Product Category Involvement and the Reaction of Belgian and Polish Consumers to Different Advertising Appeals

Citation:

Maggie Geuens and Patrick de Pelsmacker (1998) ,"Product Category Involvement and the Reaction of Belgian and Polish Consumers to Different Advertising Appeals", in E - European Advances in Consumer Research Volume 3, eds. Basil G. Englis and Anna Olofsson, Provo, UT : Association for Consumer Research, Pages: 33-41.

European Advances in Consumer Research Volume 3, 1998      Pages 33-41

PRODUCT CATEGORY INVOLVEMENT AND THE REACTION OF BELGIAN AND POLISH CONSUMERS TO DIFFERENT ADVERTISING APPEALS

Maggie Geuens, University of Antwerp, Belgium

Patrick de Pelsmacker, University of Antwerp, Belgium

INTRODUCTION

During the last decade emotional advertising has received increasing attention. Some researchers claim emotional appeals as compared to non-emotional appeals lead to more positive communication effects such as ad and brand recognition, attitudes and intentions (Aaker and Bruzzone 1985, Aaker et al. 1986, Batra and Ray 1986). Different emotional appeals may lead to different communication effects, though. Indeed, a review of the literature suggest that for instance humour, eroticism and warmth can induce different communication results. Ad and brand recall, the attitude towards the ad (Aad), the attitude towards the brand (Ab), the evaluation of the company and the purchase intention (Pi) seem to be significantly different when respondents are exposed to erotic, humorous or warm stimuli (see for an overview of the literature LaTour and Henthorne 1994, Weinberger and Gulas 1992, De Pelsmacker and Geuens 1996a, Geuens and De Pelsmacker 1996).

There are two reasons to expect different responses to advertising stimuli for consumers of Belgium and Poland. First of all, cultural differences is an aspect no marketing or advertiser researcher should neglect. For instance, Andrews et al. (1991) report significant cultural differences in the attitude consumers hold towards advertising in general. With respect to advertising the influence of cultural differences is very pervasive since "as a form of social communication, advertising is considered to be particularly reflective of culture" (Hong et al. 1987). Cross-national research regarding the content of magazine or television advertising usually shows more differences than similarities (Dowling 1980, Madden et al. 1986, Rice and Lu 1988, Cutler and Javalgi 1992. Cutler et al. 1992, Biswas et al. 1992). These cultural differences in ad content might perhaps be interpreted as indicators of differences in ad perception and ad preferences since advertising practice should reflect the best domestic strategies. A second reason to expect different responses to advertising stimuli in Belgian and Polish respondents might be situated in a different marketing and advertising tradition between the two countries. The marketing environment in Poland is considerably different from the one in Belgium. Poland is an economy in transition, in which professional marketing activity only has a short history. According to Mibner (1994) 29% of the Polish families consider advertising as a friend or as a counsellor while 32% view advertising as a seducer to spend money. Especially young and well-educated people feel sceptical towards advertising in general. As a consequence of this short marketing and advertising tradition exposure levels to advertising are much lower in Poland than in Belgium. The latter might have a positive influence on the attitude towards advertising in general and to specific advertising executions in particular. On the other hand, although De Pelsmacker and Geuens (1996b) expected to observe large differences in communication effects of different types of ads, a lot of similarities between Polish and Belgian subjects emerged. The most important difference between the two groups is that affective responses are less important than cognitive ones for the formation of attitudes toward the ad and the brand in the Polish sample, as opposed to the Belgian one.

RESEARCH OBJECTIVE AND RESEARCH HYPOTHESES

The objective of this paper is to investigate the influence of product category involvement on the responses to emotional and non-emotional advertising stimuli, and to compare these responses in a Belgian and a Polish sample. The paper focuses upon the effects of product category involvement both across advertisement types and countries.

