Cardiff University IP: 131.251.133.26 On

0 downloads 0 Views 235KB Size Report
There is no doubt that the essential features of the BPM can be found in .... A mobile phone not only provides communications services when ..... at the bank.
Journal of Customer Behaviour, 2002, 1, 19-48

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Gordon R. Foxall1

Cardiff University

Marketing’s Attitude Problem – and How to Solve It Marketing’s attitude problem stems from the view – which has traditionally underlain research throughout the social sciences – that behaviour can be predicted from measures of beliefs, attitudes and intentions regardless of situational factors. The demonstration that such measures only rarely predict action led to the search for situationally relevant factors that would strengthen predictions and restore the expected link between attitudes and behaviours. Although recent success in predicting behaviour from attitudes and intentions is the result of researchers’ giving greater attention to the influence of situations, there has hitherto been no systematic account of contextual influences on attitude—intention—behaviour relationships. However, empirical research in England and South America indicates that attitudes known to predict consumer behaviours are themselves predictable on the basis of contextual variables, notably the utilitarian and symbolic consequences of behaviour and the scope of the setting in which it takes place. The research suggests a resolution of marketing’s attitude problem and paves the way for an alternative understanding of both consumer and marketer behaviours as situationallydetermined. The article discusses the theoretical and managerial implications of the research and points the way towards further investigation.

Introduction Marketing’s attitude problem stems from the view that consumer behaviour can be predicted from measures of beliefs, attitudes and intentions regardless of situational factors.This expectation, which has underlain research in social psychology and sociology as well as marketing for decades, is problematic for two reasons. The first is that the empirical evidence simply does not bear it out (Davies et al. 2002). The second is that although attitude research has 1

Correspondence: Professor Gordon R. Foxall, Cardiff Business School, Cardiff University, Aberconway Building, Colum Drive, Cardiff CF10 3EU, Wales, UK Email: [email protected]

ISSN1477-6421/2002/0100019+29 £8.00/0

©Westburn Publishers Ltd.

20

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

improved dramatically in predictive power in recent years, notably through the incorporation of situational variables, there is no systematic account in marketing thought or practice to show how situations are implicated in the formation of attitudes and how attitudes can predict and explain actions. If this appears remote from everyday marketing experience, something that can be safely left to psychologists while practical people get on with the business of life, consider how pervasive is marketing’s dependency on the concept of attitude. Persuasive marketing communications are everywhere aimed towards changing or strengthening people’s evaluations of the brands they have bought or are likely to buy. Distribution strategies are directed towards the creation of retail images that are similarly based on ambient responses to stores, restaurants, airport lounges, and other channels that deliver not just goods but life-enhancing consumption experiences. In affluent, consumer-orientated economies, price is widely employed as a creative marketing variable: no longer communicating simply the financial burden that consumers will have to bear but augmenting the impression of quality provided by an integrated marketing mix. And the product or service itself, of course, is a skilfully-constructed bundle of attributes that reflects the ‘emotionally coloured points of view’ by which Katona (1960: 55) described consumers’ attitudes. The fact is that there is scarcely any element of either academic or practical marketing that is not closely bound up with the concept of attitude and, in particular, with the expectation that attitudes prefigure, predict and cause consumer behaviours. The challenge for marketing research is to show how, in the face of so much contradictory empirical evidence, this foundational assumption of marketing thought and practice can still be relied upon. Answering this challenge is not merely an intellectual exercise: it is one on which the credibility and practicality of managerial marketing vitally depends. It requires not just a series of isolated empirical results but an integrated framework that makes sound prediction possible and, through it, reliable managerial prescription. The suspicion that marketing research must systematically incorporate situational variables in order to predict behaviour (Foxall 1984a, 1984b) inaugurated a research programme, described in summary in this article, that is now bearing fruit for both marketing theory and managerial practice. This article addresses the question of how these factors can be incorporated into a model of consumers’ attitudes and behaviour that takes situational influences properly into consideration. The notion that attitudes provide straightforward explanations and predictions of behaviour is beguiling. After all, attitudes are generally defined as predispositions to behave in particular pro or contrary ways to given objects (Foxall, Goldsmith and Brown 1998) – the task seems only to

Marketing’s Attitude Problem

21

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

find a means of measuring attitudes sensitively in order to have no difficulty in forecasting behavioural outcomes. That task preoccupied social psychologists in particular for many decades, beginning in the 1920s, and marketing researchers for almost as long. And not surprisingly, for attitude theory and research promise not only a ready-made explanation why human behaviour takes the forms it does but also a practical technology by which consumers’ attitudes can be tracked over time and perhaps shaped by marketing managers and other change agents. Unfortunately for this approach, a whole host of research evidence had accumulated by the end of the 1960s showing that, at best, attitudes and behaviour were only weakly related (Wicker 1969). This may come as a surprise to consumer and other applied social researchers who have trained since then, Surely, in the theories of reasoned action (Ajzen and Fishbein 1980) and planned behaviour (Ajzen 1985) we have clear demonstrations that attitudes do after all prefigure behaviour and that it is perfectly legitimate to seek both explanatory models and research techniques in the prevailing social cognition paradigm that underlies almost every consumer behaviour textbook? There is a lot riding on this view. Countless dissertations, theses, and journal articles rely absolutely on the supposition of an attitude intention behaviour sequence; and a whole host of commercial studies rest no less surely though perhaps less formally on the same underlying assumption. It works to some extent – students get their degrees and marketing mangers find empirical bases for advertising campaigns and new product development programmes. But our systematic knowledge of how attitudes are related to behaviour is nowhere as certain as these pragmatic dependencies suggest (Foxall 1983). To the extent that progress has been made since then in accurately predicting consumer behaviour from these and other cognitive measures, it has come through the incorporation of direct and indirect measures of situational influence on consumer choice (Foxall 1997a, 1997b). Yet there is no generally-accepted model of consumer behaviour that relates it to the situational influences that shape it. There is no systematic understanding of how attitude—behaviour consistency is related to these situational pressures, nor any methodological basis for understanding consumer choice as situationally determined or marketing management as dependent upon such understanding. This article pursues these themes and offers a solution. It first explores the expectation that attitudes would be related to behaviour and shows why it was thwarted. It goes on to show how recent attempts to relate attitudes and actions have implicitly incorporated measures of the two situational variables on which a situationally-based model of consumer must be based: the consumption history of the buyer and the elements of the physical and social setting in which consumer behaviour takes place. These variables are

22

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

combined into a model of the consumer situation from which a typology of consumer situations is derived. The model has been tested in England and Venezuela in terms of its capacity to predict attitudes known to relate to consumer behaviours in a variety of situational contexts. Before we go further, let us remind ourselves what we mean by attitude. While there are numerous concepts and definitions of attitude, they have in common the affective side of human reactions, how people feel about a particular object, person or other entity, their favourable or unfavourable emotional or evaluative response to it. Although individuals’ physiological responses can sometimes be measured to gain some idea of their attitudinal feelings, the usual way of assessing attitudes is to ask people for their feelings and to record their verbal responses. The pencil and paper tests that are at the heart of attitude research are therefore an attempt to gauge people’s underlying attitudes, evaluations, emotional reactions second hand from what they are able to say about their feelings.

