least not shown. In 2010 PEBs, unlike in nearly all other general election campaigns since 1983, the depiction of party leaders diverged across the parties. Gordon Brown was almost entirely invisible from Labour PEBs. The Conservatives did use David Cameron, but at least one of their PEBs was a late substitution for original attack ads, responding to the so-called post TV debates ''Clegg effect,'' and as such there would similarly have been a greater invisibility of the Conservative leader too had they stuck to their original plans. Only the Liberal Democrats placed their leader more centrally in their PEB (which was screened several times), and even here the format used was anything but a traditional talking head.
In 2010, as in 2005, the major parties largely avoided negative campaigning in their PEBs preferring instead to make positive statements about their own policies and achievements. Labour avoided negatives completely, while for the Liberal Democrats and Conservatives between one in four and one in three statements were criticisms of other parties, largely the Labour government. Negative campaigning, as a means of generating voter involvement through presenting the election as a competition between political agents willing to adopt whatever tactics are necessary to win (see Thorson, Christ, and Caywood, 1991), thus remains a more prevalent technique in paid-for political advertising than in PEBs.
The use of emotional appeals increased for the Conservative and Labour across the 1980s, then dipped before rising again across the 1990s and then dipping significantly again in 2005. For the Liberal Democrats rational appeals always outnumbered emotional appeals until 1997 and the latter continued to rise into 2001. In 2005 all three parties shifted to greater use of rational appeals. In 2010, rational appeals reached their highest points for the Conservatives and Liberal Democrats, while there was a shift back on the part of Labour to greater use of emotional appeals that once again outnumber rational appeals.
In 2010, where emotion was present in PEBs, it was much less likely than in previous years to comprise fear appeals. These PEBs also exhibited a much diminished tendency to use humor, whereas in 2005, humor was used to a greater extent in PEBs than in any other campaign since 1979.
Whether the use of negativity and emotionality enhance the ''entertainment value'' of PEBs and also enhance their political impact upon the audience is a debatable point (Scammell and Langer 2006). Certainly, it is important to attract and maintain the interest of the audience. In the television environment, PEBs must compete for audience share and attention against other broadcast outputs that also use attention-grabbing and emotioneliciting techniques. As with other competing programming and advertising outputs on television, PEBs must know their audience and use the techniques known to be most effective in attracting the interest of those they seek to appeal to or influence.
A primary and traditional function of PEBs has been to reach the electorate on a large scale with informative messages about the political parties that has relevance to aiding decisions about who to vote for. Some observers have noted that PEBs do contain important information in this context (Hodess et al. 2000;Pattie and Johnston 2002), and they do continue to reach large audiences: The first Labour PEB of 2010 reached some 9 million people (Harrison 2010, p. 273). It has also been recognized, however, that PEBs have an important role in engaging members of the public who display less interest than average in politics and who may also have not made up their minds about which party they wish to vote for. Involved viewers who have made up their minds already are unlikely to be greatly influenced by PEBs (Hodess et al. 2000;Pattie and Johnston 2002). To be of interest to the politically alienated and disengaged, it may be necessary to beyond straight informational appeals and to use techniques designed to make political content a sufficiently attractive package to buy into. The shifts observed in this research in the production treatments used in PEBs reflect these kinds of developments. The techniques being used increasingly resemble those adopted by commercial advertisers. In effect, PEBs have, over time, come increasingly to resemble advertisements, demonstrating the growing role of political marketing in transforming established, some might say even old-fashioned, channels of election communication into carefully constructed components of sophisticated, dynamic modern election campaigns.