Involvement and Advertisement Size Effects on Information Processing

Jin K. Han, Columbia University
ABSTRACT - This study examines the moderating role of involvement on advertisement-size effects to predict the extent of information processed. The results indicate that, in the moderate product involvement situation, larger ads are given greater attention and thus are better recalled than smaller ads, whereas, no such difference is observed in the low involvement situation. These findings are explained in the context of previous findings on: 1) advertisement size 2) involvement, 3) distraction and attention capacity, and 4)hierarchical information processing.
[ to cite ]:
Jin K. Han (1992) ,"Involvement and Advertisement Size Effects on Information Processing", in NA - Advances in Consumer Research Volume 19, eds. John F. Sherry, Jr. and Brian Sternthal, Provo, UT : Association for Consumer Research, Pages: 762-769.

Advances in Consumer Research Volume 19, 1992      Pages 762-769

INVOLVEMENT AND ADVERTISEMENT SIZE EFFECTS ON INFORMATION PROCESSING

Jin K. Han, Columbia University

[The author is grateful to Donald R. Lehmann and the two anonymous ACR reviewers for their helpful comments and suggestions.]

ABSTRACT -

This study examines the moderating role of involvement on advertisement-size effects to predict the extent of information processed. The results indicate that, in the moderate product involvement situation, larger ads are given greater attention and thus are better recalled than smaller ads, whereas, no such difference is observed in the low involvement situation. These findings are explained in the context of previous findings on: 1) advertisement size 2) involvement, 3) distraction and attention capacity, and 4)hierarchical information processing.

INTRODUCTION

Ever since the advent of print advertising, the size of ads has been a key component of interest to both managers and scholars (i.e., Copland 1958; Finn 1988; Holbrook and Lehmann 1980; Hendon 1973; Rossiter 1981; Strong and Adams 1912; Yamanaka 1962). Apparently, there is a positive relationship between the ad size and the attentional value of the ad itself, which has direct implications to a firm's budgetary and media scheduling decisions.

The size of an ad contributes to the ad's attentional value in two ways simultaneously. One is enhancing the probability of the ad being viewed due to its sheer size, while the other is reducing the number and/or the size of other surrounding materials (i.e., other advertisements, editorials) which may detract from the ad's attentional value (Sandage, Fryburger, and Rotzoll 1979). Despite the strong focus on the advertisement size's ability to attract attention, the the size literature appears to have overlooked the effect of involvement, which is an integral component in attention allocation decisions (Kahneman 1973; Norman 1976). The extent to which the size effect may be influenced by the advertisement's surrounding may be governed by the audience's involvement with the product in the ad (Finn 1988). There are a number of studies which suggest that involvement, in conjunction with executional factors (i.e., ad size), may affect the processing of the advertised information (Batra and Ray 1983; Cohen 1983; Muncy and Hunt 1984). For example, Muncy and Hunt (1984) posit that "factors such as the media in which the communication is present, the editorial content surrounding the communication,Y have all been related to involvement" and are very likely to affect the processing of the advertisement message. Although these studies suggest the presence of potential interaction effects between involvement and external factor(s) on the extent of the message processed, no directional effects have been proposed.

Hence, the objective of this study is to explore whether the involvement factors moderate the effects of ad size in the processing of the information presented in the advertisement. Since the premise of this study is to examine how various levels of involvement affect the ad size's contribution to the ad's attentional value, we are particularly interested in observing consumer responses to ads under conditions of freely allocated attention. The conjectures presented in this study are based on the previous findings of: 1) advertisement size, 2) involvement, 3) distraction and attention capacity, and 4) hierarchical information processing.

The paper is organized in the following manner. The first section enumerates the general findings of ad size which relate increased processing (attention and comprehension) with an increase in ad size. The second section depicts the distinction between the high and the low levels of involvement. The following section integrates the findings from ad size studies and from involvement studies with the distraction/attention capacity theories in the context of hierarchical processing framework. The conceptual framework is followed by a description of an experiment that investigates whether, 1) in a low involvement condition, there is any difference between a large ad and a small ad on attention and comprehension; 2) in a moderate involvement condition, attention and comprehension are greater for a large ad compared to a small ad.