The study of the importance of involvement for consumer perception and consumer decisions has a long tradition. Although no exact definition of involvement exists there seems to be general agreement that the level of involvement can be dscribed as the degree of personal relevance or importance of the product based on inherent needs, values and interests (Zaichkowsky 1985, Park and Young 1986). In high involvement conditions, in other words when the message information is perceived as personally relevant or important, subjects appear to devote considerably more cognitive effort to the message content and they seem to process the information presented at a deeper level than when personal relevance is low (Petty et al. 1983, Park and Young 1986). With respect to persuasion and attitude formation Petty and Cacioppo (1986) distinguish two different routes, the central and the peripheral route. Which one will be used depends on the elaboration likelihood, i.e. on a person’s level of involvement or motivation and on a person’s ability to carefully process the ad. When the elaboration likelihood is high people are said to engage in an extensive processing of the message arguments trying to find out what the ad really has to offer. This kind of argument scrutinizing is typical for the central route to persuasion. However, when the motivation and/or ability is low, people are more likely to concentrate on simple cues in the message context (such as an attractive source, background music, humour, ...) without actually processing the information. The latter kind of processing refers to the peripheral route to persuasion (Petty and Cacioppo 1986, Cacioppo and Petty 1989). Although the operation of simple or peripheral cues seems quite simple on the surface, it is more complex than it seems. According to the ELM simple cues such as source attractiveness and evoked affect can serve as peripheral cues when message involvement is low, but as central cues when involvement is high. In the latter case high involvement individuals canBas a consequence of their inability to carefully process the message argumentsBtry to make inferences on the basis of the simple cues with regard to the believability and usefulness of the message. In other words, the simple cues may serve as persuasive arguments or assist in the evaluation of the arguments presented (Petty and Cacioppo 1986, Cacioppo and Petty 1989).

Predictions based on the Elaboration Likelihood Model (ELM) have been confirmed in many studies. The majority of those studies concentrated on confirming central route processing of message arguments by high involvement respondents and peripheral route processing of simple cues by low involvement subjects (Batra and Ray 1985, Park and Young 1986, Maheswaran and Meyers-Levy 1990, Miniard et. al. 1991, Andrews et al. 1992, McKenzie and Spreng 1992, Park and Hastak 1995). Some general findings are that recall and recognition are usually better for high than for low involvement respondents (Petty and Cacioppo 1986, Park and Hastak 1995) and that the quality of arguments affects persuasion of high involvement individuals but not of low involvement subjects, while it is the other way round for peripheral cues. Expertise of the message source, status of the product endorser, favourable pictures, the mere number of arguments used, pleasant music and source likeability, for instance, have been found to facilitate persuasion of low involvement subjects but not of high involvement individuals (Petty et al. 1983, Petty and Cacioppo 1986, Miniard et al. 1991, Park and Hastak 1995). Wright (1974) investigated cognitive responses and observed significant more message comments (counterarguments) under conditions of high involvement than of low involvement but significantly more source comments (derogations) for low than for high involvement respondents. In a study by Miniard et al. (1991) high involvement subjects reported significantly more product thoughts than their low involvement counterparts while the reverse was true for the number of ad thoughts. Batra and Ray (1985) found that ad likeability is an important determinant of Ab and Pi for low involvement individuals while the Ab and Pi of their high involvement counterparts seems to be based more on the quality of the arguments shown and on the amount of pro- and counterargumentation. Maheswaran and Meyers-Levy (1990) also claim that message-based cognitions are a significant predictor of the attitude towards he brand for high involvement respondents while simple evaluative thoughts form a better predictor for the attitude towards the brand of low involvement individuals. Furthermore, brand attitudes formed by central route processing are better predictors of behavioural intentions and last longer than attitudes formed by peripheral route processing. In other words, the relation between Aad (ad likeability) and Ab and between Aad and Pi seems to be stronger for low than for high involvement respondents, but the relation between Ab and Pi and between Pi and actual buying behaviour is stronger for high than for low involvement individuals (Cacioppo and Petty 1986, MacKenzie and Spreng 1992, Lord et al. 1995).

The hypothesis concerning the possibility of central route processing of simple cues in conditions of high involvement has been less extensively studied. Support for this hypothesis has been found by Lord et al. (1995) who investigated two competing hypotheses, more specifically the peripheral-cue hypothesis and the combined-influence hypothesis. The peripheral-cue hypothesis claims that Aad is formed only by peripheral processes and that Aad influences Ab and Pi only when the elaboration likelihood is low. According to the combined-influence hypothesis Aad is a function of responses to both message arguments and peripheral cues and influences Ab and Pi in case of low as well as high elaboration likelihood. Their research results lend support for the combined-influence hypothesis. Petty and Cacioppo (1980) showed respondents shampoo ads varying in quality of arguments and physical attractiveness of the endorser. The quality of the arguments presented appeared to have a greater effect on the attitudes of high than of low involvement subjects. The physical appearance of the endorser had a similar impact in both high and low involvement conditions. Petty and Cacioppo conclude that the physical appearance of the endorsers (especially their hair) may have served as a cogent product-relevant argument demonstrating the product’s effectiveness. In a subsequent study Petty and Cacioppo (1983) used ads for edge disposable razors and varied the nature of the product endorser (a celebrity or a citizen). In this case the peripheral cue could not serve as a cogent product-relevant argument and the nature of the endorser appeared to have an impact on attitudes only for low involvement respondents. Miniard et al. (1991) examine the moderating role of the level of involvement on the impact of pictures on Ab and Pi. Their conclusion is that the effect of affect-laden pictures devoid of picture-relevant information declines with increasing level of involvement while the opposite is true for product-relevant pictures. The latter results seem to suggest that not all peripheral cues can serve as central cues but only those interpretable as cogent product information.

