Content Effects: Advertising and Marketing

Patrick Weber, Wolfgang Schweiger

University of Hohenheim

This is the submitted version of the following book chapter:

Weber, P. & Schweiger, W. (2017). Content Effects: Advertising and Marketing. In. P.

Roessler (Hg.), International Encyclopedia of Media Effects. Hoboken, NJ: Wiley-Black-

well. doi: 10.1002/9781118783764.wbieme0123,

which has been published in final form in http://onlineli-

brary.wiley.com/doi/10.1002/9781118783764.wbieme0123/abstract

Author Note

Patrick Weber, Department of Communication, University of Hohenheim; Wolfgang

Schweiger, Department of Communication, University of Hohenheim.

Correspondence concerning this chapter should be addressed to Patrick Weber, Univer-

sity of Hohenheim, 540G, 70593 Stuttgart, Germany.

E-mail: p.weber@uni-hohenheim.de, Telephone: +49 711 459 24473, Fax: +49 711 459

24472
 1

Content Effects: Advertising and Marketing

Patrick Weber & Wolfgang Schweiger

University of Hohenheim

p.weber@uni-hohenheim.de, wolfgang.schweiger@uni-hohenheim.de

Word Count

5941

Abstract

The study of advertising effects usually follows two traditions. The first relates the intensity of

advertising input (mostly ad expenditures and exposures) to behavioral and market outcomes

(mostly sales and market share) using statistical modeling techniques to determine advertising

effectiveness. The second seeks to uncover individuals’ mental responses to advertising stimuli

that are associated with memory, attitudinal, and behavioral effects of advertising. This entry de-

scribes the logic of research, central concepts and theories in these two traditions and outlines

long-term trends in academic advertising research.

Advertising and the study of advertising effects

Advertising can be broadly defined as any form of paid strategic communication by an identified

sponsor that aims at informing and/or persuading receivers about an advertising object (e.g.,

product, service, brand, organization, or idea), and traditionally it is conveyed via purchased time

or space in mass media (Thorson & Rodgers, 2012). In an economic context, advertising is a

subcategory and integrated part of all the activities that promote a brand (marketing), next to,

e.g., pricing or other forms of marketing communication like personal selling. Advertising has

several important societal functions (Fennis & Stroebe, 2010, pp. 5–6; Tellis, 2004, pp. 3–5): In

markets it is an important factor in facilitating competition between firms because it enables

them to communicate efficiently with consumers and thus to compete for their attention and pref-

erences. Moreover, it is a key source of funding for mass media in many media systems, and the

advertising industry is an important employer.

In many contemporary societies advertising is omnipresent and its manifold potential con-

sequences offer a fertile ground for research into its intended and unintended, short- and long-

term, functional and dysfunctional effects on the individual, collective or organizational level.

However, the study of advertising effects is mainly driven by interest in the economic functions

that advertising is ought to fulfill. Studies on unintended or dysfunctional advertising effects like

materialism, life-dissatisfaction, or eating disorders are rare. Therefore, advertising effects stud-

ies mostly investigate either outcomes that are intended by an advertiser, or consumers’ mental

processes that are supposed to be related to these outcomes as a function of advertising input (see

Tellis, 2004, pp. 43–50).

Research typically focuses on one of three characteristics of advertising input. First, inten-

sity as the amount of advertising that is targeted to consumers. Second, media as the channel
 3

through which advertising is conveyed. Measures of these two input characteristics range from

measures of budget (e.g., a firm’s advertising expenditures) to measures for the distribution of

advertising through single media (e.g., exposure, reach) and measures of advertising contacts in

the target population. Third, advertising input is qualified regarding its creative content. Re-

search investigates the effects of a great variety of ad content, for example effects of expert and

celebrity endorsers or fear appeals in an advertisement. The outcomes that are studied in re-

sponse to such input are changes in consumer or market behavior. On the individual level, typi-

cal outcome measures include consumers’ brand choice and purchase intensity. Aggregated

measures of outcomes are accounting variables like, e.g., a firm’s sales or revenues in response

to an advertising input, and market outcome measures that quantify a firm’s performance in a

market relative to its competitors (e.g., market share). Finally, the term processes refers to the

mental responses of consumers to an advertising input. Processes are usually classified as cogni-

tive (e.g., thoughts, recall, knowledge), affective (e.g., attitude, image), or conative (behavioral

intentions). Correspondingly, the most frequently investigated individual level effects are

memory effects, attitudinal effects and behavioral effects with respect to the advertising message

and/or the advertising object.

