Content-Type: text/html

                Quality








Commercial Quality Influence on Perceptions of Television News

by

Stephen Perry

Dana Trunnell

Chris Morse

Cori Ellis

Illinois State University, Department of Communication


Submitted to Radio-Television Journalism Division
Association for Education in Journalism and Mass Communication




Contact: Stephen D. Perry, Ph.D.
Department of Communication
Campus Box 4480
Illinois State University
Normal, IL 61790
E-mail: <a href="/cgi-bin/wa?LOGON=A3%3Dind0109B%26L%3DAEJMC%26E%3D8bit%26P%3D4310218%26B%3D--%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D_-3298531%253D%253D_%26T%3Dtext%252Fplain%3B%2520charset%3Diso-8859-1" >[log in to unmask]</a>
Phone: 309-438-7339



Running Head:  QUALITY
  Commercial Quality Influence on Perceptions of Television News
Abstract
        The impact of high and low-quality commercials upon high and low-quality television newscasts were examined using Elaboration Likelihood Model and contrast effects research. This study showed some support for contrast effects. Results also suggest an interaction between news quality and the presence of commercials within newscasts in producing an emotional response. Additionally, we found that when commercials were present within the news program, participants were able to recall fewer of the news stories.

  Commercial Quality Influence on Perceptions of Television News

        Television news programming serves an important function in society. Not only does it provide information, but it also serves to entertain the viewer (Duke, 2000). Consumers of news programming have many options when choosing a particular program from which to obtain their information-Several local market news programs, national broadcast news programs, and cable news channels. Therefore, it is important to examine factors that influence viewer's perceptions of news programming. In this study we will look at the influence of commercial presence and commercial quality across two different quality level news programs.
Several studies have examined the influence of television programming upon commercial effectiveness (Celuch & Slama, 1993; France & Park, 1997; Kamins, Marks & Skinner, 1991). Recently, however, commercials have come under examination as influencing factors upon enjoyment of television programming. A few studies have been conducted to analyze the effect television commercials have on the programs they are embedded in. Finn and Hickson (1986) studied the impact of inserting arousing commercials into a slow-paced news program. The results, although not overwhelmingly conclusive, found that segments with commercials were "consistently judged to be less biased and more accurate, more responsible, and more reliable" than segments which showed no commercials (p. 371). While this is the only study to date concerning commercial influence on news, others have examined entertainment programming. Perry, Jenzowski, Hester, King and Yi (1997) found that increased humor in commercials resulted
in greater program liking in humorous television shows. They found that this held true across two different types of humorous television programs-stand-up comedy and a humorous talk show-since high humor commercials increased enjoyment for both types of shows. Most research that has studied commercial effects on programming has employed the principles of excitation transfer theory to predict the influences commercials have upon television programming (Bjorna et al., 2001; Finn & Hickson, 1986; Perry, 2001; Perry et al, 1997). This theory suggests that a high arousal state will extend into a post exposure state. Therefore, high arousal levels in commercials should extend to program enjoyment as well.
To date, studies of commercial influence have focused on arousal concepts, but an element that has been virtually ignored as a contributing factor to program enjoyment is the quality of commercials. Our study herein will further Finn and Hickson's (1986) study by examining how the quality of commercials influences viewers' perceptions of news programming while examining the competing theoretical foundations of Elaboration Likelihood Model (Petty & Cacioppo, 1986) and contrast effects (Becker & Villanova, 1995; Perry, 1998). The ELM theory suggests that people will be likely to associate quality commercials with quality programming through a peripheral processing route, and, therefore, may associate a low quality program as higher quality simply because of the high quality of the commercials inserted into the program. The contrast effects theory suggests that higher quality levels in commercials could make programs seem lower quality by comparison and vice versa.
Attitudes toward Commercials
        The audience often sees commercials as filler space or clutter between program segments. There is a significant amount of research, which suggests that television audience members view commercials negatively. Mittal (1994) explains that commercials are often described as irritating and boring, indicating a general distaste for commercial break periods. Additionally, research has shown that channel-surfing and ignoring of commercials frequently occurs during commercial breaks (Eastman & Newton, 1995; Eastman & Newton 1999).
        However, commercials are not always viewed in such a pessimistic light. For example, Mittal (1994) found that about half of viewer respondents said they agreed with the statement, "Sometimes TV commercials are even more entertaining than TV programs" (p. 43). In fact, Pavelchak, Antil & Munch (1988) suggest that in programs such as the Super Bowl, advertising is often better liked than the actual program itself. Greene (1992) indicates that there are many different attributes that contribute to audience enjoyment of television commercials. Surprisingly, entertainment functions such as amusement, fast-moving images and imagination are not as important to enjoyment of an ad as are communicative/persuasive functions such as effectiveness, conviction and images that are worth remembering.
