Consumer affective responses to direct mail ...

Journal of Marketing Communications Vol. 17, No. 5, December 2011, 337–353

Consumer affective responses to direct mail messages: The effect of gratitude and obligation Natalia Kolyesnikovaa*, Sara L. Sullivan Doddb and Coy Callisonc a Dept. of Nutrition, Hospitality & Retailing, Box 41240, Texas Tech University, Lubbock, TX, 79409-1240, USA; bCollege of Human Sciences, Texas Tech University/Texas Agrilife Extension, Box 41162, Lubbock, TX, 79409-1162, USA; cCollege of Mass Communications, Box 43082, Texas Tech University, Lubbock, TX, 79409-3082, USA

The research opens new ways of utilizing emotional content in direct mail messages. The study considered the capacity of gratitude (e.g. to personnel for services received) and obligation (e.g. to make a purchase) to act as separate affective influences upon consumer purchase intent and attitudes toward the company. Participants (N ¼ 120) were exposed to direct mail copy text in mock postcards sent from a hypothetical winery to its visitors. A total of six postcards were used (two gratitude-inducing, two obligation-inducing, and two neutral controls). The study offers strong evidence that gratitude and obligation operate differently in their impact on consumer attitudes and behaviors. Gratitude exerts a more positive influence than does obligation. Communicating through gratitude, and even neutral, messages was found to be more persuasive than those messages conveying obligation. Keywords: direct mail messages; reciprocity; affective responses

Introduction In today’s increasingly competitive global economy, producers of goods and services expend significant resources to identify their customers and communicate with them. The goal is not only to generate revenue through sales but also to build brand loyalty and positive company perceptions among consumers. It is a complex and continuing process to attract and retain customers. The companies who do it best make the most of each opportunity to interact with their consumers. In today’s marketplace, direct marketing is not optional anymore. Thomas (2007) points out that direct marketing is not a tactic, a device for accomplishing an end. Rather, direct marketing needs to be strategic, relying on a careful plan to achieve long-term goals of building brand loyalty. Thus, direct marketing is about focused, well planned communication with targeted customers. The nature and quality of direct communication between a business and its customers has been subjected to extensive research. The outcome of these scholarly efforts has been a growing recognition that consumer attitudes and purchasing behavior are formed not merely through cognition but also through affective processing. The challenge then becomes determining the most efficacious ways to craft marketing messages that will reach the targeted consumer who processes the message and, ideally, reacts with the intended response – either activated purchase or a positive perception of the company. This study explores affective facets of consumer responses to a direct mail message. We present a

*Corresponding author. Email: [email protected] ISSN 1352-7266 print/ISSN 1466-4445 online q 2011 Taylor & Francis http://dx.doi.org/10.1080/13527266.2010.481142 http://www.tandfonline.com

338

N. Kolyesnikova et al.

theoretical framework to investigate the dimensions of two affective responses that have been overlooked in marketing research – gratitude and obligation. Literature review Although the focus of this study is an impact of a remotely delivered marketing message (direct mail) on consumer intentions, building the rationale for such a study must by necessity be grounded in the literature of advertising, which in essence is research about consumer responses to strategically crafted marketing messages. If marketers want to exploit the full potential of direct mail messages, they must first understand the relationship between consumer behavior and advertising. Scholars have sought to understand and model the relationship between advertising and consumer behavior for over a century (e.g. E. St Elmo Lewis in 1898, as cited in Vakratsas and Ambler 1999). It was not until the late twentieth century that empirical evidence indicated there were multiple components to consumer attitudes toward advertising, and by extension, consumer behaviors. The effects of advertising were traditionally conceptualized through cognitive response theories; that is, consumers evaluate advertising primarily using recall and recognition of content and brand information and semantic judgments (Greenwald 1968; Wright 1973). In the 1980s, however, multiple studies identified the ability of advertising to generate affective responses strong enough to influence consumer purchasing behavior and brand perception (see Batra and Ray 1986; Edell and Burke 1987). Once affective responses were recognized as playing a role alongside cognitive responses, the challenge for researchers became how to identify, classify, measure, and validate affective responses. Defining affect as ‘emotions, moods, feelings, and drives’, Batra and Ray (1986, 235) provided a comprehensive review of affective typologies. Later, Batra and Holbrook (1990) validated a typology of affect suitable for advertising. The researchers ultimately recommended 12 different types of distinct affective responses to measure advertising effectiveness: activation (arousal), skepticism, anger, restful, bored, fear, desire, social affection, sadness, irritation, surgency (playfulness), and gratitude; with another eight words of less powerful validity (tension, confidence, dominated, drowsy, elation, moral, scornful, and soothed) (Batra and Holbrook 1990, 22). Most of these affect types have underlying dimensions, which have been individually explored, scaled, and validated for measurement purposes. Words and synonyms that Batra and Holbrook (1990) used in their typology anchored measurements of positive and negative affective responses to different kinds of advertising messages. For example, one such scale constructed to measure warmth used words such as affectionate, calm, kind, peaceful, sentimental, and touched (Burke and Edell 1989). Another scale measuring disapproval used words such as disgusted, contemptuous, distrustful, uneasy, and worried (Bhat, Leigh, and Wardlow 1998). As useful as these scales may be, certain types of affective responses have been overlooked. One of the more interesting gaps in terms of communication messages centers on gratitude and obligation, key components of reciprocity theory. Theoretical background: Reciprocity The norm of reciprocity traces its roots to Alvin Gouldner (1960), who claimed that for a social system to be stable, there must always be some mutual exchange of benefits. One implication of the theory is that people feel obligated to reciprocate positive behavior,

