Means End Chains

279

MAR WILEJ RIGHT BATCH

Base of text

Psychology & Marketing, Vol. 21(4): 279294 (April 2004)Published online in Wiley InterScience (www.interscience.wiley.com) 2004 Wiley Periodicals, Inc. DOI: 10.1002/mar.20006

Top of text

Top of CT

Base of DF

Advancing MeansEndChains by IncorporatingHeiders Balance Theoryand Fourniers ConsumerBrand RelationshipTypologyArch G. WoodsideBoston College

ABSTRACT

Meansend chain theory proposes that knowledge held inconsumers memory is organized in a hierarchy with concretethoughts linked to more abstract thoughts in a sequence progressingfrom means (i.e., product features), to psychological and socialconsequences, and finally to ends (i.e., fulfillment of personal values).This article proposes several advances in the theory. First, specificbuying and consumption situations serve as frames of referencewhen consumers are thinking about products and alternativefeatures of products and brands. Second, states of psychologicalimbalance may occur in consumers minds among linkages retrievedautomatically for featuresconsequences and consequencesvalues;thus, Heiders balance theory informs meansend chain theory andresearch. The theoretical and practical usefulness of meansendresearch increases from asking consumers to name an acceptablealternative to the product and brand used in a recent consumptionsituation, as well as an unacceptable option, and to describe thefeaturesconsequencesvalues of these options; consequently,Fourniers alternative relationships of consumerbrands (e.g., casualfriendships, marriages, enmities) become relevant for meansend

280 WOODSIDE

MAR WILEJ LEFT BATCH

shortstandard

Top of textBase of text

Base of RF

chain theory. To examine the propositions empirically, the articledescribes psychological schemas for four meansend chains thatlink two consumers recent lived consumption situations to personalvalues. 2004 Wiley Periodicals, Inc.

In consumer research on meansend chains, the meanings that a con-sumer associates with a product, service, or specific brand are repre-sented hierarchically. The lowest level depicts an objects attributesthose that are physical or concrete, and those that are abstract (e.g., abrands image). The remaining levels are consumers outcomes includ-ing functional and psychosocial consequences, followed by consumersinstrumental and terminal values (see Gutman, 1997).

Laddering is a frequently used research method for electing meansend chains. Traditionally, laddering is a metaphor representing a re-spondents answers to a series of why probes, typically starting withthe attributes that distinguish more-desired from less-desired alterna-tives. Asking for successive elicitations creates a chain of elements lead-ing from a products attributes to one or a few terminal values: Eachsuccessive concept learned becomes a subgoal for the final goal. Gutmanand Reynolds (1978) illustrate a resulting ladder elicited from a respon-dent in a study of breakfast cereal: crunchy : has body : stays withme : avoid snack : aids weight loss : improves appearance : ro-mance.

In the empirical research the present article describes, the questions,what makes you say that? and what does that lead to? are used inplace of why probes. Becker (1998) emphasizes that why requires agood answer, one thatmakes sense and can be defended. Such answersare unlikely to reveal logical flaws and inconsistencies.Becker recom-mends using How? questions.

When I asked them [such questions] gave people more leeway, wereless constraining, invited them to answer in any way that suited them,to tell a story that included whatever they thought the story ought toinclude in order to make sense. They didnt demand a right answer,didnt seem to be trying to place responsibility for bad actions or out-comes anywhere. (Becker, 1998, p. 59)

Because the scientific literature is compelling that most thinking isunconscious and consumers typically are unable to surface and reportunconscious processes (unconscious processes often drive behavior thatconsumers are unaware that they performed, e.g., see Bargh, 2002; Zalt-man, 2003), advancing theory and research that inform knowledge ofconsumer unconscious processes is worthwhile. Heretofore, the valuefor consumer theory and research of meansend chains and the lad-deringmethod relates only implicitly to unconsciousmental processesand how laddering may help to surface unconscious thoughts. For

ADVANCING MEANSEND CHAINS 281


shortstandard


Base of RF

Figure 1. Extended original balanced state for Posner and VW case study.

example, direct questioning of how a cereal being crunchy aids a con-sumers love life may cause a scoffing response from the consumer. Ask-ing the consumer a sequence of probing questions is a form of auto driv-ing (cf. Heisley & Levy, 1991) research that may enable the consumerto surface unconscious thoughts that rarely enter conscious processing.

