The Interaction of Coupons With Price and Store Promotions

Caroline M. Henderson, Dartmouth College
ABSTRACT - This research investigates two behavioral mechanisms for consumer response to simultaneous marketing stimuli: "coupon primacy" and "lowest price. Under coupon primacy, coupons inhibit consumers from responding to in-store promotional conditions. Under lowest price, consumers choose brands on the combined basis of all available promotions -- coupon and shelf price discounts. Using a logit choice model on scanner panel data for two product categories, the research concludes that there is some limited evidence to support both of these mechanisms -- depending on type of consumer. The less deal-prone exhibit coupon primacy, the more promotion sensitive tend to use a lowest price strategy.
[ to cite ]:
Caroline M. Henderson (1988) ,"The Interaction of Coupons With Price and Store Promotions", in NA - Advances in Consumer Research Volume 15, eds. Micheal J. Houston, Provo, UT : Association for Consumer Research, Pages: 364-371.

Advances in Consumer Research Volume 15, 1988      Pages 364-371


Caroline M. Henderson, Dartmouth College

[The author is grateful for funds provided by the Marketing Science Institute, the Tuck Associates Program, and the Division of Research at Harvard Business School; and for data provided by Selling Areas Marketing, Inc. and the Test Marketing Group, Inc.]


This research investigates two behavioral mechanisms for consumer response to simultaneous marketing stimuli: "coupon primacy" and "lowest price. Under coupon primacy, coupons inhibit consumers from responding to in-store promotional conditions. Under lowest price, consumers choose brands on the combined basis of all available promotions -- coupon and shelf price discounts. Using a logit choice model on scanner panel data for two product categories, the research concludes that there is some limited evidence to support both of these mechanisms -- depending on type of consumer. The less deal-prone exhibit coupon primacy, the more promotion sensitive tend to use a lowest price strategy.


Couponing continues its recent popularity in the marketing mix: 1984 coupon distribution increased 14% over 1983 to a record total of 163 billion (Nielsen Clearing House 1985). Yet there are allegations that most coupon promotions are unprofitable (Irons, Little and Klein 1983). In this context, research that can shed light on couponing's effectiveness has become more prevalent. For example, researchers have investigated prediction of redemption rates (Reibstein and Travers 1982, Ward and Davis 1978); the effect of coupons on incremental sales (Hee 1981), purchase acceleration (Neslin, Henderson and Quelch 1985), and brand switching (Dodson, Tybout and Sternthal 1978); coupon profitability (Neslin and Shoemaker 1983; Klein 1981; Irons, Little and Klein 1983); and the identification of coupon users (Teel, Williams and Bearden 1980). Other work has focused on the behavioral issues of why and how coupons can achieve results (Raju and Hastak 1979, Rothchild and Gaidis 1981, Gardner and Strang 1983, and Shimp and Kavas 1984). One important area that is not included in this list is the interaction of couponing with other elements in the marketing mix. Interaction effects are a common belief (for example, Beem and Shaffer 1981) but are not frequently researched.

Interactive effects are particularly important at this time. The tremendous increase in the use of all forms of sales promotion--particularly couponingChas led to a situation popularly known as "clutter." In many categories, coupons are offered so frequently as to be a constant presence in the market Under these conditions, couponing is no longer an isolated event but one that needs to work at all times with other marketing elements.

Such simultaneous events have implications for the effectiveness of promotion programs. For example, a consumer holding a coupon for a particular brand may be faced with an in-store promotion for the same brand (store displays or special prices, etc.). With the modern tendency toward combining promotional techniques, this multiple exposure becomes more probable. If these elements reinforce each other this may create a higher probability of choice. In this case, the coupon is more likely to be used and the brand selected because of the availability of an additional inducement. However, if only one element is required to achieve consumer response, the second element may be overlooked. In this case certain types of promotion programs involving simultaneous elements may be highly inefficient.

The objective of this research is to look at two behavioral mechanisms, which might explain consumer response to coupons and other marketing elements. This research is restricted to a particular set of elements defining the shopping environment and includes price, price cuts, special packaging, local newspaper advertising by retailers, and local newspaper advertising by manufacturers.


