Low-Involvement Isn't Low-Involving

David W. Finn, Texas Christian University
ABSTRACT - In a discussion paper written for the 1978 ACR Conference, Michael Ray pointed out that in consumer research: We borrow concepts and techniques from several sciences. Early in the process we have unrealistic expectations for a new concept. Then, because these expectations cannot possibly be met, there is a period of disillusionment. That period could lead to the concept being dropped or being refined (1979, p. 197). He went on to say that, with regard to the involvement concept, we appear to be heading toward a period of refinement rather than rejection. This paper suggests that the best route to refinement may require rejection of certain approaches to the concept.
[ to cite ]:
David W. Finn (1983) ,"Low-Involvement Isn't Low-Involving", in NA - Advances in Consumer Research Volume 10, eds. Richard P. Bagozzi and Alice M. Tybout, Ann Abor, MI : Association for Consumer Research, Pages: 419-424.

Advances in Consumer Research Volume 10, 1983      Pages 419-424

LOW-INVOLVEMENT ISN'T LOW-INVOLVING

David W. Finn, Texas Christian University

ABSTRACT -

In a discussion paper written for the 1978 ACR Conference, Michael Ray pointed out that in consumer research:

We borrow concepts and techniques from several sciences. Early in the process we have unrealistic expectations for a new concept. Then, because these expectations cannot possibly be met, there is a period of disillusionment. That period could lead to the concept being dropped or being refined (1979, p. 197).

He went on to say that, with regard to the involvement concept, we appear to be heading toward a period of refinement rather than rejection. This paper suggests that the best route to refinement may require rejection of certain approaches to the concept.

INTRODUCTION

Two recent papers (Finn 1982 ; Smith and Swinyard 1982) have cast doubt on the validity of a separate low-involvement hierarchy of behavior (Ray 1973). These papers take the view that the behavior-before-attitude sequence is an information acquisition process wherein product trial is used as a source of information in forming strong brand beliefs. Strong beliefs then result in confident brand attitudes that influence further purchase behavior. Finn declared that trial-as-search was not confined to low-involvement situations and argued that even "high-involvement" might commonly lead to the behavior-before-attitude flow. In his Single Hierarchy model (Figure 1), trial is seen as just one possible source of information for consumers to use in evaluating alternatives.

FIGURE 1

THE SINGLE HIERARCHY MODEL

Involvement explanations of the behavior/affect sequence may be on the way toward rejection, but that is only one of the many conceptualizations of involvement found in the literature. A careful look at other views of the concept may identify other approaches that are less than valid. However, even if all of the many other approaches are individually acceptable as avenues for future refinement of the concept, the profusion of different definitions of "involvement" promises to prolong today's state of confusion in the area, and some valid approaches might best reject the label "involvement" in favor of more unique aliases- This paper is preliminary exploration of many conceptualizations of involvement that relate to the Cognition stage of the Single Hierarchy and recommends whether to reject or refine each approach.

THE GENERAL PARADIGM

Involvement, as it relates to the early stages of the hierarchy, has been variously studied as: an independent variable; a dependent variable; a person specific variable; a product specific variable; a situation specific variable; a type of cognitive processing; and a behavioral variable. Figure 2 illustrates the various involvement views in relation to the first stage of the Single Hierarchy (Cognition). Notice that this discussion (and Figure 2) does not emphasize decision making. Rather, it emphasizes initial awareness, knowledge, and learning that has been described as a first step in behavior hierarchies (Ray 1973; Lavidge and Steiner 1961.) Questions about this "Cognition" or "Awareness' stage center on the issues of how consumers acquire and process information about new brands or new claims and on the impact of this information acquisition on later decision-making, but not on the decision making process itself.

FIGURE 2

A GENERAL RESEARCH PARADIGM

Link 1 of Figure 2 positions research that studies differences in people, situations, or products (PSP's) and their impact on information acquisition styles. The PSP's are the independent variables, and the expectations are that low-involvement PSP's result in passive processing while high-involvement PSP's lead to active processing.