Surfacing unconscious processes may result in tension and stress be-cause unbalanced associations occur among concepts now in a con-sumers working memory (cf. Heider, 1958). The thought, a Jewish cou-ple buys a German car (see Figure 1 and Woodside & Chebat, 2001)implies an example of such stress resulting from a surfacing of nega-tively valued concepts and imagessome held unconsciouslywith thepositive experiences associating with buying a car highly prized for itsdesign features. Advancing Heiders balance theory to person, brand,and attribute (and attribute-to-benefit-to-instrumental and terminalvalues) enables more robust meansend chain laddering research be-cause the resulting ladders may uncover and display stress and stressresolution, as well as harmonious associations.

Pieters, Baumgartner, and Allen (1988) advocate following an alter-native empirical method for meansend chains versus the more well-known laddering approach that Gutman and Reynolds (1978) demon-strate. Peiters et al. (1988) elicit peoples reasons for wanting to dosomething or to pursue a goal rather to than to name concrete featuresthat they associate with a product. Bagozzi and Dabholkar (2000) applythe approach of Peiters et al. approach for mapping persons cognitiveschemas regarding a specific target, President Clinton.Whereas the Pie-ters et al. (1998) and Bagozzi and Dabholkar (2000) refinements areuseful for describing how specific schemas accurately predict attitudes

282 WOODSIDE


shortstandard


Base of RF

and intentions regarding future events, the advances below focus onframing meansend theory and research by consumers recent lived ex-periences.

HEIDERS BALANCE THEORY

Heider (1958, Chapter 7) distinguishes two types of relations betweenseparate entities: unit and sentiment relations. Separate entities com-prise a unit when they are perceived as belonging together. For example,members of a family are seen as a unit; a person and his deed belongtogether (p. 176). U denotes the cognitive unit between two entities,and notU the fact that the two entities are segregated (p. 201). A sen-timent relation refers to the positive (L or ) or negative (DL or )feelings or valuation that one gives to an entity, such as a person, ac-tivity, or object.

These relations may be for dyads, triads, or more complex cases, butall relations are from the perceivers subjective point of view. Thus, al-though a brand may possess a given attribute or provide a specific ben-efit, if a consumer perceives the opposite, a notU relation results be-tween a brand and such an attribute or benefit.

Heider (1958) describes the two relation concepts to result in fourpossibilities between two entities: U, notU, L, and DL. He empha-sizes:

By a balanced state is meant a situation in which the relations amongthe entities fit together harmoniously; there is no stress toward change.A basic assumption is that sentiment relations and unit relations tendtoward a balanced state. This means that sentiments are not entirelyindependent of the perceptions of unit connections between entities andthat the latter, in turn, are not entirely independent of sentiments.Sentiments and unit relations are mutually interdependent. It alsomeans that if a balanced state does not exist, then forces toward thisstate will arise. If a change is not possible, the state of imbalance willproduce tension. (Heider, 1958, p. 201)

Unit and sentiment relations represent independent theoretical andempirical propositions grounded in cognitions (i.e., units) and affections(i.e., sentiments). When tension caused by imbalance arises in the mindof the individual, then the individual is likely to exercise some mentaland physical effort to eliminate the tension.

Unbalanced situations stimulate us to further thinking; they have thecharacter of interesting puzzles, problems which make us suspect adepth of interesting background . . . Stories in which the stress islaid on unbalanced situations are felt to have a deep psychologicalmeaning. Dostoevski, for instance, describes again and again feelings



shortstandard


Base of RF

Figure 2. Meansend laddering research: Updating the theory and data collectionmethod.

full of conflict resulting from just such situations. (Heider, 1958,pp. 180181)

Advancing MeansEnd Chain Laddering Research withHeiders Balance Theory

Figure 2 summarizes the application of Heiders balance theory andrelated extensions to meansend chain laddering research. Note thegrounding in Figure 2 includes prequel to action to emphasize that ifconsumer research focuses on existing consumerbrand relationships(cf. Fourier, 1998), such relationships depend on specific purchase oruse situations. A situation is defined as a conjunction of time, place,objects, and person(s). Consequently, the lived experiences and top-of-mind attributes and benefits that a consumer identifies (to him- or her-self and the interviewer) are contingent on the structure of the situationleading to brand purchase or use. Consequently, the situation informsthe attributesbenefitsvalue associations.