There are two major traditions of the scientific study of advertising effects (Tellis, 2004):

the modeling paradigm and the behavioral paradigm. Researchers working in the modeling tradi-

tion use statistical modeling techniques to relate the intensity of advertising input (mostly ad ex-

penditures and exposures) to behavioral and market outcomes (mostly sales and market share) to

determine advertising’s effectiveness. For example, the classical Anheuser-Busch field experi-

ments for the Budweiser brand in the 1960s varied the firms’ advertising expenditure (i.e., inten-
 4

sity of advertising input) and observed how the brand’s sales reacted to these variations (see Tel-

lis, 2004, pp. 74-75). The researchers found that decreasing the level of advertising input did not

lead to notable market responses (i.e., decreases in sales) in the short term. By varying expendi-

tures for TV advertising separately, they also varied the second input variable that is commonly

used in advertising research: the media through which advertising is distributed. The study

showed that only longer suspensions of TV-advertising led to a decline in brand sales.

The primary interest of researchers working in the behavioral tradition is in how consum-

ers respond to specific advertising stimuli. They mostly use experimental research designs to re-

late specific content variables of advertising to individual level responses (i.e., processes) to de-

termine the effects of advertising. For example, a recent study by Weber, Buchmann, & Wirth

(2014) explored how congruence in the use of unconventional advertising media, i.e., ambient

media, affects the attitude towards the advertised brand. In terms of advertising input, congru-

ence is a content characteristic of advertising that is defined as an overlap in the associations

evoked by the medium and the advertised brand. In an experiment, participants were exposed ei-

ther to advertising with congruent ambient media (e.g., a door of a solarium on which the slogan

and brand of a travel agency were placed) or to advertising with incongruent ambient media

(e.g., the same travel brand whose slogan was placed on a strip of a zebra crossing). The authors

compared the attitude towards the advertised brand in the experimental groups and found that it

was higher in the groups that were exposed to congruent ambient media. Since attitude is an af-

fective response to a specific advertising stimulus, the study investigated a process variable that

is typical for research in the behavioral tradition. In addition, the study gives hints to further pro-

cesses: It showed that the comparative advantage of congruent ambient media disappeared when
 5

participants were highly involved. This indicates that highly involved consumers might be moti-

vated to dissolve any incongruence so that the ambient medium makes sense, and that conse-

quently a mildly incongruent ambient medium can be as efficient as a congruent one in affecting

consumers’ attitudes.

Modeling approach to advertising effectiveness

The modeling approach is mainly based on field studies that use real market data and tradition-

ally investigate relationships between aggregate level variables. Input measures are media spend-

ings, audience or target group reach, and average contact frequency at the level of brands, cam-

paigns and media channels. Typical outcome measures are at the level of markets or market seg-

ments, mainly sales, revenues, market shares. Effects can be proven if an input variable’s change

(increase or decrease) precedes the outcome variable’s change in a given timeframe. To examine

this kind of causality, traditional studies following the modeling approach usually use aggregate

economic data to test so-called econometric models in longitudinal designs. Studies of advertis-

ing’s effects on economic indicators represent a minority in academic advertising research: In a

content analysis of 17 major advertising, marketing, and communication journals, Kim et al.

(2014) found economic effects to be focused in about 10 percent of all advertising studies pub-

lished from 1980 to 2010.

Results from modeling studies are of a high external validity for several reasons (Tellis,

2004, pp. 59–60): First, studies are dealing with real brands, products, and advertisements on real

markets. Second, they measure real-world consumer behavior, which means that no sampling bi-

ases or measurement effects like social desirability or looking-good effects might occur. Third,
 6

grand samples are used from commercial market research or in some cases even the whole popu-

lation is analyzed, e.g. in online advertising when banner contacts are registered and statistically

related to online shop visits or shopping activities.

A further advantage compared to experiments is the potential to investigate time series.

This allows analyzing the so-called advertising carryover, i.e. the question how long the effect of

a campaign lasts until it completely vanishes. This mainly applies to products that consumers

buy only once every few years (e.g., washing machines, stereos or automobiles). Other time se-

ries phenomena are wear-in and wear-out effects. Wear-in measures the time that a new cam-

paign needs to become fully effective; wear-out refers to the loss of a campaign’s effectiveness

over time, e.g. when the audience starts getting used to and bored by a specific motif or kind of

joke.