THE EFFECT OF SIZE OF A PRINT ADVERTISEMENT

The importance of print advertisement's attention value have long been acknowledged, and a number of studies have focused on the ad characteristics' ability to enhance the attentional value of print advertisements (i.e., Diamond 1968; Finn 1988; Hanssens and Weitz 1980; Hendon 1973; Holbrook and Lehmann 1980; Sparkman 1985; Strong and Adams 1912-1920; Twedt 1952; Yamanaka 1962). For example, ad characteristics have been found to account for more than 45% variance in print ads' effectiveness measures (i.e., recall, read most score), and among ad characteristics, ad size was shown to be strongly correlated (.78) with the effectiveness measures (Hanssens and Weitz 1980).

Hence, ad size, being an influential ad characteristic and also a significant cost factor, has been examined from various perspectives. Some research used experiments (i.e., Strong and Adams 1912-1920), whereas others relied on readership scores (i.e., Diamond 1968; Finn 1988; Holbrook and Lehmann 1980; Troldahl and Jones 1965) to explore the relationship between size and attention and/or comprehension. Both methods reveal that as the advertisement size increases, so does the ad's effectiveness. In fact, Hendon (1973) reports that Strong and Adams (1912-1920) and Copland (1958) found that the permanency of an advertisement impression varied directly with the square root of the increase in ad area. The positive effects of ad size is also corroborated by the findings from readerships studies. In Finn (1988)'s meta-analysis of readership studies, there is an overwhelming evidence for ad size's positive effect on cognitive activity: 14 findings on attention, 2 on comprehension, and 6 on elaboration. Additionally, Troldahl and Jones (1965) found that ad size alone explained over 40% of the variability in readership, and that the correlation between ad size and ad readership were .64 among men and .73 among women. There, however, appears to be one exception: Ferguson (1935) found that there was no relationship between the size of an ad and the ad's attentional value.

With the exception of Ferguson (1935)'s null effect, there seems to be a relatively strong consensus on the positive relationship between ad size and the attentional value of an ad. The effect of ad size on behavior, though significant, is not as strong as the effect on the attentional aspects: ad size accounted for 19 to 36% of the variance in inquiry generation prompted by advertisements (Hanssens and Weitz 1980). They explain this phenomenon with a hierarchy of effects model, which states that "the communication variables typically have a greater effect on lower order responses (awareness) than higher order responses (behavior)." In any case, it is evident that ad size has considerable impact on both awareness and behavior.

Despite the tremendous focus on the attentional aspects of advertisement sizes, however, past research on ad size has overlooked the role of involvement, which is an integral element determining the allocation of attention. The size studies have yet to account for the effects of involvement in the context of ad size effects. This concern is expressed by Finn (1988):

Y A study of the relationship between level of involvement, type of processing, recall, and recognition responses to print ads is needed to resolve the validity issue (of ad characteristics)

The consideration of involvement factors is critical when examining the effects of ad size, because the size of an ad indirectly affects the composition of the advertisement's environment. Muncy and Hunt (1984) report that a stimulus' environmental factors (i.e., media, editorials) have been found to be related to involvement and also to affect the processing. For example, Mitchell (1981) illustrates a scenario in which a consumer in a market for a new automobile devotes disproportionately more attention to the advertisement's background rather than to the information of the car, which results in the inferior encoding of the latter. Hence, there are some initial indications that involvement and ad size may jointly affect the attentional value of a print advertisement.

One theory that provides initial insight into how size effects might operate under different levels of involvement is the Elaboration Likelihood Model (ELM). According to the ELM, peripheral cues (i.e., simple acceptance/rejection cues) have a greater effect on persuasion under low involvement conditions, whereas, the quality of message (via central route) has a greater impact under high involvement conditions (Petty and Cacioppo 1979; Petty, Cacioppo, and Schumann 1983). Hence, the ELM would predict that ad size, as a peripheral cue, would likely to be more effective in low involvement conditions than in high involvement conditions. The ELM, however, considers the involvement continuum as being either high or low; therefore, a further investigation of the size effects in the context of a wider range of the involvement continuum is warranted.

INVOLVEMENT

The importance of acknowledging the differences between high and low levels of involvement is evident in many advertising studies. Kassarjian (1981) points out that:

Research on the hierarchy of effects in marketing communications is an excellent example of the fact that cognitive activity for low involvement and high involvement are simply different and that one cannot generalize research results from one situation to the other.