Still, there seems to be no consistent definition of "quality" in the literature in terms of commercials. We suggest that quality commercials employ the strategies of enjoyment as outlined by Greene as well as professionalism in the realm of production. Suggested here is that commercial quality can be a function of the money and effort spent on production.
Quality of News Programming
        In addition to conceptualizing the characteristics of a quality commercial, it is necessary to define a quality newscast. A few studies have examined viewers' perceptions of the quality of news programming. Lind (1995) found that while definitions for quality production in news are sparse, viewers do believe that many news shows can be improved. Further, viewers suggested that they feel the responsibility lies on the viewer, not the producer to demand and instigate more quality in news programming. Therefore, it seems that quality is an important measure that viewers take into consideration when choosing their news. Rosenthiel and Gottlieb (1999) have outlined a definition of quality news programming. They say that a quality program is one that:
Accurately reflect[s] their entire community, cover[s] a broad range of topics, focus[es] on what is significant, make[s] it locally relevant, balance[s] stories with multiple points of view, and rel[ies] on authoritative sources (p. 83).
Rosenthiel and Gottlieb (1999) also offer standards for a "good newscast."  These standards include airing stories that cover a wide variety of topics, have local relevance, use a number of sources and find sources with expertise on the topic being covered. The same study indicates that viewers' perceptions of a quality newscast influences ratings. For example, the lowest rating programs in the study had fewer elements of a quality program than those that received the highest ratings.
Commercial Effects on Program Perception
        Most research that looks at the interaction between programming and advertising has been interested in factors such as ad placement within commercial pods (Gutenko, 1997; Lipman, 1989, Webb & Ray, 1979) and whether advertising effectiveness is impacted by congruence of mood between the ad and program (Celuch & Slama, 1993; Gunter, Furnham & Frost, 1994). The Finn and Hickson (1986) and Perry et al. (1997) studies, however, were among the first to study the effects of commercials on programming. Both found that when arousing commercials were used it enhanced evaluation of programs. Perry (2001) again studied the effects of humorous commercials on television sitcom liking, but controlled for gender. The study found that men were more likely to find a program more enjoyable due to higher levels of humor in commercials while there was no difference in program appreciation by women. He attributed this difference to the lack of appreciation of the show by women overall, suggesting that the episode must have some desirable qualities in order to be enhanced by humorous or other arousing commercials.
Not all studies have shown a favorable connection between commercial arousal and program appreciation. Perry (1998) reevaluated his prior research, testing the effects humorous commercials have on television sitcoms. He found for the sitcom Dharma & Greg that when commercial humor was highest appreciation was lower than when more moderate levels of commercial humor were used. Similarly, Bjorna et al. (2001) reported that arousing, high-energy commercials detracted from enjoyment of the program Crocodile Hunter, though some hightening in enjoyment was found for the program Friends. When they conducted a similar experiment to test the effect of varied levels of humor in commercials on the same two programs, no significant differences were caused.
While early studies based on excitation transfer theory indicated that higher levels of commercial arousal enhanced the enjoyment of television programs, results are now less clear with results indicating that a distraction or contrast effect (competition hypothesis) may be a better explanation. Still, we see that commercials often impact the programs in which they air though the theoretical basis for such impact is unclear.
A totally different theoretical basis for explaining the impact of commercials on programming could come from the Elaboration Likelihood Model (ELM). No one has yet looked outside of the area of arousal in commercials to find an impact. It seems likely that higher quality commercials could provide cues that would lead viewers to perceive television news programming to be of higher quality due to ELM. However, no studies have looked at the impact of differing levels of quality commercials on programming. This, of course, means none have examined commercial quality impact on news. This study will compare commercial quality with viewers' perceptions of differing quality levels of news programming.
Elaboration Likelihood Model
        Elaboration Likelihood Model (ELM) predicts the probability of critical thinking about a particular stimuli. More specifically, the ELM attempts to predict how individuals will process new information that is presented to them. For example, according to ELM, people process information using both central and peripheral routes. The central processing route involves critical thinking based on motivation factors including, involvement, diversity of the argument, and level of enjoyment of critical thinking. Peripheral processing routes often occur when information is new and the respondent is less involved (Petty, Cacioppo & Goldman, 1981; Petty & Cacioppo, 1986). Thus, a viewer may see a program as lower in quality because of peripheral cues not cognitively processed. We are interested in whether that cue could be a low-quality commercial pod. Gottlieb and Swan (1990) tested the ELM in a study that examined the effect price savings had on the motivation of subjects to process a print