Nonetheless, modeling studies only allow a limited level of causality testing as changes in

dependent variables might also be induced by external variables, like market developments or

channel irregularities, overriding the effects of the independent variables under analysis. Conse-

quently, cross-lag correlations between a cause at t0 and an effect measured at t1 offer no satisfy-

ing proof of causality if they are only measured once.

In the past years, advertising research has paid increasing attention to other relevant types

of aggregated consumer behavior that became available to researchers: communicative consumer

reactions. Typical examples are the number of product-brochures ordered or telephone-calls ask-

ing for further product or service information. In online and mobile marketing, these and other

special types of consumer behavior are called conversions, including, e.g., the number of online

shop visits, download of trial licenses, newsletter-subscriptions, and contacts to other marketing

or public relations outlets promoted by an initial advertising (e.g., watching a brand’s YouTube
 7

video). Further consumers’ communicative reactions (e.g., Facebook likes, retweets, comments,

recommendations) can be registered online and analyzed as outcome variables. Consequently,

the dynamic multiplication of easily available market outcome measures in the past years has in-

creased the relevance and spread of modeling studies in advertising effects research beyond the

former focus on economic data.

New and easily available data-sources – mainly on the Internet – not only provide a broad

variety of aggregate outcome measures, they also allow for the analysis of individual level data

in field studies with longitudinal designs. Two examples: Whenever an individual user perceives

an advertising banner, clicks on it, then lands at an online shop and finally buys a product there,

this sequence of actions is automatically registered and can be analyzed on the individual level.

The same applies to Facebook, where each individual fan’s communicative action is registered to

the minute and can be connected to the stimulating advertising activities.

As the availability of more detailed output measures rose, the portion of field studies scru-

tinizing the impact of concreter input variables on these output measures also increased. Espe-

cially communication studies moved beyond general campaign input data and started to analyze

the effects of campaign, media, and copy strategies.

Table 1 provides an overview of all types of modeling studies by differentiating the levels

of aggregation for input and outcome measures.

Table 1: Classification of input and outcome measures in modeling studies

Level of Input Outcome

aggregation

high (1) (A)

• advertising budget • revenue, return-of-invest-

• advertising weight ment

• advertising elasticity

mid (2) (B)

• campaign strategy x • sales (aggregate purchase

• media strategy: advertising behavior)

frequency and scheduling, • aggregate communicative

selection and combination behavior

of media and instruments

low (3) (C)

• copy strategy • individual communicative

• individual number of con- and purchase behavior

tacts to advertisement

The highest level of aggregation stands for basic economic variables, often called KPIs

(key-performance indicators). On the input-side, classical KPIs are advertising budget and

weight, each referring to a specific brand, product, or campaign in a given timeframe. Advertis-

ing weight can be represented by the grand total of all potential advertising contacts, which can

be calculated by multiplying the number of ads published and each vehicle’s reach. Common
 9

outcomes are monetary measures like revenue and return-of-investment. Advertising elasticity is

a derived measure indicating how elastic the market reacts to changes in advertising. It measures

how effective an increase/decrease of the advertising budget or weight is (elasticity = % change

in quantity demanded / % change in advertising).

On the mid-level, operative campaign data are aggregated and analyzed. This refers to the

general campaign strategy (e.g. branding versus sales campaign) and the media strategy as input.

Typical media strategy parameters are advertising frequency and scheduling tactics (e.g. continu-

ous vs. flighting or pulsing, the selection and combination of media vehicles and advertising in-

struments (e.g., online banner versus interstitial). As outcome, sales and all other kinds of aggre-

gate consumer behavior can be examined, including the above-mentioned communicative reac-

tions.

On the lowest level, the concrete copy strategy concerning the dimension (e.g. format,

spot-length), content and design of each advertisement on the one hand and the individual num-

ber of contacts to advertisements on the other can be analyzed as input. As outcome, studies deal

with each consumer’s overt communicative and purchase behavior.

It is important to note that all levels of aggregation on the input and output side can be

combined. Traditional econometric modeling studies deal with the impact of the advertising

budget or weight on effectiveness (type 1xB) or monetary efficiency (type 1xA). Typical ques-

tion are: What advertising weight is needed to achieve the desired effects? Is there a linear corre-

lation between advertising input and outcome (‘the more the better’) or is there a threshold where

an increased weight does not result in an increased output (saturation)? Other studies connect

campaign and media strategy characteristics to monetary measures (type 2xA) or to aggregate

behavior (type 2xB). Finally, there are field studies investigating the impact of all levels of input
 10

measures on individual behavior (mostly type 2xC and 3xC). Prior to the already-mentioned

online and social media data, single-source data were used for this kind of studies containing in-

dividual-level data of consumers’ campaign and copy exposure on one hand, and consumption

and communication data on the other (e.g., Lodish et al, 1995). This also permits the investiga-

tion of cross-media effects either on an aggregate campaign level (mid input level) or on the indi-

vidual level focusing on consumers’ ad contacts on different media channels (see Assael, 2011).