Journal of Marketing Communications

339

which leads to the introduction of the two conditions that make the reciprocity rule work: gratitude and obligation. Gouldner (1960, 176) argued that gratitude and obligation operate as independent predictors of reciprocal behavior: ‘the sentiment of gratitude joins forces with the sentiment of rectitude and adds a safety-margin in the motivation to conformity’. That is, beneficiaries of favors feel indebted to repay benefits because of a compliance with social norms resulting from a state of obligation. Alternatively, a beneficiary may reciprocate out of a desire to express gratitude. We further suggest that these two components of the reciprocity theory – gratitude and obligation – are important dimensions of consumer behavior. For example, consumers may feel appreciation (gratitude) to the personnel for services provided. On the other hand, consumers may feel pressure (obligation) to make a purchase in return for services received. Over the past 40 years, numerous social studies examined the effect of gratitude and obligation on human behavior. These studies focused mainly on the relationship between receiving a favor and subsequent compliance with a request from the favor-giver. Receiving even a small favor leads to further compliance with the request of the person who provided the favor (Greenberg and Shapiro 1971; Howard 1995). This effect is robust even when the favor is unsolicited and unexpected (Regan 1971), and even when a favor-giver is not perceived as likable (Goei et al. 2003; Regan 1971). Despite the stream of empirical research on reciprocity in social psychology, little has been written about the subject in marketing. Recently, the awareness of reciprocity has been highlighted in the business-to-business context (Pervan, Bove, and Johnson 2004), and initial attempts have been made to examine reciprocity in consumer behavior (Kolyesnikova, Dodd, and Laverie 2007). Most studies on the reciprocity norm were conducted in settings with face-to-face interactions with a favor-giver and using tangible stimuli (e.g. gift, loan, payment incentives, refreshments, etc). Even when the interaction between a favor-giver and a favorreceiver did not occur in person, but via mail correspondence (Berry and Kanouse 1987; Groves, Cialdini, and Couper 1992), tangible incentives were still used to create a perceived need to reciprocate (e.g. check or cash in return for survey participation). Similarly, the few marketing studies on reciprocity relied upon practices that are based upon face-to-face interactions between customers and service providers (Kolyesnikova et al. 2007). No previous study has investigated the possibility that feelings of gratitude and obligation can be evoked in a manner other than face-to-face interaction and/or tangibility of an incentive. Moreover, it has not been investigated whether these feelings can be stimulated through textual message content (e.g. direct marketing communications). We suggest that direct marketing messages can stimulate gratitude and obligation affective responses to the extent that these feelings may influence consumer attitudes and behaviors. Given previous research suggesting that affective responses to advertising are strong predictors of consumer purchasing intentions (see Morris et al. 2002), the investigation of whether direct marketing messages can elicit feelings of gratitude and obligation is opportune. We therefore explore whether gratitude-inducing and obligation-inducing direct mail messages will be strong enough to influence consumer purchasing behavior and attitudes. Because gratitude and obligation were found to function independently for reciprocal behavior (Kolyesnikova et al. 2007; Watkins et al. 2006), we suggest that gratitude and obligation messages will evoke different consumer responses. Specifically, we hypothesize that because gratitude is usually considered a positive desirable state associated with contentment, pride and happiness (McCullough et al. 2001), gratitude messages will have a stronger effect on consumer purchase intent than will obligation messages.

340 H1:

N. Kolyesnikova et al. Gratitude-inducing messages will have a stronger effect on consumer purchase intent than will obligation-inducing messages.

Previous research also showed that affective responses toward marketing messages can predict brand attitude (Morris et al. 2002). Many researchers agree that attitudes to brands are not formed necessarily solely on the basis of product attributes. They also can be based on emotions. Thus, affective responses to a marketing message can be a significant factor in the formation of perception of a company (MacKenzie and Lutz 1989; Vakratsas and Ambler 1999). The current study investigates whether gratitude- and obligation-inducing direct mail messages are strong enough to influence consumer perceptions of the message-sender. We examine not only whether gratitude and obligation messages can trigger the desired response (i.e. purchase) but also whether that response carries over to the perception of the company. Previous research found that gratitude manipulation created an increase in positive affect (Emmons and McCullough 2003). By contrast, strong negative feelings toward an ad elicited negative feelings toward the brand (Brown, Homer, and Inman 1998). We hypothesize that gratitude messages will elicit more favorable perceptions of the company than will obligation messages. H2:

Gratitude-inducing messages will be associated with more positive, favorable overall perception of the company, while obligation-inducing messages will be associated with more negative, unfavorable perception of the company.

Method Choosing a product category capable of evoking affective response was important for this study because gratitude and obligation are considered emotional states (Emmons and Crumpler 2000; Goei and Boster 2005). Wine was chosen as the product category of interest because its consumers are frequently described as having emotional attachment and relatively high product involvement (Dodd 1997; Kolyesnikova et al. 2007; Lockshin, Spawton, and Macintosh 1997), which is critical to marketing messages that require central processing to achieve persuasion (MacKenzie and Lutz 1989). To construct a realistic communication scenario utilizing wine as the product, we considered how wineries typically communicate with their customers. Reid et al. (2001) reported that mass-oriented media are the focus for larger wineries while smaller wineries rely upon more targeted approaches. Among the wine brands with larger marketing budgets, less than 60% used direct mail to their consumers; however, approximately 85% of wine brands with smaller budgets utilized direct mail strategies. Because our study was conducted in an emerging wine region where roughly 80% of the wineries are considered small (Texas Wine Marketing Research Institute 2008) we used a scenario of a small winery sending direct mail (follow-up postcards) to tasting room visitors. Sample A total of 124 subjects were recruited to participate in the study. Because the experiment was conducted in two sessions, two groups of participants were recruited. The first group consisted of 64 former and current students of a wine appreciation class offered through a continuing education program for adults at a large public university in the southwestern United States. This group of respondents was considered appropriate for the sample because it included adults with at least some interest in the product category.