 advertising message. This research found that the ELM may have implications on how consumers view source credibility, depending on whether the subject is highly involved (central processing) or less involved (peripheral processing) and depending on whether or not the information is new. Gottlieb and Swan suggest from their findings that perceived source credibility was higher when the subject was involved on only a peripheral processing level. Bozman and Mueling (1994) also studied advertising and its relationship with the ELM. This study looked at the implementation of music backgrounds (peripheral cues) in television commercials and the subsequent effects on viewers' attitudes and behaviors. It was found that these peripheral cues were highly effective as persuasion devices.
Contrast Effects
        The competing theoretical perspective, contrast effects (competition hypothesis), which also may impinge on our study predicts that when two stimuli that are different in quantity, quality or some other measure are rated together, a contrast will exist--one stimuli will increase in measure, while the other will decrease. As Becker and Villanova (1995) point out, the contrast effect is greater and more pronounced when one stimuli is perceived to be average in measure. Becker and Villanova studied the procedure of performance appraisal and the effect of temporal delay between viewing a performance and rating it in terms of the magnitude of contrast effects. This research identified that contrast effects were influenced by both the performance rating procedure and the length of delay.
        Gutierres and Kenrick (1999) also examined contrast effects in regards to mate selection. Specifically, the authors hypothesized that women would be more adversely affected as perceived marriage partners when their mates were exposed to highly attractive women in contrast with highly dominant women. Also, men would be more adversely affected as perceived marriage partners when their mates were exposed to highly dominant men in contrast with highly attractive men. Similarly, because of a perceived connection to how their mates would respond when confronted with an attractive or dominant person of the opposite sex, the authors predict that subjects would react in accordance with the hypothesized perceived behavior of their mate when confronted with a member of the same sex. The findings suggest that there are, in fact, contrast effects present in that men perceive themselves to be less effective as marriage partners when confronted with a dominant male and females consider themselv
es worse marriage partners after being confronted by a highly attractive female.
        In relation to advertising's effect on program perception, both Perry (1998) and Bjorna et al. (2001) found support for contrast effects as described earlier. Perry called this occurrence a distraction effect. However, he could have as easily explained his findings under contrast effects.
        In summary, while the ELM implies that higher quality commercials will result in heightened perceptions of the news characteristics, we also acknowledge that under contrast effects a high quality level in commercials could actually compete with and render of lesser perceived quality the programming it is paired with and vice versa.
Research Question & Hypothesis
This study examines whether commercial quality will have an impact on viewer perceptions of television news programming. Since the advertisements and news programs we use are taken from television markets not in the immediate area of the testing location, we suggest that the information presented to subjects is new and that viewers will be less involved with the content. Therefore, this allows that viewers will respond to the effects of television commercials on news programming using peripheral processing routes as described by the ELM. If viewers only process the program through peripheral routes, ad quality will be more able to influence viewers' perceptions of news programming. However, peripheral processing does not rule out the possibility of a contrast effect. In fact, if viewers separate the news content from the advertising content, even if done subliminally, this would encourage the emergence of contrast effects.
Due to discrepant theories, the following research question is offered:
RQ1     How will the quality of commercials influence viewers' perceptions of newscasts?
        In addition to the above research question, it is also expected that, in accordance with the results of Finn and Hickson (1986), the absence of commercials will result in lower viewer perceptions of the news entertainment values and other news qualities. Therefore:
H1      Newscasts with commercials will be perceived as higher in entertainment values and news qualities than those without commercials.
Method
Eighty-four undergraduates, from a midwestern university were participants in this study. The students were enrolled in various undergraduate classes and received a small amount of extra credit for their participation. The subjects were assigned randomly to one of six experimental conditions tested across twelve experimental sessions. Sessions were allowed a maximum of twenty participants.
        Each session consisted of showing the participants a segment of A) a high quality newscast or B) a low quality newscast. Two levels of newscasts were employed in order to examine more clearly the influence of any contrast effects that might emerge. Each newscast included one of three commercial conditions; A) three high-quality commercials, B) three low-quality commercials or C) no commercials. This resulted in a 2 x 3 factorial design.
        After watching the program, the participants were then asked to complete a questionnaire giving their impressions of the news program as a whole, the individual news stories, and the commercials. In order to check the manipulation of commercial quality, commercials were evaluated as a group only rather than as individual ads.
Procedure
