A limited time offer!

urgent 3h delivery guaranteed

The Effect of Visual and Verbal Information on Attitudes

The Effects of Visual and Verbal Information on Attitudes and Purchase Intentions in Internet Shopping Minjeong Kim, Ph. D. Oregon State University Sharron Lennon, Ph.

We will write a custom essay sample on The Effect of Visual and Verbal Information on Attitudes

or any similar topic only for you

Order Now

D. University of Delaware ABSTRACT The present study investigated how different product presentation formats (visual vs. verbal) influence consumer attitudes toward product and purchase intentions in Internet shopping.

The overall results from two Web experiments simulating Internet apparel shopping showed that both visual and verbal information had significant effects on affective and cognitive attitudes toward apparel products, but only verbal information had a significant effect on purchase intention. Though the superiority of visual information was predicted based on prior literature, the results of the study supported verbal superiority. This finding provides an important implication for Internet retailers who tend to pay more attention to visual product presentation.

Although visual product presentation is also found to be important, detailed product descriptions are critical to positively influence consumer shopping experience in Internet shopping. © 2008 Wiley Periodicals, Inc. The Internet is changing almost every aspect of our daily lives, from how we communicate, learn, and play, to how we shop, buy, and consume products and services (Dertrouzos, 1997). Evolving from a new communication Psychology & Marketing, Vol. 25(2): 146–178 (February 2008) Published online in Wiley InterScience (www. interscience. wiley. com) © 2008 Wiley Periodicals, Inc.

DOI: 10. 1002/mar. 20204 medium into an innovative retailing medium, the Internet is changing the world of retailing (Klein, 1998). As the fastest growing retail channel, the growth of Internet retail sales nearly tripled that of total retail sales in 2004 (U. S. Census Bureau, 2004). Although Internet retail sales remained only 5% of total retail sales in 2005 (DMNews, 2006), its future growth is optimistic. According to Forrester Research (2004), Internet retail sales will reach over $331 billion by 2010, accounting for 13% of total retail sales in 2010.

With the rapid adoption of the Internet and the growing popularity of broadband among the general population, the future of Internet retailing is bright (“Digital Economy,” 2000). Despite the impressive growth rate and optimistic outlook, there is compelling evidence to suggest that many consumers are still reluctant to purchase via the Internet. Many Internet retailers continue to struggle with low conversion from browsers to purchasers and high shopping cart abandonment (Internet Retailer, 2005a).

The proportion of actual purchasers to total browsers has remained low, ranging between 2. 8% and 3. 2% of Web site visitors (Shop. org. & Boston Consulting Group, 2000), compared to nearly 50% of mall visitors who purchase during their visit as reported by Stillerman Jones and Co. (Sansoni, 1999). In addition, shopping cart abandonment during the Internet shopping process, especially just prior to checkout, has been prevalent among would-be Internet customers (Shop. org, 2001). Such phenomena imply that there are some factors that keep Internet shoppers from buying via the Internet.

A primary deterrent of Internet buying is the inability to physically examine items prior to purchase (Internet Retailer, 2005b; Retail Forward, 2001). According to Forrester Research, more than half the consumers who visit an Internet store do not purchase because they cannot physically inspect an item before purchasing (Internet Retailer, 2005b). Consumers need to acquire adequate product information to make a purchase decision, often by physical examination of a product, but Internet shopping does not accommodate physical product evaluations like brick-and-mortar stores do (Nitse et al. 2004). This is more problematic for certain types of products that require sensory evaluation. Holbrook and Moore (1981) suggested that products with aesthetic, sensory, or symbolic benefits (e. g. , apparel) must be experienced for adequate judgments to be made. Likewise, in Internet shopping the problem of lack of product examination is magnified for products like apparel that require sensory inspection to assure adequate fit or color co-ordinate items.

Although apparel is one of the major merchandise categories sold via the Internet (Internet Retailer, 2003), many sales opportunities are lost because of the inability to touch and feel an item prior to a purchase (Beck, 2003; Pastore, 2000). The biggest reason for not purchasing via the Internet was the uncertainty of fit and size. Such problems caused by the lack of adequate product examination further result in high product return rates (e. g. , 30%) and lost customer loyalty. The estimated loss due THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 002/mar 147 to these problems was more than $2 billion dollars in the Internet apparel industry (Beck, 2003). Responding to the insufficiency of customer experience related to product examination, Internet retailers have begun to implement innovative technologies that improve the Internet shopping experience. To simulate the store shopping experience in which physical inspection of an item is possible, several technologies that enhance visual product presentation (i. e. , 3-D images, virtual models, digital images, and zooming technology) have been introduced (Retail Forward, 2001).

