Measuring Salesperson Orientation of Consumers
Download 147.99 Kb.
|
- Navigate this page:
- Sterling A. Bone
- Measuring Salesperson Orientation of Consumers
- Development of the Salesperson Orientation of Consumers (SOC) Scale
- Scale Purification: Study 1
- Factor Structure: Study 2
- Interpretation of Factors
- Discriminant and Convergent Validity: Study 3
- Generality of the SOC: Study 4
| Measuring Salesperson Orientation of Consumers
Jeffrey S. Larson Assistant Professor of Business Management Marriott School of Management Brigham Young University 681 TNRB Provo, UT 84602 (801) 422-2266 jeff_larson@byu.edu Sterling A. Bone Assistant Professor of Business Management Marriott School of Management Brigham Young University 663 TNRB Provo, UT 84602 (801) 422-3113 sterling_bone@byu.edu September 3, 2008 Please do not cite without permission of the authors. Measuring Salesperson Orientation of Consumers Despite the prevalence of consumer interaction with salespeople, little is known about the enduring attitudes and responses of consumers toward this interaction. Drawing upon a review of the salesperson literature the authors argue that a multi-dimensional scale to measure the Salesperson Orientation of Consumers (i.e., SOC scale) is critical to the understanding of consumer heterogeneity in attitudes and behaviors toward salespeople as marketing agents. In total, five studies were run to develop and validate the four dimensions (Information Seeking, Self Presentation, Avoidance, Convinceability) of the SOC scale, each lending new insights into consumer attitudes and behaviors in sales-aided marketplaces. Across four studies, scale items were purified and the SOC scale’s reliability and construct validity were tested. In study 5, participants responded to a number of vignettes about marketplace behavior relating to salespeople and the results provide support of the predictive validity of the SOC scale. Based on the results of these studies the authors present a discussion of how a knowledge of the four dimensions of the SOC scale will aid marketing agents in ensuring positive consumer-salesperson interactions as well as aid public policy-makers in protecting consumers from the adverse consequences of high-pressure sales situations. In conclusion, a future research agenda is outlined to examine the implications of the SOC scale in other contexts and in relation to other sales and purchase variables. Measuring Salesperson Orientation of Consumers Consider the prevalence of consumer interaction with salespeople. Whether shopping for clothing, automobiles, electronics, insurance, or real estate, consumers frequently interface with salespeople, sometimes by choice and sometimes because the market requires it. According to the U.S. Census Bureau over $1.5 trillion dollars were spent in sales-assisted retail purchases in 2007 (e.g., automobile, furniture, clothing, appliances) (www.census.gov/svsd/www/artstbl.html). Involved in many of these purchases were the 14.3 million individuals (10.9% of the total U.S. Labor force) employed in sales occupations in the U.S. (http://www.bls.gov/oes/current/oes410000.htm#(3)). Not surprisingly, research on customer-salesperson relationships and sales performance has been plentiful (Brown and Peterson 1993; Churchill et al. 1985; Franke and Park 2006; Jaramillo et al. 2007; Wood et al. 2008). This research can be divided into two major categories: 1) research that focuses on the properties of specific customer-salesperson relationships and the effects of those properties (Atuahene-Gima and Li 2002; Crosby, Evans and Cowles 1990; Dixon, Spiro and Jamil 2001; Joseph and Thevaranjan 1998; Ramaswami and Singh 2003; Singh 1998; Speier and Venkatesh 2002; Verbeke and Bagozzi 2000) and 2) research on enduring salesperson characteristics and their consequents (Brown, Cron and Slocum 1998; Kohli, Shervani, and Challagalla 1998; McFarland, Challagalla, and Shervani 2006; Weitz, Sujan and Sujan 1986). More recently, researchers have begun to focus on the role of the consumer in the salesperson-customer interaction. Much of this study has grown from the acknowledgement of the Persuasion Knowledge Model of consumer behavior (Friestad and Wright 2004). For example, Campbell and Kirmani (2000) find that both cognitive capacity and the accessibility of salesperson ulterior motives underlie consumer usage of persuasion knowledge in the customer-salesperson interaction. DeCarlo (2005) similarly examines consumer suspicion of salesperson ulterior motives and finds the mediating role of persuasion-motive attributions. Kirmani and Campbell (2004) follow up their previous study of consumer usage of persuasion knowledge with a more general study of the various strategies used by consumers to manage salesperson interactions. This recent consumer-focused literature parallels the first category of the salesperson-focused stream of literature—an assessment of situation-specific consumer behavior in the sales process. Thus far researchers