Attitude Theory and Research Current emphases in attitude theory and research stem from the dire assessments of the evidence for attitude-behaviour consistency published in the late 1960s and early 1970s. Wicker (1969) showed that early attitude research (from the 1930s to the 1960s) generally revealed positive but insipid relationships between attitudes and behaviours. The extent to which attitudinal variance accounted for behavioural variance was small indeed. Rather, as Fishbein (1972) noted, the evidence favoured the prediction of attitudes from behaviour rather than the expected position. More recent work has generated much higher correlations. The accurate prediction of behaviour requires an understanding of why this change has come about. This is not an easy task since, despite the increasing cognitivist rhetoric of those attempting to explain the results of such work, the underlying methodologies are increasingly behavioural. Among several approaches to the attitude—behaviour problem, Fishbein’s work has proved especially productive of solutions and is probably the current basis of most attitude research in marketing and related areas (Fishbein and Ajzen 1975; Ajzen and Fishbein 1980). In a judgment that was to foreshadow the role of specificity in future research, Fishbein registered that the relationships probed by earlier research were generally between global measures of attitude towards an object and very particular indices of behaviour towards the object. Fishbein’s approach had other simplifying features. Wicker had argued that the prediction of behaviour required that consideration be given to numerous additional variables including personal factors (other attitudes, competing motives, activity levels) and situational factors (presence of others,

Marketing’s Attitude Problem

23

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

normative prescriptions, specificity of attitude objects, the expected and/or actual consequences of various acts). Incorporating such a wide diversity of ‘other variables’ would require a model of immense complexity. Fishbein’s emerging approach concentrated on finding the summary variables that would prove most predictive of behaviour: ‘Rather than viewing attitude toward a stimulus object as a major determinant of behavior with respect to that object, the theory identifies three kinds of variables that function as the basic determinants of behavior: (1) attitude toward behavior, (2) normative beliefs (both personal and social), and (3) motivation to comply with the norms’. (Fishbein 1967: 490)

This formulation became the mainstay of the well-known Theory of Reasoned Action (Fishbein and Ajzen (1975) which is relevant to behaviour that is under the volitional control of the actor (Figure 1). Ajzen’s Theory of Planned Behaviour (Figure 2), specifically adapted to situations in which behaviour is not entirely under voluntary control, subsequently added a further cognitive variable, perceived behavioural control, to account for the extent to which the respondent felt able to undertake the behaviour under investigation in order to achieve a goal (losing weight or taking exercise, for instance). As Figure 2 shows, perceived behavioural control may act both directly on behaviour or indirectly via behavioural intention. These theories have generated predictions of behaviour, including consumer behaviour, which are considerably more accurate than those reported in the studies reviewed by Wicker. Correlations between behavioural intentions and behaviour of 0.8 and 0.9 are by no means exceptional; meta-analyses and literature reviews have shown that relatively high correlations are a general result of employing these methodologies (Ajzen 1991; Sheppard et al. 1988; Van den Putte 1993). This improvement in results can be traced to attitude researchers’ having incorporated into their models the awareness of situational variables which Wicker argued was essential. Specifically, two situational factors have been taken into account by the theories of Fishbein and Ajzen – known as ‘deliberative theories’ since they assume that prior to behaving in a particular way, the individual reasons or deliberates in his or her actions and their consequences – which provide important clues to the development and testing of a comprehensive model of consumers’ attitudes and behaviour.

24

Gordon R. Foxall

Attitude towards the act

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Behavioural intention

Behaviour

Subjective norm

Figure 1. Outline of the Theory of Reasoned Action Attitude towards the act

Subjective norm

Behavioural intention

Behaviour

Perceived behavioural control

Figure 2 . Outline of the Theory of Planned Behaviour These two factors are situational correspondence among the cognitive and behavioural measures included in the model, and the learning history of the respondent, his or her previous behaviour with respect to the attitude object and its positive and negative, rewarding and punishing, consequences. As a prelude

Marketing’s Attitude Problem

25

to developing and testing the required model, it will be useful to trace Fishbein and Ajzen’s use of these essential constructs in greater detail.

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Situational Correspondence The deliberative theory approach indicates that in order to obtain high correlational consistency among attitude, intention and behaviour these variables must be measured at identical levels of situational specificity. The first source of evidence concerns measurement specificity. The systematic processing group of attitude theories revolves around the belief that the degrees of specificity with which attitudinal and behavioural measures are each defined must be identical if high correlations are to be found between them. Generic attitude measures are therefore consistent with multiple-act measures of behaviour towards the attitude object. It follows that the prediction of single acts is only likely to result from equally narrow measures of attitude, those that correspond exactly in level of specificity to the act to be predicted; those, moreover, that are framed as measures of the respondent’s attitude towards performing that act in closely designated circumstances (Fishbein and Ajzen 1975; Ajzen and Fishbein 1980). A second source of evidence is the quest for setting correspondence. Ajzen and Fishbein’s (1977) analysis of numerous studies of attitudinal-behavioural consistency revealed that high correlations are probable only when the measures of attitude and behaviour coincide with reference to the precise action to be performed, the target towards which the action is to be directed, the context in which the action would occur, and its timing (usually summarised in the acronym TACT). Evidence is finally available from the insistence on temporal contiguity. An important recognition was that measures of the cognitive precursors of attitude will be highly predictive only when there is maximal temporal contiguity of the behavioural and antecedent measures (Ajzen and Fishbein 1980). The greater the temporal gap between attitude or intention and the behaviour to which they refer, and hence the extent of situational intervention that potentially separates them, the lower will be their correlative consistency. It is the intention which immediately precedes behaviour that is predictive (Ericsson and Simon 1993). The implication of the tight situational compatibility required of measures of target behaviour and measures of its antecedent cognitive predictors (Ajzen and Fishbein 1977) is that situational factors are highly significant for the correlational consistency of attitudes/intentions and behaviour. Only when the situational influences governing both the prebehavioural and the behavioural variables are functionally equivalent are high correlations found. That the intertemporal period between prebehavioural and behavioural measures must be minimal if high correlations are to be found corroborates this view by pointing to the undesirability of unexpected situational demands reducing the predictive value of measured intentions.

26

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Context and situation deserve a more central place in the explanation of behaviour which is denied them by the partiality inherent in acceptance of the preeminence of the cognitive paradigm. The explanatory power of past behaviour is frequently sufficient to make cognitive variables superfluous; that a behaviour analytic theory may be capable of explaining or interpreting the evidence on attitudinal-behavioural consistency in full; and that in any case the reason for including a behaviourist perspective is to identify the consequences that past behaviour has produced to account for the consistency of that prior responding and thus to use those consequences to predict future behaviour. Behavioural History The individual’s behavioural history – what he or she has done in the past with respect to the attitude object and the consequences of such behaviour – is captured in the attitudinal variable (Aact) which measures what the respondent thinks the consequences of behaving in a particular way vis-à-vis the attitude object will be and his or her evaluation of those consequences. It is clear that the formulation of an attitude towards an act depends on having had prior experience with the attitude object and having prior experience of the consequences of such behaviour: it is a behavioural outcome. Perceived behavioural control (PBC) is a similar variable that must be strongly influenced as it gains potency by experience of self-efficacy (Bandura 1997). The importance of prior behaviour is also shown by the fact that this variable often exerts a main effect on behaviour in its own right (Bagozzi and Warshaw 1990; for a critical review see Foxall 1997a). The cognitive bias of most attitude research does not encourage researchers to investigate why past behaviour is a primary predictor of current and future action – a more behavioural approach would attribute its effects to the rewarding and punishing consequences of that prior behaviour (Foxall 1997b). Fishbein and Ajzen’s approaches are more ready to find expression of such consequences through respondents’ attitude statements.