There are a limited number of studies which have looked at the interaction between forms of promotion or between promotions and other elements of the marketing mix.

The Chevalier (1975) controlled store experiment with eight products finds an interactive effect between display and price reductions in that a combination of the two is most effective in increasing sales. Sunoo and Lin (1978), with a one-product cable panel experiment, find only a minor interaction between promotion and advertising. Woodside and Waddle (1975), again with a single product experiment, find a significant positive interaction between advertising and price specials. The Eskin and Baron (1977) experiment finds an interactive effect of free samples with advertising. Prasad and Ring (1976), also with one product, find an interactive effect between advertising and price. Dennerlein's analysis (1980) of proprietary data concludes that trade promotions are most successful when used in conjunction with other promotional elements. Wilkinson et al.'s (1982) in-store experiment shows that two out of four products studied have an interactive effect between special display and price cuts, while one product of the four has a three-way interaction among display, price, and advertising.

This set of studies appears to show that limited interactions--with respect to displays, advertising and price reductions--are significant in stimulating sales. None of this research, however, has really had the interaction of coupons and other elements as the focus, nor has modeled the manner in which multiple offers should be expected to impact behavior.

Levedahl (1984) has looked specifically at the question of the interaction of prices with coupons. His data show that on an individual brand basis, consumers face higher average shelf prices when they are using a coupon than when they are not. His explanation of this result involves a price discrimination hypothesis: average prices are raised when coupons are offered because non-coupon using consumers, the only ones who will pay full price, are less price sensitive. When coupons are not offered and the market does not self-select into coupon-using and non-coupon using segments, average prices must be lowered to reflect a greater average price sensitivity.

Another interpretation can be made of these results. It is plausible that consumers who are preparing to use a coupon are in fact less sensitive to shelf price than those who are not. This would occur if the presence of a coupon obstructed the consumer's usual concern with price. In this case, shelf prices would not be used in making a brand selection. Without a coupon, however, consumers would consider shelf price and would tend to select the brand when it was favorably priced.

Two Models Of Coupon Interactions

There are two possible types of interactions between coupons and other mix elements.

Model 1: Coupon Primacy

This model states that coupons have a "primacy" effect in blocking the impact of other purchase influences. There are two related explanations for this effect: precommitment and choice simplification. Under a theory of precommitment, taking advantage of coupons requires a degree of clipping, saving, carrying, and planning, which binds the consumer to a particular brand, regardless of competing offers. Coupons take precedence because the consumer is involved with them prior to store visiting. Thus, consumers may be following a lexicographic choice process in which coupons are the first attribute considered. This attribute is sufficient to determine choice and thus creates precommitment--the choice has been made. Only when this attribute is not relevant (the consumer holds no coupons) are other attributes such as price considered. Note that this model of purchase behavior is only concerned with choices made from an evoked set of acceptable brands. Other preference criteria would be relevant in the creation of this acceptable set but are not considered in this theory.

Alternatively, the effect may be derived from choice simplification resulting from information overloads (see Malhotra 1984a for a review). In the case where the consumer is exposed to clutterCmultiple coupons, multiple store ads, and multiple in-store inducements--confusion may result in the use of a simple heuristic as a way to avoid further information processing demands. For example, the consumer may believe that using a coupon will lead to obtaining the lowest possible price. Using this rule avoids the necessity of processing additional price-related information.

Both of these explanations would result in coupon primacy a form of serial position effect (Feigenbaum and Simon 1962). The primacy construct has been primarily researched in the communications context as order effects in verbal learning where a U-shaped curve of both primacy and frequency effects has been repeatedly confirmed (Ray, 1973). The Ray and Webb clutter studies (1978) in particular find first commercials of commercial strings most effective even in conditions of increasing clutter. Primacy, rather than recency, is a particularly attractive hypothesis as it is bolstered by the construct of cognitive dissonance. Consumers may not wish to jeopardize previous decisions by being shown a failure to achieve objectives such as "pay the lowest price." New information may have the potential to show that the investment in coupon clipping has been wasted.