Behavioral approach to advertising effects

The primary objective of the behavioral approach to the study of advertising effects is to relate

distinct qualities of advertising messages to individual consumer responses. Responses of interest

are the processing, retention and recollection of advertising information, the formation and

change of attitudes toward advertising objects, and behavioral responses (Fennis & Stroebe,

2010). The following paragraphs describe how respective individual level effects are conceptual-

ized in advertising research and they briefly present some mechanisms that are used to explain

advertising’s impact on these responses (see Fennis & Stroebe, 2010 for a detailed treatment of

these issues).

Memory effects

Memory effects of advertising are important because many advertising contacts occur in non-

purchase contexts (e.g., listening to the radio while doing domestic work) so that an effect of ad-

vertising on purchase behavior should somehow be mediated through long-term memory for ad-

vertising information (Montgomery & Unnava, 2007).

A basic distinction of memory effects is between effects on explicit and implicit memory

(see Fennis & Stroebe, 2010, pp. 82–85). Explicit memory is a form of memory that enables the

conscious recollection of events and facts. It is usually tested with recognition and (aided and un-

aided) recall tests of advertising information following advertising exposure (e.g., remembering

an advertisement, the advertised product category or brand). Implicit memory is a nonconscious

form of memory. Implicit memory effects occur when a previous exposure to advertising affects

the performance in a subsequent task without that one remembers the advertisement or is aware

of its impact on subsequent performance. Typical tasks to test implicit memory are word stem

completions and word fragment identifications. In such tests, recipients who have been exposed

to advertising are presented with some letters and are asked to identify a word that fits. Implicit

memory is shown by completions to words that were content of the advertisement. This effect

can occur when the recipient does not explicitly remember the advertising contact and is not

aware of the connection between the word identification task and advertising exposure.

It is a common assumption in advertising research that memory is a function of the depth

of processing of advertising: the deeper advertising information is processed, the better is explicit

memory (see Fennis & Stroebe, 2010, pp. 79–80 for details). A prominent model to represent the

different stages of advertising processing is Greenwald & Leavitt’s (1984) model of four levels

of audience involvement in advertising. The levels differ with respect to the attentional capacity

required to process information on each stage. The first stage, preattentive analysis, requires no

conscious attention, can occur without intention and is likely to result in storage of perceptual

and basic semantic information in implicit memory. Processing information on the second level,

with focal attention, is a prerequisite for advertising to be remembered explicitly. Conscious at-
 12

tention depends on consumer factors (e.g., current desires, attitudes) and features of the advertis-

ing message such as vividness, salience, or novelty (Fennis & Stroebe, 2010, pp. 51–55). The

third level is comprehension, i.e. representing the semantic meaning of the message. The deepest

level of information processing on which most cognitive capacity is required is elaborative rea-

soning by which the represented information is linked to other knowledge that is stored in

memory and what allows for more complex inferences.

One central question with respect to memory effects is how advertising information is re-

trieved from memory in situations subsequent to exposure (e.g., in purchase situations) and what

factors influence information retrieval. Research shows that the amount and quality of infor-

mation that can be retrieved depends on the time that passed between advertising exposure and

retrieval situation, on the relative position of a to-be-remembered information in an advertise-

ment or in a set of advertisements (primacy- and recency effects), on repetition and spacing be-

tween repeated advertising exposures, on the amount of competitive information in the exposure

situation (i.e., advertising clutter), on the presence of appropriate of cues in the retrieval situation

(e.g., at the point of purchase), and on the affective state of the individual (see Montgomery

& Unnava, 2007 for an overview).

An important process that is based on the activation of memory content is priming (see

Fennis & Stroebe, 2010, pp. 85–88). Priming is an increase of the accessibility on the mental rep-

resentation of a word or object (e.g., a brand name, product attributes) that is stored in memory

as a result of exposure to this word or object. A consequence of the activation (i.e., priming) of a

concept is that, for some time, it can influence the individual’s responses in a later unrelated con-

text without any awareness of this influence on the side of the individual (priming effects). Prim-

ing effects can include evaluative responses such as brand evaluations, or behavioral responses
 13

such as purchase decisions. Importantly, priming works with and without awareness of exposure

to the priming stimulus. This implies that shallow processing of advertising that is characteristic

for incidental advertising exposure, i.e. preattentive analysis, can be sufficient for advertising to

have substantial effects.