Each experimental condition was tested via two exposure sessions. Each exposure session was independent and conducted between 12:00 p.m. and 6:00 p.m. over a two-week period. In each session, an experimenter indicated that the subjects were going to be viewing a portion of a newscast, recorded somewhere in the United States. The subjects were then instructed to observe the program in a relaxed manner as if they were at home. Finally, the subjects were informed that they would be completing an evaluation of the program at its conclusion.
The four stories in the high quality news segment were about a prison breakout, car seizures and impounding, high housing prices, and a participant on the TV series Survivor. The high quality newscast originated from a station in the Phoenix market. The four stories included in the low quality newscast were about problems with police and fire radio transmission dependability, the local orchestra schedule, the purchase of a local golf course, and an actor from the TV series Big Brother. The low quality news segments were taken from a newscast out of the Davenport, IA/Moline, IL market.
Quality ads were determined by selecting winning commercials from the London International Advertising Awards. Low quality ads were selected from commercials aired on local television in Moline, IL. In the conditions in which commercials were present, no special attention was drawn to the commercials.
        After exposure, the participants were given one of two questionnaires, A) a seven-page questionnaire (when commercials were present), which included scales for evaluating the news overall, individual stories, as well as the commercial pod, or B) a six-page questionnaire (when commercials were absent), which included scales for evaluating the program. After receiving the questionnaires, the participants were instructed not to look ahead. They were to fill out each page and not go back to a previous page. Upon completion of the questionnaires, the experimenter collected them, and the participants were allowed to leave.
Program Evaluation
Participants answered the following questions about the program. First, the participants rated the news on a set of 14 unipolar scales for the following characteristics: upsetting, cheerful, entertaining, frightening, humorous, boring, interesting, disturbing, scary, uplifting, sad, amusing, depressing, and informative. The ratings were from 0 (not at all) to 10 (extremely). Second, the participants were asked to rate the overall quality of the newscast based on the question, "Overall, how would you rate the quality of this newscast on a scale of 0-10?" (zero to ten corresponded with low to high respectively). Third, the participants completed an open-ended question, which required them to list the subjects of the four stories that were presented in the newscast. Answers were coded as correct if any of the keywords used in the news story were present.
Next, the participants completed an evaluation of each individual news story. The evaluation consisted of two parts. The first part consisted of five questions. The questions were: "In your opinion, how important was this issue?" "In your opinion, how good was the presentation of this story?" "How likely is it, that an issue such as this, would affect you personally?" "In your opinion, how informative was the story?" and "In your opinion, how worthy of news coverage was this topic?"  Each of these questions was evaluated individually using an eleven-point scale ranging from 0 (not at all) to 10 (extremely). The second part consisted of four questions involving a five-point semantic differential scale. These four questions were selected from a set of 17 scales developed by Tannenbaum and Lynch (1960). They were chosen on the basis of Finn & Hickson's (1986) findings and on their applicability to news programming. For the four questions participants rated whether the news was timid/
bold, weak/strong, light/dark, and hard/soft by placing an "X" on a line divided into five equal segments.
Commercial Evaluation
 Once the news questions had been completed, participants answered questions about the commercials. First, participants were asked to rate how well they remembered the commercial pod, using an eleven-point scale ranging from 0 (not at all) to 10 (extremely well). Second, they were asked to rate the commercial pod using a set of 15 unipolar scales for the characteristics creative, unlikable, pleasant, depressing, visually appealing, good, imaginative, dull, captivating, informative, well produce, stupid, entertaining, involving, and boring. The ratings ranged from 0 (not at all) to 10 (extremely). Next the participants were asked to indicate their desire to change the channel during the commercial pod, using an eleven-point scale ranging from 0 (not at all) to 10 (very much). Then participants were asked to indicate the quality of the commercial pod using an eleven-point scale ranging from 0 (low) to 10 (high). Finally, the participants were asked to indicate their opinion as to th
e length of the commercial pod. The scale ranged from one minute to five minutes at half-minute intervals.
Results
Program Treatment Checks
First, the programs were tested for quality. More specifically, we tested whether the programs intended as high quality or low quality were deemed as high or low quality respectively by the participants. While the question directly asking about overall news quality did not reflect a significant difference in participant evaluations, the average evaluation of individual news stories on elements of quality described by Rosenthiel and Gottlieb (1999) did. The question ". . . how important was this issue?" resulted in participants finding the stories in the high quality newscast to be more important (M = 4.6) than stories in the low quality newscast(M = 3.8) (F (1, 78) = 8.25, p <.01). The question ". . . how good was the presentation of this story?" resulted in higher ratings for the quality newscast (M = 5.5) than for the low quality newscast (M = 4.6) (F (1,78) = 5.73, p < .05). Also, the question ". . . how informative was the story?" resulted in higher evaluations for the high qu