This model of consumer coupon behavior should be mediated by individual differences. Information overload research has advanced a number of relevant differences (see Malhotra 1984a for a review). Particularly relevant for promotion behavior should be the consumer's degree of experience in using coupons. Consumers relatively inexperienced in using coupons should be particularly vulnerable to a coupon primacy effect. Once consumers have become accustomed to using coupons, they may develop more complex decision rules--incorporating other purchase influences in a joint strategy or "best" rule--for example the decision rule "buy the cheapest."

Model 2: Lowest Price

According to a lowest price model, consumers may be more likely to consider all available options and not remain committed to Using a particular coupon. These consumers would be expected to be less motivated by feelings of wasted effort, as they would be more skilled at selection of only the best offers--high value coupons, store advertising with "hottest" specials, etc. Consumers without these skills may respond to such offers when not in possession of a particular coupon. Once precommitted, other factors should have no impact on choice. Responding-to several promotions could be part of an overall purchasing strategy to find the lowest price. Information processing studies of consumer choice heuristics (Bettman and Zins, 1977) show that some consumers are attempting this strategy. That consumers can actually determine the cheapest is not always clear. Capon and Kuhn (1982) show that only a minority of consumers are able to calculate best buys, although the Russo (1977) unit price experiment does show immediate consumer learning.

The distinction between the coupon primacy and lowest price models is analogous to the distinction made by Bettman (1979) between decision making outside and inside the store. A decision made outside the store--to buy the brand for which a coupon is available--is indeed an easier decision than one made by comparing a number of alternatives inside the store. Contrary to the Bettman conception, however, this type of decision is seen here as most appealing to those consumers with the [east experience at incorporating promotions into their decision making. The comparison of a number of brands on the basis of multiple promotional attributes -- the lowest price model -- should be appropriate for those more experienced with promotions.


The two models of coupon behavior described above are directly contradictory. However, the behavior patterns may accurately describe two alternative types of consumers. As discussed above, the two alternative models are distinguished by the consumer's desire to simplify the process of using a coupon. This desire should be more pronounced in those who are less experienced at using promotions. Experience with promotions can best be approximated by a consumer's current degree of deal-proneness. Deal-proneness will therefore provide the qualification to each model.

H1: Coupon primacy--Price paid should be higher when a coupon is used than when it is not used for less deal-prone consumers.

Consumers who are relatively less experienced with using coupons should be particularly vulnerable to simplification strategies and thus more likely to ignore shelf price or other in-store stimuli when planning to use a coupon. Note that such consumers will be generally less responsive to other promotional elements. The hypothesis test the difference in responsiveness between coupon and noncoupon purchases.

H2: Lowest price--More deal-prone consumers should make choices based on the combination of coupon discount and shelf price.

Consumers who are relatively more experienced with using coupons should, of course, be generally more sensitive to all forms of promotion. The hypothesis suggests that such sensitivities are more than additive; they are expected to interact in a positive fashion.


The two competing models are tested by comparing the effect of using a coupon on a purchase and not using -a coupon. The data are records of individual purchases collected from a "northeast scanner market"--a metropolitan area of about 250,000 population. This market contains three major grocery chains, with a total of 15 stores, all of which are part of the scanner system. This coverage amounts to 85% of the All Commodity Volume. The period covered by the data is from April to October 1981 for overlapping data series. Purchases of two product categories are analyzed: bathroom tissue ("paper") and caffeinated instant coffee ("coffee"). The categories were selected on the basis of data availability. Category boundaries are defined using preexisting partitioning data (Urban, Johnson and Hauser 1982; Vanhonacker 1979).

A data base was constructed with all category purchases and included brand choice, whether a coupon was used on the purchase, and characteristics (price, advertising, and deal packs) of all available brands. The characteristics represent the price and promotional environment available to the consumer. Other purchase influences such as qualities of the brands themselves or media advertising are not used in modeling. The available variables include the following:

Price/Unit: The price per unit variable was calculated from shelf price information provided by store-level data files. For paper the relevant price is price per roll; for coffee it is price per ounce.