Attitudinal effects

Advertising’s impact on attitudes is investigated because attitudes are believed to be important

predictors of behavior, particularly purchase decisions. Research focusses on effects on evalua-

tions of two attitude objects: the advertising object, usually a specific brand (attitude towards the

brand), and the advertisement (attitude towards the ad). There are evaluations that the individual

is consciously aware of and that can be tested using self-report measures (explicit attitudes), and

evaluations that the individual is not aware of and that influence reactions that are not under the

individual’s conscious control (implicit attitudes). Implicit attitudes are assessed by observing

such reactions in specifically designed tasks.

Advertising’s impact on attitudes can be mediated by a multitude of mechanisms that are at

the core of theories that are commonly applied in the behavioral tradition of advertising research.

Some of these describe rather cognitive processes, others explain attitudinal effects through af-

fective responses to advertising. A basic cognitive mechanism is that exposure to advertising can

increases the cognitive accessibility a preexisting attitude toward an advertised object: An adver-

tising message might simply activate the concept of a brand and associated evaluative concepts

in memory so that when attitude is tested, the existing attitude can be retrieved more easily than

when the brand was not primed by an advertising message. Moreover, an advertising message or
 14

its’ context might activate concepts beyond the brand, but the brand will be evaluated in the light

of the activated concepts (priming effect).

Advertising can affect knowledge in form of the beliefs that consumers hold about the ad-

vertised object. Classical attitude theory sees such beliefs and their evaluative implications as the

building blocks of attitudes (Fishbein, 1963). Thus, advertising can affect attitudes by associating

the object with certain attributes and thereby establish or alter beliefs about the advertised object

(belief formation and change).

Other theories conceptualize advertising’s effect on attitudes as the result of message elab-

oration, i.e. as a function of the net valence of the thoughts that consumers generate in response

to the arguments in an advertising message (systematic processing). This idea is central in the

dual process theories of persuasion that also specify the boundary conditions of this mechanism:

high motivation and/or ability to process the advertising message (see Schumann, Kotowski,

Ahn, & Haugtvet, 2012 for sources and a review of applications of dual process theories, espe-

cially the elaboration likelihood model, in advertising research). Because the favorability of cog-

nitive responses depends on the quality of arguments advertising can influence attitudes by

providing strong arguments for the advocated position.

Another way of influence of advertising on attitudes is through conveying information that

are part of judgment heuristics (heuristic cues). For example, a product’s country of origin can be

used by people relying on the country-of-origin heuristic (that is based on country stereotypes) to

make attitudinal judgments about an advertised product. Other heuristic cues, e.g. expertise of a

spokesperson or the number of arguments, can play a role in the evaluation of the validity of the

advocated position in an advertising message and the decision to accept it or not. Dual process

theories predict that such heuristic judgments can be a route to persuasion when consumers are
 15

not motivated and/or able to think about the message arguments with high effort (heuristic pro-

cessing).

Advertising’s impact on attitudes can also be mediated by affective responses. These can

range from affective metacognitive experiences accompanying ad processing to discrete emo-

tions that are induced by executional features of an advertising message. An important metacog-

nitive experience is hedonic fluency, i.e., the experience of a mildly positive emotion that is asso-

ciated with a heightened subjective ease of perceiving and processing information. Hedonic flu-

ency can serve as an information to evaluate the processed stimulus, i.e., the ad (Schwarz, 2004)

and the attitude towards the ad has a positive effect on attitude towards the brand (MacKenzie,

Lutz, & Belch, 1986). Among others, repeated exposure contributes to hedonic fluency because

when an advertising message is encountered again, existing memory content, be it implicit or ex-

plicit, eases processing. Therefore, hedonic fluency offers one explanation for mere exposure ef-

fects of advertising.

Both substantive information in an advertising message and its creative execution can elicit

emotions in the receiver. Such emotions can influence attitudes in at least two ways. During ex-

posure and/or judgment cognitive concepts can be activated that are congruent with the triggered

affective reactions (affective priming). This affects cognitive reactions to the ad, especially think-

ing about arguments, and thereby attitudes. Another possible mechanism is that affect is used as

a basis for judgment (affect-as-information): people infer their attitude towards the advertised

brand from their experienced feelings (‘how do I feel about it’ heuristic) – even if these feelings

were induced by executional features of an advertisement (like music) that are irrelevant for the

evaluation of the attitude object (misattribution of affect). Affective priming effects are more
 16

likely in systematic processing, effects through misattribution of affect in heuristic processing

(Forgas, 1995).

