Effectiveness of sales promotional tools in Malaysia: the case of low involvement products.

Ndubisi, Nelson Oly

ABSTRACT

The thrust of this paper is to evaluate the effectiveness of sales promotional strategies namely, coupon, price discount, free sample, bonus pack, and in-store display in the purchase of low involvement products by Malaysian consumers. The paper also recognises that certain demographic factors such as education and income of consumers could potentially confound the observed relationships hence, these factors were controlled. A total of 312 consumers in Kota Kinabalu, Malaysia were surveyed using structured questionnaire. The results show that price discounts, free samples, bonus packs, and in-store display are associated with product trial. Coupon does not have any significant effect on product trial. Details of the findings and their implications are discussed.

Keywords: Promotion Strategies, Product trial, Low Involvement Product, Consumers, Malaysia.

INTRODUCTION

A large body of research on consumer responses to sales promotions (e.g., Bawa & Shoemaker, 1987 and 1989; Blattberg & Neslin, 1990; Leone & Srinivasan, 1996; Huff and Alden, 1998) has accumulated over the past few decades due to the growing importance of this marketing lever. However, too much stress on coupons at the expense of other equally important promotional tools, has created the need for more work to be done in the area of investigating (together with coupon) the effects of other sales promotional tools such as free sample, bonus pack, price discount, and in-store display on product trial and repurchase behaviour, especially among Malaysian consumers, whose behavioural responses to promotional strategies are ill understood due largely to lack of research on them.

Moreover, research on the use of marketing tools in Malaysia is very scanty at best. And very little (if at all) is understood about the Malaysian customers and their purchase behaviours, especially with regards to how they respond to the various promotional strategies practised by marketers. Since the bulk of the extant literature on these relationships till date remains the Western perspective; there is an urgent need for research focusing on the Malaysian consumers and the Malaysian environment, which is unfamiliar to most readers. Since understanding the behavioural responses of Malaysian customers to sales promotion strategies is salient in customer management and in designing effective sales promotion strategies, important impetuses for this research are established.

To embark on this task, the research focuses on low involvement products (LIP), which are generally believed to be more responsive to promotional tools than high involvement products. Low involvement products are those that are bought frequently and with a minimum of thought and effort because they are not of vital concern nor have any great impact on the consumer's lifestyle (www.marketingprofs.com). Not all purchase decisions are equally important or psychologically involving for the consumer. People engage in a less extensive decision-making process, involving a less detailed search for information and comparison of alternatives, when buying low involvement goods and services than when purchasing high involvement items. Because of the differences in the decision-making process between low and high involvement products, and the high frequency of purchase of low involvement products, this study focuses on the LIPs in order to unveil promotional strategies that might be more effective in the Malaysian context.

LITERATURE REVIEW

According to Shimp (2003), sales promotion refers to any incentive used by a manufacturer to induce the trade (wholesalers, retailers, or other channel members) and/or consumers to buy a brand and to encourage the sales force to aggressively sell it. Retailers also use promotional incentives to encourage desired behaviours from consumers. Sales promotion is more short-term oriented and capable of influencing behaviour. Totten & Block (1994) stated that the term sales promotion refers to many kinds of selling incentives and techniques intended to produce immediate or short-term sales effects. Typical sales promotion includes coupons, samples, in-pack premiums, price-offs, displays, and so on.

Coupons have been used to produce trial (Robinson & Carmack 1997). According to Cook (2003), coupons are easily understood by the consumer and can be highly useful for trial purchase. Gilbert and Jackaria (2002) concurring to the popularity of coupon reported that coupon is ranked last as the promotional least widely used by consumers and least influence on product trial. Other studies (e.g. Peter & Olson 1996; Gardener & Trivedi 1998; Darks 2000; Fill 2002) have reported the importance of coupons as a sales tool.

Price promotion does influence new product trial (Brandweek, 1994). According to Ehrenberg et al. (1994) short-term peaks in sales were due primarily to purchases made by occasional users of a brand rather than by new customers. Furthermore, the study concluded that these occasional users, after taking advantage of the price reduction, would most likely return to their favourite brands in their portfolio rather than buy the promoted brand at full price. However, Shimp (2003) and Fill (2002) among other extant studies have documented a link between price promotion and product trial.