Retail Ad: Retail ad means that the product was advertised in local newspapers by the grocery store during the week of purchase. Such advertising stressed price, but the product was not necessarily discounted by the retailer. Advertised specials were defined as pertaining to all variations of a brand size.

Manufacturer Ad: This measure indicates that the brand was advertised in local newspapers by the manufacturer during the week of purchase. All such ads carried coupons.

Coupon: Coupon denotes that the purchase was made using a coupon from any source. The coupon did not need to be the one carried in a manufacturer's ad.

Price Change: Price change was defined as the difference, measured in dollars, between the current price and the average shelf price at the same store, using store scanner data, during the week preceding the purchase. (A value of -.05 indicated that the brand had been discounted five cents.) This data base was divided into relatively homogeneous deal-proneness segments, defined on the basis of previous work (see Henderson 1986 for details of the segmentation procedure). These segments, shown in Table 1, were unique to each product category.

The groups were isolated with the aid of cluster analysis on five variables: percentage of purchases made with coupons, price changes, deal packs, retail newspaper ads, and manufacturer newspaper ads. The coupon-using clusters are defined on the basis of very high (heavy coupon users) or moderately high (light coupon users) coupon usage. The multi-deal-prone group was a small segment with abnormally high responsiveness to all five forms of promotions. The cluster analysis did not yield a multi-deal-prone cluster or the paper category.


Preliminary analysis involved a cross-tabulation comparison of characteristics of coupon vs. noncoupon purchases. A multinomial logit brand choice model was then used to compare purchase influences between purchases on which a coupon was used and those on which one was not used. The cross-tabulation compares purchase attributes which pertain to the selected brand, for example, the average price paid when a coupon is used and not used. The logit model compares the relative effect of a variable, for example, the shelf prices of all available brands, on choice for coupon and noncoupon purchases (see Malhotra 1984b for a review of application of this model). This model has been widely used in promotion studies (Hee 1981; Guadagni 1980; Novich 1981; Levin 1980; Jones and Zufryden 1980) and is based on a choice theory (Luce 1959).

The model expresses choice as a function of marketing variables for each brand:



I = brand chosen (where a choice is defined as a brand/size combination)

j = purchase occasion

i = attribute (promotion variables)

k = alternatives (available brands)

Xij1 = i th attribute of alternative I chosen on occasion j

Bi = coefficient of attribute i

Xijk = i th attribute of the k th alternative on the j th occasion


The model's consumer homogeneity assumption is mitigated in this research by running the choice model for individual deal-proneness segments. Within each segment, consumers are assumed to be homogeneous. Where sample sizes permitted, the files were further divided into two samples to test the model's stability. Note that in a few instances, this research was forced to use samples that fall slightly below the minimum size (N=50) generally considered acceptable for maximum likelihood estimation (Malhotra 1984b). The IIA (Independence of Irrelevant Alternatives) assumption is dealt with by calculating an "evoked set." The components of the available choice set were allowed to vary for each consumer and for each purchase. A brand was considered unavailable if the consumer had never bought the brand or if the brand was out of stock. In this way completely brand-loyal consumers were automatically eliminated from the analysis. The independent variables were the marketing variables for each brand on each observation: price change, price/unit, deal pack dummy, retailer ad dummy, and manufacturer ad dummy. There were some small correlations between these variables--with only coffee retail ads and price changes potentially problematic at .53.

The overall significance of the model was tested with a likelihood ratio statistic and with U-squared (Hauser 1978), a measure of the amount of uncertainty in the data explained by the model. T-tests were used to test the significance of toe coefficients and the relative contribution between the coefficients of the coupon and noncoupon models. All tests were at the 90% level of significance.