Behavioral effects

Conative processes are the most proximate to actual behavior among the measures of mental re-

sponses to an advertising input. Research frequently relies on self-report measures to assess pur-

chase intention, purchase probability, buying interest, attitude toward purchasing the advertised

brand, and consideration of a brand for purchasing. In the behavioral tradition, advertising’s im-

pact on such measures is often conceptualized as mediated by attitudes (that have been changed

or formed through advertising).

According to the MODE-Model (Fazio, 1990), exactly how attitudes affect actual behavior

depends on the motivation and opportunity to engage in an effortful decision making process. If

this motivation is high (for example when buying expensive durables), the attitude toward the

behavior (e.g., attitude toward purchasing a certain brand) will be considered to form a conscious

behavioral intention (e.g., a purchase intention). The Theory of Planned Behavior (see, e.g.,

Ajzen, 2005) posits that attitudes don’t affect behavior directly but that their impact is mediated

by such behavioral intentions which are additionally influenced by subjective norms and the per-

ceived behavioral control with regard to the specific behavior. The potential impact of advertis-

ing in this process is that it might affect the beliefs on which attitude, subjective norm, and per-

ceived behavioral control are based.

On the other hand, when motivation and opportunity to make an effortful decision are low

(for example when buying everyday consumables), contact with an attitude object can lead auto-

matically to attitude consistent behavior (e.g., buying a product that one likes) – without that the
 17

consumer consciously considers his attitude, is aware of its influence, or forms an intention. The

condition for this is that an attitude towards the object is stored in memory and that this attitude

is accessible when the individual encounters the object, i.e., a brand. The potential impact of ad-

vertising in this process is that it might affect the attitude and/or general attitude accessibility

through repeated contact with the attitude object in non-purchase contexts or that it primes an ex-

isting attitude in the purchase situation, e.g., via in store advertising.

Another route of advertising’s influence on behavior is through affecting the cognitive ac-

cessibility of information that are used by consumers when forming a consideration set in pur-

chase situations (see Fennis & Stroebe, 2010, pp. 95–99). Such information include brand names

or contextual information, e.g., typical usage situations for a product. Increased accessibility of

such knowledge (for example through repeated exposure to advertising for a brand) increases the

likelihood that a brand comes to mind when the consumer deliberately thinks of buying a brand

from a certain product category. Moreover, accessibility of information can direct attention and

lead to feelings of familiarity when consumers encounter different brands in a purchase situation.

This in turn influences which brands will be considered for buying.

Resistance towards advertising

It is well recognized in advertising research that consumers are not naïve receivers of advertising

messages that are defenseless against the persuasive attempts of marketers. Consumers fre-

quently try to avoid advertising (for example through zapping TV-ads), and even if they are ex-

posed to it, they oftentimes are not motivated to process the message deeply but follow advertis-

ing at best with low attention. Another, more extreme form of resistance is to respond with reac-
 18

tance to the perceived threat to attitudinal freedom that a perceived influence attempt through ad-

vertising might cause. Reactance motivates consumers to restore this freedom, for example

through increased counterarguing a message’s arguments, by disregarding the message or by de-

valuating the communicator (see Fennis & Stroebe, 2010, pp. 184–186).

The advertising literature describes several individual level variables that are likely to

moderate the effects of advertising input on cognitive, affective, and behavioral responses. First

of all, advertising itself is an attitude object and attitudes towards advertising (see, e.g., Tan &

Chia, 2007) affect responses to specific advertising stimuli. Research shows that consumers hold

attitudes on different levels of concreteness (Attitude towards advertising in general and attitudes

towards advertising in different media), and that they differentiate between two objects when

evaluating advertising (advertising as a marketing instrument and as a social institution). These

attitudes are based on several beliefs, among others beliefs about the informational value of ad-

vertising. Negative informational beliefs can be associated with a generalized tendency to disbe-

lief advertising claims, i.e. ad skepticism (Obermiller & Spangenberg, 1998), what propels re-

sistance towards advertising. Finally, consumers possess lay knowledge about the aims and tac-

tics of marketers (persuasion knowledge) and have developed strategies to cope with persuasive

attempts, e.g. by reinterpreting an advertising message on the basis of persuasion knowledge

(Friestad & Wright, 1994).