The approach that treats active/passive processing as the independent variable expects that low-involvement learning (passive processing) results in the behavior-before-affect hierarchy and that active processing leads to the affect behavior sequence. This view is pictured as link 2 in Figure 2.

When PSP's are studied as independent variables influencing the affect/behavior sequence, as in link 3 of Figure 2, we expect that low-involvement PSP's lead to behavior-before-affect and that high involvement PSP's result in the learning hierarchy (Raw 1973.)

The mere fact that the term "involvement" is used in so many different ways is reason enough to expect that the concept should cause some confusion and disillusionment; but if our expectations about the relations among the various definitions hold, the confusion should be minimal, and a mere re-labeling would serve to reduce it. Therefore, an exploration into the reasonableness of our expectations is in order.

If the concept of "involvement" is to be useful in consumer behavior research, two things are required. First, there must actually be consistent and definable differences in involvement as a phenomenon of interest. That is, there must be such things as high- and low-involvement PSP's, cognitive processing styles, and/or behavioral outcomes. Second, we should expect to find consistently different outcomes of high- and low-involvement levels of PSP's and cognitive processing styles. For example, in links 1 and 3 of Figure 2, high-involvement PSP's should be associated with high-involvement cognitive processing and with the Learning Hierarchy of behavior (Ray 1973.) Low-involvement PSP's should be associated with low-involvement outcomes. Similarly, in link 2 of Figure 2, low-involvement processing should result in behavior without affect, and high-involvement processing should result in the affect then behavior sequence.

The usefulness of the term "involvement" for link 3-type research has been seriously questioned in the recent past. Smith and Swinyard ( 1982) and Finn (1982) have argued that in many cases, the behavior-->affect sequence may simply be an instance in which the consumer is actively collecting decision related information by trying a product. Consumers in this active information collection mode would be described as high-involved consumers by Wright (1973) and by Mitchell (1979), yet they would be displaying the low-involvement hierarchy. For example, Lutz and Reilly forced subjects into a hypothetical situation in which they needed information for an imminent decision about brand choice from a set of unknown brands (1974) -- a high involvement situation. When asked to estimate their information collection behavior, those subjects choosing among brands of low and moderate performance risk products overwhelmingly chose to just buy a brand -- a low-involvement behavior hierarchy. Subjects choosing among brands of high performance risk products intended to collect information from non-trial sources, displaying the learning hierarchy of behavior. The inherent performance risk in a product category was a better predictor of the behavior/affect sequence than involvement. If it is possible that "involvement" is not related to difference in behavior hierarchies, we have a violation of our second criterion of usefulness, and we should consider rejecting the concept for link-3 type research rather than refining it.

The Finn (1982) and Smith and Swinyard (1982) papers also support the rejection of the concept for link 2 research. Low-involvement learning is defined as the process whereby new product/brand information is learned without active participation in the learning process. Consumers gradually develop a perception of the product without being specifically aware of any learning taking place --rather like the learning of nonsense syllables (Krugman 1965). If a consumer passively learns about an expensive, high risk product for which s/he has no felt needs (as might happen when a satisfied television owner is exposed to messages about a new model of television), it is doubtful that when the need arises, brand purchase will be the most favored source of information. In this situation, such low-involvement, passive information processing would be followed by extensive search and the affect-behavior sequence. The high economic risk inherent in the product class would be a better predictor of this sequence than the extent of cognitive processing that the consumer went through when s/he first became aware of a new brand. Alternatively, a consumer with a strong felt need for information about brands of hand lotion would be expected to be an active information processor and a high involvement learner because of the importance of this product class to her needs, but because of the low cost of the product, we might expect that purchase would precede affect. In this case product-use is a relatively valuable source of information, and we would witness the apparent "Low-involvement" hierarchy. In both cases we again see a violation of our second requirement for usefulness of the involvement concept -- we should not expect a consistent influence on the dependent variable. Therefore, we should also reject the concept of involvement for link 2-type research, rather than refine it

In summary, we have no reason to expect a consistent low/ high involvement influence of either PSP's or cognitive processing styles on a purchase/attitude sequence (links 2 and 3 in Figure 2). Similar arguments have been made elsewhere (Finn 1982), and it probably is "Time to Lay the Low-Involvement Hierarchy to Rest", but we might still be able to salvage low/high involvement PSP's as an independent variable influencing low/high involvement processing (link 1 in Figure 2). To explore this possibility it will be easier if we separate the various interpretations of the concept in terms of whether involvement is viewed as a stimulus centered variable (i.e., it is a characteristic of a product), a subject centered concept (i.e., individual consumers vary in involvement), or a response centered variable (i.e., involvement is an information processing characteristic).