The brand choice model used in hypothesis 1 was expanded to include the additional variable net price-defined as shelf price per unit minus the average coupon value of all coupons redeemed on the brand during the week (corrected to a per unit basis). Net price represents the average price available to a consumer taking advantage of current coupons. Hypothesis 2 was tested by the coefficient of this variable--expected to be significant for the most deal sensitive consumers. For this analysis, the model was run on files combining both coupon and noncoupon purchases, the price variable was dropped from the model, and a new variable-- coupon shareCwas added. Coupon share accounts for the relative availability of a brand's coupons during the week of purchase and is defined as share of redemptions. Where possible, these files were subdivided by store of purchase.




The results of the comparison between coupon and non-coupon purchases are shown in Tables 2 and 3. Table 2 shows that price and price change remain virtually unchanged for both paper consumer segments irrespective of coupon use. These results fail to replicate the Levedahl findings. Both segments do show that consumers are more likely to be buying an advertised brand when they are using a coupon--although the result is not statistically significant--and are less likely to be buying a deal pack.

The results for coffee, shown in Table 3, are in the-direction predicted by the coupon primacy hypothesis. Consistently--across all three consumer groups shown--the price paid per ounce is higher when a coupon is used than when one is not used. While this finding is not statistically significant it is supported by two additional results. Consumers are more apt to respond to a price change or a retail feature when they are not using a coupon than when they are.

The results of the multinomial logit analysis are shown in Tables 4 and 5 for the coupon primacy hypothesis. There is only limited evidence to support the existence of a coupon primacy effect in brand choice. Table 4, for the paper category, shows that light coupon users have a similar pattern of purchase influences irrespective of coupon use. Heavy coupon users, those not expected to exhibit coupon primacy, do show minor differences between the two types of purchases. However, these differences are not evidence for the effect: manufacturer ads (which carry coupons) are associated with coupon purchases and shelf price/unit is statistically significant for these purchases. For the coffee category, both the heavy coupon users and the multi-deal- prone segment have a similar set of coefficients for coupon and noncoupon purchases. However, there is one example of promotion variables becoming less important when coupons are used. Light coffee coupon users appear likely to buy the retail advertised brand when not redeeming a coupon. When using a coupon, this association is not seen--there is a significant negative coefficient for retail advertising on coupon purchases. For these consumers, therefore, the use of the coupon would appear to- inhibit the "normal" effects of other purchase influences.



While the likelihood statistics are significant (using a Chi-square test) for these models, the U-squared shows that in most cases the models have modest explanatory power. These results are consistent with previous work in this area for the impact of promotion variables alone (Guadagni 1980; Hee 1981). The results are quite stable, however, as seen in the two samples generated for heavy coffee users. Eight out of the ten coefficients are well within one standard error between the two samples.

Table 6 shows strong evidence in favor of the lowest price hypothesis for the coffee category. Net price is significant for both the multi-deal segment and for store 1 (a warehouse type of grocery store) purchases for the heavy coupon-using segment. Heavy coupon users who do not shop in store 1, appear to make choices based on price changes alone. For these consumers net price--the price which incorporates coupon values--is not a significant predictor of choice. The multi-deal segment and store 1 shoppers are responding to a price variable which includes coupon values.


Despite the widespread belief in the interactive effects of coupons with other elements of the marketing mix, this research finds only limited evidence for these effects. For the most part, the results favor the null hypothesis: that coupons do not have large interactive effects with the other factors tested. Coupon and noncoupon purchases look surprisingly similar, both with respect to characteristics of the purchased brand and influences affecting brand choice. These results are analogous to previous work (Henderson 1984) profiling redemption occasions. Factors such as day of week were found to have no effect in distinguishing redemption occasions. The picture that emerges, therefore, is that coupons are largely incorporated in purchasing patterns without negating the usual influences on choice.

There is, however, some minor evidence supporting the coupon primacy hypothesis. All coffee consumers--not just those who are light coupon users-appear to be slightly less price sensitive or feature sensitive when using coupons. From this research alone, however, this result-is not really managerially relevant. Future research should test this finding in other product categories and with variables unmeasured in this study, particularly with display. If further evidence can be found for this effect, the heavy use of overlays and simultaneous events can be questioned.