With regard to free sample, another important promotional tool often used by firms, marketing managers recognize the importance of product trial and direct behavioural experience with a product, hence they often mail free samples of products to consumers so that consumers can try the products for themselves, rather than just hear about the products (Kardes, 1999). However, Gilbert and Jackaria (2002) found that a free sample as a promotional offer had no significance on consumers' reported buying behaviour, whereas Pramataris et al. (2001), Fill (2002), and Shimp (2003), have shown otherwise.

Factory bonus pack according to Lee (1963) is used to increase consumer trial of the brand. Larger package size and accompanying advertising of the offer tended to make the promotion noticeable (Gardener & Trivedi 1998). Since more of the product is included at no extra cost, consumers can be persuaded to buy the product if they feel it represents a deal that produces the greatest value for their money. According to Gilbert and Jackaria (2002), packs with "buy-one-get-one-free" may not increase brand awareness before trial purchase because the customer will only come across the product once in the store (unlike samples or coupons), however, if the promotion is noticeable it will facilitate brand recognition and brand recall for future purchases. Since an additional amount is given for free, consumers may be persuaded to buy the product if they feel it represents a fair deal that provides value for money. Ong et al. (1997) found that consumers appeared to be slightly sceptical of the bonus pack offer, but somewhat more trusting of the price and quantity claimed. In other words, believability of the bonus pack offer was weak, however, they would likely buy one bottle and not buy more than one bottle they concluded. The report speculated that this happens because consumers suspect that manufacturers do raise prices slightly in conjunction with bonus pack offerings.

Product trial involves actually trying or using a product (Kardes, 1999). According to Peter and Olson (1996), trialability refers to the degree to which a product can be tried on a limited basis or divided into small quantities for an inexpensive trial. Banks (2003) wrote that with sales promotion, brands have a chance to quickly affect consumer choice and behaviour by adding value through an on-pack offer, by achieving incremental display or by encouraging trial via sampling and/or couponing. According to Schindler (1998), a price promotion that is designed to evoke attributions of responsibility could be expected to appeal to consumers more than one that does not evoke such attributions, and thus have a greater ability to create product trial among consumers. Wayne (2002) found a link between sales promotion and product trial. Chandon, et al. (2000) indicated that sales promotion may be attractive to highly promotion prone consumers for reasons beyond price savings. These highly promotion prone consumers may switch brands to receive "special" deals that reflect and reinforce their smart shopper self-perception. They concluded that highly promotion prone consumers might try a new product that has promotion. Thomas (1993) argued that the magnitude of planned distribution and promotion expenditures (advertising, sales promotions, sales force, and so on) could affect initial trial of the brand. Based on the issues and discussion raised above, the following hypotheses are generated for verification:

H1: There is a significant positive relationship between coupon and product trial.

H2: There is a significant positive relationship between price discount and product trial.

H3: There is a significant positive relationship between free sample and product trial.

H4: There is a significant positive relationship between bonus pack and product trial.

H5: There is a significant positive relationship between in-store display and product trial.

All told, it is important to jog the readers mind that none of the extant literature cited above is about Malaysia. This observation corroborates one of the key impetuses for the conduct of this study, that is, the dearth of research on the Malaysian customers' behavioural responses to promotional tools, which has resulted in a poor understanding of the effectiveness of various promotional tools in the same context.

METHODOLOGY

In this study, five consumer promotion tools- coupons, discount, samples, bonus packs, and in-store display were investigated for their impact on consumer purchase behaviour. Items from Garretson and Burton's (2003) study of consumer proneness towards sales promotion were adapted in the measurement of proneness to coupon, price discount, free sample, bonus pack, and in-store display. Trial behaviour of consumers were measured with items adapted from Gilbert and Jackaria (2002).

