Consideration Set Size and Familiarity With Usage Context

Philippe Aurier, Universite Montpellier 2
Sylvie Jean, Universite Montpellier 2
Judith L. Zaichkowsky, Simon Fraser University
ABSTRACT - This study proposes a theoretical framework and an operationalization of the concept of consideration set in relation with usage context. Taking into account the consumer usage situation enables us to analyze the influence of two main components on consideration set size: context of consumption and familiarity (depth and breadth). In an application to the French beverage market, we show that consideration set size varies significantly across consumption contexts and is positively correlated to the breadth of familiarity. Moreover, we find an inverted U relation between depth of familiarity and consideration set size.
[ to cite ]:
Philippe Aurier, Sylvie Jean, and Judith L. Zaichkowsky (2000) ,"Consideration Set Size and Familiarity With Usage Context", in NA - Advances in Consumer Research Volume 27, eds. Stephen J. Hoch and Robert J. Meyer, Provo, UT : Association for Consumer Research, Pages: 307-313.

Advances in Consumer Research Volume 27, 2000      Pages 307-313


Philippe Aurier, Universite Montpellier 2

Sylvie Jean, Universite Montpellier 2

Judith L. Zaichkowsky, Simon Fraser University


This study proposes a theoretical framework and an operationalization of the concept of consideration set in relation with usage context. Taking into account the consumer usage situation enables us to analyze the influence of two main components on consideration set size: context of consumption and familiarity (depth and breadth). In an application to the French beverage market, we show that consideration set size varies significantly across consumption contexts and is positively correlated to the breadth of familiarity. Moreover, we find an inverted U relation between depth of familiarity and consideration set size.


The interest of consumer knowledge on the consideration set concept continues to increase over the years thanks to a wide of variety of articles published on this theme. SinceHoward (1963) and Howard and Sheth (1969), a lasting research tradition has attempted to define the individual decision process by emphasizing its working memory functioning and its limits. Being confronted with a wide range of brands, the consumer selects various options, creating and stocking up on brands and products that are part of a purchase goal (Brisoux and Laroche, 1983). These studies stressed the limited nature of the consumer’s cognitive ability and its consequence: the restricted size of this set, initially called the "evoked set".

More recently, authors have investigated the influence of memory and, more particularly, the brand accessibility (Alba and Chattopadhyay, 1985; Nedungadi, 1990). Additionally, categorization theory brought a new light on the role of goals linked to consumption. Here the consumer selects a subset of brands according to the considered usage context (Barsalou, 1985; Alba and Hutchinson, 1987; Bettman and Sujan, 1987; Shocker et. al., 1991; Park, 1993; Sinha, 1994). Thus, Nedungadi (1990) defines the consideration set as: "the set of brands brought to the consumer’s mind in a particular choice occasion". On an empirical level, Ratneshwar and Shocker (1991), controlling for two usage situations, showed that the content of this set varies according to the context of consumption.

Size is the aspect of the consideration set which has been widely investigated. However, most researchers use a static view of the consideration set since they do not explicitly take the consumption context into account. This is so, even when research emphasized the role of context on the choice process and the considered brands (Park, 1993; Warlop and Ratneswhar, 1993; and Sinha, 1994). Another important aspect, when studying the consideration set size, is the influence of familiarity with the context. In this respect, there are too few studies to reach a consensus. Hence, the purpose of this study is twofold: first, to control the consumption context thought by an individual when making his choice and analyze its influence on the consideration set size and secondly, to study the impact of familiarity (in terms of depth and breadth) on the consideration set size.


1.1 Consideration set size

If being considered constitutes a necessary condition for being purchased, then consideration set size plays a crucial role on consumer behavior, choice probabilities and then on marketing strategy. To explain the consideration set size, initially called "evoked set", research first looked at the impact of consumer characteristics (according to social, demographic, geographical, and cultural variables). Campbell (1969), Ostlung (1973), and Gronhaug (1974) showed that among social-geographical variables, only education is positively related with the evoked set size. The impact of education was also studied by May and Homans (1976), who used cognitive style theories to show that consumers with higher education use more abstract information and form a broader evoked set size. In a study of car buyers, Lapersonne et al. (1995) showed that the consumer tends to limit his consideration set to the former brand, and this is all the more so with age and education, but earning a low income.