For example, Lands’ End launched My Virtual Model (MVM) Visualization technology in 1998 to allow online shoppers to experience products in the virtual dressing room. However, even though big Internet retailers are making substantial financial commitments to adopt these new technologies to improve consumer experience of online product evaluation, their effects are largely unknown. It is a common belief that new technologies will improve the Internet shopping experience, but available evidence does not support this common belief.

In their panel study with Internet shoppers, Retail Forward found that the Internet shoppers did not perceive 3-D images to be important to their shopping satisfaction, although they considered them a nice feature to have on the Web site. In addition, one recent study found that survey respondents perceived 3-D images and virtual models to be unimportant features in Internet apparel shopping, whereas large pictures and close-ups were perceived to be important features (Kim, Kim, & Lennon, 2006).

Regarding virtual models, although Lands’ End reported the positive impact of MVM on conversation rates and average order value (Direct Marketing, 2001), Lane Bryant, the nation’s largest plus-size retailer, removed MVM technology after using it only for a few years because their customers no longer used it (Lane Bryant, 2005). Likewise, the effects of various product presentation technologies are largely unproven, despite the magnitude of financial commitment required in adopting new technologies. Product nformation plays an important role in consumer purchase decisions (Kim & Lennon, 2000; Mitchell & Boustani, 1994). Particularly due to the inability to physically evaluate products in Internet shopping, product presentation offered by Internet retailers plays a critical role in satisfying consumer needs for adequate product information for purchase decisions (Fiore, Jin, & Kim, 2005; Nitse et al. , 2004; Then & Delong, 1999). Despite its importance in Internet shopping, very little has been learned about how different online product presentation formats influence consumer decision making in Internet shopping.

Therefore, the purpose of this study was to examine how different online product presentation formats influence consumer attitudes toward the product and purchase intention in Internet shopping. This study focused on the two most basic forms of product presentation; visual (picture) and verbal (text) as the first step in the line of research that would investigate various presentation techniques including 3-D images, virtual models, and zooming technology in the future. Research objectives for this study were 148 KIM AND LENNON Psychology & Marketing DOI: 10. 002/mar Stimuli Information Processing Consumer Responses Imagery Information Processing Affective Attitude Purchase Intention CONTEMPORARY TORN STRETCH DENIM JACKET INFUSED WITH STRETCH FOR A TOUGH LOOK, AND DECORATED IN RHINESTONE ZIG-ZAGS FOR A GIRLY EDGE. ZIP FRONT. HITTING AT THE HIP. 96% COTTON/14% ELASTIC. Discursive Information Processing Cognitive Attitude Figure 1. The conceptual model for the effect of visual and verbal information on attitudes and purchase intentions in Internet shopping. hreefold: (1) to examine the effects of visual and verbal information on consumers’ attitudes toward the product, (2) to examine the effects of visual and verbal information on consumers’ purchase intentions, and (3) to evaluate the relative importance of visual and verbal information in product presentation in Internet shopping. Although Internet retail sales remain a small fraction of total retail sales and are considerably less than once predicted, Internet retailing is becoming more important in the retail industry, and consumer demands for Internet shopping are increasing.

The findings of this research will provide useful information that Internet retailers can use to develop more effective product presentations and thus satisfy consumer needs for adequate product evaluation in Internet shopping. LITERATURE REVIEW In this section, a conceptual model is developed to explain how visual and verbal information influence consumer attitudes toward a product and further influence Internet purchase intentions (see Figure 1).

Visual versus Verbal Information1 Information presented in visual and/or verbal form is a fundamental element of the consumer information environment, especially in a nonpersonal marketing context such as advertising or non-store retailing. 1 Visual information is limited to pictorial representation of a product and verbal information is limited to textual information about a product in this study. THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 149 Information is available in visual form, verbal form, or more frequently as a combination of both forms.

In Internet shopping, product information is most often presented as a combination of both visual and verbal forms. There have been two distinct approaches to studying the effects of visual versus verbal information in consumer and advertising research. One approach focused on the effects of visual and verbal information on memory (Guenther, Klatzby, & Putnam, 1980; Lutz & Lutz, 1977; Shepard, 1967; Starch, 1966), and the other approach focused on the effects of visual and verbal information on consumer judgments or attitudes (Childers & Houston, 1984; Edell & Staelin, 1983; Hirschman, 1986; Holbrook & Moore, 1981; Kisielius & Roedder, 1983).

The first research stream has generally supported the idea that visual information is superior to verbal information in recall and recognition. Using print advertisements as stimuli, Starch (1966) found that people remembered a print advertisement with a picture better than one without a picture. Shepard (1967) also found that a photo from advertisements was more easily recognized and remembered over time. Subsequent research (Hirschman & Solomon, 1984; Guenther et al. , 1980) provided additional support for the superiority of visual information.