The Development of a Comprehensive Model It is not enough to have identified the factors that have made the prediction of behaviour more accurate, notably situational correspondence and behavioural history. The evidence that these methodological refinements increase the precision of our explanations of behaviour lies scattered in the reports of dozens if not hundreds of empirical investigations which have taken them deliberately but unsystematically intro consideration. There is no guarantee that further attempts to do so will automatically succeed; nor – and this is vitally important in an applied field such as marketing – that the results of research based hopefully but haphazardly on the inclusion of these factors

Marketing’s Attitude Problem

27

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

will yield productive applications. ‘Situational correspondence’ and ‘behavioural history’ are useful terms in so far as they summarise the methodological improvements introduced by deliberative attitude theory and research. But they are scarcely a guide to action for either marketing research or marketing management. The question is how they relate to the realities of consumer situations in marketing-orientated economies and it must be answered in a way that makes these factors concrete basis for research and action. For this we require a more formal representation is required that relates situations, attitudes and behaviour reliably. The Behavioural Perspective Model (BPM) is an attempt to incorporate the situational variables that recent attitude theory has only implicitly recognized and to do so in such a way that consumers’ attitudes and behaviours can be predicted and understood. Since the derivation and refinement of the model has been described in some detail (Foxall 1990, 1993. 1994, 1996, 1997b, 1998a, 2002), the following account is designed to be succinct. A Model of the Consumer Situation The numerous ways of deriving theories of consumer behaviour differ according to the extent to which they rely on description of marketing phenomena in themselves and the extent to which they include concepts and measures borrowed from the social sciences. This is no place to debate the pros and cons of these avenues to our greater understanding of consumer behaviour and marketer response, but it is likely that none has an absolute advantage over the others. The selection of one or other depends largely upon the purposes of the investigator. However, a relative advantage of models built on some systematic knowledge base such as is provided by psychology is that the reasoning on which the model rests and the findings by which it is evaluated can be evaluated according to pre-existing, tried and tested canons of judgement. There is no doubt that the essential features of the BPM can be found in simple description of consumer behaviour and the marketing system: the capacity to specify the model in such terms is vital to its relevance to its relevance to marketing (Foxall 1997a). That it is developed in the context of behavioural psychology has the additional benefit, however, that the nature and scope of the resulting theory can be gauged and the findings can be related to those produced in differing contexts by other researchers. Behavioural psychology relates the rate at which a behaviour is performed to the consequences it has attracted in the past: some consequences result in the behaviour or response becoming more frequent and are known as positive rewards or reinforcement; others decrease the probability of the response and are known as punishers. Because such behaviour is conceptualised as operating upon the environment to produce

28

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

consequences, it is known as operant behaviour, the process in which the consequences come to influence the behaviour as operant conditioning, and the behavioural psychology which studies this process as operant psychology. Formally, the rate of response (R) is seen as contingent on the nature of the consequences it generates, positive reinforcers (Sr) or aversive stimuli (Sa): R

Sr/a.

In addition, some prebehavioural stimuli signal the likelihood of positive or aversive consequences arising as a result of performing a particular behavioural act. These ‘discriminative’ stimuli (Sd) can be regarded as signals or cues. The full paradigm then becomes Sd

R

Sr/a.

The capacity of so simple a formulation to account for complex human behaviour such as consumer brand choice and marketer behaviour is surprising perhaps but real for all that as a range of theoretical and empirical work attests (Barnes-Holmes et al 2000; Berry 1968, 1969; Crowell et al. 1988; Foxall 1998b,1999a; Foxall and James, in press a, b; Hantula et al. 2001; Rajala and Hantula 2000; Green and Neistat 1983; Green et al 1984; Kunkel and Berry 1970; for a critical review see Foxall 2001). The BPM is an elaboration of this formulation, based on the descriptive core of marketing studies that portrays the realities of consumer choice and managerial marketing. The Consumer Situation The central explanatory component of the BPM is the consumer situation which exerts a direct influence on the shaping and maintenance of consumer behaviour in specified surroundings (Figure 3). The consumer situation is defined as the intersection of the consumer behaviour setting and the consumer’s; learning history. A consumer behaviour setting comprises the stimuli that form the social (including regulatory) and physical (including temporal) environment. These initially neutral stimuli are transformed into the discriminative stimuli that signal the probable outcomes of approach and avoidance responses in the setting by their intersection with the consumer’s pertinent history of reinforcement and punishment. It is this learning history that adds meaning to the otherwise neutral setting stimuli by investing them with the consequences of previous approach-avoidance behaviours in similar circumstances.

Marketing’s Attitude Problem Consumer behaviour setting

Utilitarian reinforcement

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

X

Consumer’s learning history

29

Consumer behaviour

Aversive consequences

Symbolic reinforcement

Figure 3. Outline of the Behavioural Perspective Model The resulting discriminative stimuli define the scope of the consumer behaviour setting, its capacity to facilitate or inhibit consumer responses such as browsing, choosing, and purchasing (approach) or delaying, deferring, and leaving the setting without purchasing (avoidance). A relatively open consumer behaviour setting is one in which several responses are available to the consumer who have discretion over which is chosen; behaviour under these circumstances may take several forms and its topography is difficult to predict. A relatively closed setting is one in which the consumer is denied such wider discretion; the consumer’s behaviour is determined by agents (e.g. retail designers) who control the setting but are not themselves subject to its contingencies. Such behaviour is relatively prescribed and is easier to predict. Classes of Consumer Behaviour The consequences signalled by the discriminative stimuli that compose the consumer situation are of three kinds: utilitarian reinforcement, symbolic reinforcement, and punishment. Utilitarian reinforcement consists in the tangible functional and economic benefits which stem from purchase, ownership and consumption. The driver of a Lada, for instance, is principally concerned with the utilitarian benefits that all cars provide: the most obvious is getting from A to B, door-to-door transportation. Symbolic reinforcement, on the other hand, is more likely to involve a lifestyle statement by which the consumer seeks to convey his or her social status or to bolster esteem and/or reported feelings of self-esteem. The driver of a Mercedes or a Bentley or a Porsche, clearly gets from A to B in it but, in addition, gains the social esteem and status provided by friends and

30

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

acquaintances who admire these prestige products and from members of the general public who see them driving around in a socially-desirable vehicle. The social status and esteem that driver is accorded are the symbolic rewards of consumption. Most products have an element of both the instrumental and the symbolic. A mobile phone not only provides communications services when and where the consumer wants them; because it is a Nokia and therefore has interchangeable coloured cases, it may also signal to that consumer’s social group that he or she is cool. Similarly, a Harley-Davidson motor cycle not only provides fast transportation: it is also the basic means of belonging to a group of bikers. And so on. Finally, punishing or aversive consequences – in everyday language, costs – are an inevitable outcome of consumer behaviour which always meets with results that tend to diminish its rate of future enactment as well as reinforcers which tend towards increasing this rate. Four operant classes of consumer behaviour can be theoretically distinguished depending on the pattern of relatively high/relatively low utilitarian reinforcement and relatively high/relatively low symbolic reinforcement which maintains the responses of which these classes are composed (Figure 4). Accomplishment is consumer behaviour maintained by relatively high levels of both utilitarian and symbolic reinforcement; hedonism, by relatively high utilitarian and relatively low symbolic reinforcement; accumulation, by relatively low utilitarian and relatively high symbolic reinforcement; and maintenance, by relatively low levels of both. High Utilitarian Reinforcement

Low Utilitarian Reinforcement

High symbolic reinforcement

ACCOMPLISHMENT

ACCUMULATION

Low symbolic reinforcement

HEDONISM

MAINTENANCE

Figure 4. Classes of Consumer Behaviour The Eightfold Way Adding the dimension of consumer behaviour setting scope to this operant classification of consumer behaviour gives the eightfold categorization of the contingencies that may control consumer behaviour

Marketing’s Attitude Problem

31

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

shown in Figure 5. Each of these eight contingency categories may accommodate numerous functionally-defined consumer situations in which behaviour is maintained by a specified matrix of structural factors. In so far as consumer choice is understood as functionally determined by the environment, Figure 5 proposes an exhaustive categorization of such contingencies. Consumer behaviour is expected to vary depending on these structural components of the consumer situation. The derivation of the labels which Figure 5 gives these eight contingency categories is discussed further in Foxall (1990).