Another aspect is the influence of enduring involvement on the consideration set size. However, no consensus seems to emerge from this premise. A group of researchers (Jarvis and Wilcox (1973), Rothschild and Houston (1977), Belomax and Javalgi (1989), Lapersonne et al. (1995) found involvement negatively related to set size. More involvement with the product category might result in learning about the brands (Zaichkowsky, 1985a); more attributes might be taken into account and more exacting levels might be sought on those brands (Rothschild and Houston, 1977); and, as a consequence, more restricted considration sets are evoked. Conversely, other authors established a positive link when the product category investigated was heterogeneous in terms of price and quality (Gronhaug, 1974; Divine, 1995). Moreover, when involvement was linked with stimulation and variety seeking (Petty and Cacioppo, 1990), it enabled more cognitive effort, which led to more important set sizes. Finally, a latter research group found no link with involvement and set sizes (Brisoux and ChTron, 1990; Eliot and Warfield, 1993).

1.2 Consumption Context

Taking the consumption context into account enables one to get a more dynamic and realistic approach of the consideration set. Belk’s research carried out in 1974 and 1975 showed that 50% of the variance in food selection can be accounted for the usage situation and its interaction with other variables, in particular, the person x situation interaction.

Empirical studies rarely take this variable into account when studying the consideration set formation. If research did take explicitly the context into account (Ratneshwar and Shocker, 1991; Graonic and Shocker, 1993; Warlop and Ratneshwar, 1993), they did not study its effect on the consideration set size. Thus in spite of a dynamic conceptualization of the consideration set concept (Nedungadi, 1990), researchers still refer to the oldest, static notion of the evoked set. As a consequence, our first proposition will show that the size of this set varies according to consumption contexts.

H1: The average size of the consumer’s consideration set in a category of products varies according to consumption contexts.

We will simply try to show that a main effect due to the situation exists (without considering the individual dimension). In other words, the consideration set size can vary according to context families and, as a consequence, the use of the static "evoked set" defined for the whole product category, whatever the context, is not a realistic hypothesis.

1.3 Familiarity with the consumption context

Another outstanding aspect resulting from the interaction between an individual and the consumption context is the impact of familiarity with the context on consideration set, particularly it’s size. In this respect, theory and empirical results do not provide clear evidence.

A first explanation to the contradictory character of empirical results can be the coexistence of two types of familiarity: depth and breadth. Depth, the most commonly studied dimension, is the mere result of the accumulation of experiences with the product category or the consumption situation (Alba and Hutchinson 1987; Punj and Srivanasan, 1989). Breadth characterizes the variety of experiences in the product category in terms of purchased products and consumption contexts (Zaichkowsky, 1985b). Punj and Srivanasan (1989) showed that the depth of the experience (viewed as the number of purchases in the category) tends to decrease slightly the consideration set size, whereas the breadth of the experience (viewed as the number of different brands purchased) tends to increase it. Thus, in this research we will study the influence of depth and breadth of familiarity on the consideration set size.

Another possible explanation of the inconsistency of results is the lack of consensus on the object of familiarity that can be product category, brand, purchase context, or consumption context. It is then possible to distinguish 1) empirical studies which consider familiarity with the product category (in terms of consumer expertise) and 2) empiricalstudies which consider familiarity with the consumption context.

In terms of familiarity with the product category, greater familiarity might lead the consumer to know and try more products. Conover (1983) and Alba and Hutchinson (1987) suggested that consumer familiarity is linked to the existence of a more complex cognitive structure that leads to wider consideration sets. Johnson and Lehmann (1997) have shown that consideration set size increases, as the consumer becomes more experienced, when this set is constructed in terms of products or brands. But, conversely, the development of procedural knowledge associated with more familiarity would lead the individual to restrict his/her consideration set for efficiency motives. Thus, as time elapses, familiarity would enable one to remove unsatisfactory products, so to reduce his/her set size, particularly when it is constructed in terms of brands (Raju and Reilly, 1980; Van Trijp et al., 1996).