Figure 1 shows the research model. Questionnaire was used for the study. The population of the study consists of consumers in Kota Kinabalu, Malaysia. The sample points were supermarkets in Kota Kinabalu area. The survey instrument was self-administered to customers using a mall intercept technique. Some respondents who could not answer on the spot were given a copy of the questionnaire (to be answered at home) with a postage paid return envelope. A five point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree) was used for the construct's dimensions. A total of 420 questionnaires were distributed and only 312 were returned, which represents a response rate of 74%. The Multiple Regression Model was employed to predict the relationships in the construct. The schema of the research model is shown as Figure 1.

[FIGURE 1 OMITTED]

The regression assumptions with respect to autocorrelation (independent of residual), normality (residual is normally distributed), homoscedasticity of error terms, multicollinearlity and linearity of independent variables were verified before making any interpretation of the statistical results.

RESULTS AND DISCUSSION

Demographic Profile

Out of the 312 usable questionnaires returned by the respondents, 59.6% were female respondents, and 40.4% were male. Various income levels (in Malaysia Ringgit) were represented, for example below 24,000 was 60.9%, 24,000-47,999.99 (29%), 48,000-71,999.99 (9%), and so on. Chinese consumers made up 51%, Kadazan-Dosuns (22%), Malays (16%), Indians and others (11%). The ages of the respondents were as follows: below 20 (13%), 20-39 (62.8%), 40-59 (22%), and 60 and above (2.2%). The rate of married respondents was 46%, while singles represented the balance of 54%. With respect to education background, 50.6% had secondary school education and less, 26% had high school and diploma qualifications, and the rest (23.4%) were degree and postgraduate degree holders.

Psychometric Properties of the Instrument

Factor Analysis was performed on the questionnaire items in order to establish their suitability for performing the subsequent multivariate analyses. The results presented are based on parsimonious sets of variables, guided by conceptual and practical considerations, namely acceptance of factor loadings of .50 and above (Hair et al, 1998), and cross loadings mostly below .20. In very rare cases where cross loadings slightly exceeded .20, loadings were much higher than .50 to justify acceptance. The orthogonal rotation was employed for this analysis (Hair et al. 1998). High communality values were recorded for each variable, indicating that the total amount of variance an original variable shares with all other variables included in the analysis is high. Overall, the results in Table 1a and 1b show that the construct measures are valid. Put differently, the measures define the concept of study very well. Table 1 shows the factor loadings and cross-loadings, Eigenvalues, and Variance of the dimensions under examination.

A total of 29 items loaded on 6 factors. Factors 1 to 5 contain items measuring free sample, bonus pack, price discount, in-store display, and coupon. Five items were used in each case (save for Factor 4) for example: (1) If a brand offers--(free sample/bonus pack/price discount/in-store display/coupon); that could be a reason for me to buy it, (2) When I buy a brand that offers--(free sample/bonus pack/price discount/coupon), I feel I am getting a good buy; (3) I have favourite brands, but most of the time I buy a brand that offers--(free sample/bonus pack/price discount/in-store display/coupon); (4) One should try to buy a brand that offers--(free sample/bonus pack/price discount/in-store display/coupon); and (5) compared to most people, I am more likely to buy brands that offer free--(free sample/bonus pack/price discount/in-store display/coupon). Factor 4 has four items only, because of the omission of item 2, which is considered irrelevant with respect to in-store display. Factor 6 has five items measuring trial, for example, coupon enables me to buy a product, which I have not tried before, price discount makes me to buy a product, which I have not tried before, etc. Total variance explained is 66% and item loadings are quite high, hence, there is high validity for the constructs measures.

Although the observed patterns of item loadings were similar for both Varimax (adopted in this study) and Oblique rotation (alternative technique), providing grounds to assume that the instruments are consistent, the internal consistency of the instruments were further tested via reliability analyses. Cronbach's alpha test was used to ensure the reliability of the variables. For sales promotional tools, the results indicate acceptable values: coupon (α=0.81), price discount (α= 0.86), free sample (α= 0.87), bonus pack (α=0.88), and in-store display (α= 0.87). The Cronbach's alpha value for product trial is 0.81. Mean score for all dimensions are as follows: coupon (2.99), price discount (3.67), free sample (3.08), bonus pack (3.28), in-store display (2.84), product trial (3.22).