341

The expectation was that the socio-demographic background of this group would be similar to the profile of a wine consumer in general – middle-aged, educated, and with higher incomes (Chaney 2001; Dodd and Bigotte 1997). Because recent research has noted a worldwide trend where younger consumers have started to develop an interest in wine (Fountain and Charters 2004; Howard and Stonier 2002), we also included younger consumers to increase the representativeness of the sample. Sixty undergraduate students (seniors) enrolled in a hospitality management class at the same southwestern US university were recruited for participation. The use of these two groups of subjects helped to ensure the respondents’ interest and involvement in the topic (Helgeson, Voss, and Terpening 2002; van Kenhove, Wijnen, and de Wulf 2002) and reduced the chance of uninformed responses (Graeff 2007). Out of the 124 questionnaires collected four were discarded due to incomplete participation, yielding for analysis a final sample size of 120 participants. Research design The experiment employed a within-subjects repeated measures design. This method was chosen because it generally increases statistical power and reduces error variance associated with individual differences (Reeves and Geiger 1994). Stimulus materials The selection of stimulus materials for the experiment was guided by a copy pretest. For the pretest, 62 people whose demographic characteristics were similar to the study participants were recruited. Forty participants were undergraduate students majoring in retailing, and 22 participants were older (ranging between 25 and 60 years old). None of the individuals from the copy pretest were included in the formal experiment. Participants in the copy pretest viewed a total of 18 messages. Only the textual portion of the messages was presented in the pretest. The subjects were instructed to rate each message in terms of feelings of either gratitude or obligation. Because the terms gratitude and obligation may be ambiguous, an explanation of the feelings was included – feel thankful (appreciated) and feel obliged (like I owe), respectively. Based on the means and standard deviation scores, two messages were determined to be most evocative of gratitude and two most evocative of obligation.1 To conceal the purpose of the experiment, two control postcards were employed. These cards did not load strongly on either of the two factors in the pretest and were further edited to express neither gratitude nor obligation. The stimulus materials were graphically designed to appear as typical direct mail postcards. A total of six postcards were used in the formal study (i.e. two gratitudeinducing, two obligation-inducing, and two neutral controls). All six postcards looked identical, varying only in the text message. For precision control, all other factors, such as card size, design, graphics, font style, font size, were held constant across all cards. Procedures The experiment was administered in a large conference room of a business building. As the participants arrived at the facility, they were given random seating assignments in order to minimize the risk of interaction between participants who may have been acquainted. The study was introduced to participants as a market pretest of a communication campaign. The instructions to the participants were as follows:

342


Please imagine that you have visited a winery in the past. Today in your mailbox, you received a postcard from that winery. Your task is to examine the message on the postcard and answer a series of questions after each card. As you look at the postcard, please remember, we are interested in your evaluation of the message in the postcard, not the wine shown or the graphic design of the postcard.

For the presentation purposes, the postcards and immediate measures were placed in binders, coded for random rotation. The participants were instructed to work at their own pace but were asked to evaluate one card at a time in the order that they were placed in the binder. The binder ended with a one-page questionnaire collecting demographic data. Each binder contained six postcards, and all the participants were exposed to all six cards, but the order of the postcards in the binders was systematically varied using the Latin square method for a total of 16 different rotations. This enabled the estimation of the main effect of each message without possible confounding interactions due to the sequence of exposure. This procedure also efficiently reduces the number of orders and has been used effectively in media related studies (Reeves and Geiger 1994). To minimize the possibility of fatigue and because the company was held constant across all repetitions, attitude toward the company was measured only once after exposure to the first systematically varied postcard in each respondent’s folder. The participants did not see the neutral cards as the first cards in their binders. Comparison of only two types (gratitude and obligation) was conducted for the attitude toward the company hypothesis. Measures The study included two main dependent measures: purchase intent (PI) and attitude toward the company (Ac). After viewing each postcard, the participants responded to the key measures of interest. In addition, previous consumer attitude studies (Bruner, Hensel, and James 2005) guided the design of a new affective response attitude toward the ad scale. The new instrument measured the degree of gratitude and obligation feelings experienced by participants when viewing a direct mail message. Purchase intent Purchase intent (PI) was measured with three items: (a) likelihood of purchasing, with ratings on an 11-point scale response from 0 (very unlikely) to 10 (very likely); (b) the number of bottles of wine the respondent would purchase; and (c) the dollar amount the respondent would spend on the purchase. Because the three items measured purchase intent from different perspectives, the ratings on the items could not be grouped together and averaged. Data analysis then included the investigation of each PI measure separately. Attitude toward the company The attitude toward the company scale (Ac) was adapted from the attitude toward the advertiser scale used in several studies (Lohse and Rosen 2001; MacKenzie and Lutz 1989; Muehling 1987; Simpson, Horton, and Brown 1996; Speed and Thompson 2000). The scale stems used by the previous research varied between advertiser, sponsor, and the company (Bruner et al. 2005). To adjust to the circumstances of our study, we used the word winery for the stem. Specifically, four 11-point scale items measured participant perceptions of the winery as not reputable/reputable, not likeable/likeable, unfavorable/favorable, and positive/negative (reverse-coded). Two additional 11-point scale items were added to determine whether the overall attitude toward the company carries over


343

to consumer future behavioral intentions: (a) How likely are you to visit this winery in the future? (very unlikely/likely); and (b) How likely are you to recommend this winery to other people? (very unlikely/likely). Attitude toward the direct mail message (gratitude and obligation) To determine whether the respondents made the cognitive distinction between the different types of the manipulated messages, we embedded attitude toward the message (Ams) measurement. The instruction statement This postcard made me feel was followed by 16 items on an 11-point Likert-type scale, anchored between 0 (not at all) and 10 (completely). The structure of the measurement was adapted from the affective response to ad scale originally developed as the Reaction Profile for TV Commercials by Wells, Leavitt, and McConville (1971). Because our primary interests were feelings of gratitude and obligation, we developed a new scale to measure the degree of these feelings experienced by consumers when viewing a direct mail message. Preliminary analyses Data reduction The multi-item scales were initially subjected to factor analyses. To confirm measurement of attitude toward the company, four items were subjected to principal component analysis. One factor with an eigenvalue greater than one emerged accounting for 79.40% of the total variance. The solution could not be rotated due to the extraction of only one factor. The factor loadings (the lowest was .72) and a high level of inter-item consistency (a ¼ .90) warranted the construction of a composite measure, labeled attitude toward the company (Ac), by averaging the rating scores across the four items. To determine the underlying structures of the newly developed Ams scale, the 16 items in the scale were subjected to a principal component analysis with varimax rotation. An examination of factor loadings, eigenvalues greater than one, and the scree plot suggested two factors, labeled gratitude and obligation. The two factors accounted for 80.25% of the total variance (42.07% and 38.18%, respectively). Of the 16 items entered in the analysis, no single item loaded on both factors. The lowest loading was .81. The gratitude factor showed high loadings on eight items which measured positive affective responses to messages in the postcards, namely grateful, recognized, thankful, valued, acknowledged, positive, pleased, and honored. The obligation factor was composed of the remaining eight items which measured negative affective responses to messages in the postcards, namely uncomfortable, obligated, indebted, reluctant, guilty, pressured, irritated, and owing. Correlation analysis revealed that the items on each factor showed a high degree of internal consistency indicating that the eight items on each factor were measuring a single, similar construct (gratitude a ¼ .95; obligation a ¼ .93). Scores for the items that loaded highly on a given factor were averaged. Manipulation check After testing the semantic differentiation of the messages on the basis of factor loadings, manipulation checks were performed for all six postcards to ensure that participants did indeed perceive the intended appeal of the three different conditions: gratitude; obligation; and neutral.