STIMULUS CENTERED VIEW

The notion that involvement is a characteristic of a product has been put forward by Hupfer and Gardner (1971) and by Houston and Rothschild (1978). Here, we presume that there are products that are either trivial and unimportant or more serious and important. Hupfer and Gardner, for example, found a general level of agreement among people in the level of perceived importance of products as opposed to issues and concluded that, in general, products are less involving than issues, but there are differences among products (1971). Houston and Rothschild contended that a "situation is high in involvement when most or all people who interact with the situation develop a high level of concern for their subsequent behavior in the situation" (1978, p. 184). Characteristics of products that lead to different situational involvement levels, according to them, are cost, risk and elapsed time of consumption.

To evaluate the usefulness of the concept of involvement in these stimulus centered approaches, let's apply the same two criteria mentioned above: 1) consistent and definable differences in involvement, and 2) consistently different outcomes of high- and low-involvement Products.

When involvement is defined according to cost and elapsed time of consumption, the first criterion is easily met, and we are on our way to accepting the involvement concept rather than rejecting it. Products do differ in cost and consumption times, and these differences are easily measured.

The second criterion for usefulness queries whether high levels of product involvement are expected to have a different impact than low levels. Since this paper is positioned at the early stages of the Single Hierarchy, the expected impact of high- and low-involvement products is on the extent of cognitive processing of new brand information. The question to evaluate is: does product type (in terms of cost, risk, and interpurchase time) affect the nature of the learning process (active or passive) a consumer goes through when exposed to new information? Very little empirical research is available to assist us in answering this question, but one study reported by Krugman (1966) may help. Krugman measured the number of connections people made between the content of an advertisement and the content of their own life. He found that the extent of this processing (involvement) varied only slightly between messages for airlines and messages for margarine. The stimulus centered approach would require us to define "airlines" as a high-involvement product and "margarine" as low-involvement, and our second criterion would not be met. But this outcome should not be surprising.

Information about new brands or new claims of existing brands should be of most interest to those consumers who have a reason to process that information and of least interest to those consumers who have no reason to collect that information -- regardless of the type of product. For example, many people are owners of refrigerators -- a high cost, long interpurchase time product. According to the product-type definition of involvement, this product category fits into the high-involvement group. Yet, those consumers who are perfectly satisfied with the refrigerator they now own would be expected to pay little attention to advertisements for refrigerators. This low attention would restrict the number of "connections" made during the message. So, in general, we would not expect high-involvement products to lead to high-involvement cognitive Processes.

Similarly, a consumer who is dissatisfied with the available brands of hand lotions (a low cost, rapidly consumed product) and is hoping to find a better brand, might be more attentive to information about new brands and to actively process messages about them. Here we'd see a low-involvement product (as defined by this stimulus centered view) leading to a high-involvement outcome, contrary to expectations. If these hypothetical examples are sound, we would conclude (along with Mitchell, 1979) that involvement is a consumer specific variable that might be high or low regardless of the cost, risk, or interpurchase time of a given product, and reject the stimulus centered view of involvement.

SUBJECT CENTERED VIEW

The subject centered view of involvement recognizes that people differ in terms of "involvement" type variables. The expectations are that high involved receivers will actively process new information about a brand while low-involved receivers will exhibit a passive processing strategy. Strictly speaking, a discussion of high- or low-involved people should focus on general differences among people. Some people may be higher or lower in overall interest in new information than others. Such an approach is acceptable by definition. That is, people who are generally more thoughtful about information are, by definition, processing information more thoroughly than those with a low need for information, and this approach would fit our two criteria. However, the subject centered studies found in the involvement literature put consumers into situations with specific products and study their involvement in those situations. The assumption is that at a specific Point in time (rather than "in general"), some consumers are more involved with a product/situation combination than others. To evaluate the usefulness of this view it is necessary to further partition the definitions of consumer involvement into the various conceptualizations that are found in the literature.