The implication for research design is that meansend chains datafor a given respondent should be collected for multiple situations someof which may be relevant to the focal brand. A focal brand is defined asthe one that the consumer names as used most often or identifies as theone used most recently; alternatively, a focal brand may be defined asthe specific target chosen by the researcherthe primary focus of thestudy for which the researcher asks a respondent to describeits prod-

284 WOODSIDE


shortstandard


Base of RF

uct features and consequences in experiencing such product featuresand how these experiences relate to personal values. Asking informantsindividually to identify a focal brand for a given product category is aless reactive procedure comparatively; the researcher naming a targetbrand indicates possible self-validation problemsinformants reportperceptions about the brand mentioned by the researcher that may notrelate much to the informants past or future lives (cf. Feldman&Lynch,1988). Thus, asking the informant to report on her most recent con-sumption behavior and collecting meansend data with respect to Sit-uation X as well as for Situation Y may be more relevant for the inform-ant and reduces the occurrence of self-validity problems.

Figure 2 includes the assignment of sentiments (positive and negativesigns) to associations in the resulting laddering steps. In Figure 2, notethat Product P and Brand B associate negatively with Attribute I, andlack of I relates negatively with Benefit B (e.g., assume Benefit B is aninformants comment, good tasting mouthwash, for a mouthwash at-tribute I). Figure 2 advancesmeansend chains by suggesting collectingdata on the first alternative that comes to mind for the consumer forthe recent lived experience in the focal situation. For example, If youhad not consumed Brand B in Situation X, what might you have con-sumed instead?

Figure 2 also asks for reject alternative information and implies theproposition that the respondent is easily able to surface a product/brandthat would have been unacceptable for consumption in his or her recentlived experience. Such rejection brand data are then followed bymeansend probes to surface brandattributesvalues associations for unac-ceptable alternatives for use in the given situation. The brand strategistmight ask here: how often, and for what situations, does my productand brand surface as the reject alternative?When rejection occurs, whatattributes and personal values surface in the respondents mind?

FOURNIERS CONSUMERBRAND RELATIONSHIPTYPOLOGY

Fournier (1998) extends the two-party social relationship metaphor toencompass the consumer and brands. She develops the consumerbrand relationship proposition, including anthropomorphizing thebrand as an active relationship partnerat the level of consumers livedexperiences with their brands. Her proposal of 15 consumerbrand re-lationships include arrangedmarriages, casual friend, marriages of con-veniences, committed partnerships, best friendships, compartmental-ized friendships, kinships, rebounds, childhood friendships, courtships,dependencies, flings, enslavements, enmities, and secret affairs.



shortstandard


Base of RF

Advancing MeansEnd Chain Laddering Research byApplying Fourniers ConsumerBrand Relationships

Using the advanced meansend chain laddering research that Figure2 summarizes serves to complement and inform Fourniers consumerbrand relationship typology: Multiple consumerbrand relationshiptypes are uncovered by collecting data within given situations of con-sumerbrand lived experiences, acceptable alternatives to these livedexperiences, as well as rejected alternatives. Equally valuable are thedata uncovering association streams of the three laddering paths (done,alternative, and rejected) to terminal values.

APPLYING THE ADVANCES IN MEANS-END CHAINLADDERING RESEARCH

The following four case studies illustrate applications of advances inmeansend chain laddering research. The four cases report two livedexperiences for each of two subjects (Ss).

Method

In order to collect the data to examine the propositions, two subjects,Eric and Peter, were chosen to participate in two separate interviewseach. Each subject is a senior in college in his early 20s living in anapartment on campus at a large university in the eastern U.S. The in-terviews were conducted at the residences of the respondents at twodifferent times during the day. The two situations under considerationfor this study were the consumption of a beverage in the morning afterwaking up and the beverage consumption between eight oclock andmidnight on a Friday night. For themeansend chain laddering processdescribed previously, each respondent was asked a series of 34 questionsfor each situation. The questions were designed to elicit responses thatwould provide the perceived attributes, benefits, and consequences ofthe products consumed.