344


The results indicated that gratitude-inducing messages elicited more gratitude feelings; that is, were perceived significantly higher on the gratitude items on the Ams scale (M ¼ 7.39, SD ¼ 1.44) than the obligation-inducing messages (M ¼ 2.63, SD ¼ 1.92) and the control messages (M ¼ 3.16, SD ¼ 2.12), F(2, 238) ¼ 313.03, p , .001. Likewise, the obligation-inducing messages were perceived significantly higher on the obligation items on the Ams scale (M ¼ 5.46, SD ¼ 2.04) than the gratitude-inducing messages (M ¼ 1.29, SD ¼ 1.32) and the control messages (M ¼ 1.31, SD ¼ 1.39), F(2, 238) ¼ 326.89, p , .001. Results Description of the sample Of the 120 participants, 50.8% were female and 49.2% were male. The average age of the participants in the first group was 45 years old; more than 40% of the participants reported their annual household income to exceed $80,000; and only one-fourth of the group did not hold a college degree. Participants in the second group were between ages 19 and 27; with the mean being 22 years old; more than one-third (34.6%) of the participants in this group reported their annual income to be lower than $20,000; and 91.2% in this group had a level of college education, but no degree. Both the older and the younger groups were homogenous in terms of inter-group education and income; to avoid using attendant measures for analyses, education and income were not included in hypotheses testing. Purchase intent hypothesis Hypothesis 1 predicted that gratitude messages would have a stronger influence on customer purchase intent than would obligation messages. Hypothesis 1 was supported. The results were similar across the three measurements of purchase intent. Specifically, ANOVA revealed a significant main effect for the message type on the self-reported scores of likelihood of purchasing, F(2, 232) ¼ 243.78, p , .001; on the number of wine bottles likely to be purchased, F(2, 232) ¼ 101.41, p , .001; and on the dollar amount likely to be spent at the winery, F(2, 232) ¼ 88.84, p , .001. Post-hoc tests were conducted to follow up the significant main effects. The mean scores of gratitude, obligation, and control messages were significantly different across all three measurement of purchase intent with gratitude consistently yielding more positive scores, followed by the control and then obligation. Figure 1 shows the mean responses to purchase intent measurements. To analyze the effect of age on purchase intent, the sample was divided into groups using a median split. The comparison was made between respondents aged 26 and younger and respondents aged 27 and older. The main effect of age was significant for only one measure of purchase intent – self-reported scores of purchase likelihood, F(1, 116) ¼ 15.58, p , .001. Older consumers reported that they were less likely to purchase wine (M ¼ 4.14, SD ¼ 1.10) than younger consumers (M ¼ 5.06, SD ¼ 1.58). However, for the other two measurements of purchase intent, the main effect of age was not significant. For the number of bottles likely to be purchased, F(1, 116) ¼ 0.41, p ¼ .53; for the dollar amount to be spent, F(1, 116) ¼ 2.69, p , .10. Outside of the main effect, the analysis revealed a significant interaction of age and message type. Older consumers were more responsive to the gratitude-inducing messages in their likelihood of purchasing (M ¼ 7.50, SD ¼ 1.60) than were younger consumers (M ¼ 6.98, SD ¼ 1.78), F(2, 232) ¼ 14.53, p , .001. In response to the gratitude messages, older consumers also were willing to buy more wine bottles (M ¼ 3.28, SD ¼ 2.29) than


345

8

Puchase likelihood

7

7.24a

6 5 4.18c

4 3 2.37b

2 1 0 Gratitude

Obligation

Control

3.5

Bottles ordered

3

2.87a

2.5 2 1.5 1.20c

1 0.63b

0.5 0 Gratitude

Obligation

Control

$60

Dollars spent

$50

$47.72a

$40 $30 $20

$17.90c

$10

$8.97b

$0 Gratitude

Obligation

Control

Figure 1. Mean responses to purchase intent measurements. Note: Means with different superscripts differ significantly at p , .05 by Tukey post-hoc tests.

younger consumers (M ¼ 2.45, SD ¼ 1.72), F(2, 232) ¼ 11.47, p , .001; as well as spend more money (M ¼ 50.92, SD ¼ 38.54) than the younger group (M ¼ 44.53, SD ¼ 43.03), F(2, 232) ¼ 6.02, p , .001. Obligation messages, on the contrary, seemed to be more motivating for younger consumers in terms of purchase likelihood (M ¼ 3.18, SD ¼ 2.17) than for older consumers (M ¼ 1.56, SD ¼ 1.36). After viewing the obligation messages,