Behaviour-Setting Scope Closed Open 2 ACCOMPLISHMENT Fulfilment Status Consumption 4 Inescapable Popular HEDONISM entertainment entertainment 6 Token-based ACCUMULATION consumption

3

5 Saving and collecting

8 Mandatory MAINTENANCE consumption

1

7 Routine purchasing

Figure 5. The BPM Contingency Matrix Testing The Model Rationale Empirical tests of this model in the context of consumers’ attitudes consist in the prediction of consumers’ verbal responses to descriptions of consumer situations representative of each of the eight feasible categories of environmental contingencies shown in Figure 5. (The testing of the model in this way has been the subject of several papers which have reported in detail the rationale and methodology of the investigations in England and South America, commenting in detail on the theoretical and applied significance of the findings: see Foxall 1997; Foxall and Greenley 1998, 1999a, 2000; Foxall and Soriano, in press; Soriano and Foxall, in press, a, b). The range of verbal

32

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

responses is suggested by Mehrabian and Russell’s (1974) verbal measures of the three affective reposes to environments which they argue are exhaustive: pleasure, arousal and dominance. This empirical approach derives from work on the classical conditioning of attitudes (Staats and Staats 1958; Lott and Lott 1968). Classical conditioning is the process in which a neutral stimulus (such as a metronome) is repeatedly paired with a stimulus (say, food) that naturally elicits an unconditioned response (UR, salivation). Over time the metronome acquires similar eliciting properties to those of the unconditioned stimulus (US, food in this case). The neutral stimulus becomes a conditioned stimulus (CS) capable of eliciting a conditioned response (CR) which is similar in form to the original UR (Eagly and Chaiken 1993). Hence CS (metronome) US (food)

CR (salivation) UR (salivation)

The same conditioning principle applies to the production of covert responses such as the affective reactions we call attitudes. A positive attitudinal reaction to a brand name might be conditioned, for instance, by pairing it in advertisements with an adjective that already elicits a favourable emotional response: CS (brand name) US (positive adjective)

re (implicit positive evaluation) re (implicit positive evaluation)

Mehrabian and Russell assess pleasure in terms of the environment’s being: happy as opposed to unhappy; pleased as opposed to annoyed; satisfied as opposed to unsatisfied; contented as opposed to melancholic; hopeful as opposed to despairing; and relaxed as opposed to bored. Arousal is assessed by verbal reactions to an environment which show the respondent to be: stimulated as opposed to relaxed; excited as opposed to calm; frenzied as opposed to sluggish; jittery as opposed to dull; wide-awake as opposed to sleepy; and aroused as opposed to unaroused. Finally, dominance is reflected in verbal appraisals in which the respondent feels: controlling as opposed to controlled; influential as opposed to influenced; in control as opposed to cared-for; important as opposed to awed; dominant as opposed to submissive; autonomous as opposed to guided. These three responses mediate more overt consumer behaviours such as a desire to affiliate with others in the setting, desire to stay in or escape from the setting, and willingness to spend money and consume (Mehrabian 1979; Mehrabian and Riccioni 1986; Mehrabian and de Wetter 1987; Mehrabian and Russell 1975; Donovan and Rossiter 1982; Donovan et al. 1994; Russell and Mehrabian 1976, 1978). Russell and Mehrabian (1976) argue that desire to purchase increases with the pleasantness of the setting and, since arousal has

Marketing’s Attitude Problem

33

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

a curvilinear relationship with approach behaviour, that such desire is maximised in settings which evoke an intermediate level of arousal. These verbal descriptions of affective response and overt behaviours are predictable from the structural dimensions of consumer situations as defined by the BPM. The strength of reported pleasure, for instance, is an index of the utilitarian reinforcement signalled as contingent upon approach behaviour by the discriminative stimuli that make up components settings of operant classes that feature high levels of this variable. The verbal behaviour that is probable responses to the promise of such reinforcement would, by definition, reflect the economic, instrumental benefits which classical economists labelled hedonic. Four of the items composing the factor which Mehrabian and Russell label pleasure largely describe the satisfaction or utility which this term denotes: happy, pleased, satisfied and contented. The strength of reported arousal is a measure of the information rate of an environment which increases with the novelty, complexity, intensity, unfamiliarity, improbability, change, mobility and uncertainty of the setting (Mehrabian and Russell 1974). Most of these are channels of feedback from the environment to the individual; while information rate is not coterminous with symbolic reinforcement, it includes the feedback on performance in which symbolic reinforcement consists, and the verbal responses nominated arousal by Mehrabian and Russell can be expected to increase in strength as this structural dimension increases in importance. Finally, dominance indicates the verbal responses whose strength will predictably increase with the degree of openness of the consumer behaviour setting. The consumer who reports feeling controlling, influential, in control, important, dominant and autonomous is likely to be found in an open as opposed to a closed setting.

Hypotheses The first three hypotheses accordingly examine the probability that pleasure will be higher for consumer behaviours defined in terms of relatively high utilitarian reinforcement; arousal will be higher for consumer behaviours defined in terms of relatively high symbolic reinforcement; and dominance will be higher for consumer behaviours enacted in relatively open settings. Hence: H1: Pleasure will be higher for accomplishment and hedonism than for accumulation and maintenance. H2: Arousal will be higher for accomplishment and accumulation than for hedonism and maintenance. H3: Dominance will be higher for the consumer behaviours occurring in open setting than for those occurring in closed settings.

34

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

These three hypotheses tested are summarised in Figure 6 which depicts the eight distinct patterns of emotional response made feasible by the combination of pleasure, arousal and dominance and, in line with the theoretical reasoning we have advanced, uniquely associates each pattern with a specific contingency category (CC). Mean scores for each emotional variable which are depicted in upper case letters are expected to exceed mean scores for the same variable shown in lower case. In the case of pleasure, for instance, the prediction is that mean scores for each of CCs 1, 2, 3 and 4 (Accomplishment and Hedonism) will be significantly higher than each of the mean scores for CCs 4, 5, 6 and 7 (Accumulation and Maintenance). The hypotheses relating to arousal and dominance are similarly delineated in Figure 6 by means of upper and lower case letters.