Relationship among Constructs

Table 3 shows the results of the regression analysis used to determine the relationship between the promotional strategies and product trial. Standardized beta coefficients are reported all through, as standardized regression coefficients allow for a direct comparison between coefficients as to their relative explanatory power of the dependent variable (Hair et al. 1998).

The above results show that coupon, price discount, free sample, bonus pack, and in-store display contribute significantly (F = 25.22; p = .000) and predict approximately 30% of the variations in product trial. The 30% explanation is considered good for a behavioural science research. Further examination of the results shows that price discount (t = 2.334; p = .020), free sample (t = 3.483; p = .001), and in-store display (t = 4.322; p = .000) are significantly associated with product trial at 5% significance level. Bonus pack is moderately associated with product trial (t = 1.900; p = .058). Hence there is enough evidence to accept hypotheses 2, 3, 4 and 5. The results indicate that in-store display is the strongest predictor of product trial followed by free sample, price discount and bonus pack. There is no significant relationship between coupon and product trial (t = 0.401, p = 0.69) at 5% significance level, which leads to rejection of hypothesis 1. Therefore, it is conclusive that coupon is not a strong determinant of product trial among the respondents. This may be because of the sparse use of coupon as a promotional strategy by marketers in Malaysia. As a most rarely used promotional tool in Malaysia, many consumers may not be familiar with it compared to other promo-tools.

The results of this study provide some useful information on the impact of the five promotional tools on consumer buying behaviour (product trial). With respect to consumer proneness to sales promotions, the results show that in-store display plays a significant role in shaping consumer product trial reaction. Moreover, the results of this study show that free sample and price discount play significant roles in influencing consumer product trial behaviour. This finding is consistent with the views of Blackwell et al. (2001). Another sales promotional tool that has important effect is bonus pack. Bonus pack is instrumental in increasing consumer trial of a brand, thus, the more of the product included at no extra cost, the greater the likelihood of consumers buying the product for trial. Although, the effect of bonus pack on product trial is lower than other promotional tools such as in-store display, free sample, and price discount, bonus pack remains a useful marketing tool.

Contrary to some earlier findings (e.g. Banks 2003; Blackwell et al. 2001), coupon in this study does not have significant effect on product trial. This could be as a result of the respondents' poor familiarity with the use of coupons. In fact in Malaysia, the use of coupons as a promotional strategy is not as common as the use of other promotional tools. Marketers in Malaysia very seldom use coupons, resulting in the tool's unpopularity among Malaysian consumers. Zajonc (1980) had earlier shown that exposure to a stimulus enhances a person's attitude toward it.

Control

The control procedures applied in this study include the following: (1) examination of the role of familiarity with each promotional tool on the impact (or lack of it) of the tool on product trial; and (2) examination of potential confounding effects of respondents' education and income levels.

Firstly, to examine whether consumer familiarity with particular promotional tool is what explains its effectiveness, the study controlled for this factor. From the result in Table 4 below, it can be said that the weak impact of coupon on trial is attributable to the unfamiliarity of Malaysian customers with coupon. This may have resulted from the seldom use of this tool by marketers in Malaysia.

Further analysis confirms that familiarity is a key issue in coupon-trial relationship. Taking coupon use at below median and above median (inclusive), the corresponding values were assigned 1 and 2 respectively. Similarly, taking familiarity with coupon at below median and above median (inclusive), 0 and 1 were assigned. By plotting the graph of these dimensions, the resulting levels of product trial for different levels of coupon usage and familiarity with the tool are as shown in Figure 2.

Figure 2 shows that customers, who are familiar with coupon, increase product trial as more coupon are offered. In other words, the behavioural responses of highly coupon-familiar customers are greater than the responses of those who are not familiar with the tool.