Hence, the underlying cognitive processes for the high and the low levels of involvement appear to be the key explanatory factor for the involvement-level specific responses. Specifically, the substantial support for this notion comes from the studies focusing on the attentional and processing aspects of involvement (i.e., Batra and Ray 1983; Celsi and Olson 1988; Cohen 1983; Greenwald and Leavitt 1984; Harris 1987; Mitchell 1981). The consensus of the findings is that involvement mediates the intensity of attentional effort devoted to a stimulus: low levels of involvement elicit minimal or negligible attentional effort - as the involvement level heightens, the attentional level follows suit. The reason for the frugal rationing of attentional effort at low involvement levels has been attributed to non-arousal (Mitchell 1979; Woodside 1983), lack of motivation (Burnkrant and Sawyer 1983), and mental laziness (Harris 1987). In any case, it is evident that relatively little (substantial) cognitive activity takes place under low (high) involvement conditions. This contention is believed to be observable not only in attentional efforts but also in the comprehension measures as well (Cohen 1983; Greenwald and Leavitt 1984; Mitchell 1981), which is consistent with the hierarchical information processing framework.

Hence, at very low levels of involvement (preattentive level), we would expect no significant effects of ad size since the viewers will not even put in the effort to process the advertisement. According to the ELM, however, ad size, as a peripheral cue, should be most effective at low levels of involvement. The conflicting predictions can be resolved by considering what the ELM means by "low" levels of involvement. The central premise of our study is to observe how consumers behave under natural newspaper viewing settings without specifically being instructed to evaluate a particular ad in the newspaper.

FIGURE A

HIERARCHICAL ADVERTISING PROCESSING MODEL

Consequently, in our framework, the subjects who do put in the effort to process the target ad will considered as at least being moderately involved, whereas, the ELM studies specifically instruct all the subjects to evaluate the target stimulus, consequently making the subjects at least moderately involved. Hence, the ELM's "low" level of involvement is not equivalent to our definition of low (which is preattentive); rather, the ELM's "low" is closer to our definition of moderate. Henceforth, the study will refer to low as preattentive.

Unlike the cognitive processes at low levels of involvement, the processing at moderate levels of involvement may encounter the constraints of attention capacity. Attention capacity is analogous to the related concept of STM (short-term memory) store - once a stimulus occupies part of the attention-capacity space, the availability of "free" attention-capacity space necessarily decreases for another stimulus, which results in reduced cognitive processing for the latter stimulus (Kahneman 1973; Norman 1976). Greenwald and Leavitt (1984) relate this phenomenon to high involvement: "Because high levels of involvement are demanding of a limited resource (attention capacity), involvement in one message is necessarily limited when capacity is allocated to some other message." This notion is also in line with distraction effects (i.e., Batra and Ray 1983; Cohen 1983; Osterhouse and Brock 1970).

The key implication is that attention capacity limitation may operate in different directions for the target ad at moderate levels of involvement. Specifically, when categorically considering the higher levels of involvement continuum as moderate and high, attention capacity limitation may work against or in favor of the target ad, respectfully. Under moderate involvement conditions, the primary attention may be devoted to the target ad's environmental materials rather than to the ad itself (Mitchell 1981; Muncy and Hunt 1984), thereby potentially being subjected to attention capacity limitation (Norman 1976). Consequently, the size effects should be apparent at moderate levels of involvement. Using our definition of moderate, this conjecture is also in line with the ELM.

HYPOTHESES

H1: At a low (preattentive) involvement level, there will be no significant difference in attentional effort and comprehension measures between a larger ad and a smaller ad.

H2: At a moderate involvement level, there will be a significant difference in attentional effort and comprehension measures between a larger ad and a smaller ad.

In sum, we hypothesize an interaction effect between ad size and involvement for both attentional effort and comprehension measures.

METHOD

Research Design

A 2x2 factorial design (involvement: moderate, low x ad size: larger or smaller) was employed. The sample consists of 60 undergraduates of a large Eastern university. We differentiated the involvement levels among the subjects by choosing the ad message as an announcement for a meeting concerning an increase in the graduations requirement effective with the sophomore class. This method of involvement manipulation has been shown to differentiate involvement successfully in past studies (i.e., Apsler and Sears 1968). Furthermore, Rossiter (1981) posits that "personal reference may stimulate reader involvement with the headline," therefore, the headline in the announcement explicitly contained the words "Class of '92 and '93" to ensure the proper manipulation of involvement.