Behaviour-Setting Closed Open PLEASURE AROUSAL dominance PLEASURE HEDONISM arousal dominance pleasure ACCUMULATION AROUSAL dominance Pleasure MAINTENANCE arousal dominance

ACCOMPLISHMENT

PLEASURE AROUSAL DOMINANCE PLEASURE arousal DOMINANCE pleasure AROUSAL DOMINANCE pleasure arousal DOMINANCE

Figure 6. Summary of Hypotheses for Affective Variables Six additional hypotheses were tested in connection with the expected levels of approach and avoidance characteristic of each of the operant classes of consumer behaviour, and of open as opposed to closed consumer behaviour settings. Approach-avoidance was measured by Mehrabian and Russell (1974, 1978) as a single, bidimensional construct. However, it is unlikely that approach and avoidance form a single dimension of consumer behaviour. Rather, consumer behaviour in a particular situation is the outcome of two opposing learning histories: the strength of approach is a function of the individual’s learning history with respect to prior approach behaviour and their consequences, while the strength of avoidance/escape is a function of his or her history with respect to prior avoidance/escape responses and their consequences (Alhadeff 1982). The effective strength of consumer behaviour

Marketing’s Attitude Problem

35

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

is, therefore, a vector quantity, each of whose components should be separately measured and assessed in relation to the structure of the situations in which it occurs. The following hypotheses, therefore, refer to approach and avoidance separately, as well as to a composite measure, aminusa, which captures Mehrabian and Russell’s single-variable approachavoidance, which was included for purposes of comparison. In that behaviour would be expected to increase with the total quantity and quality of reinforcement available to reinforce it, approach was expected to be higher for Accomplishment, which is characterised by relatively high utilitarian and relatively high symbolic reinforcement than for other operant consumer behaviour classes. On this basis Maintenance would exhibit the lowest level of approach, and the remaining operant classes, Hedonism and Accumulation, intermediate levels. Hence: H4: Approach will be highest for accomplishment, followed by hedonism and accumulation, followed by maintenance.

Avoidance was expected to exhibit the converse pattern: H5: Avoidance will be highest for maintenance, followed by hedonism and accumulation, followed by accomplishment.

Aminusa was expected to follow a similar pattern to that of approach: H6: Aminusa will be highest for accomplishment, followed by hedonism and accumulation, followed by maintenance.

The influence of consumer behaviour setting scope was predicted as follows: H7: Approach will be higher for open (as opposed to closed) consumer behaviour settings. H8: Avoidance will be higher for closed (as opposed to open) consumer behaviour settings. H9: Aminusa will be higher for open (as opposed to closed consumer behaviour settings.

Method In England, 561 consumers rated eight consumer situations derived from the model on Mehrabian and Russell’s measures. The four sets of consumer situations which were used are illustrated here by a single set (Table 1): for the complete set of consumer situations, see Foxall (1997c). The reasoning

36

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

supporting the selection of these consumer situations as representative of the theoretical range of low/high utilitarian reinforcement, low/high symbolic reinforcement and the closed—open continuum can be found in Foxall (1990) and the research on their validity and reliability is reported in Foxall (1999b). Data were gathered at some 60 sample points in the North, Midlands and South of England. Respondents were both male and female in approximately 1:1 ratio and were aged from 20 to 50 years. They belonged predominantly to ABC1 occupational gradings.

Table 1. Sample Consumer Situations from the English Study Contingency Category 1

Consumer Situation Luxury shopping

2

Gambling in Casino

3

Watching tv

4

Inflight entertainment

5

Saving up

6

Frequent flier scheme

7

Grocery shopping Paying taxes

8

Description You are wandering from department to department in a store such as Harrods, looking for an expensive treat for yourself which you feel you deserve and which you can well afford. You are playing roulette in an exclusive casino. Many people around you are gambling and enjoying themselves. You are watching a fast-moving entertainment program on TV: a sports program, a quiz show, a soap -- whatever you often watch. You use your remote control to switch channels to see similar shows. You are on a transatlantic flight, travelling economy class. You are reading an interesting book. The flight attendants close the blinds, subdue the lighting and announce that a movie is about to be shown. You are saving up to buy a major item. Each week you deposit cash in your savings account. You have just received notice of the amount of interest to be added to your account. You have just bought a number of items which you chose specifically because they confer frequent-flier points. You make a note of how close you are to getting your goal of a free flight. You are doing your weekly grocery shopping in a large supermarket. You are comparing the ex-VAT price with what you must actually pay for a consumer durable such as a home computer.

Marketing’s Attitude Problem

37

Table 2. Consumer Situations: Venezuelan Studies CC* 1

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

2

3

4

5

6

7

8

Sub-study 1

Study 1

You are on holiday, let’s say, on a Mediterranean cruise or on a tour of the Fijian Islands in the Pacific Ocean. You are on a training course as part of your job or to learn a new technique (for your job or for a hobby). This involves attendance at class and at practical sessions. You are at a party. You hear your favourite music. Around you, people are talking in a lively manner.

You are at the cinema watching documentaries and advertisements while you wait for the beginning of the film you went to see. You are flipping though the latest issue of a magazine collection that you buy every month (the collection will turn into an encyclopedia or manual on something that interests you).

You are using the credit cards from a particular bank because they give you points for free travel (for every 2,000 Bolivares charged to the card, you get one point; and after reaching a certain number of points you will be able to exchange them for air tickets). You check on how many points you have and how many you still need to get the trip that you would like. You are doing your weekly shopping at a large supermarket. You go round the supermarket with your cart, placing products in it. You are waiting at an airport terminal for your flight to leave. You know that you are going to be at the terminal for a good while.

Sub-study 2

Study 2

You are showing off your new Mercedes Benz sports car to your family and friends. You are playing roulette in an exclusive casino. Around you, there are a lot of people playing and enjoying themselves.

You are watching an entertaining program on TV: a sportscast, a game show, a soap opera, or whatever program you generally watch regularly. You use the remote control to change channels and see similar programs. While you wait on the phone, you are listening to background music.

You are saving to buy something quite important. Each fortnight you deposit money in a special savings account that earns interest at 5% more than the inflation rate and is tax-free. You have just received a credit note from the bank stating the amount of interest that will be added to your savings. You are collecting “loyalty points” when you buy at a certain supermarket (when you reach a certain number of points, you will have the right to exchange them for products, or request a discount, for the equivalent amount of Bolívares, on your next purchase).

On the way to work, you call at the newsstand to buy a newspaper, just as you do every day. You are waiting in line to deposit a check at the bank.

*CC = Contingency Category The Venezuelan sample consisted of 260 consumers. Of the 32 consumer

38

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

situations used in the English study, 16 could be tested readily in Venezuela (Table 2). Other items such as listening to buskers while queuing for the cinema and paying into a Christmas club, used with some of the English respondents but not shown in Table 1, would have been unfamiliar consumer situations to Venezuelans among whom even DIY jobs are not popular activities. The socio-economic and demographic characteristics of the sample matched those of the English sample. In both national contexts, the respondents rated each of eight consumer situations using the Mehrabian-Russell model to measure their emotional reactions and behavioural responses. Each person was asked to rate all eight consumer situations. In the case of the measurement of affect, Mehrabian and Russell’s (1974) scales for the measurement of pleasure, arousal and dominance were used without modification. Approach and avoidance were measured by six of Mehrabian and Russell’s eight statements for these items (those to do with thinking out a difficult task and working in the situation were felt to be inappropriate to consumer behaviour). The six statements selected were, for approach: ’How much time would you like to spend in this situation?’ ’Once in this situation, how much would you enjoy exploring around?’ and ’To what extent is this a situation in which you would feel friendly and talkative to a stranger who happens to be near you?’ and for avoidance: ’How much would you try to leave or get out of this situation?’ ’How much would you try to avoid any looking around or exploration in this situation?’ and ’Is this a situation in which you might try to avoid other people, avoid having to talk to them?’