[FIGURE 2 OMITTED]

Potential Confounding of Education and Income Divide

There are several important demographic variables that could potentially confound observed relationships depending on the nature of the study. These are level of education, age, gender, and income (Minton & Schneider 1980; Praeger 1986; Kite 1996; Morris & Venkatesh 2000). With respect to responding to promotional tools, the most important covariates are those whose inclusion might theoretically eliminate observed moderation effects, such as income and education. This is because high-income earners may be less interested in few dollar savings than their lower income counterpart; hence, they may be less responsive compared to the low-income group. In addition, since education and income are positively correlated, similarly more educated consumers may be less responsive to promotional tools than their less educated counterpart. Thus, in this research it is necessary to evaluate and control for possible confounding effect of education and income.

Before the control mechanism was applied, the two demographic variables, which originally had more than two groups, were recoded into two groups for ease of understanding. Thus, educational levels were re-grouped into non-graduates and graduates, and income into low-mid and high-income earners. In order to introduce the recoded demographic dimensions into the regression model, dummy variables were created for the groups (Hair et al. 1998). In creating the dummy variables, the first step was to determine the number of dummy variables, which is simply k-1, where k is the number of levels of the recoded variable. In this instance 1 (2-1) dummy variable was created as follows: non-graduate (0), graduate (1); and low-mid income (0), high-income (1). The results of the controlled hierarchical regression analyses are presented in Table 5 below. These results are compared with the uncontrolled results in Table 3 above.

The results in Table 5 show that education and income levels are not confounds. In all the cases, the demographic variables have no significant relationship with the dependent dimension. By introducing each demographic variable in the 2nd stage of the hierarchical regression, it was found (as shown in Table 5) that the significant impacts witnessed in the price discount-trial, sample-trial, bonus pack-trial, and display-trial relationships (see Table 3) remain significant, and the non-significant effect in the coupon-trial relationship remains insignificant after controlling for potential confounds. This shows that there is no confounding effect. If education or income is confounding the results, when controlled, the significant effects will become non-significant, and the non-significant effect will become significant.

LIMITATIONS AND FUTURE RESEARCH

Like any empirical research, there are few limitations to this study. First, only one product category (low involvement products) was considered in this study (which is a pioneer in the Malaysian context), leaving out high involvement products, which are somewhat noted for their poor responses to promotional tools. An interesting future research direction is to estimate the effectiveness of these promotional tools in high involvement product situations. Since some scholars believe that high involvement products are not as responsive as low involvement products to promotional tools, it is needful to verify this view in Malaysia. This future research will help to increase present knowledge in this area by providing empirical support for or refuting the above supposition. Additionally, future research may take a comparative approach between the high and low involvement products to see if they equally or differentially respond to promo tools, and if the former is statistically less responsive than the latter in the Malaysian context.

There is still an urgent need to investigate the impact of other promotional tools on product trial because research in this area is still inconclusive. Beside the five promotional tools (i.e., coupon, price discount, free sample, bonus pack, and in-store display) that were examined in this study, future research may investigate other types of sales promotions (e.g., contests, refund) on product trial. In addition, studies that utilize data compiled by retailers that track buying and sales promotion participation habits across various tools will add much value since it is based on hard data rather than perceptions.

CONCLUSIONS AND IMPLICATIONS

This research has important implications on theory. The framework provides new insights into the understanding of sales promotional strategies and their impacts on Malaysian customers' behavioural responses in low involvement product setting. In addition, it helps to explain the role of familiarity with sales promotion tools. Malaysian consumers respond more to free sample, price discount, in-store display, and bonus pack than coupon. A plausible explanation for the weak influence of coupon is poor familiarity with the tool. This research shows the linkages among various promotional tools and product trial, and thereby helps to better understand how Malaysian consumers respond to various promotional tools offered by marketers. This is an important contribution to the body of knowledge in this field and in Malaysia in particular, being one of the pioneer studies in this area in Malaysia.