An equal number of moderate involvement subjects (30 underclassmen) and low involvement subjects (30 upperclassmen) were randomly selected. The sample was selected by making room-to-room visits in the undergraduate dormitories (underclassmen dormitory for moderate involvement group and upperclassmen dormitory for the low involvement group). Since all of the moderate involvement subjects were selected from a freshman dormitory, there were no sophomores in the moderate involvement group. Similarly, since all of the low involvement subjects were selected from a senior dormitory, there were no juniors in the low involvement group.

VARIABLES

Ad Size. There are two ad sizes (larger and smaller) in this experiment which varies the amount of distraction salient for the target ad. The larger ad is a full-page ad, and the smaller ad is a quarter-page ad with editorials around it on the same page. (See experimental procedure for more details.)

Involvement. Two levels of involvement (moderate and low) are used in this study. Moderate (Low) involvement is defined by the information in the ad that is personally (ir)relevant to the viewer. As previously mentioned, we implemented Apsler and Sears (1968)'s method to manipulate involvement. The ad contains information about a new graduation requirement effective with the freshmen and the sophomore class. Therefore, the underclassmen are in the high involvement group, whereas the upperclassmen are in the low involvement group.

Dependent Measures. Two measures are used to measure the subjects' responses. One is the amount of time the subject spends when viewing the ad. This measure acts as a proxy for attentional effort (Celsi and Olson 1988). The other measure is the number of items correctly recalled from the information presented in the ad. The ad contains eight key information (i.e., time and place for the meeting, speaker, language requirement). Hence, recall serves as a measure for the level of comprehension (i.e., the extent of the information processed) (Celsi and Olson 1988; Petty et al. 1983).

EXPERIMENTAL PROCEDURE

A mock school newspaper was prepared for the experiment. The newspaper consists of eight pages: pages one through eight contain all text (i.e., articles and editorials) and no advertisements other than the target ad. The target advertisement was placed after the fourth page. The mock newspaper was varied across treatments as follows: one version contains a full-page size of the target advertisement, and the other version contains a quarter-page size of the advertisement with several current-affair articles filling up the rest of the page as the attention-competing materials. The advertisement was an announcement for a forum for discussing an increase in graduation requirement. The advertisement states that the new policy will go into effect starting with Class of '92. The choice of Class of '92 differentiates the involvement level of the ad: underclassmen are in the moderate involvement group, and upperclassmen are in the low involvement group. During each room visit, the subjects were asked to view the mock newspaper as a prototype of a new school newspaper. The subjects were instructed to view the paper and give an evaluation of the newspaper. The purpose of this instruction was to prevent the subjects from guessing the research objective. While the subject was viewing the newspaper, s/he was timed for the viewing of the advertisement. The time measure provides a measure for the subject's attentional effort (Celsi and Olson 1988). A stopwatch, which measures the attentional effort, was kept from the subject's view (i.e., in a pocket). After the viewing of the paper was over, the subject was asked to fill out an evaluation form. One part of the evaluation form was a recall test concerning the information in the advertisement. The recall test serves as a measure for the amount of advertisement information processed (Celsi and Olson 1988; Petty et al. 1983). When this procedure was done, the subject was given a debriefing and thanked for the cooperation.

RESULTS

Manipulation Check

Due to the design of the study, a successful manipulation of the involvement levels is inherent in achieving the research objective. We used the subject matter of the announcement to vary the involvement. In order to check the extent of the involvement manipulation, a manipulation check question was included in the questionnaire. The question asked the subjects to rate their interest in the announcement on a 5-point scale (1=low,5=high). The t-test result (t=1.99; df=39; p=0.05) shows that the mean interest of the moderate involvement group (2.80) was significantly greater than that of the low involvement group (1.86).