In terms of the familiarity with the context, more familiarity with the choice context makes it easier to recall a memory-based solution (top down process) and contributes to diminish the consideration set size. Conversely, when the situation is unfamiliar the set may be constructed out appropriately on the basis of the product benefits (bottom up process) and, as a consequence, the size may increase. This proposition is also consistent with prior results from Warlop and Ratneshwar (1993), Bettman and Sujan (1987) and Sinha propositions (1994).

Therefore we can deduce the hypothesis of an inverted U relationship between depth and consideration set size because the type of processing (bottom-up vs. top-down) and the information processing capacity have opposite effects on the size of the consideration set. When depth is low, individuals may use a bottom-up process for the construction of their consideration set, resulting in more consideration but counterbalanced by their lower cognitive capacity due to less familiarity and less expertise. Conversely, when depth is high, individuals may use a top-down process for the construction of their consideration set with, as a consequence, a reduction of the set, even if they have a greater cognitive capacity due to more expertise. As a consequence, the consideration set size might be maximum at a moderate level of familiarity. This kind of relationship is similar to the results of Johnson and Russo (1984) who identified an inverted U shaped curve between familiarity with the product category and prepurchase information search, because of the existence of opposite effects between ability to process and motivation to find information. Thus, we shall test the two following propositions that summarize our interpretation of the literature:

H2: the breadth of experience (familiarity) with the products and consumption situations associated with a category of products positively influences the consideration set size.

H3: the depth of experience (familiarity) with a consumption situation associated with a product category affects the consideration set size in a inverted U shaped relationship.


2.1 Product class, population and sample selection

The research was implemented on the beverage market in France. Beverages constitute an interesting product category where consumption contexts play an important role in beverage selection. The data collection was conducted in three stages. In the first qualitative stage, we used the product x usage situation developed by Stefflre (1971) and Day, Shocker and Srivastava (1979) to identify a typology of 22 usage contexts and 20 products which are representative of the "beverage universe". The methodology was successively implemented during tree semi-structured groups and ten individual interviews. In the second stage, a first version of a quantitative questionnaire was administrated to a convenience sample of 72 people. Respondents had to fill in a table presenting consumption contexts as rows (22) and products as columns (20). For each row corresponding to a particular context situation, they had to mark off the cells corresponding to the product they would personally consume in such a situation. Each cell was then coded one if the product was considered for the corresponding situation and zero otherwise. A correspondence analysis of this table, aggregated across respondents, was then used to reduce the original list of 22 usage contexts and 20 products to a more tractable table of 12 situations (see Table 1 and Appendix B) by 16 products (see Appendix A). In the third stage, a final version of the questionnaire was developed comprising this reduced table and administrated to a sample of 413 men representative of the French male population. [Only men were studied with respect to the main objective of the original research which was wine selection.] Quota samples were selected by controlling for gender, age (15 to 65 years old), profession and area of residence. Men were interviewed face to face at home by a professional market researcher.

2.2 Operationalization of concepts

Dependent variable:

The consideration set size, for a given usage context, is our dependent variable. As we indicated, respondents filled in a table with 12 usage contexts as rows and the 16 products as columns. For each context (row), the respondent had to indicate which products he would consume, personally, for this particular usage situation. The obtained table is row conditional and allows us to identify his consideration set for each usage context. Consideration set size was established by calculating for each usage context and each consumer, the number of products considered. The main characteristics of this consideration set are:

BIt corresponds to the definition proposed by Shocker et al. (1991), because it is formed when the individual is not in his/her final context of purchase, in opposition to the choice set, which corresponds to the set of alternatives that have been retained when the choice becomes effective.

BThe unit of analysis is the product (for instance water or beer) and not the brand, as is usually the case. As noted by Johnson and Lehmann (1997), the consideration set formation has rarely been studied in terms of products.

BAccessibility of the products is not taken into account as all the information (products x situation) is presented in the Table. What we believe people are responding to, is what drinks they would consume in each situation. They may even be responding with what drinks they have consumed in each situation.