The results also have important implications for practitioners. One of the major implications of this research is that firms can increase sales by offering the right promotional tools to attract trial customers. Therefore organisations should carefully plan their promotional strategies, and allocate promotional budget over the different promotion tools, giving preference to the more effective tools. Promotions that emphasize in-store display, free sample, price discount, and bonus pack are likely to be more effective than coupon.

Second, the findings indicate that in-store display proneness has the strongest effect on product trial compared to other sales promotional tools. Attractive in-store display practices are necessary to gain the greatest sales from product trial. Third, (as shown in the results) bonus pack, free sample, and price discount significantly affect product trial, albeit the determinant power of bonus pack is the lowest among other promotional tools. Thus, one of the ways to improve the determinant power of bonus pack is to keep a regular pack along side with a bonus pack on the shelves, in order to enable consumers to make comparison. Such opportunity for a comparative observation will help to enhance the credibility of the tool and consumers' confidence in it. With regard to free sample and price discount, sellers should continue to apply them because of their robust influences on product trial.

Fourth, the findings show that coupon have no significant effect on product trial. This is largely due to consumers' unfamiliarity with the tool. Thus, it is suggested that manufacturers and retailers should use more of coupons in their promotional efforts, with longer redemption period, prior to which they should create greater awareness of the benefits of coupons and how they could be redeemed. This will help ignorant customers to be better informed about coupons and their uses. Another probable reason for the poor influence of coupon may be because coupons provide less shopping convenience benefits, require more skill and effort than buying a product on sale. For example keeping the coupon and redeeming it before expiring date, searching for a product that has coupon, matching coupons with brands, etc can be cumbersome and time consuming. In the other hand, price discount, free sample, bonus pack, and in-store display can provide greater shopping convenience benefit.

Lastly, it is important to note that the outcome of this research, that is, the observed significant and non-significant relationships among the independent and dependent variables are not confounded by respondents' educational and income levels. In other words, these observations hold true, irrespective of the level of education or income of the respondents, hence the results are generically applicable to all income and educational groups.

REFERENCES

Banks, P. (2003). Store was set to rage on. Ireland's Marketing Monthly. 14(8), online, available at: http://www.marketing.ie/sep03/article4.htm.

Bawa, K. & Shoemaker, R.W. (1987). The effects of a direct mail coupon on brand choice behavior. Journal of Marketing Research. 24, 370-376

Bawa, K. & Shoemaker, R.W. (1989). Analyzing Incremental Sales from a Direct Mail Coupon Promotion. Journal of Marketing. 53 (3), 66-76.

Blackwell, R.D., Miniard, P.W. & Engel, J.F. (2001). Consumer Behavior, (9th Edition). Fort Worth, TX: Harcourt College Publishers.

Blattberg, R.C. & Neslin, S.A. (1990). Sales Promotion: Concepts, Methods, and Strategies. N.J.: Prentice Hall. Brandweek, (1994). Promotional influence spurs buyers to try something new. Brandweek. 35 (12): 32-34.

Chandon, P., Laurent, G., & Wansink, B. (2000). A Benefit Congruency Framework of Sales Promotion Effectiveness. Journal of Marketing. 64 (4): 65-81.

Cook, A. (2003). How to cash in on the coupon craze. Incentive Business. Jun/Jul: 3

Darks, M.C. (2000). Calling all Coupon Clippers. Business News New Jersey. 13(13): 16.

Ehrenberg, A.S.C., Hammond, K. & Goodhardt, G.J. (1994). The after-effects of price related consumer promotions. Journal of Advertising Research. 34 (4): 11-21.

Fill, C. (2002). Marketing Communications: Contexts, Strategies and Applications, (3rd Edition). Italy: Pearson Education Limited.

Gardener, E. & Trivedi, M. (1998). A communication framework to evaluate sale promotion strategies. Journal of Advertising Research. 38 (3): 67-71

Garretson, J.A., & Burton, S. (2003). Highly Coupon and Sale Prone Consumers: Benefits Beyond Price Savings. Journal of Advertising Research. 43 (2), 162-172.

Gilbert, D.C. & Jackaria, N. (2002). The Efficacy of Sales Promotions in UK Supermarkets: A Consumer View. International Journal of Retail & Distribution Management. 30 (6): 315-322.