346


younger consumers were willing to buy more wine bottles and spend significantly more money (M ¼ .94, SD ¼ .75; M ¼ 14.15, SD ¼ 13.02, respectively) than older consumers (M ¼ .33, SD ¼ .40; M ¼ 3.79, SD ¼ 15.30, respectively). The effect of gender on purchase intent was also tested, but no significance was found, for the likelihood of purchase F(1,16) ¼ 2.14, p ¼ .15; for the number of bottles F(1,116) ¼ .02, p ¼ .89; and for the total dollars F(1, 16) ¼ .07, p ¼ .79. The interaction effects were not significant either. With regards to purchase intents, males and females felt equally in response to different message types – for purchase likelihood, F(2, 232) ¼ 2.50, p ¼ .84; for the number of bottles, F(2, 232) ¼ .33, p ¼ .72; and for the total dollars, F(2, 232) ¼ . 13, p ¼ .88. Attitude toward the company hypothesis Hypothesis 2 predicted that gratitude-inducing messages would affect consumers’ perceptions of the company as more favorable than would obligation-inducing messages. Hypothesis 2 was also supported. Because the participants did not see the control cards as the first cards in their binders, comparison of only gratitude and obligation messages was conducted to test hypothesis 2. Data analysis revealed that participants had more favorable perceptions of the winery after viewing the gratitude messages (M ¼ 6.55, SD ¼ 2.11) than the obligation messages (M ¼ 3.18, SD ¼ 1.95), F(1, 112) ¼ 115.82, p , .001. Similar results were obtained when future behavioral intents were tested. Specifically, participants were more willing to visit the winery in the future after the exposure to the gratitude cards (M ¼ 5.98, SD ¼ 2.40) than after the obligation cards (M ¼ 2.59, SD ¼ 2.14), F(1, 112) ¼ 74.62, p , .001. Likewise, after seeing the gratitude cards, participants were more likely to recommend the winery to other people (M ¼ 6.03, SD ¼ 2.35) than after the obligation cards (M ¼ 2.70, SD ¼ 2.26), F(1, 112) ¼ 71.76, p , .001. The data were further analyzed to examine the effects of demographic characteristics. Although the main effect of age was not significant for the attitude toward the company, F(1, 112) ¼ 2.59, p ¼ .11, and for the future intent to visit and recommend the winery, F(1, 112) ¼ .05, p ¼ .83; F(1, 112) ¼ .13, p ¼ .72, respectively, a significant interaction effect of age and message type was found. The interaction was similar to that in hypothesis 1 – that is, the effect of the gratitude messages was stronger on older consumers. In contrast, the effect of obligation was stronger on younger consumers. When exposed to the gratitude messages, older consumers had more favorable attitudes toward the company (M ¼ 7.48, SD ¼ 1.85) than the younger group (M ¼ 5.59, SD ¼ 1.95), F(1, 112) ¼ 54.87, p , .001. In response to the gratitude messages, older consumers also were more willing to visit the winery in the future (M ¼ 6.97, SD ¼ 1.96) than were younger consumers (M ¼ 4.97, SD ¼ 2.43), F(1, 112) ¼ 22.46, p , .001; as well as more likely to recommend the winery to others (M ¼ 7.10, SD ¼ 1 90) than younger consumers (M ¼ 4.93, SD ¼ 2.28), F(1, 112) ¼ 23.15, p , .001. On the contrary, younger consumers had more favorable perceptions of the company in response to the obligation messages (M ¼ 4.57, SD ¼ 1.49) than the older group (M ¼ 1.74, SD ¼ 1.18). After viewing the obligation messages, younger consumers were more likely to visit the winery in the future (M ¼ 3.48, SD ¼ 1.67) than were the older group (M ¼ 1.67, SD ¼ 1.67); as well as to recommend the winery to other people (M ¼ 3.58, SD ¼ 1.18) than older consumers (M ¼ 1.80, SD ¼ 1.67). The effect of gender was also tested, but similarly to the results for hypothesis 1, no significant main or interaction effects were found. Males and females did not differ


347

significantly in their attitudes toward the company, F(1, 112) ¼ .29, p ¼ .59; in their intentions to visit the winery in the future, F(1, 112) ¼ .06, p ¼ .81; and in recommending the company to others F(1, 112) ¼ .05, p ¼ .83. Males and females did not react differently to different message type either – for attitudes toward the winery, F(1, 112) ¼ .30, p ¼ .58; for future visits, F(1, 112) ¼ .06, p ¼ .80; and for recommendations of the company, F(1, 112) ¼ .50, p ¼ .48. Discussion The norm of reciprocity historically has been applied as a one-dimensional expression of the extent to which an individual feels the need to repay a perceived benefit received from another person. More recently, social science researchers have shown reciprocal behavior to be multidimensional; specifically, reciprocal behavior was found not to have a single, undiluted source. An act of reciprocity may originate in feelings of obligation, but it also may stem from feelings of gratitude. This distinction is important for two reasons. First, the source of the need to reciprocate carries a strong influence upon the nature and quality of the reciprocal response, such as purchasing. Second, when reciprocal behavior is the goal, the medium for stimulating such behavior becomes critical to triggering the desired response. The application of these two principles is readily apparent in human endeavors on a macro level, such as governmental politics and international diplomacy. Applications on a micro level – such as the way children interact with each other on school playgrounds, the manner in which employees engage with coworkers at the office, or the approaches individual consumers bring to their marketplace behavior – may be more subtle but are no less meaningful. The recognition and exploitation of affective response in face-to-face communication practices is hardly innovative. However, marketing of products and services to modern consumers has been extended far beyond what can be accomplished in personal encounters between sales staff and customers. Messages are delivered to consumers through a variety of media and with increasing levels of technological sophistication. Even so, there remains a class of products which can be effectively promoted through targeted direct messages. Considering that most wineries in the USA are relatively small businesses that rely on tasting room sales and experiences to build customer relationships (Barclay 2006), the ability to maximize the effectiveness of any marketing opportunity is critical. Service in the tasting room is one opportunity; strategically crafted direct mail communication that follows the tasting room experience is another. Traditionally, direct mail messages are decision-focused (immediate purchasing behavior) rather than attitude-focused (knowledge, preferences, intent) (Smith and Berger 1998). Direct follow-up postcards sent from a business to its customers would fall somewhere between these two message types. They are personal messages which should nurture the relationship between the company and the customer as well as encourage a timely purchasing intention. The follow-up postcard is a unique opportunity for the company to evoke pleasant memories of the consumer experience and to build on the relationship with a customer. Thus, the content of that targeted message should be carefully considered. The purpose of the current experiment was not merely to determine whether a direct marketing message (as opposed to a face-to-face interaction) could evoke feelings of gratitude and obligation. Rather, the purpose was to discover if these two dimensions of reciprocal feeling could be stimulated by a textual message to the extent that respondents