Results Because the results have already been published in detail (see Foxall 1997c; Foxall and Soriano 2002; Soriano and Foxall 2002b), this section simply summarises them in graphical form. The point is not to present the results of the individual studies but to emphasise that in combination they support the hypotheses tested and underpin the novel approach to the problem of reconciling consumers’ attitudes and behaviour which is our theme. As Figure 7 summarises, pleasure means were found to be significantly higher for consumer situations maintained, according to the BPM, by high levels of utilitarian reinforcement than for those maintained by relatively low levels of utilitarian reinforcement: i.e., for Accomplishment and Hedonism compared with Accumulation and Maintenance. Our studies indicate that each of the pleasure means for CCs 1, 2, 3 and 4 significantly exceeds each of the pleasure means for CCs 5, 6, 7 and 8.

Marketing’s Attitude Problem

39

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Behaviour-setting scope Closed Open ACCOMPLISHMENT

PLEASURE

PLEASURE

HEDONISM

PLEASURE

PLEASURE

ACCUMULATION

pleasure

pleasure

MAINTENANCE

pleasure

pleasure

Figure 7. Summary of the Results for Pleasure The arousal means proved significantly higher in those operant classes of consumer behaviour theoretically characterised by relatively high levels of symbolic reinforcement, namely Accomplishment and Accumulation, than in Hedonism and Maintenance. The results are summarised in Figure 8. Pleasure means for CCs 1, 2 5 and 6, taken individually, significantly exceed each of the pleasure means for CCs 3, 4, 7 and 8.

Closed ACCOMPLISHMENT HEDONISM ACCUMULATION MAINTENANCE

Behaviour-setting scope Open

AROUSAL

AROUSAL

Arousal

Arousal

AROUSAL

AROUSAL

arousal

Arousal

Figure 8. Summary of the Results for Arousal Behaviour-setting Closed Open ACCOMPLISHMENT HEDONISM ACCUMULATION MAINTENANCE

dominance

DOMINANCE

dominance

DOMINANCE

dominance

DOMINANCE

dominance

DOMINANCE

Figure 9. Summary of the Results for Dominance

40

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Finally as far as affect is concerned, as Figure 9 sums up, the dominance were significantly higher in consumer behaviour settings which according to the theory are relatively open than in those which are relatively closed. Dominance means for CCs 1, 3, 5 and 7 individually exceed those for CCs 2, 4, 6 and 8 at the conventional level of statistical significance. The behavioural variables also formed the expected patterns: each of the approach means for accomplishment and hedonism exceeds each of those for accumulation and maintenance; each of the avoidance means for accumulation and maintenance exceeds each of those for accomplishment and hedonism; and each of the aminusa means for accomplishment and hedonism exceeds each of those for accumulation and maintenance. These results are graphically summarised in Figure 10. Behaviour-setting scope Closed Open ACCOMPLISHMENT

HEDONISM

ACCUMULATION

MAINTENANCE

APPROACH avoidance AMINUSA APPROACH avoidance AMINUSA Approach AVOIDANCE aminusa Approach AVOIDANCE aminusa

APPROACH avoidance AMINUSA APPROACH avoidance AMINUSA approach AVOIDANCE aminusa approach AVOIDANCE aminusa

Figure 10. Summary of the Results for Behaviour(1) Behaviour-setting scope Closed Open ACCOMPLISHMENT

HEDONISM

ACCUMULATION

MAINTENANCE

approach AVOIDANCE Aminusa approach AVOIDANCE Aminusa approach AVOIDANCE Aminusa approach AVOIDANCE Aminusa

Figure 11. Summary of the Results for Behaviour (2)

APPROACH avoidance AMINUSA APPROACH avoidance AMINUSA APPROACH avoidance AMINUSA APPROACH avoidance AMINUSA

Marketing’s Attitude Problem

41

In addition, the scope of the consumer behaviour setting was found to exert an influence on reported behaviour. As Figure 11 summarises, each of the approach means for the open settings significantly exceeds each of those for the closed settings; the avoidance means for the closed settings similarly significantly exceed those for the open settings; the aminusa means for the open setting significantly exceed those for the closed settings.

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Discussion Predicting Attitudes and Behaviour with Sensitivity to Situations The research described in this article provides hope that the problem of integrating situational and attitudinal variables in order to predict consumer behaviour is being solved. The development of an integrative model of attitude—situation—behaviour not only substantiates the view that all three are closely related but indicates how they are linked. This in turn gives confidence that managerial prescriptions based on the findings of research that supports the model will be more reliable than nostrums based on ad hoc investigations. Further research, which is inescapable in the context of the sort of research programme of which the project reported here is a key component, should also benefit from being conducted within an assimilative framework of conceptualisation and analysis. But before the implications for either managerial practice or further research can be considered, it is important to determine whether the model itself can accommodate the findings and their theoretical and methodological significance. Theoretical and Methodological Implications The BPM has been developed principally as a model of consumer choice within a research programme that aims to evaluate the capacity of operant psychology to contribute to consumer and marketing research. The underlying assumption is therefore that consumer behaviour must be portrayed as an operant response in order that this approach can be appraised and, if found successful and useful, applied. Each of the classes of consumer behaviour and each of the contingency categories involved in the research reported above has been conceptualised and constructed on the basis of reasoning derived from operant psychology. But the concept of attitude investigated has been primarily based on classical conditioning. Operant psychologists, while accepting the reality of classical conditioning, tend to de-emphasise it in favour of the operant approach and have not generally sought to integrate the two conditioning principles into a single framework of behaviour analysis. Yet it is obvious that if we are to make sense of the results of this research and to apply them we must undertake exactly that task of integration. Staats (1996) proposes just such an integration. A stimulus which is

42

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

capable of eliciting an emotional response (and which thus forms the basis of classical conditioning) is also able to function as a reinforcer (the central stimulus function in operant conditioning). In addition, the same stimulus can act as a directive (or, in operant terms, discriminative) stimulus which elicits approach or avoidance responses with respect to the attitude object. Consumers’ experiences in situations like those employed in the studies described here can thus be expected to elicit emotional responses (notably pleasure, arousal and dominance); elements of those situations, which provide utilitarian and symbolic reward for such consumer behaviours as browsing, search and evaluation, purchasing and consuming, are the stimulus elements of the behaviour setting that elicit approach and avoidance both verbal and non-verbal. Positive verbal responses are indicative of favourable attitudes towards consumer environments while negative verbal responses evidence unfavourable attitudes. The incorporation of Staats’s model into the BPM framework (Foxall and Soriano 2002) relates situational cues to attitudinal responses and attitudes to behaviour. Managerial Relevance These results are potentially of immense managerial significance in that they identify the structural components of consumer environments that relate first to utilitarian and symbolic reinforcements and secondly to the emotional reactions of pleasure, arousal and dominance that underlie consumer behaviours in, for instance, retail settings. Prior to this research, prescriptions for marketing managers based on Mehrabian and Russell’s approach consisted in somewhat vague ideas about increasing the amount of pleasure generated by sales and consumption settings, or avoiding extreme levels of arousal (e.g., Donovan et al. 1994; Russell and Mehrabian 1976). It is now possible, however, to point directly to the kinds of combination of reward which retail and other consumer environments must provide in order to engender the required emotional responses in consumers. Furthermore, although Mehrabian and Russell’s approach provides for three emotional responses, previous research has found empirical evidence for just two: pleasure and arousal. Dominance has not been substantiated by other researchers and has tended to drop away. Our investigations are the first to identify dominance as a central emotional response on the part of consumers and to relate it significantly to a comprehensive spectrum of consumer situations. Moreover, they have achieved this with consistency over a diverse range of consumer situations and for two distinct cultural and linguistically-separate cultures. Careful examination of the sources of pleasure, arousal and dominance which the examples of consumer situations employed in the research link to specific combinations of utilitarian and symbolic reinforcement and