This study shows an interesting pattern to the coupon primacy findings: the hypothesis is more clearly supported in the coffee category than in the paper category. As a potentially more salient purchase because of its higher shelf price, these results are not surprising. Certain coffee buyers may be so eager to take advantage of the coupon that their selection is largely guided by the coupon's availability. Makers of high-priced products might be particularly sensitive to this consumer tendency. High-valued coupons may be a sufficient inducement in themselves and may not require any Supporting programs.

On the other hand, other coffee consumers seem well able to integrate coupon usage with price considerations. The defining characteristics of such consumers are deal usage and store of purchase. This situation lends some modest support to the effectiveness of trying to achieve a lowest price position among a set of competing brands through the use of coupons. Manufacturers should note, however, that only a subset of consumers appear to be following a lowest price shopping strategy. This would argue for the selective use of high valued couponsCappropriately targeted to those segments who use such coupons in making choices. High valued coupons to other groups may be an inefficient use of promotional budgets. In designing effective promotional programs, manufacturers should recognize the importance of interactions between the program's elements. This study advocates the importance of conceptualizing and testing the behavioral mechanisms which can explain interactions on an individual level. Further research is needed to specifically test the relevant behavioral mechanisms. For example, although this study asserts that commitment to a coupon may prevent response to other promotions or prices, the exact casual ordering could be investigated in a laboratory setting. In addition, more complex choice models including variables beyond prices and promotions would enrich our understanding. Finally, -research is needed to separate the exact psychological mechanism involved--for example, precommitment or choice simplification involved in coupon primacy. Only further exploration of these effects--particularly for individual product categories--can help sharpen our understanding of consumer response in a cluttered environment.


Beem, E.R. and HJ. Shaffer (1981), Trigger to Customer Action - Some Elements in a Theory of Promotional Inducement, Marketing Science Institute.

Bettman, J.R. (1979), An Information Progressing Theory of Consumer Choice, Addison-Wesley.

Bettman, J.R. and M.A. Zins (1977), "Constructive Processes in Consumer Choice," Journal of Consumer Research, September.

Capon, N. and D. Kuhn (1982), "Can Consumers Calculate Best Buyer The Journal of Consumer Research, March.

Chevalier, M. (1975), "Increase in Sales Due to In-Store Display", Journal of Marketing Research, November.

Dennerlein, I.M. (1980), "Trade Promotions Net Two-Way Benefit", Advertising Age, May 5.

Dodson, J.A., A.M. Tybout, and B. Sternthal (1978), "Impact of Deals and Deal Retraction on Brand Switching", Journal of Marketing Research, February.

Eskin, G.J. and P.H. Baron (1977), "Effects of Price and Advertising in Test Market Experiments", Journal of Marketing Research, November.

Feigenbaum, E.A. and H.A. Simon (1962), "A Theory of the Serial Position Effect", The British Journal of Psychology, Volume 53.

Gardner, M.P. and R.A. Strang (1983), "Consumer Response to Promotions: Some New Perspectives", in T.C. Kinnear, ed., Advances in Consumer Research, Volume XI.

Guadagni, P.M. (1980), Market Response Measurement Using the Multinomial Multiattribute Logit Choice Model, unpublished M.S. thesis, Massachusetts Institute of Technology.

Hauser, J. (1978), "Testing the Accuracy, Usefulness, and Significance of Probabilistic Choice Models: An Information Theoretic Approach", Operations Research, May-June.

Hee, J.R. (1981), Determining Manufacturers' Couponing Strategies, unpublished M.S. thesis, Massachusetts Institute of Technology.

Henderson, C.M. (1984), "Modeling the Coupon Redemption Decision", in E.C. Hirschman and M.B. Holbrook, ed., Advances in Consumer Research, Volume XII.

Henderson, C.M. (1986), "Sales Promotion Segmentation: Refining the Deal-Proneness Construct", working paper.

Irons, H.W., J.D.C. Little, and H.L. Klein (1983), "Determinants of Coupon Effectiveness", in F.S. Zufryden, ed., Advances and Practices of Marketing Science 1985, Proceedings of the 1983 ORSA/TIMS Marketing Science Conference, University of Southern California.