In both the Belgian and the Polish study described in this paper 12 print ads are tested, 3 warm, 3 erotic, 3 humorous and 3 non-emotional or neutral. All ads tested are characterized by a low information content. The neutral ads merely picture the product accompanied by a headline. Emotional ads can be defined as ads in which a brand name or a picture of a brand is shown, as well as a picture which induces emotional feelings, and possibly a headline or a short text (Kroeber-Riel, 1984). This definition applies to the emotional ads tested in this study (see also Geuens and De Pelsmacker, 1996). In view of the foregoing we expect that the affectively oriented ads have more impact on low than on high involvement subjects (the affect-laden pictures may serve as simple cues) while the neutral ads have more influence on high than on low involvement respondents (the picture of the product accompanied by a headline may serve as product information). In view of the foregoing the following hypotheses can be advanced:

*    H1a

Ad and brand recognition are higher for high than for low involvement respondents as a consequence of central versus peripheral ad processing (Park and Hastak, 1995).

*    H1b

For low involvement subjects we expect that emotional ads will lead to a higher ad and brand recognition than neutral ads because emotional ads are expected to attract more attention and to enhance more peripheral processing than neutral ads. Recognition for high involvement individuals, however, will not be influenced by the type of ad appeal used.

*    H2a

Regarding cognitive responses we expect them to be more pronounced for high than for low involvement respondents, except for cognitions about the ad execution which should be more pronounced for low than for high involvement respondents (Miniard et al. 1991).

*    H2b

Neutral ads, as compared to emotional ones, should result in more pronounced cognitive responses as a consequence of more information oriented pictures but only for high involvement subjects. For low involvement subjects cognitive responses should not differ for the different types of ad executions.

*    H3

Since low involvement individuals rely more on peripheral cues, one might expect affective responses to be stronger for low than for high involvement respondents and this should be especially true for emotional ads. High involvement subjects are not expected to experience emotional feelings with a varying degree of intensity for the different ad types.

*    H4

Aad, Ab and Pi are expected to be enhanced by the emotional cues but only for low involvement respondents. For high involvement individuals ad execution should not give rise to differences in Aad, Ab and Pi. This hypothesis is based on the research results of Miniard et. al. (1991) who tested ads showing favourable and unfavourable pictures and of Andrews et. al. (1992) who found similar conclusions for the use of distinctive versus non distinctive ads (colour ads between a majority of black and white ads).For emotional ads we expect low involvement subjects to report a more positive Aad, Ab and Pi than their high involvement counterparts while the reverse should be true for neutral ads.

*    H5

Predominantly affective responses determine Aad, Ab and Pi for low involvement respondents. Predominantly cognitive responses determine Aad, Ab and Pi for high involvement subjects (Lord et. al., 1995). If this hypothesis is rejected, the results lend support to the combined-influence hypothesis rather than the peripheral-cue hypothesis.

*    H6

In congruence with the ELM we expect the correlation between Aad and Ab and between Aad and Pi to be stronger for low than for high involvement respondents while the reverse is true with regard to the correlation between Ab and Pi (Batra and Ray 1985, Petty and Cacioppo 1986, MacKenzie and Spreng 1992, Lord et al. 1995).

Since nothing much is known about differences in responses between Belgian and Polish consumers, explicite hypotheses cannot be advanced, and the analysis as to differences between the two nationalities is rather exploratory.