Hair, J.F., Jr., Anderson, R.E., Tatham, R.L., Black, W.C. (1998). Multivariate Data Analysis. (5th Edition), New Jersey: Prentice Hall, Inc.

Huff, L.C. & Alden, D.L. (1998). An Investigation of Consumer Response to Sales Promotions in Developing Markets: A Three Country Analysis. Journal of Advertising Research. 38 (3), 47-56.

Kardes, F.R. (1999). Consumer Behavior: Managerial Decision Making. New York: Addison Wesley Longman.

Kite, M. (1996). Age, Gender, and Occupational Label, Psychology of Women Quarterly, 20, 361-374.

Lee A. (1963). Sales Promotion Effectiveness Can Be Measured. Journal of Marketing. 27 (4): 69-70

Leone, R.P. & Srinivasan, S.S. (1996). Coupon Face Value: Its Impact on Coupon Redemptions, Brand sales, and Brand Profitability. Journal of Retailing. 73 (3): 273-89.

Marketingprofs (2004). High Involvement and Low Involvement Products, available at http://www.marketingprofs.com. Site visited on 19/1/05.

Minton, H.L. & Schneider, F.W. (1980). Differential Psychology, Prospect Heights, IL: Waveland Press.

Morris, M.G. & Venkatesh, V. (2000). Age Differences in Technology Adoption Decisions: Implications for a Changing Workforce, Personnel Psychology, 53, 375-403.

Ong, B.S., Ho, F.N. & Tripp, C. (1997). Consumer perceptions of bonus packs: An exploratory analysis. Journal of Consumer Marketing. 14 (2): 102-112.

Percy, L., Rossiter, J.R., & Elliott, R. (2001). Promotion Tactics. Strategic Advertising Management. Oxford: Oxford University Press

Peter, J.P. & Olson, J.C. (1996). Consumer Behavior and Marketing Strategy, (4th Edition). New York.: Irwin.

Praeger, K. (1986). Identity Development, Age, and College Experience in Women, The Journal of Genetic Psychology, 147(1), 31-36.

Pramataris, K.C., Vrechopoulos, A.P., & Doukidis, G.I. (2001). The transformation of the promotion mix in the virtual retail environment: An initial framework and comparative study, available at http://www.eltrun.aueb.gr/papers/iec.htm. Site visited on 28/11/04

Robinson, W.A., & Carmack, L.L. (1997). Best Sales Promotions, (6th Volume). New York: NTC Business Books.

Schindler, R.M. (1998). Consequences of perceiving oneself as responsible for obtaining a discount: evidence for smart-shopper feelings. Journal of Consumer Psychology. 7 (4): 371-392.

Shimp, T A. (2003). Advertising, Promotion, and Supplemental Aspects of Integrated Marketing Communications. (Sixth Edition). United States of America: Thomson South-Western.

Thomas, R.J. (1993). New Product Development: Managing and Forecasting for Strategic Success. U.S.A: John Wiley & Sons, Inc.

Totten, J.C., & Block, M.P. (1994). Analyzing Sales Promotion Text & Cases: How to Profit from the New Power of Promotion Marketing, (2nd Edition). U.S.A.: The Dartnell Corporation.

Wayne, M. (2002). Hitting your target with direct mail coupons. Marketing Magazine. 107 (22): 19-1/2p.

Zajonc, R.B. (1980). Feeling and Thinking: Preferences Need No Inferences, (citation) Huff, L.C., & Alden, D.L. An Investigation of Consumer Response to Sales Promotions in Developing Markets: A Three Country Analysis. Journal of Advertising Research. 38 (3), 47-56.