ality newscast (M = 5.1) than for the low quality newscast (M = 4.4) (F (1,78) = 4.34, p < .05). Finally, the measure asking ". . . how worthy of news coverage was this topic?" showed that the stories in high quality condition were more worthy of coverage (M = 4.7) than the stories in the low quality condition (M = 3. 6) (F (1, 77) = 16.4, p < .001)  Therefore, it seems clear that participants did perceive a difference in news quality between the newscasts.
Commercial Treatment Checks
Factor analysis (principal component, varimax rotation) was used to determine which of the 15 eleven-point unipolar interval scale characteristics variables measured similar aspects of commercial quality. Two factors emerged. The adjectives creative, visually appealing, good, imaginative, and well produced were positively loaded with each other. All factors loaded at .82 or higher and accounted for 53.8% of the variance. The ratings of the five positively correlated variables were averaged to obtain a composite measure labeled "commercial production quality."  Interitem consistency was high (alpha = .97).
A second factor emerged as well. The adjectives dull, stupid, and boring were also positively loaded with each other. All factors loaded at .80 or higher and accounted for 33.3 % of the variance. The ratings of the three positively correlated variables were averaged to obtain a composite measure labeled "commercial irritation," which arguably suggests the commercials were low in quality. Again interitem consistency was high (alpha = .89).
An ANOVA was performed on the composite measure "commercial production quality" using the independent variable ad condition. The "commercial production quality" measure varied (F (1, 59) = 65.3, p < .001) as a main determinant of the quality of the commercial pod. Respondents rated the production of the commercials as was expected with the high quality ad pod rated as significantly higher (M = 7.1) than the low quality ad pod (M = 2.9).
        Further, an ANOVA was also performed on the composite measure "commercial irritation" using the independent variable ad condition. The "commercial irritation" measure varied, ( F (1, 59) = 31.6, p < .001) as a secondary determinant of the quality of the commercial pod. Respondents rated the low-quality commercial condition to be more irritating (M = 6.4) than the high quality condition (M = 2.8). These two ANOVAs confirm that the manipulation of commercial quality was successful.
Test of Hypotheses
Three variables that seemed to measure information quality components of the newscast were subjected to a confirmatory factor analysis. The variables interesting, informative, and overall quality loaded highly on one factor with all variables loading at .80 or higher. The factor accounted for 68% of the variance and had interitem reliability of alpha = .75. A composite variable created from averaging the three measures was labeled "information quality."
Variables that seemed to measure emotional components of a viewer's response to news were tested by a confirmatory factor analysis to confirm that they indeed measured one construct. The factor analysis revealed that the adjectives upsetting, frightening, disturbing, scary, sad, and depressing were positively loaded with each other. This factor explained 69.7% of the variance with all factors loading at .74 or higher. The ratings of the six variables were averaged to obtain a composite measure of "emotions."  Consistency was high: alpha = .91.
        Variables that seemed to measure the enjoyment of the news were also subjected to a confirmatory factor analysis. These variables, however, measured more than one construct as two factors emerged from the analysis. The first factor involved the adjectives, entertaining, and interesting, which were positively loaded with each other and negatively loaded with the adjective boring. This factor explained 38.6 % of the variance with all factors loading at .73 or higher. The ratings of the two positively correlated variables and the inverse of the negatively correlated variable were averaged to obtain a composite measure of "news enjoyment."  Interitem consistency was high at alpha = .75. The second factor loaded highly on the variables cheerful and uplifting. This factor was disregarded however due to a low alpha of less than .5.
All three of the composite measures were subjected to analysis of variance using news condition and commercial condition as independent variables. Since the analysis of variance for the "emotion" measure only approached significance on the interaction (F (2, 74) = 3.0, p = .054), an analysis of covariance was used with gender as a covariate (see Perry, 2001, who found gender to affect commercial influence on program liking). The dependent variable "emotion" varied significantly, (F (2, 72) = 3.1, p <. 05) as an interaction between news condition and commercial condition when gender was controlled for. With regards to high quality news, respondents felt a similarly high amount of negative emotions resulting from the program upon inclusion of high and low-quality commercials (M = 3.3 and M = 3.4 respectively). However, upon the removal of commercials, respondents felt a decrease in negative emotion, with a mean of (M = 2.0). Low quality news, on the other hand, resulted in low negative emotion as a result of the program upon inclusion of high (M = 1.8) and low quality (M = 1.5) commercials. However, upon the removal of commercials, respondents showed an increase in negative emotions (M = 3.1) as a result of watc
hing the news stories as shown in Table 1. Similar tests of  "information quality" and "news enjoyment" showed no significant effects based on news or commercial conditions.
        Further analysis was performed on the retention level of the stories in the newscast versus the presence of commercials. An analysis of variance was performed using the recall of individual stories as the dependent variable and news condition and ad condition as independent variables. The "recall" varied, (F (2, 78) = 3.4, p < .05) as a main effect of the commercial condition. A Student-Newman-Keuls post hoc analysis showed that respondents remembered fewer of the stories when either high quality or low-quality commercials were present (M = 2.9) compared to when commercials were absent altogether (M = 3.4).