METHODOLOGY

The research design in the two studies is almost identical (see De Pelsmacker and Geuens 1996a, 1996b). In the first qualitative stage of the study 43 and 48 alcohol ads (for the Belgian and Polish study respectively) were selected from recent issues of popular national magazines and were presented to a jury of 10 Belgian/Polish students. Each of them was asked to place each stimulus in one of 3 categories of warmth, eroticism and humour. On the basis of these rankings, 12 stimuli were selected: the three stimuli with the highest frequency in the most warm, c.q. erotic, c.q. humorous category, and three non-emotional stimuli, having the highest frequency in the least warm, erotic and humorous categories. A manipulation check proved that in both studies the ads were perceived as intended (De Pelsmacker and Geuens 1996a, 1996b). The ads in the humorous, erotic and warm categories are hereafter defined as "emotional", the ads in the neutral category as "non-emotional". Although we did not control for potentially biasing variables such as the budget spent and the media strategy, there is sufficient reason to assume that the four sets of three ads per stimulus category are comparable in a number of important ways. As mentioned before, all ads were taken from recent issues of the largest magazines in the concerning countries. All of them are ads for well-known , top 3 brands, and most of them are part of long-running multi-media campaigns. The actual quantitative study was carried out on a sample of 115 Belgian and 100 Polish undergraduate and graduate students at the University of Antwerp and Gdansk. The 12 stimuli were presented in a random order for each student, and a number of 7 category Likert scales and semantic differentials, measuring various communication effects, had to be completed:

Advertisement recognition. Students indicated whether or not they recognized the ad shown.

Brand reconition or correct brand attribution. Students were asked to attribute a brand name to an ad on which the brand name was covered.

Cognitive reactions. The measurement of cognitive reactions is based on suggestions by Deighton et al. (1989) and Coulson (1989). Students had to indicate their degree of agreement with 5 cognitive single item statements: the urge to formulate counterarguments, to think of reasons not to buy the brand, to pay more attention to the ad than to the product, to think more about the product than about the brand, and the extent to which the ad leads to a more positive impression of the brand.

Ad evoked feelings. A 16-item semantic differential scale was used (Brooker and Wheatley 1994) to measure the feelings evoked by the ads. In the Belgian sample factor analysis resulted in 5 basic dimensions: irritation, interest, insult, carefreeness and cheerfulness. The Polish study showed 6 basic dimensions: interest, irritation, cheerfulness, carefreeness, adventurousness and contemplativeness. In subsequent analysis we only use those single items that are most representative for each factor.

Attitude towards the ad. A 6-item Likert scale (Baker and Kennedy 1994) was used. In subsequent analyses the average score over all 6 items was used.

Attitude towards the brand. A 4-item Likert scale was used (Coulson 1989; Homer 1990). Again the average score over all 4 items will be used.

Purchase intention (PI). A 3-item Likert scale was used (Duncan et al. 1985; Homer and Yoon, 1992). In the analysis the average score will be used.

In both the Belgian and the Polish study, product category involvement has been measured by means of a 10-item scale developed by Zaichkowsky (1994). In both cases this scale proved to be very reliable (Cronbach alpha =.89 for the Belgian study, Cronbach alpha =.82 for the Polish study). On the basis of the summated score for the 10 items three groups of equal size were defined: high, moderate and low involvement individuals. Subsequently, the results of low involvement subjects are compared with those of their high involvement counterparts. One might suspect a possible confounding effect of Ab on Aad, apart from the effect of involvement as we measure it. On the other hand , the correlation between Aad and Ab is very low (see hereafter). Therefore we can conclude that this confounding effect will be limited.

RESULTS

With regard to hypothesis 1a, we expected ad and brand recognition to be higher for high than low involvement respondents and to be higher for emotional than neutral ads but the latter only in case of low involvement subjects. For the Belgian sample, there were no significant differences between high and low involvement respondents nor between emotional and neutral ads. For the Polish sample, only in case of humorous ads a significant difference in ad and brand recognition for low and high involvement respondents emerges but the direction is contrary to expectations. Furthermore, for low as well as for high involvement subjects emotional ads as compared to neutral ads lead to significant higher recognition levels. In other words, hypotheses 1a and 1b have to be rejected for both studies.

Hypothesis 2 assumes that cognitive responses such as "the urge to formulate counterarguments" are more pronounced for high than for low involvement respondents while "thinking of the ad execution raher than of the product" is more typical for low than for high involvement subjects. Analyses of variance [Since the main effects of execution type are reported in De Pelsmacker and Geuens 1996a and 1996b, they are not discussed in this paper.] for the Belgian sample (table 1) reveals a significant main effect of level of involvement for "devoting more attention to the ad than to the product" and "thinking more about the product than about the brand". None of the interaction effects between level of involvement and type of advertising stimulus are significant. The results of independent samples t-test shows that for all types of stimuli these 2 cognitive responses are more pronounced for low than for high involvement respondents which confirms hypothesis 2a.