Jones, J.M. and F.S. Zufryden (1980), "Adding Explanatory Variables to a Consumer Purchase Behavior Model: An Exploratory Study", Journal of Marketing Research, August.

Klein, R. (1981), "Using Supermarket Scanner Panels to Measure the Effectiveness of Coupon Promotions", Proceedings: Third ORSA/TIMS Market Measurement Conference, N.Y.U.

Levedahl, J.W. (1984), "Marketing Price Discrimination and Welfare: Comment", Southern Economic Journal, January.

Levin, A.R. (1980), A Market Response Model Based on UPC Scanner Data, unpublished M.S. thesis, Massachusetts Institute of Technology.

Luce, R.D. (1959), Individual Choice Behavior, John Wiley and Sons.

Malhotra, N.K. (1984a), "Information and Sensory Overload", Psychology and Marketing, Vol. 1, No. 314.

Malhotra, N.K. (1984b), "The Use of Linear Logit Models in Marketing Research", Journal of Marketing Research, February.

Neslin, S.P., C.M. Henderson, and J. Quelch (1985), "Consumer Promotions and the Acceleration of Product Purchases", Marketing Science, Spring.

Neslin, S.A. and N.W. Shoemaker (1983), "A Model for Evaluating the Profitability of Coupon Promotions", Marketing Science, Fall.

Nielsen Clearing House (1985), "What Consumers Think of Coupons", A.C. Nielsen Company.

Novich, N. (1981), Price and Promotion Analysis Using Scanner Data: An Example, unpublished M.S. thesis, Massachusetts Institute of Technology.

Prasad, V.K. and L.W. Ring (1976), "Measuring Sales Effects of Some Marketing Mix Variables and Their Interactions", Journal of Marketing Research, November.

Raju, P.S. and M. Hastak (1979), "Consumer Response to Deals: A Discussion of Theoretical Perspectives", in J.C. Olsen, ed., Advances in Consumer Research, Vol. VII, Association for Consumer Research.

Ray, M.L. (1973), "Psychological Theories and Interpretations of Learning", in S. Ward and T. Robertson, eds., Consumer Behavior: Theoretical Sources, Prentice-Hall.

Ray, M.L. and P.H. Webb (1978), Advertising Effectiveness in a Crowded Television Environment, Marketing Science Institute.

Reibstein, D.J. and P.A. Traver (1982), "Factors Affecting Coupon Redemption Rates", Journal of Marketing, Fall.

Rothchild, M.L. and W.C. Gaidis (1981), "Behavioral Learning Theory: It's Relevance to Marketing and Promotion", Journal of Marketing, Spring.

Russo, J.E. (1977), "The Value of Unit Price Information", Journal of Marketing Research, May.

Shimp, T.A. and A. Kavas (1984), "The Theory of Research Action Applied to Coupon Usage, Journal of Consumer Research, December.

Sunoo, D. and L. Lin (1978), Sales Effects of Promotion and Advertising", Journal of Advertising Research, October.

Teel, J.E., R.H. Williams and W.O. Bearden (1980), "Correlations of Consumer Susceptibility to Coupons in New Grocery Product Introductions", Journal of Advertising, Vol. 9, #3.

Urban, G.L., P.L. Johnson and J.R. Hauser (1984), "Testing Competitive Market Structures", Marketing Science, Spring.

Vanhonacker, W.R. (1979), Brand Switching and Market Partitioning: An Empirical Study of Brand Competition, unpublished Ph.D. dissertation, Purdue University.

Ward, R.W. and J.E. Davis (1978), "Coupon Redemption", Journal of Advertising Research, August.

Wilkinson, J.B., J.B. Mason and C.H. Paksoy (1982), "Assessing the Impact of Short-Term Supermarket Strategy Variables", Journal of Marketing Research, February.

Woodside, A.G. and G.L. Waddle (1975), "Sales Effects of In-Store Advertising", Journal of Advertising Research, June.

Wright, P. (1975), "Consumer Choice Strategies Simplifying vs. Optimizing", Journal of Marketing Research, February.