Trends in advertising research

Long-term trends in the academic study of advertising are revealed by the already mentioned

content analysis of research articles from 1980 to 2010 by Kim et al. (2014). The analysis shows
 19

that advertising effects is the second most frequent topic area of advertising research, with adver-

tising practice (e.g., management issues) being the most frequently studied topic in general. No-

tably, research that originates from the communication discipline is more likely to address adver-

tising effects than research that is published in pure advertising journals.

The majority of research focuses on individual level effects; economic effects, i.e. market

outcomes that are the focus in the modeling tradition, are concerned only in a small proportion of

research. Interest in individual responses to advertising input dipped between 1980 and 1990 but

became dominant since then. Among the individual consumer responses to an advertising input,

cognitive effects are examined most frequently, followed by affective effects (such as attitudes)

and behavioral effects.

The media of interest to advertising researchers changed over the last 30 years. Overall,

print is the most studied medium, followed by television. But since 1995, advertising research on

the Internet has steadily increased and surpassed television by 2005. However, print advertising

in 2010 still was the dominant focus of research. Since 2005, a considerable, yet inconsistent cor-

pus of academic research on mobile advertising and marketing exists, still lacking a broadly ac-

cepted definition, conceptualization and framework of the phenomenon (Varnali & Toker, 2010).

The role of theory in advertising research increased over the last three decades (Kim et al.,

2014). Until mid-1990s, research was mainly driven by empirical results or did not use a theory.

With the beginning of the third millennium, theory became more prevalent in advertising re-

search, and by 2005 a clear majority of studies was driven by theory. The main theoretical driv-

ers of advertising research originate from the discipline of psychology: The most frequently used

theories and frameworks are dual-process models, information processing theory, and congruity

theory. Involvement, interactivity, and source credibility are the three most applied constructs.
 20

Frequently referred concepts from the marketing literature are brand equity and integrated mar-

keting communication. Advertising specific theoretical approaches that are most frequently used

in research are hierarchy of effects models, the concept of attitude toward advertising, and ad-

vertising effectiveness models. The most referred contributions from the communication disci-

pline following interactivity are the uses and gratifications approach and models of media. These

findings show that hierarchy of effects models that conceptualize advertising’s effects as a fixed

sequence of cognitive, affective, and behavioral effects like the classical AIDA-model (propos-

ing a sequence of attention, interest, desire, and action) are no longer the dominant theoretical

tools to understand advertising effects. Sometimes they are used in the modelling tradition of ad-

vertising research that observes relationships on the aggregate level to represent assumptions

about how advertising works on the individual level (see Tellis, 2004). However, research in the

behavioral tradition that explicitly focusses on the individual uses more contemporary psycho-

logical approaches to understand advertising’s effects (see Fennis & Stroebe, 2010).

The dominant research orientation in the study of advertising is empirical and follows a

quantitative approach. The empirical quantitative approach has dominated advertising research

from the 1980s on but the gap between nonempirical and empirical, as well as between qualita-

tive and quantitative research started to widen in the 1990s. Correspondingly, the methods pre-

dominantly used in advertising research are quantitative techniques, most prominently laboratory

experiments, surveys, and content analyses. A focus on individual level effects that are assessed

empirically is also evident in the commonly researched sampling units: The most frequently

sampled units are adults and students, followed by secondary data sets and advertisements.

Among the nonquantitative methods the most commonly applied is the systematic (literature) re-

view.
 21

Among the implications of research findings that are typically discussed, practical implica-

tions for marketers and advertisers are most commonly addressed by advertising researchers. Far

less discussed are theoretical and methodological implications, and consequences for public pol-

icy or education are even more seldom addressed. This focus on the practical implications of re-

search results reflects the predominant interest in the economic function of advertising that moti-

vates and justifies research. Therefore it seems that the dominant function of current advertising

research is to support the effective and efficient realization of advertising’s economic function.

Future research on the effects of advertising, especially research that originates in the discipline

of communication, could develop alternative and distinct perspectives on advertising effects to

be able to contribute knowledge about the manifold consequences that advertising can have on

the individual, organizational and societal level.

SEE ALSO: Media Effects: Levels of Analysis; Uses and Gratifications: Basic Concept; Uses

and Gratifications: Type of Media; Priming; Dual Process Models of Persuasion; Fear Appeals in

Strategic Communication; Attention/Awareness; Interactivity; Modality/Medium Characteristics

References and Further Readings

Ajzen, I. (2005). Attitudes, personality and behavior (2. ed.). Maidenhead: Open Univ. Press.