348


differentiated between them and responded differently to them (i.e. whether significant differences in attitudes and purchasing behavior/intent could be detected and measured). It is important for enterprises such as small wineries to build brand recognition alongside a brand relationship. Each opportunity to strengthen that relationship must be strategically exploited. If a small winery is going to the expense of creating, producing, and distributing a direct mail message to consumers who have indicated they are open to an ongoing relationship, then that message must be carefully crafted to evoke the response that will sustain and expand that relationship. The statistical significance of the manipulation check showed that consumers are able to discriminate between messages designed to evoke either gratitude or obligation. This is an important finding because it confirms the dimensionality of the gratitude and obligation constructs. The support for hypothesis 1 suggested that participants were more willing to make a purchase, buy more bottles of wine, and spend more money in response to gratitude messages. Obligation-inducing messages were the least motivating for purchasing. Furthermore, the results for hypothesis 2 suggested that the textual content of direct mail messages can influence consumers’ attitudes toward the company. Messages designed to evoke feelings of gratitude were more likely to lead to the formation of a favorable perception of the company, which, in turn, could lead to future business and positive word-of-mouth recommendations. In response to obligation messages, however, consumers seemed to form more negative, unfavorable perceptions of the company, and were less willing to visit the winery in the future or recommend the products to others. The results reported here are consistent with the findings of other scholars who found that feelings of gratitude were more likely to stimulate reciprocal behavior in nonconsumer-oriented settings (Goei and Boster 2005; Watkins et al. 2006). Our findings provide strong empirical support for the use of gratitude-inducing messages targeted at consumer behavior. Obligation-inducing messages are significantly less likely to exert a positive influence upon either the consumer’s purchase intent or the consumer’s attitude toward the company. Cotte, Coulter, and Moore (2005) found that consumers who evaluated an advertising message designed to create feelings of guilt had their responses mitigated by the extent to which they perceived the advertiser’s intention to manipulate their emotions. The more manipulation perceived, the less positive the response to the ad. Smith and Berger (1998, 38) found that messages ‘that encourage people to maintain or enhance [a] positive feeling state are likely to generate more attention and interest, leading to more favorable brand attitudes and intent to purchase’. Both assertions are consistent with the response in this study. Messages designed to make consumers feel obligated to respond are not as effective as messages designed to make consumers feel grateful. Consumers who feel obligated are not only less likely to make a purchase than grateful consumers, but they also will spend less on the purchases they do make and will have a less positive perception of the company. The use of control messages in this study revealed another important insight into the value of targeted messages. Conventional marketing wisdom posits that a negative message is more effective than a neutral message; the theory being that to leave consumers with a negative impression is better than leaving no impression at all (see Brown et al. 1998). Contrary to this position, our findings suggest that neutrally worded messages in fact have the capacity to generate higher levels of purchase intent and more positive attitudes toward the company than messages that pressure the consumer.


349

The results of this study also suggest that consumer attitudes and behavioral intentions are influenced by gratitude- and obligation-inducing messages differently according to demographic characteristics. Previous research on gratitude and obligation virtually ignored demographic differences among participants. This deficiency can possibly be attributed to the fact that much research on the topic has employed undergraduate students as participants. Student samples typically do not have demographic heterogeneity, and perhaps for this reason have not been suitable for a more thorough investigation of how demographic profiles influence results. Our sample, however, varied in age, gender, education, and income. As shown by the reported interaction effects, the effect of gratitude messages was stronger on older consumers, while younger consumers were more responsive to obligation messages. We interpreted this finding as reflective of two separate possibilities. First, older consumers attend more to marketing messages that are emotionally meaningful and resonate with their own emotional goals (Fung and Carstensen 2003). The positive feelings associated with gratitude messages could be described as emotionally meaningful. Most individuals do not consciously form negative emotional goals and thus may be more likely to reject messages that create negative feelings. Secondly, although product knowledge and purchasing savvy were not explicitly tested in this study, the implied relative lack of product experience of younger consumers may have played a role in their susceptibility to the social reciprocity requirements implied by an obligation message. Implications Despite – or perhaps because of – the unique knowledge and involvement they bring to the product, the consumers in this study appear to incorporate affective responses in their attitudes toward direct mail messages, the company, and purchasing intentions. It seems reasonable then for marketing decision-makers to deliberate on the influence of gratitude and obligation on visitors’ purchasing decisions not only at the cash register but also post purchase. Because feelings of gratitude appear to exert a significant positive influence on buying intentions and attitudes toward the company, the incorporation of gratitudeinducing messages into direct mail interactions with customers deserves serious consideration. Moreover, any use of obligation-inducing messages should be carefully scrutinized. Although creating a sense of obligation may have a short-term impact on sales, obligation may also create customer resistance to a long-term relationship with the company. Where the message deviates from gratitude or other positive affective response, it would be better for messages to err on the side of neutrality than to cross over to obligation. Limitations and recommendations By directing respondents to a focused evaluation of specially created messages, a sensitivity artifact is introduced (Brown et al. 1998). Such an artifact may heighten positive evaluations of positive messages and negative evaluations of negative messages in unexpected ways. We attempted to control for this in several ways such as using a study scenario that obscured the primary purpose of the evaluation request; using more than one message for both treatments; incorporating two affect-neutral messages; exposing all respondents to each message; and rotating the sequence of exposure. The use of a repeated-measures design may also run the risk of carry-over effects. However, the study