Researchers further found that memory was enhanced when there was a certain amount of redundancy or correspondence between visual and verbal information (Childers & Houston, 1984; Son, Reese, & Davie, 1987). The second research stream focused on the influence of visual and verbal stimuli on attitudinal responses (Holbrook, 1985; Mitchell & Olson, 1981). Using print advertisements, Mitchell and Olson found that the visually oriented advertisement was more effective in generating a positive attitude toward the brand and more effective in communicating attributes of the product advertised than the verbally oriented advertisement.

They concluded that visual information led to more changes in beliefs about the product and thus created more positive attitudes and purchase intentions than verbal information. Imagery versus Discursive Information Processing Different forms of information lead to different information processing (e. g. , dual coding hypothesis by Paivio, 1971; left-right hemisphere specialization by Geschwind, 1979; sequential vs. simultaneous processing modes by Das, Kirby, & Jarman, 1975) (see Figure 1). Visual stimuli2 evoke imagery information processing, whereas verbal stimuli evoke discursive information processing.

Previous research focused more on discursive processing through an examination of how words or numbers are pooled together in working memory to signify or resolve problems 2 Visual (verbal) information is also referred to as visual (verbal) stimuli. The terms “information” and “stimuli” are used interchangeably in this study. 150 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar (e. g. , Bettman, 1979), whereas increasing research attention has shifted to the role of imagery information processing (Childers & Houston, 1982, 1984; Childers, Houston, & Heckler, 1985; Rossiter & Percy, 1983; Smith, Houston, & Childers, 1984).

Imagery information processing evoked by visual stimuli represents sensory or perceptual information in working memory. Imagery processing sometimes includes multi-sensory dimensions—including sight, taste, smell, and other sensations—or involves a single dimension such as sight, whereas discursive processing by verbal stimuli tends to be detached from inner sensory experience (MacInnis & Price, 1987). Therefore, discursive information processing becomes less concrete than imagery processing due to its lack of sensory experience of information in working memory (MacInnis & Price, 1987).

Overall, prior research findings support the superiority of imagery information processing. In a consumer research context, researchers found superior effects for imagery information processing as opposed to discursive processing (Cautela & McCullough, 1978; MacInnis & Price, 1987). MacInnis and Price posited that both discursive and imagery processing can be activated to frame problems, and the way in which a problem is presented with visual or verbal information can have a remarkable impact on problem solving.

In brand evaluation, discursive processing may lead to an implicit or explicit summary of brand attributes and features based on some combination rules, whereas imagery processing may lead to a holistic evaluation of the brand. They further speculated that imagery information processing leads consumers to expect a higher likelihood for decision outcomes than discursive processing because imagery makes it easier to visualize decision outcomes and visualization makes an event look more real (MacInnis & Price, 1987). Prior research findings further supported the effects of imagery processing on purchase intentions and purchase timing.

In clinical contexts, Cautela and McCullough (1978) found that imagery processing was more influential in affecting behavioral intentions than discursive processing. Staats and Lohr (1979) posited that imagery could affect behavior by eliciting an emotional response. Images that create positive emotions elicit approach responses, whereas images that create negative emotions elicit avoidance responses. MacInnis and Price (1987) posited that imagery processing may generate a stronger emotional or more concrete sensory experience than discursive processing, which in turn increases desire for the product.

They further postulated that the emotions evoked by elaborated imagery processing may reduce the delay between purchase consideration and actual purchase, thus affecting purchase timing. Moreover, imagery processing can enhance the consumption experience compared to discursive processing because the sensory experience evoked by imagery processing allows consumers to attain some of the enjoyment, satisfaction, or stimulation that would derive from actual consumption (Holbrook & Hirschman, 1982; Lindauer, 1983).

THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 151 Dual Coding Theory One consideration in the study of human cognition is the process of verbal coding. Verbal coding proponents assert that verbal coding is vital to perceptual processing and believe that visual information is identified by naming it (Bruner, 1957; Glazner & Clark, 1963). A second approach is imagery coding, and this approach argues that both visual and verbal information are stored most efficiently as nonverbal images.