TABLE 1

MAIN AND INTERACTION EFFECTS ON COGNITIVE RESPONSES (BELGIAN STUDY)

For the Polish sample a somewhat different pattern emerges (table 2). Analyses of variance shows a significant main effect of involvement for "the inclination to think of counterarguments" consisting of a higher proneness for high than for low involvement individuals and for "thinking about the product rather than about the brand" which receives significant higher agreement scores of low than of high involvement respondents. Since thinking of a product category in general usually involves less cognitive effort than thinking of the merits or pros and cons of a specific brand, the latter result might perhaps also be interpreted as a confirmation of hypothesis 2. Furthermore, emotional ads in general lead to a reinforcement of a positive impression of the brand significantly more for low than for high involvement individuals. This might indicate a greater carry-over effect of ad likeability for low than for high involvement subjects, again supporting hypothesis 2a.

We expected that in case of low involvement, cognitive responses would not differ for emotional and non-emotional ads. Surprisingly, low involvement Belgian subjects seem to think significantly more of reasons not to buy the brand when they are exposed to emotional ads as compared to non-emotional ones. For low involvement Polish respondents emotional ads as compared to non-emotional ads lead to less counterarguments, more attention to the ad than to the brand and a greater reinforcement of a positive impression of the brand. Emotional ads as compared to neutral ones lead high involvement Polish individuals to think more of the ad execution than of the product while the opposite holds for thinking of reasons not to buy the brand. For both Belgian and Polish high involvement subjects a positive impression of the brand is reinforced significantly more when they are exposed to emotional ads as compared to non-emotional ones. Therefore, hypothesis 2b has to be rejected.

Hypothesis 3 assumes stronger affective responses in low than in high involvement respondents. Furthermore, in case of low involvement stronger affective responses are expected to occur for emotional than for non-emotional ads while different ad appeals should not evoke different affective responses for high involvement individuals. This hypothesis could not be confirmed in the Belgian case: affective responses hardly differed as a consequence of the level of involvement. Furthermore, emotional ads as compared to non-emotional ads induce significantly more positive as well as negative feelings in both low and high involvement respondents (table 3).

Analysis of variance for the Polish respondents shows interaction effects between the level of involvement and ad execution for the feelings "irritated", "interested" and "cheerful" (table 4). A more detailed analysis reveals that low involvement individuals as compared to their high involvement counterparts are significantly more interested in and less irritated by emotional ads but less interested in and more irritated by neutral ads. Surprisingly, low involvement respondents as compared to high involvement individuals feel less carefree when they are exposed to emotional ads although only for warm ads the difference reaches significance. Comparing emotional and non-emotional ads (table 5) results in significantly less irritation and significantly higher levels of cheerfulness, interest and carefreeness in favour of the emotional ads and this for low as well as high involvement individuals. Contrary to te Belgian results, in the Polish study positive feelings seem to be enhanced while negative feelings seem to be reduced by emotional ads and this for both involvement groups. The only exception is the induction of an adventurous feeling which is more pronounced after viewing a non-emotional ad than after being exposed to an emotional ad but only for high involvement respondents. In summary, hypothesis three is totally rejected for the Belgian study. For the Polish sample the first part of hypothesis 3 concerning differences in affective responses for different levels of involvement is confirmed. However, with respect to different affective responses for emotional and non-emotional ads in low but not in high involvement respondents the hypothesis could not be confirmed.

According to hypothesis 4 emotional ads as compared to non-emotional ads should lead to a more positive Aad, Ab and Pi for low than for high involvement subjects while the opposite is expected for non-emotional stimuli. Furthermore, Aad, Ab and Pi should be more enhanced by emotional than by non-emotional appeals in case of low involvement while the type of ad execution should not give rise to differences in Aad, Ab and Pi in case of high involvement. Hypothesis 4 cannot be confirmed for the Belgian sample since no significant differences emerge for high and low involvement respondents. Furthermore, low as well as high involvement subjects hold a significantly more positive attitude towards emotional than towards non-emotional ads. In case of high involvement, the latter is also true for the attitude towards the brand (table 6).