Interest/Importance

Some subject centered views define involvement in terms of the level of interest that a consumer has in a product category, or how important it is to him/her. For example, Freedman defined involvement as "a general level of interest in or concern about an issue without reference to a specific position" (1964, p 290), and Day argued that involvement "describes the general level of interest in the object" (1974, p 131). Unfortunately, there is not a great deal of empirical work available that tests the effect of importance or interest levels on acquisition of new information. One study by Holbrook and Maier (1978) indicated that increasing importance of product attributes leads to more extensive information acquisition, providing some limited support for our second criterion. Because of this lack of empirical work in this specific area, we revert to "armchair analysis" to evaluate the usefulness of this subject centered approach.

A person who has a desire to keep abreast of the latest information in a product class (e.g., a wine connoisseur) can be described, almost by definition, as interested. Likewise a person engaged in active search for information about a product class (e.g., trying to learn about stereo component systems) is naturally interested, and people are expected to differ in the amount of interest they have. Both of these definitions of interest seem to satisfy our first criterion of usefulness -- that definable differences exist.

As stated above, the Holbrook and Maier (1978) findings suggest that information acquisition styles may differ with differences in importance, but beyond that it can be hypothesized that people who are highly interested in a product category will attend to and actively process new information about that product class. This describes high involvement consumers evincing high involvement learning. In the same manner, uninterested people might be expected to pay little heed to messages about the related product class, and be more likely candidates for the gradual learning effect of television repetition that characterizes low-involvement learning. The "interest/importance" definition of subject centered views of involvement, then, has the potential to meet both requirements of usefulness -differences among people most assuredly exist; and these differences will probably lead to differences in cognitive processing. The "interest/ importance" approach therefore, cannot be easily rejected as an indicant of involvement.

Interest in a product or topic, as postulated above, seems very closely related to the interpretation of involvement as a goal related characteristic of the consumer.

Goals and Consequences

This subject centered view of involvement presumes that consumers are more or less involved with a product category insofar as it is related to some achievable end, or concerns information that will impact directly on them. For example: Wright sees variations in involvement with advertising messages as "stemming from the receiver's perception of the relevancy of that content to some impending problem he faces" (1973, p 55); Bowen and Chafee conceive of involvement "as a direct outgrowth of the potential benefits or rewards the product holds for the consumer" (1974, p 615): and Mitchell conceptualizes involvement as an individual level, internal state variable related to "the location of the particular stimulus and/or situation in a goal hierarchy at a particular point in time" (1979, p 194). The usefulness of these views relies on the same two criteria as above.

First, regarding actual differences, it is clear that at a given point in time some consumers will have strong needs for information or for solutions to existing problems, and that others will have weak or no needs. Also, the configuration of a population, in terms of goals and problems, will not stay constant over time. For example, the social psychological component of situational involvement defined by Houston and Rothschild (1978) may be, by itself, high or low in involvement, but only some people in a social system will find themselves in a high involvement situation (e.g., gift giving) at a given point in time. The differences are differences among consumers in relation to the felt needs they have, and the first criterion for usefulness is met.

The expected impact of high- and low-involvement, so defined, is on the extent and type of cognitive processing. To satisfy our second criterion, consumers hearing messages about topics that are high on a goal hierarchy should have higher levels of attention and be more active processors of the message than consumers exposed to messages about topics that are lower on a goal hierarchy. Likewise, consumers exposed to messages about topics that will have a direct consequence for them should be more active processors of the message than receivers exposed to messages about less consequential products or topics. Evidence f or these outcomes exist in a study by Petty and Cacioppo (1981, Experiment #1). They manipulated involvement by having students listen to a tape advocating the administration of comprehensive exams at their university. High-involvement subjects were told that the exams would be instituted during their undergraduate days. Low-involvement subjects believed that the exam policy was for ten years later. Results indicated that high-involvement "motivates diligent processing of the content of a message" (p 22). We cannot recommend rejection of the " involvement" concept for this subject centered approach either.