Marketing’s Attitude Problem

43

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

consumer setting scope enables retailers and other managers of marketing outlets to construct the reward packages that will produce the optimum levels of emotional response which empirical research shows links directly to sales volumes. Some obvious universals are apparent from these examples and the findings presented here: luxurious surroundings and the lack of a narrow task orientation are closely linked to pleasure and dominance, while routine chore-like consumer behaviours involving mandatory responses are associated with low levels of all three emotional responses. Since marketing environments differ so much in terms of their peculiar structural characteristics, each retail manager would need to carry out specific research in order to make the most of the technology outlined in this article; it seems clear, nevertheless, that the theoretical and empirical work reported here has practical significance for marketing mangers which goes well beyond the prescriptions to be found in the numerous research papers that employ Mehrabian and Russell’s approach to environmental psychology or, for that matter, retailing textbooks. Further Research The pragmatic question remains whether the differences in attitudinal response captured by this response can be expected to translate in the course of marketing reality to consistent consumer responses. This, after all, is the central issue in attitude research. There are two responses to this. First, consumer behaviour, admittedly measured through self-reports, has been shown by this research to vary consistently with the attitudes expressed by the respondents. This provides no more than an initial indication of attitudinal—behavioural consistency but the revealed relationships have the merit of being entirely those predicted by the model. Secondly, there is every indication from previous consumer research which has employed Mehrabian and Russell’s approach to show that pleasure, arousal and dominance are directly related to such consumer behaviours as time spent in the consumer environment and amount of money spent there (Donovan and Rossiter 1982; Donovan et al. 1994). All of this gives confidence that the measures of attitude-in-situation we have employed are predictive of consumer behaviour. Research which can examine this proposition in detail is now the priority.

Conclusion This article has taken a novel approach to the problem of attitudinal— behavioural consistency which formally incorporates situational variables. The approach relies on an assiduous reading of the failings of early attitude theory and research which demonstrated only weak relationships between measures of attitudes and measures of behaviour but which gained markedly

44

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

in predictive value once two kinds of environmental influence were taken into account. The success of the Theories of Reasoned Action and Planned Behaviour stem from their inclusion of close situational correspondence in the cognitive and behavioural measures they employ and the behavioural history of the respondent, i.e., the consequences which similar behaviour has had for them in the past. The Behavioural Perspective Model is a formalisation of these developments. Through the adoption of Mehrabian and Russell’s work in the testing of this model, a psychometrically sound instrument has been employed to show how consumers’ emotional responses to the environments they confront are distributed. Without any of these elements, the research programme described would have been unconvincing. Careful attention to the deficiencies of attitude research, especially in the form of Wicker’s milestone review, indicated not only the sources of deficiency but also the need for a suitable model of situational influences on choice. The BPM has provided such a model and its taxonomy of situations in terms of pattern of reinforcement and behaviour setting scope has proved a successful testing ground for Mehrabian and Russell’s framework for the conceptualisation and analysis of emotional response to environments. The consequent framing of attitude as a classically-conditioned affective response provided a means of framing hypotheses by means of which both the BPM and Mehrabian and Russell’s approach could be empirically examined with positive consequences for attitude—behaviour research in marketing and psychological research. Much remains to be done. Although there is every indication from marketing research that pleasure, arousal and dominance mediate consumer spending, the current focus of research attention is on the replication of these results within the framework described here. Nevertheless, in a field of endeavour often beset by ad hoc testing of superficially derived ideas from casual observation, the theoretically informed avenue of investigation which has been the subject of this article promises much for both marketing theory and managerial practice.

References Ajzen, I. (1991), “The Theory of Planned Behavior”, Organizational Behavior and Human Decision Processes, 50, pp.179-211 Bagozzi, R. P. and Warshaw, P. R. (1990), “Trying to consume”, Journal of Consumer Research, 17, pp.127-140 Bandura, A. A. (1997), Self-Efficacy: The Exercise of Control, New York: W. H. Freeman and Co Barnes-Holmes, D., Barnes-Holmes, Y. and Smeets, P. M. (2000), “A Derived Transfer of Emotive Functions as a Means of Establishing Differential

Marketing’s Attitude Problem

45

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Preferences for Soft Drinks”, The Psychological Record Berry, L. L. (1969), “The Components of Department Store Image: A Theoretical and Empirical Analysis”, Journal of Retailing, 45. pp.3-20 Berry, L. L. (1968), Banking, Marketing and the Image Concept Arizona Business Bulletin, 15, pp. 239-44 Crowell, C. R., Anderson, D. C., Abel, D. M. and Sergio, J. P. (1988), “Task Clarification, Performance Feedback, and Social Praise: Procedures for Improving the Customer Service of Bank Tellers”, Journal of Applied Behavior Analysis, 21, pp. 65-71 Davies, J. D., Foxall, G. R. and Pallister, J. G. (2002), “Beyond the intentions behaviour mythology: an integrated model of recycling behaviour”, Marketing Theory, 2, pp. 29-113 Donovan, R. J. and Rossiter, J. R. (1982), “Store atmosphere: An experimental psychology approach”, Journal of Retailing, 58, pp. 34-57 Donovan, R. J., Rossiter, J. R., Marcoolyn, G. and Nesdale, A. (1994), “Store atmosphere and purchasing behaviour”, Journal of Retailing, 70 (3), pp. 283-294 Eagly, A. H. and Chaiken, S. (1993), The Psychology of Attitudes. Fort Worth, TX: Harcourt Brace Jovanovich. Foxall, G. R. (1983), Consumer Choice. London: Macmillan; New York: St. Martin’s Foxall, G. R. (1984a), “Consumers’ intentions and behaviour: A note on research and a challenge to researchers”, Journal of the Market Research Society, 26, pp. 231-241 Foxall, G. R. (1984b), “Predicting consumer choice in new product development: attitudes, intentions and behaviour revisited”, Marketing Intelligence and Planning, 2, pp. 37-52 Foxall, G. R. (1990), Consumer Psychology in Behavioural Perspective, London and New York: Routledge Foxall, G. R. (1993), “Situated consumer behaviour: a behavioural interpretation of purchase and consumption”, Research in Consumer Behaviour, 6, (Eds.) Costa, J. A. and Belk, R. W., Greenwich, CT: JAI Press, pp.113-52 Foxall, G. R. (1994), “Behaviour analysis and consumer psychology”, Journal of Economic Psychology, 15, pp.5-91. Foxall, G. R. (1996), Consumers in Context: The BPM Research Programme. London and New York: Routledge Foxall, G. R. (1997a), “Explaining consumer behaviour: from social cognition to environmental control”, International Review of Industrial and Organisational Psychology, 12, edited by Cooper, C. L. and Robertson, I. Chichester: Wiley. pp. 229-287 Foxall, G. R. (1997b), Marketing Psychology: The Paradigm in the Wings. London: Macmillan Foxall, G. R. (1997c), “Affective responses to consumer situations”, International