The Polish results are more in congruence with the first part of hypothesis 4. Table 7 shows the ANOVA-results indicating a significant main effect of involvement for Ab and Pi and significant interaction effects between involvement and ad execution for Aad, Ab and Pi. A more detailed analysis shows that in general Aad, Ab and Pi are significantly higher for low than for high involvement individuals when exposed to emotional ads , while the reverse is true for the attitude towards neutral ads. As for the Belgian sample, low involvement respondents seem to appreciate emotional ads significantly more than neutral ads (table 8). Contrary to the Belgian sample and more in congruence with hypothesis 4, in case of high involvement only Aad differs significanlty for emotional and non-emotional ads. In summary, while the hypothesis 4 has to be rejected for the Belgian sample it seems to be confirmed for the Polish respondents.

TABLE 2

MAIN AND INTERACTION EFFECTS ON COGNITIVE RESPONSES (POLISH STUDY)

TABLE 3

AFFECTIVE RESPONSES TO EMOTIONAL AND NON-EMOTIONAL APPEALS (BELGIAN SAMPLE)

TABLE 4

MAIN AND INTERACTION EFFECTS ON AFFECTIVE RESPONSES (POLISH SAMPLE)

According to hypothesis 5 affective responses predominantly determine Aad, Ab and Pi in case of low involvement while cognitive responses are important determinators in high involvement situations. The Belgian results once again do not confirm this hypothesis. Stepwise regression analyses are carried out in which a functional relation is estimated between Aad on the one hand and affective and cognitive responses on the other. The same is done for Ab but with inclusion of Aad and for Pi with inclusion of Aad and Ab. Aad is predominantly explained by the feelings of interest and irritation and to a lesser extent by the enhancement of a positive impression of the brand and this for low as well as for high involvement respondents. Also for Ab affective responses are the most explanatory variables (Aad, a feeling of cheerfulness and to a lesser extent irritation and enhancement of a positive impression of the brand) in case of low as well as high involvement. The most explanatory variable for purchase intention is Ab.

TABLE 5

AFFECTIVE RESPONSES FOR EMOTIONAL AND NON-EMOTIONAL APPEALS (POLISH SAMPLE)

TABLE 6

AAD, AB AND PI FOR EMOTIONAL AND NON-EMOTIONAL APPEALS (BELGIAN SAMPLE)

TABLE 7

MAIN AND INTERACTION EFFECTS ON ATTITUDES AND INTENTIONS (POLISH SAMPLE)

In the Polish case, cognitive responses seem to be more important than for the Belgian sample. With respect to Aad and Ab cognitive responses are important contributors although also affective responses matter and this for both involvement categories. In general , affective as well as cognitive responses matter for low as well as for high involvement subjects and for Belgian and Polish respondents which provides more support for the combined influence hypothesis than for hypothesis 5 since the assumption that affective responses are more important in case of low involvement while cognitive responses matter more in case of highinvolvement could not be confirmed.

As to the last hypothesis, we expect stronger correlations between Aad and Ab and between Aad and Pi for low than for high involvement individuals and the reverse with regard to the correlation between Ab and Pi. The Belgian results largely confirm the part of the hypothesis concerning the Aad-Ab and Aad-Pi correlations. Regarding the correlations between Ab and Pi hypothesis 6 has to be rejected (table 9). The results for the Polish sample are shown in table 10. As can be seen they largely reveal the expected pattern. Once again, hypothesis 6 seems to be confirmed more for the Polish than for the Belgian sample in that the correlation between Aad on the one hand and Ab and PI on the other is greater in low involvement than in high involvement subjects.

TABLE 8

AAD, AB AND PI FOR EMOTIONAL AND NON-EMOTIONAL APPEALS (POLISH SAMPLE)

TABLE 9

CORRELATIONS BETWEEN AAD-AB, AAD-PI AND AB-PI (BELGIAN SAMPLE)

TABLE 10

CORRELATIONS BETWEEN AAD-AB, AAD-PI AND AB-PI (POLISH SAMPLE)