The respondents were also asked to provide alternative products thatthey might have chosen in each situation, as well as rejected productsthat would not be considered. The subjects were also asked questionspertaining to prior actions to the consumption of the beverage and aboutthe situation itself, such as where it occurred and if anyone else wasaround. With the responses gathered from each subject an extendedmeansend chain was created for each subject in each situation withthe use of the theoretical model shown in Figure 2.

The extendedMEC begins at the bottomwith the prequel to the actiondiscussed. This prequel is the event(s) that is believed to influence the

286 WOODSIDE


shortstandard


Base of RF

action to occur. From this prior event there must be some sort of rec-ognition, or immediate feelings or thoughts that act as drivers to act.Stemming from the driver is the action itself, or the product usage. Theconsumer describes of the product (what it was, how much was con-sumed, and what brand), and provides a brief description of the situa-tion (when and where the action occurred, others present at the time).From these steps the entire MEC develops and builds up to the terminalvalues.

Identifying a specific lived consumption experience and situation pro-vides theoretical ground for developing meansend chains based on theproposition that the consumption of a specific product and brand is ac-ceptable in only certain situations. In fact, the same brand and productform may associate positively to one situation and negatively to alter-native situations for the same consumer. Also, different attributes andbenefits may come to mind for consumers of the same brand, dependingon the usage situation that the consumer is thinking about (cf. Bearden& Woodside, 1978; Belk, 1974). Consequently, resulting meansendchains may be expected to vary contingent on the focal situations beingdescribed in consumers lived experiences.

From the description of the product and the situation comes the threedistinct series of the extended MEC. On the left is the reject alternativeand on the right is the alternative that would be considered. The centerof the model focuses on the product that was actually used or consumed.The separate attributes of the product or brand are laid out in order toget an idea of the benefits or consequences of each. This same question-ing process is used to collect information for the alternative and rejectproducts. After the benefits of the attributes is where the MEC beginsto probe for the instrumental beliefs and values of the respondent. Thesebeliefs are often unconsciously stored thoughts that the consumer maynot connect but that actually play a role in their decision-making pro-cess. Some benefits may lead to the same values, and benefits from thealternative and the reject products can also tie into these same values.The model shows that the alternative and reject products may tie inwith the entire meansend chain, which is why they should be consid-ered when one is looking at the unconscious thinking of consumers. Fi-nally, the chain concludes with terminal values or their core values andbeliefs in the Ss life.

FINDINGS

Beverage Consumed on Monday Morning

For this situation each respondent answered questions about the firstbeverage they consumed on the day that the interviews took place. Eric



shortstandard


Base of RF

Figure 3. Meansend chain laddering findings for Erics first beverage on Mondaymorning, 210 February 2003.

went to bed around 1:00 a.m. and did not have anything to drink beforegoing to sleep. After waking up at 10:00 a.m. on Monday morning witha dry mouth and morning breath he went down to the kitchen in hisapartment and poured a 16oz. glass of Tropicana orange juice. Ericproceeded to have another glass of orange juice while he ate a bagelalone at the table. The attributes that he associates with Tropicanaorange juice are that it tastes sweet, kills morning breath, and containsvitamin C (see Figure 3).

No benefit was linked with the sweet taste, but sweet taste is anattribute that Eric remarked that he prefers. The benefit of killingmorning breath was having better breath throughout the day, whichwas also a benefit that he did not believe any alternative drinks offered.Eric explained that he had always been told by his mother that heshould drink a lot of orange juice to prevent getting a cold, which is thebenefit of the product containing vitamin C. By preventing illness hebelieves that he would not have to visit the doctor, which was a positiverelationship because he does not like paying the extra money or dealingwith doctors in general. As an alternative Eric suggested that he alsolikes to drink Gatorade because it quenches his thirst and keeps himhydrated. Again, by staying hydrated he believes that he will be able tostay in better shape and remain healthier, which ties in with the ben-efits of the vitamin C in orange juice.