350


was carefully crafted in an effort to avoid fatigue and testing effects while also providing adequate statistical power and use of human subjects. In addition, longer-held attitudes undoubtedly are more stable and less prone to simple manipulation. But consumers often encounter situations where little pre-formed attitude exists toward a brand/product. In the instance of a person who may have no brand loyalties or few/no favorites in the category, attitudes although loosely held may still drive consumer behavior, and if marketing techniques can influence these attitudes in these instances, then purchase decisions may be affected. The design employed in the study allowed for the effect of no pre-existing attitude to be spread equally across the manipulations. Future study should investigate exactly what it is that is underlying why gratitude leads to purchase intentions. It is possible that liking/non-liking may result from gratitude/obligation, which in turn leads to purchase intent. But the findings of the current study provide foundation for subsequent investigation. Two further limitations included using a single product category (wine) and a single message medium (direct mail postcards). The choice of wine offered plausibility because of its unique emotional characteristics. The use of direct mail communication between a winery and its visitors optimized the ecological validity of the study. In tandem with the limitations, the recommendations for future study of gratitude and obligation as affective responses to marketing messages include replicating the study with different product categories and a wider band of marketing media. Furthermore, given that wine involvement can make a significant difference in terms of wine consumer behavior (Lockshin et al. 1997), replication with other product categories exemplified by high product involvement is recommended.

Conclusions This research opens new ways of utilizing emotional content in direct mail messages. Using the context of winery visits and wine consumers, this study considered the capacity of gratitude and obligation to act as separate affective influences upon consumer attitudes toward the company and purchase intent. Unlike other affective responses that have been studied by social science researchers, gratitude and obligation frequently have been overlooked or treated as the same construct. This study offers strong evidence that gratitude and obligation operate differently in their impact on consumer attitudes and behaviors. Gratitude exerts a more positive influence than does obligation. Communicating through gratitude, and even neutral, messages was found to be more persuasive than those messages conveying obligation. For the academy, this research signals the need to explore in greater depth the persuasive capacity of different communication media. For the business practitioner, the results of this study serve as a road map on producing more effective direct mail messages for customers.

Note 1. Example of gratitude-inducing message: ‘We appreciate a good neighbor, a good friend, and, of course, a good wine. At our winery, we truly appreciate good customers like you. Come by anytime. For you, our door is always open.’ Example of obligation-inducing message: ‘We count on you to recognize our hard work. We press the best grapes to make our wines. Now we press ourselves to get them ready. The time is ripe. Where are you?’


351

Notes on contributors Natalia Kolyesnikova is an Assistant Professor at Texas Tech University where she has a joint appointment between the Department of Nutrition, Hospitality, and Retailing and the Texas Wine Marketing Research Institute. Dr Kolyesnikova’s research interests are mostly in the realm of wine consumer behavior, wine tourism, and brand development. Sara L. Sullivan Dodd holds a joint appointment as Assistant Professor in the College of Human Sciences at Texas Tech University and Extension Specialist with Texas AgriLife Extension in the Texas A&M University System. Her research interests are multidisciplinary and include consumer behavior, behavioral economics, leadership theory, and development. Coy Callison (PhD, University of Alabama) is an Associate Professor and Associate Dean for Graduate Studies in the College of Mass Communications at Texas Tech University. Dr Callison’s research interests include source and message factors influencing media persuasiveness.

References Barclay, V. 2006. Tasting room survey report. Wine Business Monthly May: 42– 8. Batra, R., and M.B. Holbrook. 1990. Developing a typology of affective responses to advertising. Psychology and Marketing 7, no. 1: 11– 26. Batra, R., and M.L. Ray. 1986. Affective responses mediating acceptance of advertising. Journal of Consumer Research 13, no. 2: 234– 49. Berry, S.H., and D.E. Kanouse. 1987. Physician response to a mailed survey: An experiment in timing of payment. Public Opinion Quarterly 51, no. 1: 102– 14. Bhat, S., T.W. Leigh, and D.L. Wardlow. 1998. The effect of consumer prejudices on ad processing: Heterosexual consumers’ responses to homosexual imagery in ads. Journal of Advertising 27, no. 4: 9 – 28. Brown, S.P., P.M. Homer, and J.J. Inman. 1998. A meta-analysis of relationships between ad-evoked feelings and advertising responses. Journal of Marketing Research 35, no. 1: 114– 26. Bruner, G.C., P.J. Hensel, and K.E. James. 2005. Marketing scales handbook: A compilation of multi-item for consumer behavior and advertising. Vol. IV Chicago, IL: Thomson. Burke, M.C., and J.A. Edell. 1989. The impact of feelings on ad-based affect and cognition. Journal of Marketing Research 26, no. 1: 69 – 83. Chaney, I.M. 2001. Opinion leaders as a segment for marketing communications. Marketing Intelligence and Planning 19, no. 5: 302– 8. Cotte, J., R.A. Coulter, and M. Moore. 2005. Enhancing or disrupting guilt: The role of ad credibility and perceived manipulative intent. Journal of Business Research 58, no. 3: 361– 8. Dodd, T.H. 1997. Techniques to increase impulse wine purchases in a restaurant setting. Journal of Restaurant and Foodservice Marketing 2, no. 1: 63 – 73. Dodd, T.H., and V. Bigotte. 1997. Perceptual differences among visitor groups to wineries. Journal of Travel Research 35, no. 3: 46 – 51. Edell, J.A., and M.C. Burke. 1987. The power of feelings in understanding advertising effects. Journal of Consumer Research 14, no. 3: 421– 33. Emmons, R.A., and C.A. Crumpler. 2000. Gratitude as a human strength: Appraising the evidence. Journal of Social and Clinical Psychology 19, no. 1: 56 – 69. Emmons, R.A., and M.E. McCullough. 2003. Counting blessings versus burdens: An experimental investigation of gratitude and subjective well-being in daily life. Journal of Personality and Social Psychology 84, no. 2: 377– 89. Fountain, J., and S. Charters. 2004. Younger wine tourists: A study of generational differences in the cellar door experience., Paper presented at the International Wine and Tourism Conference, Margaret River, Australia, May 2 – 5, 2004. Fung, H.H., and L.L. Carstensen. 2003. Sending memorable messages to the old: Age differences in preferences and memory for advertisements. Journal of Personal and Social Psychology 85, no. 1: 163– 78. Goei, R., and F.J. Boster. 2005. The roles of obligation and gratitude in explaining the effect of favors on compliance. Communication Monographs 72, no. 3: 284– 300. Goei, R., L.L. Massi Lindsey, F.J. Boster, P.D. Skalski, and J.M. Bowman. 2003. The mediating roles of liking and obligation on the relationship between favor and compliance. Communication Research 30, no. 2: 178– 97.