have not attempted to measure enduring customer characteristics or examine how such enduring characteristics affect customer-salesperson interactions. For example, Campbell and Kirmani (2000) manipulate cognitive capacity rather than measure it, and Kirmani and Campbell (2004) find that consumer strategy usage is more related to the properties of the customer-salesperson relationship than to any enduring consumer characteristic (though they do find that consumer experience with persuasion affects strategy usage). We develop the Salesperson Orientation of Consumers (SOC) scale to aid researchers and practitioners alike in understanding the many enduring consumer characteristics that affect the consumer-salesperson interaction. We define the Salesperson Orientation of Consumers (SOC) as the enduring disposition of a consumer to engage in particular salesperson-related thoughts and behaviors across a variety of marketplace encounters with salespeople. Previous research on consumer behavior regarding salespeople has focused on how this behavior changes as a result of the properties of a particular interaction. By developing the SOC scale, we hope to identify and measure the salesperson-related behaviors that are determined more by consumer traits than the interaction properties. The reliable measurement of such enduring dispositions will aid managers to design effective salesperson interactions for consumers across the SOC spectrum. A theoretical basis for the hypothesized generality of the SOC scale across various types of salesperson interactions comes from recent thought on evolutionary foundations of marketplace metacognition (Wright 2002; Whiten and Byrne 1991). According to this school of thought, social intelligence developed before nonsocial intelligence in evolutionary history. A key domain of social intelligence is marketplace exchange, which survives most directly in modern society in the consumer-salesperson interaction. Since consumer marketplace metacognition is a basic domain of intelligence, consumer heterogeneity in this domain should lead to consisten differences across several types of salesperson interactions. We show empirical evidence of the generality of the SOC in the scale development section. The remainder of this paper presents the generation and validation of the 18-item Salesperson Orientation of Consumers (SOC) scale. Across five studies, we generate and validate the scale. The resulting scale has four subscales, each lending new insights into consumer attitudes and behaviors in the marketplace. Development of the Salesperson Orientation of Consumers (SOC) Scale We followed widely-accepted procedures for creating and validating scales (Churchill 1979; Hinkin 1998) to develop the Salesperson Orientation of Consumers (SOC) scale. First, an initial set of items were drafted. The original list of potential scale items came from several sources. A literature review on consumer-salesperson interaction turned up a small set of papers employing the Consumer Susceptibility to Salesperson Influence (CSSI) scale (Goff, Bellenger and Stojack 1994; Goff and Walters 1995). Their scale takes several forms across the papers, but most useful for our purposes was the 13-item scale from Goff, Bellenger and Stojack (1994) which measured consumers’ orientation toward automobile salespeople. We adapted their scale items to address general salesperson orientation, rather than car-salesperson-specific orientation. A second source of preliminary scale items came from a two-stage brainstorming process. In the first stage, we hypothesized various factors that consumers could use to describe their feelings and orientations toward salespeople, including factors from the literature and some of our own hypothesis. After brainstorming these factors with ourselves and by informal consultation with various colleagues, we generated several items to measure each of these factors. We then removed redundant measures, which resulted in a pool of 35 items. Scale Purification: Study 1 The 35-item scale was administered to three different samples. In Study 1a, 154 masters of accounting students completed the scales for partial course credit at a private Western university. In Study 1b, 159 students and staff members at a private East Coast university completed the scale along with a battery of other questionnaires in return for ten dollars. In Study 1c, 297 undergraduate students completed the scale for extra credit in a marketing course at the same university as Study 1a. The order of the scale items was randomized between subjects. The data from all three studies were subjected to separate exploratory factor analyses with a varimax rotation. In all three studies, the results suggested a four-factor structure with similar loadings across the 35 items. Individual scale items that either 1) did not load strongly on any of the four dimensions or 2) did not load consistently across all three studies were dropped. This left the 18 items shown in Table 1.