The final test of commercial and news quality influence re-examined Finn & Hickson's (1986) findings using the semantic differential scales. Matching semantic scales for each story were averaged across the four stories viewed. Initial ANOVAs showed main effects based on news quality, but that finding is less meaningful than findings related to advertising quality. Therefore we controlled for news viewing habits using both the number of times per week participants watched the news and the time of day (morning, mid-day, evening, night) that they most often watched as covariates.
The resulting ANCOVAs showed a main effect of commercial quality on the perception of whether the news was "timid/bold" (F (2, 73) = 3.24, p < .05). The analysis revealed there was a contrast effect with low-quality commercials leading to evaluation of the news stories as more "bold" (M = 2.9 on a scale of 1 to 5). When high-quality commercials were shown and when commercials were omitted the news stories were assessed as more "timid" (M = 2.5 for both). A main effect also emerged for news quality (F (1, 73) = 4.35, p < .05) with the high quality news rated as more "bold" (M = 2.8) and low quality news rates as more "timid" (M = 2.5).
The second ANCOVA for the dependent variable "weak/strong" resulted in a main effect only on news quality (F (1, 73) = 4.14, p < .01). The high quality news was rated as more "strong" (M = 2.8) while the low quality news was thought to be "weaker" (M = 2.4). The ANCOVA for the dependent variable "hard/soft" resulted in significant effects based on news condition (F (1, 73) = 7.12, p > .01) with low quality news being rated more "soft" (M = 3.7) than high quality news (M = 3.3).
        In the fourth ANCOVA over the dependent ratings of light/dark, by replacing the covariate "frequency of watching news" with the participants' "primary media source for news" (i.e. television, newspapers, internet, other), the commercial condition comes close to repeating the effect it had on the timid/bold ratings, though significance is at a weaker level (F (2,72) = 2.62, p < .10). The low-quality commercials move viewers toward perceiving the news as more dark (M = 2.54) while those who saw no commercials rated the news more toward the light rating (M = 2.22). Perceptions of viewers who saw high-quality commercials fell in the middle at M = 2.34. In this ANCOVA, the main effect of news condition (F (1, 72) = 18.0, p < .001) is still significant with high quality news being perceived as darker (M = 2.6) than low quality news (M = 2.1).
  Pre vs. post commercial pod effects on news perception. With the commercial break in the newscast segments falling between the second and third stories in both the high and low quality conditions, it was possible to analyze whether the commercials made a difference in perception of news stories that came before the break vs. those that came after the break. A contrast effect would be most likely to show up in news stories shown after the commercials. Therefore, ratings for the stories that appeared before the commercial break were averaged for each of the four semantic differential scales and for one additional question, "How likely is it that this issue such as this would affect you personally?" Five similar composite measures were created for the news stories following the commercial break. Each of these measures was then subjected to analysis of variance. The commercial quality condition was used as the only independent variable since news quality should have no bearing on be
fore and after commercial differences.
None of the pre-commercial news story ratings were influenced by the commercial quality. Responses were significantly different, however, for the post-commercial rating of whether the topics of the news stories would affect them personally (F (2, 81) = 3.28, p < .05). While no viewers thought that being personally affected was very likely on a scale from zero (0) to ten (10), viewers who saw the low-quality commercials gave higher estimations of their likelihood of being affected (M = 2.3) compared to viewers who saw no commercials (M = 1.2). The estimates of those who saw high-quality commercials (M = 1.6) did not differ significantly.
Discussion
        As we analyze the first research question, it is clear that commercial quality influence on newscasts in this study is moderate at best. However, mild support for contrast effects was present. News was perceived to be bolder when shown in combination with low-quality commercials and more timid when shown with high-quality commercials or when no commercials were included. News seems to have also been seen as "darker" in quality with low-quality commercials when contrasted with newscasts lacking commercials. Similarly, an analysis of the perception of only those news stories aired after the commercial break showed that when low-quality commercials were included, the perceived likelihood that the viewer would be affected by the story was greater than when no commercials were shown. The consistency between these three findings, though commercial quality failed to result in other discernable effects, suggests that when the quality of the commercials is considered, low-quality commerci
als can improve perceptions of the news program in which they are aired.
        The elaboration likelihood approach to this research was not supported. The quality level of the award winning commercials did not subconsciously indicate to viewers that the newscast was of higher or lower quality compared to the other two commercial conditions. The newscast was not found to be more enjoyable based on the quality and presence of the commercials, nor did commercial quality and presence influence the reported emotional response to the stories in the news.