Having good breath is important for Eric because one of his main

288 WOODSIDE


shortstandard


Base of RF

goals is to feel comfortable in his surroundings. By not having badbreath he felt that he could socialize with others and not worry aboutothers smelling his breath.

Coffee is the product that Eric reports that he rejects because of caf-feine, calories, and causing him to have to go to the bathroom. The con-sequences of all of these attributes negatively connected with the valueof feeling comfortable because each one posed a problem with comfort.

Interestingly there were two other benefits that surfaced in this in-terview stemming from staying healthy and staying in shape. Eric feltthat a result of staying in shape was that this would make him moreattractive to females, which seemed to be a common theme with the fourdifferent situations. Saving money came up as a core value because inthe interview he explained that he bought Tropicana orange juice eventhough it tended to be more expensive. He was willing to pay the pricepremium for the brand but overall he is looking to save money.

Peter spent Sunday night watching movies and eating popcorn withhis girlfriend until about 2:00 a.m. when they went to bed. When hewoke up around 11:00 on Monday morning his mouth was very dry fromthe popcorn and he had bad breath. He got out of bed and went to therefrigerator in the bedroom and pulled out the gallon jug of orange juice.He drank three 10-ounce servings directly from the jug while his girl-friend and roommate were asleep in the room. Peter does not recall thebrand because he bought whatever was cheapest at the store, but hedoes know that it was not a major brand like Tropicana. His alternativeproduct was apple juice and the rejected product was cranberry juice(see Figure 4).

The attributes that Peter described for the orange juice were that itwas refreshing, it killed germs, and that it contained vitamin C. In theinterview he remarked that he was not sure if the orange juice actuallykilled germs but this was what he guessed gave him better breath afterdrinking it. As a result of having better breath he said that he wouldactually be able to kiss his girlfriend because she did not like to tastehis bad breath. Besides the fact that Peter does not like the taste ofcranberry juice in general he also felt that it stained teeth with its redcoloring, and caused saliva buildup that made him spit. These conse-quences can be negatively linked with Peter being able to kiss his girl-friend. All of this leads up to the core value of romance and the feelingof closeness that Peter believes is important.

Peter mentioned that he believed that vitamin C prevents colds andwould help him to stay healthy. He also felt that the alternative, applejuice, could help to stay healthy because it is good for you. Stayinghealthy is important to Peter because another core value is that he likesto keep up with his responsibilities in life. He does not like to fall behindin schoolwork or to miss days at his job.

Both respondents chose the same beverage for quite similar reasons



shortstandard


Base of RF

Figure 4. Meansend chain laddering findings for Peters first beverage on Mondaymorning, 210 February 2003.

but it did not turn out that their core values were also the same. Therewas the similarity of staying healthy but that is one of the only simi-larities. Also each respondent had a different relationship with theirbrand of choice. Using the relationships developed by Fournier (1998),Eric is in a committed partnership with the Tropicana brand, whereasPeter is a casual friend/buddy with his brand of orange juice. Eric isin a long-term, voluntary union with Tropicana even though it is higherpriced, because it is a brand that he knows and trusts. Peter buys morebased on price, so as long as a certain brand is the cheapest he willcontinue to purchase it, but if a competitor suddenly becomes cheaperhe will now buy that brand. There is no long-term commitment in thebrand relationship.

Beverage Consumed Last Friday Night between 8:00 P.M.and Midnight

Both Eric and Peter describe themselves as heavy drinkers, drinking 10beers on average four or five nights a week. This consumption has in-creased somewhat over the last couple of years but all throughout theyhave been beer drinkers. Erics situation on Friday night began whenhe invited friends over for pizza and to watch the game on TV (Figure5). He had gone out earlier and bought two 30-packs of Busch Light forthe evening, and his friends were expecting to drink when they arrived.

290 WOODSIDE


shortstandard


Base of RF

Figure 5. Meansend chain laddering findings for Erics beverage consumption onFriday night, 27 February 2003.

As the night progressed the group began to play drinking games andconsume even greater amounts of beer. By midnight Eric recalls havingabout 12 cans of beer before going to sleep.