Independent variables:

Familiarity, in terms of depth and breadth, characterizes the interaction between the person and usage context. Depth represents the cumulated relationship with a particular consumption situation and was operationalized as the frequency with which the respondent is in this situation. This frequency was measured using the following 6 point declarative scale: "never" (0)BAless than once a month" (1)BAseveral times a month" (2)BAseveral times a week" (3)BAalmost every days" (4)BAseveral times a day" (5). Even if the level "several times a day" is meaningless for a number of situations (for instance during a party with friends), the same scale was used for all contexts in order to make direct comparisons across situations. Breadth represents the variety of the consumption experiences at the product category level (here, the beverage universe). It was operationalized s a composite indicator (the simple sum) taking into account the number of products consumed at least at a level of "several times a week" and the number of usage contexts frequented at least at a level of "several times a week". This indicator characterizes the variety of behavior relative to the whole beverage universe, in terms of products and usage contexts, all situations combined. As a consequence, each respondent is characterized by a unique breadth score whatever the context.


In order to understand the structure linking products to situations we first applied a correspondence analysis to the aggregated matrix (over consumers) of consideration set relative to the 12 usage contexts (rows) and 16 products (columns). Two mains factors explain 65% of the variance and structure the beverage universe according to a first dimension consisting of individual versus socialized contexts and a second dimension consisting of meals (i.e., to sit and eat) versus leisure contexts (i.e., not during meals, see Appendix B). These results were confirmed applying a hierarchical clustering to the same matrix.

3.1 Consideration set size and usage context (hypothesis H1)

Table 1 presents the simple distribution of consideration set size. [All situations combined, we have a (413 individuals x 12 contexts=) 4956 sample size (consideration sets). When analysis will be implemented at the context level, sample size will be equal to 413.] As we can see, 44% of consideration sets are of size equal to one (from far the mode of the distribution!) and 72% are of size less or equal to two. This result supports general findings about the low size of consideration sets and is confirmed in Table 2 where we can see that the mean consideration set size is 2.1. However, we can also observe important variations between situations. In particular, socialization of situations tends to increase size, either in a meal context or in a leisure context. For instance, we can observe a mean consideration size of 2.55 in the context "before an everyday meal at home" increasing up to 2.71 in the context "before a meal with friends, at home". Identically, we go from 2.33 for the context "during a regular meal at home" up to 2.62 for the context "during a meal with friends, at home". A one-way Anova shows significant differences of the mean consideration set size between situations (p<0.001) and, as a consequence, hypothesis H1 cannot be rejected.

3.2 Breadth of familiarity and consideration set size (hypothesis H2)

All situations combined, we observe a positive (r=0.092) and significant correlation (at .0001) between breadth of familiarity and consideration set size. Thus, at the aggregate level, we cannot reject our H2 hypothesis (Table 2). This result is still true at the context-specific level: all correlations between breadth and consideration set size are positive and eight of 12 are significant (at a 0.05 level or less). Familiarity breadth in a category (here beverages) positively influences consideration set size, in almost all contexts. Consumers who have a more varied experience with products and contexts, consider on average more products, whatever the situation. This result confirms findings by Punj and Srinivasan (1989).

3.3 Depth of familiarity and consideration set size (hypothesis H3)

All situations combined, we observe a positive (r=0.30) and significant (p<0.001) correlation between breadth and depth. These two dimensions of familiarity, even if they are separate constructs, are not independent.

All situations combined, we observe a positive (r=0.06) and significant correlation (p<0.001) between depth and consideration set size. However, in respect of each situation, we observe very contrasted results: correlations are positive in six situations and negative in six situations. All negative correlations, ut two, are not significant. For these situations the direction of correlation suggests a simplification by routine (efficiency) tendency, but it is not significant. This is similar with Punj and Srivanasan (1989) results. Conversely, four of the six positive correlations are significant at 10% or less ("during a lunch, at the canteen"; "during work breaks"; "during the day at the terrace of a cafe"; "after a physical effort or sport"). Moreover, one can observe that these four situations demonstrate the lowest levels of familiarity depth.

To investigate our H3 hypothesis, we classified consideration sets (individual x situation) into six groups, on the basis of the depth of familiarity level, going from zero up to five (remember that each individual has a level of depth and a consideration set size for the 12 contexts). We then calculated the mean consideration set size for each level of depth (Table 3).