46

Gordon R. Foxall

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Review of Retail, Distribution and Consumer Research, 7, pp.191-225 Foxall, G. R,. (1998a), “Radical behaviorist interpretation: Generating and Evaluating an account of consumer behaviour”, The Behavior Analyst, 21, pp.321-354 Foxall, G. R. (1998b), “The Marketing Firm”, Journal of Strategic Marketing, 6, pp.1-20 Foxall, G. R. (1999a), “The substitutability of brands”, Managerial and Decision Economics, 20: pp.241-257 Foxall, G. R. (1999b), “The Behavioural Perspective Model: consensibility and consensuality”, European Journal of Marketing, 33: pp. 570-596 Foxall, G. R. (2001) “Foundations of consumer behaviour analysis”, Marketing Theory, 1,pp.165-198 Foxall, G. R. (Ed.), (2002), Consumer Behaviour Analysis: Critical Perspectives. London and New York: Routledge Foxall, G. R. and Greenley, G. E. (1998), “The affective structure of consumer situations”, Environment and Behavior, 30, pp.781-798 Foxall, G. R. and Greenley, G. E. (1999), “Consumers’ emotional responses to service environments”, Journal of Business Research, 46, pp.149-58 Foxall, G. R. and Greenley, G. E. (2000), “Predicting and Explaining Responses to Consumer Environments: An Empirical Test and Theoretical Extension of the Behavioural Perspective Model”, The Service Industries Journal, 20, pp.3963 Foxall, G. R., Goldsmith, R. E. and Brown, S. (1998), Consumer Psychology for Marketing. Second edition. London and New York: International Thompson Business Press. Foxall, G. R. and James, V. K. (in press, a), “The behavioural basis of consumer choice: a preliminary analysis”, European Journal of Behavior Analysis, 2 Foxall, G. R. and James, V. K. (in press, b), “The behavioral ecology of brand choice: what and how do consumers maximize?” Psychology and Marketing Foxall, G. R. and Soriano, M. (in press), “Situational influences on consumer attitude formation and attitudinal—behavioral consistency”, Journal of Business Research Greene, B. F. and Neistat, M. D. (1983), “Behavior Analysis in Consumer Affairs: Encouraging Dental Professionals to Provide Consumers with Shielding from Unnecessary X-Ray Exposure”, Journal of Applied Behavior Analysis, 16, pp.13-27 Greene, B. F., Rouse, M., Green, R. B. and Clay, C. (1984), “Behavior Analysis in Consumer Affairs: Retail and Consumer Response to Publicizing Food Price Information”, Journal of Applied Behavior Analysis, 17, pp. 3-21 Hantula, D. A., DiClemente, D.F. and Rajala, A. K. (2001), “Outside the Box: The Analysis of Consumer Behavior”. In: Hayes, L., Austin, J. and

Marketing’s Attitude Problem

47

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Flemming, R. (Eds.), Organizational Change Context Press: Reno, NV. pp. 203-233 Katona, G. (1960), The Powerful Consumer. New York: McGraw-Hill Kunkel, J. H. and Berry, L. L. (1968), “A Behavioural Conception of Retail Image”, Journal of Marketing, 32, pp.21-27 Lott, A. J. and Lott, B. E. (1968), “A learning theory approach to interpersonal attitudes”. In Greenwald, A. G., Brock, T. C. and Ostrom, T. M. (Eds), Psychological Foundations of Attitudes. San Diego, CA: Academic Press. pp. 6788 Mehrabian, A. (1979), “Effect of emotional state on alcohol consumption”, Psychological Reports, 44, pp.271-82 Mehrabian, A. and Riccioni, M. (1986), “Measures of eating-related characteristics for the general population: relationships with temperament”, Journal of Personality Assessment, 50, pp.610-629 Mehrabian, A. and Russell, J. A. (1974), An Approach to Environmental Psychology, Cambridge, MA: MIT Press Mehrabian, A. and Russell, J. A. (1975), “Environmental effects on affiliation among strangers”, Humanitas, 11, pp.219-30 Mehrabian, A. and de Wetter, R. (1987), “Experimental test of an emotionbased approach to fitting brand names to products”, Journal of Applied Psychology, 72, pp.125-30 Rajala, A. K. and Hantula, D. A. (2000), “Towards a Behavioral Ecology of Consumption: Delay-Reduction Effects on Foraging in a Simulated Internet Mall”, Managerial and Decision Economics, 21, pp.145-158 Russell, J. A. and Mehrabian, A. (1976), “Environmental variables in consumer research”, Journal of Consumer Research, 3, pp.62-3 Russell, J. A. and Mehrabian, A. (1978), “Approach-avoidance and affiliation as functions of the emotion-eliciting quality of an environment”, Environment and Behavior, 10, pp. 355-87 Russell, J. A. and Pratt, G. (1980), “A description of the affective quality attributed to environments”, Journal of Personality and Social Psychology, 38, pp.311-22 Sheppard, B. H., Hartwick, J. and Warshaw, P. R. (1988), “The theory of reasoned action: a meta-analysis of past research with recommendations for modifications and future research”, Journal of Consumer Research, 15, pp.325343 Soriano, M. and Foxall, G. R. (in press, a), “Validation of a Spanish translation of Mehrabian and Russell’s emotional scales”, Journal of Consumer Behaviour Soriano, M. and Foxall, G. R. (in press, b), “Emotional responses to consumer environments: An empirical examination of the behavioural perspective model in a Latin American context”, Journal of Consumer Behaviour Staats, A. (1996), Behavior and Personality. New York: Springer

48

Gordon R. Foxall

Staats, A. and Staats, C. K. (1958), “Attitudes established by classical conditioning”, Journal of Abnormal and Social Psychology, 57, pp.37-40 Van den Putte, B. (1993), On the theory of reasoned action. Unpublished doctoral dissertation. University of Amsterdam

About the Author

Delivered by Publishing Technology to: Cardiff University IP: 131.251.133.26 On: Tue, 01 Jul 2014 09:29:51

Gordon Foxall is a Distinguished Research Professor at Cardiff University. His chief research interests lie in psychological theories of consumer choice and consumer innovativeness and their relationships to marketing management and strategy. He has published some 16 books and over 250 articles and papers on these and related themes and has previously held professorial appointments at the Universities of Strathclyde, Birmingham and Keele. His books include the popular text, Consumer Psychology for Marketing (co-authored with Ron Goldsmith and Stephen Brown) which has been translated into Korean, Russian, Polish and Chinese, and monographs such as Consumer Choice and Marketing Psychology. In addition, he has published in numerous scholarly journals including Journal of Consumer Research, Journal of Business Research, International Journal of Research in Marketing, Psychology and Marketing and Journal of the Academy of Marketing Science. His undergraduate work was at the University of Salford and he is also a graduate of the Universities of Birmingham (PhD industrial economics and business studies) and Strathclyde (PhD psychology). In addition, he holds a higher doctorate of the University of Birmingham (DSocSc). He is a Fellow of both the British Psychological Society (FBPsS) and the British Academy of Management (FBAM) and was recently elected an Academician by the Academy of Learned Societies for the Social Sciences (AcSS).