        The hypothesis that newscasts with commercials would be perceived to be higher in entertainment values and news qualities compared to newscasts without commercials was not supported. When comparing our results with Finn and Hickson (1986) our findings are similar only for the emotional response to the low quality newscast. In the low quality condition, the presence of commercials resulted in less negative emotional response to the program. When commercials were removed, the program invoked much higher levels of negative emotion. This is similar to Finn and Hickson's participants who responded more positively to the MacNeil/Lehrer News Hour when it contained commercials. Thus, we found only slight support for our hypothesis.
On the other hand there is more to contradict the hypothesis. When the high quality newscast was shown, responses were in the opposite direction. When the news contained commercials, the emotional responses were more negative than when commercials were absent. Thus, we find that the influence of commercials may be dependent on qualities of the newscast. Another negative impact of commercials was in recall of the news stories. When commercials were present, recall was lower than when no commercials were included. Clearly, the impact of disruptive information (the commercials) is likely to impact memory and recall.
Other Findings
        Designing a newscast that controls for all variables except quality is likely to be impossible. The elements that determine whether a newscast is high or low quality are varied and many. Therefore, we avoided hypotheses that predicted variations in perception based upon news quality. The findings we attribute to news quality are a bit tenuous since other more specific variations may account for the differences, but they are still worth mentioning.
        High quality newscasts regularly accounted for heightened perceptions of the newscast on the semantic differential measures. High-quality news was perceived to be more "bold," "strong," "hard," and "dark," while low-quality news was more "timid," "weak," "soft," and "light." Even though these differences were found, the newscasts did not result in differing perceptions of news enjoyment or of information quality. Differences in the emotional response to news were influenced by news quality, but only as an interaction with commercial presence. When either high or low quality commercials were used, respondents showed increased perceptions of negative emotions after viewing high-quality news vs. low quality news. When no commercials were present, perceptions of emotion in the two news conditions were not significantly different but means were in the opposite directions.
Conclusions for Television Journalism
        From a journalism standpoint, this research indicates that commercials may make a difference in how news is perceived. The fact that local news features local advertisers and that in many instances, these advertisers have produced only low budget, low-quality commercials seems to play to the strength of perceptions of the newscast. The newscast will get better reviews from viewers overall because the commercials are poor in quality. What has variously been called contrast, competition, or even a distraction effect comes into play when the commercial quality is lower. As the viewer makes comparisons between the ad and the adjacent news material, this leads the viewer to perceive stronger characteristics of the newscast. Interestingly, the assumption in the viewer is apparently that all content should be expected to show quality since the no-commercial condition and the high quality commercial condition lead to nearly the same perceptions of the newscasts. Thus, for the most part e
ven the award winning ads were not significantly different from the absence of commercials.
        One exception to this seems to have been where the very presence of commercials caused effects regardless of quality. Commercials interrupted information acquisition from news stories, a problem that should be of concern to journalists as they attempt to communicate with their audiences. Furthermore, the presence of commercials influenced emotional evaluation of news. News stories are often emotional in nature provoking sad, disturbing, upsetting, and other emotional responses. Arguably, this may be a perception that a news director would want for the newscast. If emotional response is a goal, then the well run (quality) newscasts may benefit from the presence of commercials. If instead such emotion gets in the way of proper evaluative judgements of news information and leads to negative feelings about the station's newscast, than commercial presence, the very lifeblood of the station, is a detriment to the local news station's image, identity, and journalistic success.
        Future research should examine how other characteristics of commercials interact with the newscast. Finn and Hickson's (1986) look at news used commercials that were arousing - some because of humor and some for other unspecified reasons. Bjorna et al. (2001) used the concept of commercial energy levels finding an interaction based on the type of program in which the ads were shown. The concept of energetic commercials should be applied to news. Also, humorous commercials have been studied for their effect on humorous programs but have not been extended to news. Other options would include evaluating the impact of commercials based on the seriousness of the subject matter. Many commercials shown during news are for serious "products." For example, political commercials and those for medicine and health care products seem to be advertised heavily during news programs. Commercial seriousness could, therefore, be studied for its influence. Finally, commercial quality may contrast wi
th perceptions of other types of programming. Perhaps entertainment programs may also be perceived differently based on the low or high quality of the ads contained in the programs. Much could still be studied along these lines.
 Table 1
Emotional Perceptions of News as a Function of Newscast Quality and Commercial Quality.
Commercial Quality
 News                   High                    Low                     None            Total
Quality                 ____________________________________________________________