CONCLUSIONS

In summary, although not all hypotheses concerning the influence of the degree of involvement can be confirmed, a number of results are in the expected direction. Both in the Belgian and the Polish sample, the hypotheses on the cognitive responses and the carry-over effect of the attitude towards the ad on the attitude towards the brand and the purchase intention are confirmed. Some support is also found in both samples for the combined-influence hypothesis as opposed to the peripheral-cue hypothesis. Neither in the Polish sample nor in the Belgian one, product category involvement seems to affect ad and brand recognition, which is contrary to expectations. Additionally, in the Polish sample the effects of the degree of involvement on affective reactions and on Aad, Ab and Pi are also largely as hypothesized. The general conclusion is that the influence of product category involvement on communication effects is largely as expected, and even more so in the Polish environment. Furthermore, the impact of product category involvement on communication effects is highly similar for warm, erotic and humorous stimuli. The fact that, contrary to expectations, emotional ads generate more positive results such as higher ad and brand recognition and a more positive Aad, Ab and Pi for low as well as for high involvement individuals might perhaps be due to the nature of the ads studied. The neutral ads hardly contained any product arguments nor any peripheral cues. The peripheral cues available in the emotional ads might have induced central processing in high involvement respondents and peripheral processing in low involvement respondents to a greater extent than did the cues in the neutral ads. If the latter is true, this leads to more processing for the emotional than for the non-emotional ads and as a consequence to more positive communication effects. If the neutral ads had contained more product arguments the results might perhaps have been even more in congruence with the ELM-predictions. Evidently, further research using ads for different types of products and different target groups should be carried out to corroborate these results. Some of the hypotheses might be more validly tested using neutral ads with more product arguments. Finally, more cross-cultural validation of differences in consumers’ reactions to different types of advertising stimuli is certainly called for.

REFERENCES

Aaker, D.A.and D.E. Bruzzone (1985), "Causes of Irritation in Advertising", Journal of Marketing, 49 (Spring), 47-57.

Aaker, D.A., D.M. Stayman and M.R. Hagerty (1986), "Warmth in Advertising: Measurement, Impact and Sequence Effects", Journal of Consumer Research, 12 (4), 365-381.

Andrews, C.J., S.H. Akhter, S. Durvasula and D.D. Muehling (1992), "The Effects of Advertising Distinctiveness and Messager Content Involvement on Cognitive and Affective Responses to Advertising", Journal of Current Issues and Research in Advertising, 14, 1, 45-58.

Andrews, J.C., S. Lysonski and S. Durvasula (1991), "Understanding Cross-Cultural Student Perceptions of Advertising in GeneralBImplications for Advertising Educators and Practitioners", Journal of Advertising, 20, 2, 15-28.

Baker, S.M. and P.F. Kennedy (1994), "Death By Nostalgia: A Diagnosis of Context-Specific Cases", Advances in Consumer Research, 21, 169-174.

Batra, R and M.L. Ray (1985), "How Advertising Works at Contact", in Psychological Processes and Advertising Effects, eds. L.F. Alwitt and A.A. Mitchell, Lawrence Erlbaum Associates, Inc., Hillsdale, New Jersey, 13-43.

Batra, R. and M.L. Ray (1986), "Affective Responses Mediating Acceptance of Advertising", Journal of Consumer Research, 13 (September), 203-214.

Biswas, A., J.E. Olson and V. Carlet (1992), "A Comparison of Print Advertisements from the United States and France", Journal of Advertising, 21, 4, 73-81.

Brooker, G. and J.J. Wheatley (1994), "Music and Radio Advertising: Effects of Tempo and Placement", Advances in Consumer Research, 21, 286-290.

Cacioppo, J.T. and R.E. Petty (1989), "The Elaboration Likelihood Model: The Role of Affect and Affect-Laden Information Processing in Persuasion", in Cognitive and Affective Reactions to Advertising, eds. P. Cafferata and A. Tybout, Lexington Mass., Lexington Books, 69-89.

Cutler, B.D. and R.G. Javalgi (1992), "A Cross-Cultural Analysis of the Visual Components of Print AdvertisingBThe United States and the European Community", Journal of Advertising Research, 32, 1, 71-80.

Cutler, B.D., R.G. Javalgi and M.M. Erramilli (1992), "The Visual Components of Print Advertising: A Five-Country Cross-Cultural Analysis", European Journal of Marketing, 26, 4, 7-20.

Coulson, J.S. (1989), "An Investigation of Mood Commercials", in Cognitive and Affective Reactions to Advertising, Eds. Cafferata, P. and A. Tybout, Lexington Mass., Lexington Books, 21-29.

Deighton, J.D. and J. McQueen (1989), "Using Drama to Persuade", Journal of Consumer Research, 16, 335-343.

De Pelsmacker, P. and M. Geuens (1996a), "The Communication Effects of Warmth, Eroticism and Humour in Alcohol Advertisements" Journal of Marketing Communications, 2, 4, 247-262.

De Pelsmacker, P. and M. Geuens (1996b), "Different Markets, Different Communication Strategies? A Comparative Study of the Communication Effects of Different Types of Alcohol Ads in Belgium and Poland", University of Antwerp (RUCA), Workingpaper 3.