As suggested by our H3 hypothesis, we can observe an inverted U curve between consideration set size and depth of familiarity. For the lowest levels of depth (0), the consideration set size is at its’ minimum (1.57) and then increases quickly with depth to reach it’s maximum (2.37) at a moderate level of depth (2). After this point, the consideration set size decreases (up to 2.17) when depth increases (up to 5). One-way anova and mean comparison results support the inverted U-shaped relationship, as indicated in Table 3 and our H3 hypothesis cannot be rejected.

Thus, when depth of familiarity is low, individuals may have small consideration sets because of their low information processing capacity, even if most of them use a piecemeal (bottom up) process for the construction of their consideration set. At moderate levels of familiarity depth, information processing capacity increases and is still associated to a piecemeal process to build the consideration set, resulting in the largest consideration sets. Here, the consumer is perhaps exploring behaviors that have positive effects on the size of his consideration set. With higher levels of familiarity depth, people use their cumulated experience and expertise to select the most appropriate (memory-based) solution to their problem, which results in a diminution of their consideration set size.

Thus, the conflicting results emerging from the literature relative to the relationship between depth of familiarity and consideration set size can be due to the fact that depth of familiarity for a particular context has not clearly been taken into account. The consumer is using different information processing strategies depending on his/her level of familiarity.




The first contribution of this paper is to show that the consideration set size (and as a consequence it’s content and the choice process associated), varies from one consumption context to another, particularly according to its socialized or individual character in our beverage application. We conclude that studying consideration set size without a reasonable specification of the consumption contexts seems to be unrealistic. Our results were obtained in the study of a large consumption universe (beverages), comprising 16 product categories and 12 usage situations identified and selected in qualitative and quantitative interviews. Thus, the context of consumption has been explicitly taken into account and in a realistic way.

The second contribution is related to the effect of familiarity with the situation on the consideration set size. Familiarity was investigated according to two dimensions: depth and breadth. Over all situations, we clearly assessed that breadth of the experience positively influences the consideration set size. However, for depth, we identified an inverted U relationship between consideration set size and depth of familiarity: when depth is moderate, then the set size is at its maximum. When depth is high, cognitive capacity is associatedwith larger sets (May and Homans, 1976) but it is counterbalanced by the use of memory-based consideration sets associated with lower sets. When depth is low, the use of situation-specific consideration sets is counterbalanced by low cognitive capacity associated with lower sets. This result is similar to the relation characterized by Johnson and Russo (1984) and Brucks (1985) in their studies between consumer knowledge and prepurchase information search.

Two main conclusions emerge from this research. First, the consideration set size cannot be studied apart from the consumption context: its familiarity level and its main characteristics in terms of socialization, hedonism, and so on. Without taking the context into consideration we venture to make groundless generalizations, like finding a positive or a negative correlation between familiarity depth and the number of considerations. Second, the level of depth of familiarity with the context plays an outstanding part influencing the mode of the information processing (memory-based vs. context-specific).


In this research, we have operationalized consideration set in terms of products (vs. brands), defined in the broad sense. This has the advantage to study consumption context with precision and to take into account the heterogeneity of the real life consideration sets, but it cannot give an account of storage effect since products were presented to the consumer. Thus, we did not consider the product recall and/or consumer brands before the setting up of consideration sets. As a consequence, this research is more concerned with choice behavior than purchase behavior.

Managerial Implications

This study shows that the consumer’s consideration set size with the consumption context familiarity represents an important indicator for business. Thus, the inverted U relation between consideration set size and familiarity with the context suggests different business strategies for market penetration. The first to enter a market should take advantage of the low familiarity level, which correspond to reduced consideration set sizes. This may be used to prevent new competitive entry by focusing the consumer’s attention on a particular consumption context. Thus, the consumer is apt to build consideration sets that have ready and simplified solutions in stock. New brands should enter the market with new consumption contexts and in their turn these brands will probably integrate into the consumer’s consideration sets. Advertisements can improve product consideration using the link between the product and context familiarity. It will enhance recall or recognition and consideration.

Areas for future research

Future research may carry out more accurate analyses on the influence of situation characteristics on the formulation of consideration sets by controlling the main characteristics of familiarity, generality level, and socialization. This research might be repeated by integrating the study of the contents of the consideration set at the brand level. This might set individuals in situations not helped by recall, a crucial factor of the consideration set formation.