Hypothesis Testing

The hypotheses predicted that there will be no significant difference in attentional effort between a larger ad and a smaller ad under low involvement. On the other hand, if the involvement level is in the moderate range, we expect the attentional effort to be greater for a larger ad than a smaller ad. The average results are given in the Table. ANOVA indicates that the interaction effect between ad size and involvement is not fully significant [F=2.67; df(1,56); p=0.10]. Further analysis of contrast effects, however, show that the findings are directionally in line with the hypotheses for low and moderate involvement conditions, respectfully: no difference was observed in the low involvement condition in attention effort between a larger ad and a smaller ad (t=0.17; df=28; p>0.05), whereas the larger ad received more attention than the smaller ad in the moderate involvement condition (t=2.22; df=28; p<0.05). See Figure B.

Similarly, the hypotheses predicted that there will be no significant difference in recall between a larger ad and a smaller ad under low involvement. If the involvement level is in the moderate range, we expect the recall to be greater for a larger ad than a smaller ad. ANOVA indicates that the interaction effect between ad size and involvement is indeed significant [F=4.30; df(1,56); p=0.05], hence supporting the hypotheses for low and moderate involvement conditions. Further analysis of contrast effects show additional support the hypotheses for low and moderate involvement conditions, respectively: no difference was observed in the low involvement condition in recall between a larger ad and a smaller ad (t=0.17; df=28; p>0.05), whereas the larger ad was recalled better than the smaller ad in the moderate involvement condition (t=2.48; df=28; p<0.05). See Figure C.

TABLE

THE EFFECTS OF SIZE AND INVOLVEMENT ON EFFORT AND RECALL

DISCUSSION

The main objective of this study was to observe consumer responses to a target advertisement under a freely-attention-allocating, casually-newspaper-viewing setting, with ad size and involvement as the variables of interest. Despite the ad size studies' preoccupation with print ads' attentional value, strangely enough, most size studies have neglected to explicitly consider involvement factors, which play a critical role in the allocation of attention. Hence, this study set out to answer such questions as: "Do size effects exist at all levels of involvement? If not, how do involvement effects mediate the effects of advertisement size? The findings of this study provide some preliminary answers to these questions.

At the lower end of the involvement spectrum, we expected to see no significant difference in the amount of attention given, consequently in the number of items recalled between a full-page ad and a quarter-page ad. The rationale for this conjecture is that there is well documented evidence of consumers choosing not to expend any attentional effort on materials personally irrelevant to them. Therefore, an ad, large or small, will have a very small probability of being processed at very low levels of involvement. This seems to be the case with the seniors in the sample. Since the ad message is intended for underclassmen, the seniors appear to pay hardly any attention to the message. The seniors' lack of interest is evident in their indications of their interest in the message (i.e., 1.86 where 1=low, 5=high). Further evidence of low involvement comes from the response measures. In terms of the time spent viewing the ad, the seniors spent on average a little over one second (1.32 seconds for the full-page ad, 1.24 seconds for the quarter-page ad). Consequently they recalled negligible amount of information (0.53 item for the full-page ad, 0.60 item for the quarter-page ad).

At moderate levels of involvement, consumers are believed to somewhat actively process information due to personal interest, relevance, etc. Hence, we hypothesized that, at moderate levels of involvement, there would enough difference in the intensity of processing from low levels of involvement such that the effect of ad size will be apparent. The premise for this conjecture is that, although it is likely that there is active processing of information, there is a fairly good chance that the primary attention may allocated to some other material (i.e., other ads, editorials) in the vicinity. Therefore, the processing of the target ad may be subjected to distractions and/or attention capacity limitation. Thus, we expect the size of an ad to reduce the presence and/or the number of potentially distracting materials, thereby, enhancing the attentional value of the target ad.

For the freshmen in the sample, their involvement seems to be in the moderate range. Their interest level indication, significantly greater that of the seniors, is at the 2.80 mark (1= low, 5= high). The means for the attentional effort and the recall measures seem to further support the moderate involvement position for the freshmen. The attentional effort for the full-page ad (2.34 seconds) was significantly greater than that for the quarter-page ad (1.02 seconds)(t=2.22; df=28; p<0.05). Similarly, the recall for the full-page ad (1.87 items) was significantly greater than that for the quarter-page ad (0.67 item). Additionally, the amount of time spent and the number of items recalled seem to further suggest that the freshmen's involvement was moderate.