The attributes suggested for Busch Light beer are a smooth taste, itcontains alcohol, and that it contains calories. The fact that the beveragecontained alcohol seemed to be the most prominent reason for consum-ing the beer. The benefit of the alcohol is that it gets the person drunk,which makes it easier to socialize, but much like the calories in the beerthe alcohol is also not healthy. The calories lead to getting fat and beingless attractive to females. The alternative product to drink on a Fridaynight for Eric was vodka. Vodka was an alternative because it wasstronger than beer, which would get him drunker, but this leads to hav-ing a hangover. The negative consequence of being so very drunk leadsto being unattractive to females and conflict with the core values offeeling good both physically and emotionally.

The rejected product for Eric was wine, no specific brand, just winein general. He reports not liking the taste of wine and he also perceiveswine as a girly drink. The bitter taste of the wine causes him to feelsick to his stomach, which has a negative relation to the core value offeeling good physically. The alcohol in beer, which makes it easier tosocialize, has a positive relationship with the value of having fun. Ericstressed that at his age he was more concerned with having a good timethan with worrying about too many responsibilities, but he did see thismentality changing in the near future because of graduation.



shortstandard


Base of RF

Figure 6. Meansend chain laddering findings for Peters beverage consumption onFriday night, 27 February 2003.

The situation for Peter was different because of the setting but forthe most part the MEC looked similar to that of Eric (see Figure 6).Peter had taken a test earlier in the afternoon that he did not believehad gone very well. After the test he went down to the local bar, MaryAnns, for happy hour. He met up with friend at the bar and they beganto drink beer. By 8:00 p.m. Peter was still at the bar and he was stilldrinking bottles of Busch Light beer. He had about eight beers between8:00 p.m. and the time he left the bar at 10:00 p.m.

The attributes that Peter describes in Busch Light beer are that it iscarbonated, and that it contains alcohol and calories. He could not ex-plain what he thought the benefits of carbonation are, but this could beinvestigated in a follow-up interview if one was performed. Alcohol wasthe major attribute that Peter described in beer. The alcohol led to get-ting drunk and acting immature, which was negatively related to hiscore value of thinking rationally and responsibly. Peter mentioned rumas the rejected beverage because of a bad experience that he had in thepast. Due to this bad experience he cannot drink any rum because itmakes him vomit.

Calories were not a major concern for Peter, but he did acknowledgethat they are not healthy. Because of this he would need to work out tobe more attractive to females. Wine was given as an alternative to beeras a beverage to consume on a Friday night. Peter felt that wine is moreof a sophisticated drink and makes him look more mature. He felt thatthis would make him more attractive to femalesin this case, one ofhis core values.

292 WOODSIDE


shortstandard


Base of RF

Figure 7. The multiple mental processes in research on marketingbuying thinking.

The brand relationships for both Peter and Eric are about the samefor Busch Light beer. Both respondents remarked that Busch Light beerwas popular with them because it was cheaper than most beers but thatit was not the cheapest. This is a casual friendship/buddy type of re-lationship because there is not a long-term commitment with the brand.The consumers are loyal but they do buy other brands and will likelynot stay with this brand after graduation because of their increasedfinancial situation. Another brand relationship can be seen between Pe-ter and rum. This is an enmities relationship because he has a desireto avoid the product due to its negative effects.

APPLYING ADVANCED MEANSEND CHAIN LADDERINGTO LEARN UNCONSCIOUS PROCESSES

Research on consumer thinking processes (see Zaltman, 2003) indicatesmultiple levels of mental processing. Figure 7 summarize five levels ofthinking processes that serve to inform meansend chain laddering re-search.

Level 1 represents conscious thinking that is verbalized between twoor more parties. Level 2 thinking includes conscious handling ofthoughts before and after verbalizing thoughts and surfacing thoughts.



shortstandard


Base of RF

Level 3 thinking includes surfacing thoughts that are mostly not underconscious control. Level 4 thinking represents unconscious thinking be-tween two or more persons. Level 5 thinking represents unconsciousprocessing, including spreading activation of relationships among con-cepts that the individual is unaware are occurring.