Alba, Joseph W. and Amitava Chattopadhyay (1985), "Effects of Context and Part-Category Cues on Recall of Competing Brands", Journal of Marketing Research, 22, 3, 340-349.

Alba, Joseph W and Wesley J. Hutchinson (1987), "Dimensions of Consumer Expertise", Journal of Consumer Research, 13, 2, 411-454.

Barsalou, Laurence W. (1985), "Ideals, Central Tendency, and Frequency of Estimation as Determinants of Graded Structure in Categories," Journal of Experimental Psychology: Learning, Memory and Cognition, 11, 4 , 629-654.

Belk, Russell W. (1974), "An Exploratory Assessment of Situational Effects in Buyer Behavior", Journal of Marketing Research, 11, 2, 156-163.

Belk, Russell W. (1975), "Situational Variables and Consumer Behavior Effects in Buyer Behavior", Journal of Consumer Research, 2, 3, 157-164.

Belonax, Joseph J. and Rajshekhar, Javalgi G. (1989), "The Influence of Involvement and Product Class Quality on Consumer Choice Sets", Journal of The Academy of Marketing Science, 17, 3, 209-216.

Bettman, James R. and Mita Sujan (1987), "Effects of Framing on Evaluation of Comparable and Noncomparable Alternatives by Expert and Novice Consumers", Journal of Consumer Research, 14, 2, 141-154.

Brisoux, Jacques E. and Michel Laroche (1983), "Evoked Set Formation and Composition: An Empirical Investigation Under a Routinized Response Behavior Situation", Advances in Consumer Research, 10, ed. Richard Bagozzi and Alice Tybout, Association for Consumer Research, 357-361.

Brisoux, Jacques E. and Emmanuel J. ChTron (1990), "Brand Categorization and Product Involvement", Advances in Consumer Research, 17, ed. Marvin Goldberg, Gerald Gorn and Richard Pollay, Association for Consumer Research, 101-109.

Brucks, Merrie (1985), "The Effects of Product Class Knowledge on Information Search Behavior", Journal of Consumer Research, 12, 3, 1-16.

Campbell, Brian M. (1969), "The Existence of Evoked Set and Determinants in Brand Choice Behavior", unpublished doctoral dissertation, University of Columbia.

Conover, J. N. (1983), "Familiarity and the Structure of Product Knowledge", Advances in Consumer Research, 10, ed. Richard Bagozzi and Alice Tybout, Association for Consumer Research, 494-497.

Day, George S., Allan D Shocker and Rajendra Srivastava K. (1979), "Customer-Oriented Approach to Identifying Product-Markets", Journal of Marketing, 43, 8-19.

Divine, Richard L. (1995), "The Influence of Price on the Relationship between Involvement and Consideration Size", Marketing Letters, 6, 4, 309-319.

Elliott, Michael T. and Anne E. Warfield (1993), "Do market mavens categorize brands differently?", Advances in Consumer Research, 20, ed. Leigh McAlister and Michael Rothschild, Association for Consumer Research, 202-208.

Graonic, Milos D. and Shocker Allan D. (1993), "On the Transferability of Feature/Level Preferences Across Competing Products the same Purposes" Advances in Consumer Research, 20, ed. Leigh McAlister and Michael Rothschild, Association for Consumer Research, 389-393.

Gronhaug, Kjell (1973-1974), "Some Factors Influencing the Size of the Buyer’s Evoked Set", European Journal of Marketing, 7, 232-241.

Howard, John A. (1963), Marketing Management: Analysis and Planning, Irwin, Illinois.

Howard, John A. and Jedish Sheth (1969), The Theory of Buyer Behavior, Mc Graw-Hill, New York.

Jarvis, Lance P. and James B. Wilcox (1973), "Evoked Set Size, Some Theoretical Foundations and Empirical Evidence", American Marketing Association, 236-240.

Johnson, Michael D. and Donald D. Lehmann (1997), "Consumer Experience and Consideration Sets for Brands and Product Category", Advances in Consumer Research, 24, ed. Merrie Brucks and Deborah J. MacInnis, Association for Consumer Research, 295-300.