352


Gouldner, A.W. 1960. The norm of reciprocity: A preliminary statement. American Sociological Review 25, no. 2: 161– 78. Graeff, T.R. 2007. Reducing uninformed responses: The effects of product-class familiarity and measuring brand knowledge on surveys. Psychology and Marketing 24, no. 8: 681– 702. Greenberg, M.S., and S.P. Shapiro. 1971. Indebtedness: An adverse aspect of asking for and receiving help. Sociometry 34, no. 2: 290– 301. Greenwald, A.G. 1968. Cognitive learning, cognitive response to persuasion, and attitude change. In Psychological foundations of attitudes, ed. A.G. Greenwald, T.C. Brock, and T.M. Ostrom, 147– 70. New York: Academic Press. Groves, R.M., R.B. Cialdini, and M.P. Couper. 1992. Understanding the decision to participate in a survey. Public Opinion Quarterly 56, no. 4: 475– 95. Helgeson, J.G., K.E. Voss, and W.D. Terpening. 2002. Determinants of mail-survey response: Survey design factors and respondent factors. Psychology and Marketing 19, no. 3: 303– 28. Howard, D.J. 1995. Chaining the use of influence strategies for producing compliance behavior. Journal of Social Behavior and Personality 10, no. 1: 169– 85. Howard, R., and J. Stonier. 2002. Marketing wine to generation X. Australian and New Zealand Wine Industry Journal 17, no. 3: 78 – 82. Kolyesnikova, N., T.H. Dodd, and D.A. Laverie. 2007. Gratuity purchasing at wineries: An investigation of the determining factors. International Journal of Wine Business Research 19, no. 4: 239– 56. Lockshin, L., A.L. Spawton, and G. Macintosh. 1997. Using product, brand, and purchasing involvement for retail segmentation. Journal of Retailing and Consumer Services 4, no. 3: 171– 83. Lohse, G.L., and D.L. Rosen. 2001. Signaling quality and credibility in yellow pages advertising: The influence of color and graphics on choice. Journal of Advertising 30, no. 2: 73 –85. MacKenzie, S.B., and R.B. Lutz. 1989. An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. Journal of Marketing 53, no. 2: 48 – 65. McCullough, M., S.D. Kilpatrick, R.A. Emmons, and D.B. Larson. 2001. Is gratitude a moral affect? Psychological Bulletin 127: 249– 66. Morris, J.D., C. Woo, J.A. Geason, and J. Kim. 2002. The power of affect: Predicting intention. Journal of Advertising Research 42, no. 3: 7 – 17. Muehling, D.D. 1987. Comparable advertising: The influence of attitude-toward-the-brand on brand evaluation. Journal of Advertising 16, no. 4: 43 – 9. Pervan, S., L. Bove, and L.W. Johnson. 2004. Conceptualizing the role of reciprocity in relationship marketing: An examination of its supporting virtues. Paper presented at the Australian and New Zealand Marketing Academy Conference, Wellington, New Zealand, November 29 – December 1, 2004. Reeves, B., and S. Geiger. 1994. Designing experiments that assess psychological responses to media messages. In Measuring psychological responses to media messages, ed. A. Lang, 165– 81. Hillsdale, NJ: Erlbaum. Regan, D.T. 1971. Effects of a favor and liking on compliance. Journal of Experimental Social Psychology 7, no. 6: 627– 39. Reid, M., T. Johnson, M. Ratcliffe, K. Skrip, and J. Wilson. 2001. Integrated marketing communications in the Australian and New Zealand wine industry. International Journal of Advertising 20, no. 2: 239– 62. Simpson, P.M., S. Horton, and G. Brown. 1996. Male nudity in advertisements: A modified replication and extension of gender and product effects. Journal of the Academy of Marketing Science 24, no. 3: 257–62. Smith, G.E., and P.D. Berger. 1998. Different message-framing for different direct response marketing goals: Choice versus attitude formation. Journal of Interactive Marketing 12, no. 2: 33 – 48. Speed, R., and P. Thompson. 2000. Determinants of sports sponsorship response. Journal of the Academy of Marketing Science 28, no. 2: 226– 38. Texas Wine Marketing Research Institute. 2008. The Texas wine marketing industry statistics. http://www.depts.ttu.edu/hs/texaswine/industry_facts.php. Thomas, A.R. 2007. The end of mass marketing: Or, why all successful marketing is now direct marketing. Direct Marketing: An International Journal 1, no. 1: 6 – 16.


353

Vakratsas, D., and T. Ambler. 1999. How advertising works: What do we really know? Journal of Marketing 63, no. 1: 26 – 43. Van Kenhove, P., K. Wijnen, and K. de Wulf. 2002. The influence of topic involvement on mail-survey response behavior. Psychology and Marketing 19, no. 3: 293– 301. Watkins, P.C., J. Scheer, M. Ovnicek, and R.L. Kolts. 2006. The debt of gratitude: Dissociating gratitude and obligation. Cognition and Emotion 20, no. 2: 217– 41. Wells, W.D., C. Leavitt, and M. McConville. 1971. A reaction profile for TV commercials. Journal of Advertising Research 11, no. 6: 11 – 17. Wright, P. 1973. The cognitive processes mediating acceptance of advertising. Journal of Marketing Research 10, no. 1: 53 – 62.

Copyright of Journal of Marketing Communications is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.