Other empirical findings (see Bargh, 2002; Wegner, 2002) support acore proposition about how the mind thinks: The most processing occursunconsciously (indicated by the thicker line 5 in Figure 7). Such uncon-scious thinking influences consumers actions in ways mostly unknownto the consumers themselves.

Such observations are likely to cause stress among consumer re-searchers who mainly apply the currently (early 21st century) domi-nating logic of asking fixed-point (e.g., 7-point scale items) questionsbecause such questioning mostly reaches only Level 1 processing, thatis, verbalized thoughts (cf. Zaltman, 2003). Some deeper form of ques-tioning is necessary to reach into Level 25 mental processing. Theadvances in meansend chain laddering research that this article hasproposed may help achieve this objective.

LIMITATIONS AND SUGGESTIONS FOR FUTURERESEARCH

The four MEC laddering cases serve to inform theory development andare not intended as representative of college students in general or forspecific segments of beverage consumers. This report of empirically ap-plying the advances in the MEC laddering is limited by not taking thestep of describing the results in follow-up interviews with the two casestudy Ss and having the Ss confirm and elaborate on the researchersinterpretations (Hirschman, 1986, argues for such a validation step).Such an additional auto driving step will likely prove useful for achiev-ing further advances in MEC laddering theory and research.

REFERENCES

Bagozzi, R. P., & Dabholkar, P. A. (2000). Discursive psychology: An alternativeconceptual foundation to meansend chain theory. Psychology &Marketing,17, 535586.

Bargh, J. A. (2002). Losing consciousness: Automatic influences on consumerjudgment, behavior, and motivation. Journal of Consumer Research, 29,280285.

Bearden, W. O., & Woodside, A. (1978). Consumption occasion influence onconsumer brand choice. Decision Sciences, 9, 274281.

Becker, H. S. (1998). Tricks of the trade: How to think about your researchwhile youre doing it. Chicago: University of Chicago Press.

294 WOODSIDE


shortstandard


Base of RF

Belk, R. (1974). An exploratory assessment of situational effects in buyer be-havior. Journal of Marketing Research, 11, 156163.

Feldman, J. M., & Lynch, J. G. Jr. (1988). Self-generated validity and othereffects of measurement on belief, attitude, intention, and behavior. Journalof Applied Psychology, 73, 421435.

Fourier, S. (1998). Consumers and their brands: Developing relationship theoryin consumer research. Journal of Consumer Research, 24, 343373.

Gutman, J. (1997). Meansend chains as goal hierarchies. Psychology & Mar-keting, 14, 545560.

Gutman, J., & Reynolds, T. J. (1978). An investigation of the levels of cognitiveabstraction utilized by consumers in product differentiation. In J. Eighmey(Ed.), Attitude research under the sun (pp. 5764). Chicago: American Mar-keting Association.

Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley.Heisley, D. D., & Levy, S. J. (1991). Autodriving: A photoelicitation technique.

Journal of Consumer Research, 18, 257272.Hirschman, E. C. (1986). Humanistic inquiry in marketing research: Philoso-

phy, method, and criteria. Journal of Marketing Research, 23, 237249.Pieters, R., Baumgartner, H., & Allen, D. (1995). A meansend chain approach

to consumer goal structures. International Journal of Research inMarketing,12, 227244.

Wegner, D. M. (2002). The illusion of conscious will. Cambridge, MA: BradfordBooks.

Woodside, A. G., & Chebat, J. C. (2001). Updating Heiders balance theory inconsumer behavior. Psychology & Marketing, 18, 475496.

Zaltman, G. (2003). How consumers think. Cambridge, MA: Harvard BusinessSchool Press.

The data collection by Eric Goodwin, Boston College, of the four cases thisarticle describes is gratefully acknowledged. The helpful comments of the P&Mreviewers, Fran Hruska, and Rajan Nataraajan on earlier drafts of this articleare also gratefully acknowledged.

Correspondence regarding this article should be sent to: Arch G. Woodside,Carroll School of Management, Boston College, 450 Fulton Hall, 140 Common-wealth Avenue, Chestnut Hill, MA 02467 ([email protected]).

Means End Chains

Documents

Transcript of Means End Chains