To test the new 18-item scale, 532 undergraduates from the same Western university completed the scale in exchange for extra credit in a marketing course. The data were then submitted to the same factor analysis with varimax rotation as previously performed to check that the factor loadings concurred with the data collected from the 35-item scale. The resulting factor structure aligned precisely with the one predicted. Table 1 displays the scale items along with the factor loadings and factor corrected item-to-total correlations. --------------------------------------------------- Insert Table 1 about here ---------------------------------------------------
Insert Table 2 about here ---------------------------------------------------
Of course, consumers are likely to differ in their beliefs about salesperson motivations. While some may believe that salespeople are experts with information that will help them make better choices, others may distrust salesperson motives. The first factor in the SOC scale captures the consumer tendency to seek and value the information provided by salespeople. On the seven-point scale, the average score on this factor was 3.5, under the scale middle-point of 4, indicating that consumers on average tend to be slightly skeptical of the help they receive from salespeople. This factor of the scale has relation to the Consumer Susceptibility to Interpersonal Influence (CSII) scale. The CSII scale measures the extent to which consumers actively seek to enhance their image in the eyes of others through the acquisition of products and brands. The second dimension of the scale, informational influences, measures the extent to which consumers use others’ behavior or opinions to make product choices. This dimension should be positively related to Information Seeking behavior with salespeople. The difference in these measures comes from the fact that the Information Seeking factor of the SOC measures this behavior related to salespeople, while the CSII scale measures this behavior related to friends and family.
While the self monitoring scale measures the ability of people to adapt their behavior according to social cues, the Self Presentation factor of the SOC scale measures their concern rather than their behavior, thus it is unclear whether one should expect a positive relationship between these two measures. However, we can clearly hypothesize that the normative dimension of the CSII scale should have a positive relation to this SOC subscale. The normative CSII measures the extent to which consumers attempt to boost their image in the eyes of others through their purchasing. The same people who are concerned with impressing their friends through their purchases might also be concerned with impressing salespeople with their purchasing. Avoidance. Previous research has shown that consumer heterogeneity in tolerance for salesperson interaction can lead to a differentiated retail equilibrium (Chu, Gerstner and Hess 1995). The existence of this third SOC subscale provides some empirical evidence in support of their model specification. Consumers vary in the extent to which they seek to avoid salesperson interaction. The average score on the Avoidance subscale was 4.3, slightly above the scale midpoint. The average person finds salesperson interactions unpleasant, but not overwhelmingly so. Convinceability. Despite universal contempt for high pressure selling tactics, salespeople continue to use them. Asch’s conformity experiments and Milgram’s fake-shock experiments illustrate the powerful effect of social pressure (Asch 1952; Milgram 1974). It should come as no surprise then that salespeople can often use social pressure to extract sales from not-entirely-willing consumers. Of course, consumers vary in their susceptibility to such sales tactics. This fourth and final SOC subscale measures the extent to which consumers are influenced by salespeople. Of course, since the scale is self-reported, a high score on this factor would necessitate not only a high level of Convinceability but also a recognition of such tendencies. This dual requirement is reflected in the average subscale score of 2.6, well below the scale midpoint. This aspect of the subscale likely makes it less accurately measured in an absolute sense, but it remains a useful measure, as we demonstrate in the next section. Table 3 presents the inter-item correlations for the four SOC subscales. Except for the Avoidance factor, the factors are positively correlated with one another. Those who value the salespeople’s information also worry about the impression they give to salespeople and tend to be convinced by them. Those who are easily convinced by salespeople also tend to worry about the impression they give to salespeople. Those who are high on the avoidance scale tend not to value the information they receive from salespeople, but still worry about the impression they give to salespeople. There is no significant correlation between Avoidance and Convinceability. This pattern of correlations raises some question as to the dimensionality of the SOC, which we address in study 3. --------------------------------------------------- Insert Table 3 about here ---------------------------------------------------