Commitment

The position that involvement is defined in terms of the level of commitment that a consumer exhibits with respect to his position on an issue is an old one. Sherif and his colleagues (Sherif, Sherif, and Nebergall 1965) have referred to this as ego-involvement; Houston and Rothschild incorporated commitment into their definition of Enduring Involvement (1978); and Lastovicka (1979) used a commitment scale as one measure of involvement. Most communication studies using the commitment concept have been more concerned with acceptance/rejection of counterattitudinal messages than with information acquisition styles. That is, they seem to be centered later in the hierarchy than the awareness stage (perhaps at the evaluation stage.) This literature has changed very little since Hovland, Harvey, and Sherif (1957) demonstrated that high commitment leads to wider latitudes of rejection than low commitment and the resulting assimilation and contrast effects. When we begin to evaluate the commitment view in terms of influence on the Cognition (or first awareness) stage however, we must again rely on "armchair analysis".

The first criterion asks whether people can be expected to differ in their levels of commitment to particular brands. Since this seems to be the basis of the brand loyalty literature, the answer has to be "yes."

Given that it is reasonable to expect that people differ in terms of commitment, the next question is whether we should expect different attention levels and tendencies to actively or passively process information about new brands. The answer to this question is problematic. On the one hand, we do expect perceptual vigilance to influence recognition of favored brand names (Spence and Engel 1969) resulting in higher attention levels for them. However, this active processing does not appear to be a general trait of a high-commitment consumer because, on the other hand, lower attention levels are expected for unfavorable brands. Furthermore, we have no basis for generating hypothesis about what happens in the face of information about an unknown brand. On the surface, the commitment view of involvement seems to remain an open issue. However, it is true that there is an extensive literature on Brand Loyalty that is also concerned with the commitment issue and in the interest of reducing redundancy and reducing the confusion in the involvement literature, I recommend rejection of the label "involvement" for this subject centered view in favor of the label "brand loyalty."

RESPONSE CENTERED VIEW

The position that "involvement" is represented by active participation in information processing typifies the response centered view. This definition is similar to the non-academic use of the term, as in: "Will country A become 'involved' in the affairs of country 3? ", and is illustrated by Krugman's characterization as the extent of personal involvement with the medium and the message (1965). Another example is Leavitt, et al's separation of involvement levels according to the extent of cognitive processing and encoding elaboration (1981).

We have already partially evaluated the usefulness of this view when we discussed involvement as a dependent variable influenced by product types and by levels of consumer involvement. So far our conclusion has been to expect consumers' involvement to impact on cognitive processing styles. However, there is no reason to expect that involvement is the only determinant of active/passive processing styles. As Mitchell has pointed out, win acquiring information from advertisements, the modality, the structure and content of the advertisement may also have an effect on these processes" (1981, p 194). That is, the cause of differential processing styles may be unrelated to "involvement". However, given that a consumer is or is not an active participant in the cognitive process, what impact might this have on other cognitive variables? Investigations into this question represent the "Response Centered View". This approach is not diagrammed in Figure 2, but it can be thought of as the study of involvement as an intervening process within the cognitive stage. Figure 3 diagrams the response centered paradigm.

As indicated in Figure 3, research in this area positions cognitive activities ( like attention, active/ passive processing, etc.) as the independent variable that influences other cognitive variables like memory, recognition and recall. For example, Krugman (1966) argued that active processing of advertising material should lead to low-involvement learning of a nonverbalizable nature. Also Mitchell (1981) and Gardner, Mitchell and Russo (1978) instructed subjects to process either brand or style related content of advertising messages and measured the impact on the ability to retrieve information from long-term memory. Finally, Leavitt, et al (1981) pondered the potential effects of low- and high-involvement levels of attention on learning.