FIGURE B

INTERACTION EFFECTS OF AD SIZE AND INVOLVEMENT ON EFFORT

FIGURE C

INTERACTION EFFECTS OF AD SIZE AND INVOLVEMENT ON RECALL

The contrast analyses at the low and the moderate levels of involvement suggest an interaction between involvement and ad size, which is consistent with our hypothesis. ANOVA confirms the contrast analysis for the recall measure [F=4.30; df(1,56); p=0.05] but not for the attentional effort [F=2.67; df(1,56); p=0.10]. The lack of interaction in the attentional effort may perhaps be attributed to fact that there was more room for measurement error in timing the viewing of the ad in comparison to counting the number of correctly recalled information.

In sum, this study was an exploratory look at how ad size effects might operate at two levels of involvement: low and moderate. The conjectures derived from past involvement and ad size studies suggest an interaction effect, where size effect is not apparent at low levels but present at moderate levels of involvement. At these levels of the involvement continuum, a significant interaction between ad size and involvement was found for recall but not for attentional effort (however, there is directional support for attentional effort). Thus, although we do not have a direct support, the results lend some support for the distraction/attention capacity limitation explanation at the moderate involvement level.

Hence, one implication of this study is as follows. Given a budget for an ad campaign, one needs to decide on the involvement level of the target audience first. For a low involvement audience, an advertiser should perhaps spend more for frequency rather than for the size of the ad. For moderate involvement audiences, however, size should be a key consideration in addition to the frequency aspect.

LIMITATIONS AND FUTURE RESEARCH

The premise of this study was examining how involvement and ad sizes influence the processing of advertisements. As dependent measures, attentional effort and correct number of recalls were used to determine an ad's effectiveness. Other measures of effectiveness, especially the evaluation of the ad, warrant further research. The advertisement used in this study contains no pictorial materials, therefore, the findings cannot be generalized to all advertisements. Further, the mock newspaper did not carry other advertisements besides the target one. Future studies could incorporate how such factors affect the processing of the target ad. Additionally, other significant advertisement components (i.e., color, bleed, location) need further looking into.

Most importantly, the biggest limitation of this study is that the data does not cover the whole range of the involvement spectrum (i.e., extremely high levels of involvement). The ELM and other studies from distraction/attention capacity area suggest that the effects of ad size may not be apparent at extremely high levels of involvement. Hence, there might be a curvilinear relationship: no effects of ad size at extremely low and high levels but present at moderate levels. Hence, future studies should gather data to cover the whole range of involvement.

Lastly, the findings should be viewed in the light that the results may be study specific. We cannot discount the possibility that the results may have been due to a confound of category involvement.

REFERENCES

Batra, Rajeev and Michael Ray (1983), "Operationalizing Involvement as Depth and Quality of Cognitive Response," Advances in Consumer Research X, Richard P. Bagozzi and Alice M. Tybout (eds.), Ann Arbor, MI: Association for Consumer Research, 309-313.

Burnkrant, Robert E. and Alan G. Sawyer (1983), "Effects of Involvement and Message Content of Information-Processing Intensity," in Information Processing Research inAdvertising, ed. Richard J. Harris, Hillsdale, NJ: Lawrence Erlbaum Assoc., 46-64.

Celsi, Richard L. and Jerry C. Olson (1988), "The Role of Involvement in Attention andComprehension Processes," Journal of Consumer Research, 15 (Sept.), 210-224.

Cohen, Joel B. (1983), "Involvement and You: 1000 Great Ideas," Advances in Consumer Research X, Richard P. Bagozzi and Alice M.Tybout (eds.), Ann Arbor, MI: Association for Consumer Research, 325-328.

Copland, Brian D. (1958), The Study of Attention Value: A Review of Some Available Material, London: Business Publications Limited.

Diamond, Daniel S. (1968), "A Quantitative Approach to Magazine Advertisement Format Selection," Journal of Marketing Research, 5 (November), 376-386.

Ferguson, Leonard W. "The Importance of the Mechanical Features of an Advertisement," Journal of Applied Psychology, Vol. 19, No. 5, 521-526.

Finn, Adam (1988), "Print Ad Recognition Readership Scores: An Information Processing Perspective," Journal of Marketing Research, 25 (May), 168-177.

Greenwald, Anthony A. and Clark Leavitt (1984), "Audience Involvement in Advertising: Four Levels," Journal of Consumer Research, 11 (June), 581-592.