High                            3.31Aa                  3.43Aa                  2.07b           2.94

Low                             1.80Ba                  1.46Ba                  3.08b           2.12

Total                           2.55                    2.45                    2.58            2.51


Note: Means not sharing lower case superscripts differ at p < .05 across commercial quality condition (horizontally). Means not sharing upper case superscripts differ at p < .05 across news quality condition (vertically).



References

        Becker, G. A. & Villanova, P. (1995). Effects of rating procedure and temporal delay on the magnitude of contrast effects in performance ratings. Journal of Psychology Interdisciplinary & Applied, 129, 157-167.
        Bjorna, H., Karsai, F., Vicary, R., Wagner, R., & Perry, S. D. (2001, April). The impact of commercial excitation on program appreciation. Paper presented to the Management and Sales Division of the Broadcast Education Association, Las Vegas, NV.
Bozman, C. S. & Mueling, D. (1994). The directional influence of music backgrounds in television advertising. Journal of Applied Business Research, 10, 14-19.
Celuch, K. G. & Slama, M. (1993). Program content and advertising effectiveness:  Hypothesis for cognitive and affective sources of involvement. Psychology & Marketing, 10, 285-299.
Duke, P. (2000). Television:  'One vast ubiquitous library.'  Virginia Quarterly Review, 76, 225-235.
Eastman, S. T. & Newton, G. D. (1995). Delineating grazing:  Observations of remote control use. Journal of Communication, 45, 77-95.
        Eastman, S. T. & Newton, G. D. (1999). Hitting promotion hard:  A network response to channel surfing and new competition. Journal of Applied Communication Research, 27, 73-85.
Finn, S. & Hickson, T. M. (1986). Impact of arousing commercials on perception of TV news. Journalism Quarterly, 63, 369-371.
France, K. R. & Park, C. W. (1997). The impact of program affective valence and level of cognitive appraisal on advertisement processing effectiveness. Journal of Current Issues and Research in Advertising, 19, 1-21.
Greene, W. F. (1992). Observations:  What drives commercial liking?  An exploration of entertainment vs. communication. Journal of Advertising Research, 32, 65-68.
Gottlieb, J. B. & Swan, J. E. (1990). An application of the Elaboration Likelihood Model. Journal of the Academy of Marketing Science, 18, 221-229.
        Gunter, B., Furnham, A. & Frost, C. (1994). Recall by young people of television advertisements as a function of programme type and audience evaluation. Psychological Reports, 75, 1107-1120.
        Gutenko, G. (1997, June). Recall of television advertising messages as influenced by commercial pod position and commercial/program affect congruity of dissonance. Paper presented to the Annual Meeting of the Canadian Communication Association, St. John's, Newfoundland, Canada.
        Gutierres, S. & Kenrick, D. (1999). Beauty, dominance, and the mating game:  Contrast effects in self-assessment reflect gender. Personality & Social Psychology Bulletin, 25, 1126-1135.
        Kamins, M. A., Marks, L. J. & Skinner, D. (1991). Television commercial evaluation in the context of program induced mood:  Congruency versus consistency effects. Journal of Advertising, 20, 1-12.
        Lind, R. A. (1995). How can TV news be improved?:  Viewer perceptions of quality and responsibility. Journal of Broadcasting and Electronic Media, 39, 360-376.
        Lipman, J. (1989). Stuck in the middle. Wall Street Journal,  (March 22, 1989).
        Mittal, B. (1994). Public assessment of television advertising:  Faint praise and harsh criticism. Journal of Advertising Research, 34, 35-53.
        Pavelchak, M. A., Antil, J. H. & Munch, J. M. (1988). The Super Bowl:  An investigation into the relationship among program context, emotional experience, and ad recall. Journal of Consumer Research, 15, 360-367.
        Perry, S. D. (2001). Commercial humor enhancement of program enjoyment:  Gender and program appeal as mitigating factors. Mass Communication & Society, 4, 103-116.
        Perry, S. D. (1998, April). Commercial humor influence on sitcom liking:  A Re-evaluation. Presented to the Mass Communication and Society Division of the Broadcast Education Association (BEA), Las Vegas, NV.
        Perry, S. D., Jenzowski, S. A., Hester, J. B., King, C. M., & Yi, H. (1997). The influence of commercial humor on program enjoyment and evaluation. Journalism and Mass Communication Quarterly, 74(2), 388-399.
        Petty, R. E. & Cacioppo, J. T. (1986). Communication and Persuasion:  Central and Peripheral Routes to Attitude Change. New York:  Springer-Verlag.
        Petty, R. E., Cacioppo, J. T. & Goldman, R. (1981). Personal involvement as a determinant of argument-based persuasion. Journal of Personality and Social Psychology, 41, 847-855.
        Rosenthiel, T. & Gottlieb, C. (1999). Quality brings higher ratings, but enterprise is disappearing. Columbia Journalism Review, 38, 80-89.
        Tannenbaum, P. H.,  & Lynch, M. D. (1960). Sesationalism: The concept and its measurement. Journalism Quarterly, 37, 381-392.
Webb, P. & Ray, M. (1979). Effects of TV clutter. Journal of Advertising Research, 19,  7-12.