FIGURE 3

THE RESPONSE CENTERED PARADIGM

Applying the same two criteria of usefulness as above, it is difficult to reject the response centered view. By definition, active processing strategies are different from passive processing strategies, and our first criterion is present. Our second criterion requires that these different processing strategies be expected to result in different outcomes. Mitchell and his colleagues (Gardner, Mitchell and Russo 1978; Mitchell 1980, 1981) have demonstrated that differences in involvement (defined in terms of the extent and kind of information processing) do impact on recall. Generally, subjects instructed to process brand related information are able to retrieve new product related information faster and more accurately than subjects instructed to execute a nonbrand processing strategy. Evidence and arguments also exist (Krugman 1965, 1966; Leavitt, et al 1981) that lead to the conclusion that passive processing requires more repetition than active processing to impact on memory and awareness. We cannot, on this basis, reject the usefulness of the involvement construct for this link-4 type research (Figure 3).

SUMMARY AND CONCLUSION

The thoughts expressed in this paper have targeted on the influence of involvement on the acquisition of new information--specifically on the awareness (or Cognition) stage of a behavior hierarchy. Table 1 provides a summary of the points raised here regarding the usefulness of the term "involvement" in relation to each of the many conceptualizations of the term. In short, the term should be rejected as a label for product types, levels of consumer commitment, and variations of the Affect and Conation stages in the hierarchy-of-effects. There is reason to further consider the usefulness of the concept regarding the impact of consumer interest levels, of consumers' recognized goals and consequences, and of information processing activity levels. Because each of these variations of the concept is different from the others, it would, perhaps, be best to either reject the concept of involvement altogether, or to restrict it to refer to only one type of variable.

TABLE 1

SUMMARY OF RECOMMENDATIONS

If we look more closely at the subject centered views of involvement that meet our criteria of usefulness, it is not difficult to find a common thread. Consumers who are "interested", because of connoiseurship or a desire to sample a complete repertoire of brands, share the felt need for information about the product class with goal oriented consumers. The term "involvement" however, does not capture the essence of this need recognition. Whereas" involvement" has come to be understood to mean "importance" of a product class, or commitment to a brand, "relation to needs" (a possible substitution label) can be strong or weak regardless of the level of perceived importance. For example, we can visualize a mother who is very concerned about toothpaste and tooth brushing (it is important to her); but if she has found an excellent (in her mind) brand of toothpaste that meets all of her needs (a high level of commitment), she would be a highly "involved" consumer with no felt needs. She might not be expected to actively process new commercial information about toothpaste brands, and her level of felt needs (low) would be a much better predictor of this than her involvement level (high). Alternatively, another parent in the same situation might be ever searching for an even better brand since toothbrushing is so important to her, and she would be an active processor of new information. In both hypothetical cases" involvement" would be high, but "need for information" would be the better predictor of extent of processing.

Consumers with unsatisfied product-related problems (the " goals and consequences" category), consumers with a desire to keep abreast of news in a product category ("interest"), and consumers engaged in active search for information about a product class (goal related interest), have the common characteristic of felt needs. If we categorize consumers according to the relation of the advertising content to their felt needs, we will add to our understanding of the process. The term "involvement" in these subject centered views should be rejected in favor of a needs-related label.

The "extent of cognitive processing" interpretation of involvement is the only one left. This definition of involvement as "extent of participation in" the activity seems to be completely unrelated to the "gut feel" interpretation that involvement has something to do with concern or importance or reflection of self and might probably be rejected out-of-hand in favor of labels like "active processing" and "passive processing". But it is hard to part completely with an old friend like involvement -- should we, or shouldn't we?

REFERENCES

Bowen, L. and S. H. Chafee (1974), "Product Involvement and Pertinent Advertising Appeals," Journalism Quarterly, 51 (Winter), 613-621.

Day, G. S. (1974), "Attitude Stability, Changeability, and Predictive Ability," in J. U. Farley, J. A. Howard, and L. W. Ring, (eds.), Consumer Behavior: Theory and Application, Boston: Allyn and Bacon, 130-146.

Finn, David W. (1982), "It is Time to Lay the Low Involvement Hierarchy to Rest," in B. Walker, et al. (eds.), An Assessment of Marketing Thought and Practice, Proceedings of the 1982 Marketing Educators' Conference, 99-103.