Johnson, Eric J. and Edward J. Russo (1984), "Product familiarity and learning new information", Journal of Consumr Research, 11, 3, 542-550.

Lai, Albert (1994), "Consumption Schemata: Their Effects on Decision Making", Advances in Consumer Research, 21, ed. Chris Allen and Deborah Roedder-John, Association for Consumer Research, 489-494.

Lapersonne, Eric, Gilles Laurent and JeanBJacques Le Goff (1995), "Consideration Set of Size One: An Empirical Investigation of Automobile Purchases", International Journal of Research in Marketing, 12, 55-66.

May, Frederick E. and Richard E. Homans (1976), "Evoked Set and the Level of Information Processing in Product Comprehension and Choice Criteria", Advances in Consumer Research, 3, ed. Beverlee B. Anderson, Association for Consumer Research, 172-175.

Nedungadi, Prakash (1990), "Recall and Consumer Consideration Sets: Influencing Choice Without Altering Brand Evaluations", Journal of Consumer Research, 17, 3, 263-276.

Ostlung, Lyman E. (1973), "Evoked Set Size: Some Empirical Results", American Marketing Association, 226-230.

Park, C. Whan (1993), "Context Effects on Consumer Choice, Brand Awareness and Decision Making", Advances in Consumer Research, 20, ed. Leigh Mc Alister and Michael Rothshild, Association for Consumer Research, 395-396.

Petty, Richard E. and John T. Cacioppo. (1990), "Involvement and Persuasion: Tradition Versus Integration", Psychological Bulletin, 107, 3, 367-374.

Punj, Girish and Narasimhan Srinavasan (1989), "Influence of Expertise and Purchase Experience on The Formation of Evoked Sets", Advances in Consumer Research, 16, ed. Thomas Srull, Association for Consumer Research, 507-514.

Raju, P. S. and Reilly M. D. (1980), "Product familiarity and Information Processing Strategies: An Exploratory Investigation", Journal of Business Research, 187-212.

Ratneshawar, Snirivasan and Shocker Allan D. (1991), "Substitution in Use and the Cognitive Structure of Product Categories", Journal of Marketing Research, 28, 3, 85-96.

Ratneshwar, Snirivasan, Pechmann Cornella and Shocker Allan D. (1996), "Goal-Derived Categories and the Antecedents of Across-Category Consideration", Journal of Consumer Research, Vol. 23, December, 240-250.

Rothschild Michael L. and Michael J. Houston (1977), "The Consumer Involvement Matrix: Some Preliminary Findings", B. A. Greenberg and D. N. Bellinger (eds.), Proceedings: AMA Educator’s Conference, 95-98.

Shocker, Allan D., Moshe Ben-Akiva, Bruno Boccara and Prakash Nedungadi (1991), "Consideration Set Influences on Consumer Decision-Making and Choice: Issues, Models, and Suggestions", Marketing Letters, 2: 3, 181-197.

Sinha, Indrajit (1994), "A Conceptual Model of Role of Situational Type on Consumer Choice Behavior and Consideration Sets", Advances in Consumer Research, 21, ed. Chris Allen and Deborah Roedder John, Association for Consumer Research, 477-482.

Stefflre Volney (1971), New Products and New Entreprises: A Report of an Experiment in Applied Social Science, Irvine, CA: University of California.

Van Trijp Hans C.M., Hoyer Wayne D. and Inman Jeffrey J. (1996), "Why Switch? Product Category-Level Explanations for True Variety-Seeking Behavior", Journal of Marketing Research, 33, 3, 281-292.

Warlop, Luk and S. Ratneshwar (1993), "The Role of Usage Context in Consumer Choice: A Problem Solving Perspective", Advances in Consumer Research, 20, ed. Leigh Mc Alister and Michael Rothshild , Association for Consumer Research, 377-382.

Zaichkowsky, Judith L. (1985a), "Measuring the Involvement Construct", Journal of Consumer Research, 12, 4, 341-352.

Zaichkowsky, Judith L. (1985b), "Familiarity: Product Use, Involvement or Expertise?", in Elizabeth C. Hirshman and Morris B. Holbrook (eds.), Advances in Consumer Research, Vol 8, Association for Cosumer Research, 708-712.