Hanssens, Dominique M. and Barton A. Weitz (1980), "The Effectiveness of Industrial Print Advertisements Across Product Categories," Journal of Marketing Research, 17 (August), 294-306.

Harris, Greg (1987), "The Implications of Low-Involvement Theory for Advertising Effectiveness," International Journal of Advertising, 6, 207-221.

Hendon, Donald W. (1973), "How Mechanical Factors Affect Ad Perception," Journal of Advertising Research, 13 (August), 39-45.

Holbrook, Morris B. and Donald R. Lehmann (1980), "Form Versus Content in Predicting Starch Scores," Journal of Advertising Research, 20 (August), 53-62.

Houston, Michael J. and Mitchael L. Rothchild (1978), "Conceptual and Methodological Perspectives in Involvement," in Research Frontiers in Marketing: Dialogues and Directions, ed. S. Jain, Chicago, IL: American Marketing Association, 184-187.

Kahneman, Daniel (1973), Attention and Effort, Englewood Cliffs, N. J.: Prentice-Hall.

Kassarjian, Harold H. (1981), "Low Involvement: A Second Look," in Advances in Consumer Research, Vol. 8, Kent Monroe, ed., Ann Arbor, MI: Association for Consumer Research, 31-34.

Krugman, Herbert E. (1965), "The Impact of Television Advertising: Impact without Involvement," Public Opinion Quarterly, 29 (Fall), 349-356.

Mitchell, Andrew A. (1979), "Involvement: A Potentially Important Mediator of Consumer Behavior," in Advances in Consumer Research, Vol 6, William L. Wilke, ed., Ann Arbor, MI: Association for Consumer Research, 191-195.

Mitchell, Andrew A. (1981), "The Dimensions of Advertising Involvement," in Advances in Consumer Research 8, K. Monroe, ed., Ann Arbor, MI: Association for Consumer Research, 25-30.

Muncy, James A. and Shelby D. Hunt (1984), "Consumer Involvement: Definitional Issues and Research Directions," in Advances in Consumer Research, Vol. 11, Thomas C. Kinear, ed., Ann Arbor, MI: Association for Consumer Research, 193-196.

Norman, Donald A. (1976), Memory and Attention, 2nd ed., New York: John Wiley.

Osterhouse, Robert A. and Timothy C. Brock (1970), "Distraction Increases Yielding to Propaganda by Inhibiting Counterarguing," Journal of Personality and Social Psychology, 15, 344-358.

Petty, Richard E. and John T. Cacioppo (1979), "Issue Involvement Can Increase or Decrease Persuasion by Enhancing Message-Relevant Cognitive Responses," Journal of Personality and Social Psychology, 37 (October), 1915-1926.

Petty, Richard E., John T. Cacioppo, and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-144.

Rossiter, John R. (1982), "Predicting Starch Score," Journal of Advertising Research, Vol. 21, 5, 63-68.

Rothchild, Michael L. (1984), "Perspectives on Involvement: Current Problems and Future Directions," in Advances in Consumer Research, Vol. 11, Thomas C. Kinnear, ed., Provo, UT: Association for Consumer Research, 216-217.

Sandage, C. H., Vernon Fryburger, and Kim Rotzoll (1979), Advertising Theory and Practice, Homewood, IL: Irwin-Dorsey Limited.

Sparkman, Richard M., Jr. (1985), "Cost Effectiveness of Advertising," International Journal of Advertising, 4 (2), 131-141.

Starch, Daniel (1966), Measuring Advertising Readership and Results, New York: McGraw-Hill Book Company.

Troldahl, Verling C. and Robert L. Jones (1965), "Predictors of Newspaper Advertisement Readership," Journal of Advertising Research, 5 (March), 23-23.

Twedt, Dik Warren (1952), "A Multiple Factor Analysis of Advertising Readership," Journal of Applied Psychology, 36 (June), 207-215.

Woodside, Arch G. (1983), "Message-Evoked Thoughts: Consumer Thought Processes as a Tool for Making Better Copy," in Advertising and Consumer Psychology, Larry Percy and Arch G. Woodside, eds., Lexington, MA: D.C. Heath and Co.: 133-147.

Yamanaka, Jiro (1962), "The Prediction of Ad Readership Scores," Journal of Advertising Research, 2 (March), 18-23.

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