Freedman, J. L. (1964), "Involvement, Discrepancy and Change," Journal of Abnormal and Social Psychology, 69, 290-295.

Gardner, Meryl P., Andrew A. Mitchell, and J. Edward Russo (1978), "Chronometric Analysis: An Introduction and an Application to Low Involvement Perception of Advertisements," in H. K. Hunt (ed.), Advances in Consumer Research, Vol. 5, 581-589.

Holbrook, Morris B. and Karl A. Maier (1978), "A Study of the Interface Between Attitude Structure and Information Acquisition Using A Questionnaire Based Information-Display Sheet," in H. K. Hunt (ed.), Advances in Consumer Research, Vol. 5, pp.9 3-98.

Hovland, Carl I., O. J. Harvey, and Muzafer Sherif (1957), "Assimilation and Contrast Effects in Reactions to Communication and Attitude Change." Journal of Abnormal and Social Psychology, 55, 244-232.

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

Hupfer, N. T. and D. M. Gardner (1971), "Differential Involvement with Products and Issues: An Exploratory Study," in D. M. Gardner (ed.), Proceedings: 2nd Annual Conference of the Association for Consumer Research, 262-269.

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

Krugman, Herbert E. (1966), "The Measurement of Advertising Involvement," Public Opinion Quarterly, 30, 583-596.

Lastovicka, John L. (1979), "Questioning the Concept of Involvement Defined Product Classes," in W. L. Wilkie (ed. ), Advances in Consumer Research, Vol. 6. 174-179.

Lavidge, Robert J. and Gary A. Steiner (1961), "A Model For Predictive Measurements of Advertising Effectiveness," Journal of Marketing, 25 (October), 59-62.

Leavitt, Clark, Anthony G. Greenwald, and Carl Obermiller (1981), "What is Low Involvement Low In?" in K. B. Monroe (ed.), Advances in Consumer Research, Vol. 8, 15-19.

Lutz, Richard J. and Patrick J. Reilly (1974), "An Exploration of the Effects of Perceived Social and Performance Risk on Consumer Information Acquisition," in S. Ward and P. Wright (eds.), Advances in Consumer Research, Vol. 1, pp.393-405.

Mitchell, Andrew A. (1979), "Involvement: A Potentially Important Mediator of Consumer Behavior," in W. L. Wilkie ( ed. ), Advances in Consumer Research, Vol. 6, 191-196.

Mitchell, Andrew A. (1981), "The Dimensions of Advertising Involvement," in K. B. Monroe (ed.), Advances in Consumer Research, Vol. 8, 25-30.

Petty, Richard E. and John T. Cacioppo (1981), "Issue Involvement as a Mediator of the Effects on Attitude of Advertising Content and Context," in K. B. Monroe (ed.), Advances in Consumer Research, Vol. 8, 20-24.

Ray, Michael L. (1979), "Involvement and Other Variables Mediating Communication Effects as Opposed to Explaining All Consumer Behavior," in W. L. Wilkie ( ed. ), Advances in Consumer Research, Vol. 6, 197-199.

Ray, Michael L. (1973), "Marketing Communication and The Hierarchy-of-Effects," in Peter Clarke (ed.), New Models for Mass Communication, Beverly Hill: Sage Publications, pp. 147-176.

Sherif, M. C., C. Sherif, and R. Nebergall (1965), Attitude and Attitude Change: The Social Judgment Involvement Approach, Philadelphia: W. B. Saunders.

Smith, Robert E. and William R. Swinyard (1982), "Information Response Models: Integrated Approach," Journal of Marketing, 46 (Winter), 81-93.

Spence, Homer E. and James F. Engel (1969), "The Impact of Brand Preference on the Perception of Brand Names: A Laboratory Analysis," in Philip R. McDonald ( ed. ), Marketing Involvement in Society and the Economy. Proceedings of the 1969 Fall Conference of the American Marketing Association, 267-271.

Wright, Peter L. (1973), "Cognitive Processes Mediating Acceptance of Research, 10 (February), 53-62.

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