Student Researcher: Carmen Stavrositu. This paper is based on a MA thesis project.

Faculty Supervisor: Dr. S. Shyam Sundar

Introduction

Interstitials, the generic term for pop-up and pop-under ads, have long gained their reputation as some of the most intrusive and frustrating Web ads. Consequently, several concerns have been prompted with regard to their impact on user perceptions and attitudes. It is posited in the present paper that, when conceptualizing Websites as the information sources, perceived ad intrusiveness is likely to influence perceptions not only of the ads themselves, but also of the credibility of the Website hosting interstitials and that, more importantly, this influence is likely to be moderated by interstitials’ relevance to Website content. One possible theoretical mechanism that could account for the proposed pattern, the “affect-as-information paradigm,” (Schwarz and Clore, 1983) suggests that instead of scrutinizing the attributes of the object under consideration, people often apply the “how-do-I-feel-about-it” heuristic, and make mood congruent judgments. A second theoretical account resides in the distraction hypothesis (Buller, 1986), positing that communication irrelevant distractions shift receivers’ focus away from a persuasive message and the source of this message fostering attitude change, while communication relevant distractions foster receivers’ attention to the persuasive message and its source, thus inhibiting attitude change, as receivers are able to produce more counterarguments.

Based on these two alternative theoretical perspectives, the following hypotheses were formulated:

Hypotheses

The affect-as-information paradigm suggests that when conceptualizing Websites as sources of information and interstitials as interruptions, frustration and irritation induced by interstitials have the potential of influencing, in turn, subsequent judgments related to the perceived source of this negative affect (i.e., the respective Website). Further, it can be argued that relevant interstitials have the potential of providing extra information to users’ primary online activity and therefore will not be perceived as intrusive or interruptive, but rather helpful, increasing the level of perceived informativeness. However, lack of relevance is very likely to intensify interstitials’ intrinsic intrusive nature, therefore irrelevant interstitials will have a more negative effect on user perceptions. We expect this pattern to emerge only for pop-up ads, but not for pop-under ads, given that generally they do not intrude upon users’ browsing activity.

H1: The relevance of interstitials will be positively associated with Asite for pop-up ads but will not make any difference for pop-under ads.

The distraction hypothesis perspective would suggest that interstitials can be conceptualized as distractions. Thus, when relevant, interstitials are likely to elicit negative perceptions of Website (source) credibility, as users will scrutinize it more, but when irrelevant, focus will be shifted away from the Website, and thus perceptions of its credibility will be more positive:

H2: The relevance of interstitials will be negatively associated with Asite for pop-up ads but will not make a difference for pop-under ads.

In terms of attitudes toward the ad (Aad), both theoretical perspectives lead to the same prediction, such that:

H3: The relevance of interstitials will be positively associated with Aad.

Method

A factorial experiment was designed in order to test for these hypotheses. All participants (N = 44) were randomly assigned to one of four experimental conditions. In order to create the four conditions, one news story (on the Kick Butts Day, a national campaign encouraging activism among youth against the dangers of tobacco), as well two ads (one relevant – anti-smoking PSA and one irrelevant – advertising cheap stocks), were selected from existent Websites and decided upon based on a pretest. Thus, four experimental conditions were created to reflect the relevance manipulation for the two types of interstitials, as follows: relevant pop-up ad, irrelevant pop-up ad, relevant pop-under ad, irrelevant pop-under ad (see below). Subsequent to their exposure to the Website and completion of the experimental task (reading the news story), perceptions of the ads and the Website, as well as memory measures were elicited via a paper-and-pencil questionnaire.

Pop-Up Conditions

Relevant

Irrelevant

Pop-Under Conditions

Results

All 16 items measuring Asite were subjected to a factor analysis in order to determine if these items would load under distinct factors. Since no such pattern emerged, subsequent analyses were conducted on each of the 16 Asite items as dependent variables.

The dependent measures used for assessing Aad were divided into 6 indices: Informative, Quality, Exciting, Annoying, Indifference, and Acceptance.

H1: Not Supported. Congruence between the ad and Website content led to higher believability of the perceived source of the ads, i.e. the Website. However, this effect did not depend on the type of interstitial (i.e. pop-up or pop-under), as the affect-as-information paradigm suggests.

H2: Partially Supported. In terms of objectivity, another strong predictor of credibility, it was revealed that for pop-under ads it is congruence between ad and Website content leading to positive perceptions of Website credibility, while for pop-up ads, it was the incongruence between the two. This prompts us to discount the affect-as-information hypothesis and partially support the distraction hypothesis, which suggests that for pop-up ads, irrelevance is more effective in eliciting positive ratings of Website credibility, while for pop-under ads relevance doesn’t make any difference.

H3. Supported. Participants perceived relevant ads to be more informative, exciting and of better quality than irrelevant ads. Further, they also reported paying more attention to the relevant ads.

Conclusion

As expected, interstitials’ relevance proved to be a key factor in shaping attitudes not only toward the interstitials themselves, but also toward the Website hosting them. Participants found the interstitials to be more informative, exciting, and of better quality when they were relevant to the Website content than when they were not. They also seemed to heed the ads significantly more when they were relevant, rather than irrelevant and also when they took the form of pop-up, rather than of pop-under ads. Moreover, in terms of perceived credibility of the Website, it was shown that ads’ relevance to Website content led to higher perceived credibility of the Website hosting these ads. From a practical perspective, the reported findings indicate that advertisers, as well as news organizations need to consider more carefully ad placement issues, as inappropriate advertisements (i.e., irrelevant, too aggressive) might hinder persuasion effectiveness and also negatively affect users’ trust in the perceived source of these advertisements.

Media Effects Research Laboratory
306, James Building
University Park, PA 16801