Table 1a: Factor Results for the Independent Dimensions Loadings and
Cross Loadings

 F1 F2 F3 F4 F5 F6

F1--Free Sample
Free sample 1 .723 .219 .152 .172 .079 .161
Free sample 2 .742 .170 .189 .130 .130 .151
Free sample 3 .716 .129 .205 .169 .108 .066
Free sample 4 .746 .165 .155 .120 .138 .117
Free sample 5 .757 .222 .172 .144 .088 .135
(Eigenvalue = 8.98; Variance = 30.95%)

F2--Bonus Pack
Bonus pack 1 .120 .801 .163 .062 .009 .124
Bonus pack 2 .105 .848 .083 .011 .053 .103
Bonus pack 3 .233 .703 .257 .118 .124 .087
Bonus pack 4 .269 .710 .143 .127 .309 .064
Bonus pack 5 .254 .734 .211 .156 .137 .075
(Eigenvalue = 2.76; Variance = 9.52%)

F3--Price Discount
Price discount 1 .147 .216 .706 .088 .102 .161
Price discount 2 .135 .129 .726 -.049 .154 .184
Price discount 3 .126 .175 .739 .086 .193 .062
Price discount 4 .275 .065 .726 .081 .190 .018
Price discount 5 .177 .204 .792 .091 .162 .017
(Eigenvalue = 2.18; Variance = 7.52%)

F4--In-store Display
In-store display 1 .095 .113 .112 .760 .019 .087
In-store display 2 .213 .069 .066 .819 .065 .033
In-store display 3 .193 .074 .014 .789 .179 .151
In-store display 4 .203 .073 .049 .847 .128 .150
(Eigenvalue = 2.04; Variance = 7.03%)

F5--Coupon
Coupon 1 .024 .098 .057 -.046 .729 .258
Coupon 2 .139 .119 .054 -.031 .726 .148
Coupon 3 .095 .052 .197 .212 .671 -.074
Coupon 4 .136 .126 .206 .137 .691 -.001
Coupon 5 .098 .065 .315 .169 .733 .000
(Eigenvalue = 1.67; Variance = 5.78%)

F6--Trial
Trial 1 .085 .055 .121 .038 .253 .724
Trial 2 .124 .082 .149 .026 .057 .787
Trial 3 .234 .051 .046 .116 -.001 .782
Trial 4 .154 .172 .045 .212 .012 .731
Trial 5 -.095 .104 .047 .049 .022 .534

(Eigenvalue = 1.47; Variance = 5.08%)

Total Variance = 66%

Table 2: Descriptive and Reliability Analysis Results

 Cronbach's
 No. of Alpha
Variables Items Mean S/D Coefficient

Coupon 5 2.99 0.77 0.81
Price Discount 5 3.67 0.75 0.86
Free Sample 5 3.08 0.81 0.87
Bonus Pack 5 3.28 0.77 0.88
In-store Display 4 2.84 0.84 0.87
Product Trial 5 3.22 0.73 0.81

Table 3: Promotional Tools and Product Trial

Independent variables Beta coefficients t-value p-value

Constant 5.932 0.000
Coupon 0.023 0.401 0.689
Price discount 0.143 2.334 0.020
Free sample 0.218 3.483 0.001
Bonus pack 0.114 1.900 0.058
In-store display 0.234 4.322 0.000

[R.sup.2] = .29

4F = 25.218

Sig. F = .000

Table 4: The Role of Familiarity on the Effectiveness of Tools

Variables B Sig.

Coupon * Familiarity 0.779 0.017
Price Discount * Familiarity 0.068 0.874
Sample * Familiarity -0.28 476
Bonus Pack * Familiarity -0.457 0.236
In-store Display * Familiarity 0.356 0.237

[R.sup.2] 0.289

Table 5: Testing for Potential Confounding of Demography

Interaction Terms
with Education
controlled Dependent Variable--Trial

 beta p-value

Coupon .041 .477
Price Discount .138 .025
Sample .238 .000
Bonus Pack .098 .064
In-store Display .245 .000
Education .075 .143

Interaction Terms
with Income
controlled Dependent Variable--Trial

 beta p-value

Coupon .022 .693
Price Discount .145 .019
Sample .223 .000
Bonus Pack .112 .064
In-store Display .236 .000
Income .029 .563

Nelson Oly Ndubisi, Monash University Malaysia