Assael, Henry. (2011). From silos to synergy. A fifty-year review of cross-media research shows

synergy has yet to achieve its full potential. Journal of Advertising Research, 51(1), 42-58.
 22

Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The mode model as

an integrative framework. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology

(pp. 75–109). San Diego: Elsevier.

Fennis, B. M., & Stroebe, W. (2010). The psychology of advertising. Hove: Psychology Press.

Fishbein, M. (1963). An investigation of the relationships between beliefs about an object and

the attitude toward the object. Human Relations, 16, 233–240.

Forgas, J. P. (1995). Mood and judgment: The affect infusion model (AIM). Psychological Bul-

letin, 117(1), 39–66. doi:10.1037/0033-2909.117.1.39

Friestad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope with per-

suasion attempts. Journal of Consumer Research, 21, 1–31.

Greenwald, A. G., & Leavitt, C. (1984). Audience involvement in advertising: Four levels. Jour-

nal of Consumer Research, 11(1), 581–592.

Kim, K., Hayes, J. L., Avant, J. A., & Reid, L. N. (2014). Trends in advertising research: A lon-

gitudinal analysis of leading advertising, marketing, and communication journals, 1980 to

2010. Journal of Advertising, 43(3), 296–316. doi:10.1080/00913367.2013.857620

Lodish, L.M., Abraham, M., Kalmenson, S., Livelsberger, J., Lubetkin, B., Richardson, B., &

Stevens, M.E. (1995). How T.V. advertising works: A meta-analysis of 389 real world split

cable T.V. advertising experiments. Journal of Marketing Research, 32(2), 125-139.

MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of attitude toward the ad as a me-

diator of advertising effectiveness: A Test of competing explanations. Journal of Marketing

Research, 23, 130–143.

Montgomery, N. V., & Unnava, H. R. (2007). The role of consumer memory in advertising. In

G. J. Tellis & T. Ambler (Eds.), The Sage Handbook of Advertising (pp. 105–119). Los Ange-

les, Calif.: Sage.

Obermiller, C., & Spangenberg, E. R. (1998). Development of a scale to measure consumer

skepticism toward advertising. Journal of Consumer Psychology, 7(2), 159–186.

Schumann, D. W., Kotowski, M. R., Ahn, H.-Y., & Haugtvedt, C. P. (2012). The elaboration

likelihood model. A 30-year-review. In S. Rodgers & E. Thorson (Eds.), Routledge communi-

cation series. Advertising theory. New York: Routledge.

Schwarz, N. (2004). Metacognitive experiences in consumer judgment and decision making.

Journal of Consumer Psychology, 14(4), 332–348.

Tan, S. J., & Chia, L. (2007). Are we measuring the same attitude? Understanding media effects

on attitude towards advertising. Marketing Theory, 7(4), 353–377.

Tellis, G. J. (2004). Effective advertising: Understanding when, how, and why advertising works.

Thousand Oaks, Calif: Sage.

Thorson, E., & Rodgers, S. (2012). What does "Theories of Advertising" mean? In S. Rodgers &

E. Thorson (Eds.), Routledge communication series. Advertising theory (pp. 3–17). New

York: Routledge.

Varnali, K. & Toker, A. (2010). Mobile marketing research: The-state-of-the-art. International

Journal of Information Management, 30, 144-151.

Weber, P., Buchmann, S., & Wirth, W. (2014). Schlechtes Ambiente? Die Bedeutung des Kon-

sumenten-Involvements für die Wirkung moderat inkongruenter Ambient-Media [Bad ambi-

ence? The role of consumer involvement for the effectiveness of mildly incongruent ambient
 24

media]. In H. Schramm & J. Knoll (Hg.), Innovation der Persuasion. [Innovation of persua-

sion] (pp. 200-217). Cologne: Herbert von Halem.

Author Mini-Biography:

Patrick Weber holds a PhD form the University of Zurich, Switzerland, and is senior researcher

and lecturer of communication at the University of Hohenheim, Germany. His research interests

include media reception and effects, persuasion and advertising, and public communication in

online media. He recently co-edited a comprehensive handbook of advertising research in Ger-

man language.

Wolfgang Schweiger is professor of communication at the University of Hohenheim, Germany.

He holds degrees (M.A., Dr. phil., habil.) in communication studies, political science, and law

from the University of Munich. His research interests cover a broad range of issues from corpo-

rate communication and evaluation, media use and effects research, online research, risk commu-

nication, and empirical methods.