Dowling, G.R. (1980), "Information Content in U.S. and Australian Television Advertising", Journal of Marketing, 44, Fall, 34-37.

Duncan, C.P., J.E. Nelson and N.T. Frontczak (1984), "The Effects of Humour on Advertising Comprehension", Advances in Consumer Research, 11, 421-433.

Geuens, M. and P. De Pelsmacker (1996), "Feelings Evoked by Warmth, Eroticism and Humour in Alcohol Advertisements" in Proceedings of the 25th EMAC Conference B Marketing for an Expanded Europe, Eds. J. Beracs, A. Bauer and J. Simon, Budapest Unversity of Economic Sciences, Budapest, 471-489.

Homer, P.M. (1990), "The Mediating Role of Attitude Towards the Ad: Some Additional Evidence", Journal of Marketing Research, 27, February, 78-86.

Homer, P.M. and S.G. Yoon (1992), "Message Framing and the Interrelationships Among Ad-Based Feelings, Affect, and Cognition", Journal of Advertising, 21, 1, 19-33.

Kroeber-Riel, W. (1984), Emotional Product Differentiation by Classical Conditioning, Advances in Consumer Research, 11, 538-543.

Latour, M.S. and T.L. Henthorne (1994), "Female Nudity in Advertisements, Arousal and Response: A Parsimonious Extension", Psychological Reports, 75, 1683-1690.

Lord, K.R., M. S. Lee and P.L. Sauer (1995), "The Combined Influence Hypothesis: Central and Peripheral Antecedents of Attitude toward the Ad", Journal of Advertising, 24, 1, 73-85.

MacKenzie, S.B. and R.A. Spreng (1992), "How Does Motivation Moderate the Impact of Central and Peripheral Processing on Brand Attitudes and Intentions?", Journal of Consumer Research, 18, (March), 519-529.

Madden, C.S., M.J. Caballero and S. Matsukubo (1986), "Analysis of Information Content in U.S. and Japanese Magazine Advertising", Journal of Advertising, 15, 3, 38-45.

Maheswaran, D. and J. Meyers-Levy (1990), "The Influence of Message Framing and Issue Involvement", Journal of Marketing Research, 27, (August), 361-367.

Miniard, P.W., S. Bhatla, K.R. Lord, P.R. Dickson and H.R. Unnava (1991), "Picture-based Persuasion Processes and the Moderating Role of Involvement", Journal of Consumer Research, 18, (June), 92-107.

Mibner, P. (1994), "Aktuelle Daten aus Polen. Die Werbung nberzeugt noch nicht", Absatzwirtschaft, 11, 128-129.

Park, J.W. and M. Hastak (1995), "Effects of Involvement on On-Line Brand Evaluations: A Stronger Test of the ELM", Advances in Consumer Research, 22, 435-439.

Park, C.W. and S. Young (1986), "Consumer Response to Television Commercials: The Impact of Involvement and Background Music on Brand Attitude Formation", Journal of Marketing Research, 23, (February), 11-24.

Petty, R.E. and J.T. Cacioppo (1980), "Effects of Issue Involvement on Attitudes in an Advertising Context", in Proceedings of the Division 23 Program, eds. G.G. Gorn and M.E. Goldberg, Montreal, Canada: Amarican Psychological Association, 75-79.

Petty, R.E. and J.T. Cacioppo (1986), "The Elaboration Likelihood Model of Persuasion", Advances in Experimental Social Psychology, 19, 123-205.

Petty, R.E., J.T. Cacioppo and D. Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement", Journal of Consumer Research, 10, September, 135-146.

Rice, M.D. and Z. Lu (1988), "A Content Analysis of Chinese Magazine Advertisements", Journal of Advertising, 17, 4, 43-48.

Weinberger, M.G. and C.S. Gulas (1992), "The Impact of Humour in AdvertisingBA Review", Journal of Advertising, 21 (4), 35-59.

Wright, P.L. (1974), "The Cognitive Processes Mediating Acceptance of Advertising Responses", Public Opinion Quarterly, 38 (Summer), 192-205.

Zaichkowsky, J.L. (1985), "Measuring the Involvement Construct", Journal of Consumer Research, 12 (December), 341-352.

Zaichkowsky, J.L. (1994), "The Personal Involvement Inventory: Reduction, Revision and Application to Advertising", Journal of Advertising, 23, 4, December, 59-70.

----------------------------------------