We measured several other scales along with the SOC scale to test discriminant and convergent validity. Study 3 was administered to 265 undergraduate students who completed the scales for extra credit in a marketing course. In addition to the SOC scale, we administered the Marlowe-Crowne Social Desirability Scale (MCSD) (Crowne and Marlow 1964), the consumer susceptibility to interpersonal influence scale (CSII) (Bearden, Netemeyer, and Teel 1989), and the revised form of the self monitoring scale (Lennox and Wolfe 1984). To test the association between the SOC and these established scales, we ran a regression for each established scale using the four SOC subscales as covariates. We present the Type III Sum of Squares results for each regression in Table 4. --------------------------------------------------- Insert Table 4 about here --------------------------------------------------- Previous research has advocated a test of the confounding of responses by social desirability response bias (Mick 1996). Consistent with prior literature, we use the MCSD to test for response bias. The first three SOC subscales showed no association with the MCSD (all p’s > .3). Curiously, Convinceability shows a significant positive relationship with socially desirable responding (t(260) = 3.59, p = .0004). A first-blush interpretation of this result could be that the fourth SOC subscale is susceptible to a social desirability bias. More thoughtful consideration of this factor rules against this interpretation. It is difficult to imagine how being easily convinced by salespeople could be viewed as socially desirable. A more logical interpretation is that socially desirable responding is what leads many people to be more easily influenced by salespeople. We hypothesized that the CSII scale would relate to the Information Seeking and Self Presentation subscales. This was confirmed by our first regression, which used the overall CSII score as the dependent variable. Both Information Seeking and Self Presentation significantly related to CSII, while Avoidance and Convinceability did not. However, we had more specific hypotheses regarding the CSII subscales. We expected that the informational subscale of the CSII would relate to the Information Seeking subscale, while the normative CSII subscale would relate to Self Presentation. As expected, the normative CSII subscale had the strongest association with Self Presentation (t(260) = 4.69, p < .0001). Information Seeking was also positively associated with normative CSII (t(260) = 3.90, p = .0001), which was not expected but is not surprising given the positive correlation between Information Seeking and Self Presentation. More surprising was the positive association between Convinceability and normative CSII (t(260) = 2.02, p = .0445), which could be explained by one aspect of normative CSII that expresses the tendency to purchase “the brand they expect me to buy”. In high-pressure sales situations, the salespeople make their expectations clear, and those high on Convinceability tend to oblige them. Also as hypothesized, Information Seeking had the strongest association with informational CSII (t(260) = 5.73, p < .0001). In line with previous findings, Self Presentation was also positively related to informational CSII (t(260) = 2.06, p = .0401). Here, the surprising result was a strong positive association between Avoidance and informational CSII (t(260) = 3.58, p = .0004). We attempt no explanation of this surprising result. Finally, we examine the association of self monitoring and the SOC subscales. While the connection between the ability to modify self presentation and Self Presentation was obvious, it was less obvious whether this connection implied that we should expect an association between the two measures. No association was found (t(260) = .06, p > .9). The ability to modify self presentation is not associated with self presentational concern related to salespeople. However, a positive association between self monitoring and Information Seeking was found (t(260) = 4.28, p < .0001) and a negative association between self monitoring and Convinceability was found (t(260) = -3.86, p = .0001). Though we could propose explanations for these findings, they would be post hoc and speculative, so we remain silent on the subject. Generality of the SOC: Study 4 We hypothesize that the SOC generalizes across various types of salesperson interactions. To empirically test this hypothesis, we created three versions of the SOC scale, each specific to a particular domain. The three types of sales interactions we chose were car sales, door-to-door sales, and clothing sales. Each item was rewritten to apply to the particular context being tests. For example, “I value the opinion of salespeople” was changed to “I value the opinion of car salespeople”. Note that a few items required more drastic wording changes, as one does not encounter a door-to-door salesperson upon entering a store. The three domain-specific versions of the SOC scale were administered to 312 participants recruited from the cafeteria of a private western university. At the time of the administration, classes were not being held, thus most of the participants were attending a conference being held on the campus. The participants were 35% male, ranged in age from 14 to 83, and were from 27 states and four countries. The data were subjected to a multiple group analysis to determine whether the factor structure was common among the three versions of the scale. The results confirmed a common factor structure across the three versions. Both the NNFI (.94) fit index and the CFI (.95) again exceeded the recommended .90 benchmark. The RMSEA (.064) and RMR (.097) were also within accepted boundaries, indicating a satisfactory fit of a model constrained to a common factor structure across the groups. Thus the data supports the view that the structure of consumers’ orientations toward salespeople is common across various types of salesperson interactions. Directory: emp emp -> United Nations Children’s Fund (unicef) Phnom Penh, Cambodia Vacancy Nº cp/15/006 Terms of Reference Individual Consultancy: Monitoring emp -> For immediate release july 2, 2002 emp -> The skill of multi-model seasonal forecasts of the wintertime North Atlantic Oscillation emp -> Football genius tim green louisiana young readers’ choice nominee 2010 emp -> Quotation for Buy Back Empties Toners & Cartridges (2007-07-31) emp -> Federal Supervisory Training Fact Sheet Introduction emp -> Blender to jme textures and Transparency emp -> The National Parks – America’s Best Idea emp -> Area 6 Piedmont Workforce Network Employment Resource Guide Download 147.99 Kb. Share with your friends:
|