A third approach to cognition is dual coding theory, first proposed by Paivio (1971, 1986). This approach explains that visual information and imagery information processing evoked by visual stimuli are superior to verbal counterparts. The dual coding theory views cognition activities as a result of two mental subsystems, a verbal system (processing verbal events) and an imaginal system (processing nonverbal events). These two subsystems are thought to be separate but interconnected components of human cognition. Each subsystem is linked to particular sensory systems through epresentational connections, and an associative network exists within each subsystem. Each subsystem is also associated with referential connections among them. The verbal system facilitates sequential processing whereas the imaginal system facilitates parallel processing of information. According to Paivio (1971, 1986), encoding of information in memory is done as a verbal form or nonverbal/pictorial form. When a person encounters a visual stimulus like a picture, an imaginal code is activated, whereas the verbal code will be activated when the person encounters a verbal stimulus like text.

These two independent and distinct codes form three discrete levels of processing for incoming stimuli. The most basic level of processing is called “representational processing,” and this processing involves the direct activation of either the verbal or imaginal systems, depending on whether incoming stimuli are visual or verbal. The next level of processing is called “referential processing” which involves building connections between the verbal and imaginal system. These connections between the two subsystems allow for evocation of imaginal responses from verbal stimuli or vice versa.

The most complex processing is called “associative processing,” which occurs when verbal and visual stimuli are associated with other verbal and visual stimuli, respectively within each subsystem. When a stimulus is received, it first goes through the representational processing, where either a verbal code or imaginal code is activated. Next, the stimulus passes through referential processing, where the visual cue is named or images are created for verbal cues. In the final step, the stimulus is processed at the associative level, where connections may be established between the verbal and imaginal codes and previously stored information.

The dual coding model further assumes that the verbal information is sequentially processed, whereas visual information is simultaneously processed and encoded as both images and verbal traces. Coding redundancy (i. e. , two 152 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar codes are better than one) accounts for the picture superiority effect. This dual process results in superior memory responses to visual stimuli (Paivio & Csapo, 1973). Prior research findings in both psychology and consumer research generally support the picture superiority effect (Paivio & Foth, 1970; Peterson & McGee, 1974; Purnell & Solman, 1992).

The picture superiority effect is generally attributed to the mental imagery elicited by visual stimuli (Paivio, 1969). Paivio (1971, pp. 135–136) defines mental visual imagery as a “memory code or associative mediator that provides spatially parallel information that mediates overt responses without necessarily being consciously experienced as a visual image. ” Many media, especially advertising, rely greatly on visual and verbal information to present the advertised product. Albeit not exactly the same, the commercial Web sites present a picture of a product with verbal descriptions in a similar manner as in advertising.

Recent e-commerce research supports the idea that the Internet works as an advertising medium (Joines, Scherer, & Scheufele, 2003; Joint & Waterhouse, 2003). Singh and Dalal (1999) contended that the commercial Web pages perform the same function as advertisements: to inform consumers of the product and to encourage consumers’ positive attitudes and behaviors toward the product. To date most prior research on the effects of visual and verbal information has been conducted in an advertising context and has generally supported the superior effects of visual information to verbal counterparts.

Given the similarities between advertising and the Internet in product presentation, this study expects that the superior effects of visual information found in advertising are likely to hold in Internet shopping. Dual Processing Models of Attitudes A number of attitude researchers have proposed two characteristics of attitudes. The first characteristic proposed is that an attitude is a function of responses to the attitude object (Eagly & Chaiken, 1993) and the second characteristic is related to the evaluative nature of an attitude categorized as either good or bad.

In spite of pervasive findings of the influence of affect on attitudes (Forgas, 1992: Schwarz, 1990), this view does not posit an affective component to attitudes. There has been disagreement among attitude researchers regarding where to place the affective component. Some researchers have attempted to distinguish affect from attitude by differentiating affect as a more temporary feeling state versus attitude as a more constant and general evaluation (Petty & Cacioppo, 1983), whereas others have expanded the scope of attitude to include all mental phenomena generating positive or negative evaluations (Eagly & Chaiken, 1993; Greenwald, 1968).

Some researchers have embraced affect as a component of attitude but distinguished it from the cognitive component of attitude, which is a deliberate, conscious, and propositional THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 153 thought process (Crites, Fabrigar, & Petty, 1994; Millar & Tesser, 1989). A more recent approach is to embrace both affective and cognitive responses as components of attitudes, called “dual processing of attitudes” (Chen & Chaiken, 1999; Epstein & Pacini, 1999; Koriat & LevySadot, 1999).

The cognitive component of attitude represents the deliberate, conscious, and propositional thought process, whereas the affective component of attitude represents immediate evaluation and emotional responses to the attitude object. Among many viable explanations, the heuristic-systematic model (Chen & Chaiken, 1999) identifies two basic modes (systematic vs. heuristic) by which people form attitudes and make social judgments. Systematic processing involves a relatively extensive and logical processing of judgment-relevant information, whereas heuristic processing involves the activation and use of judgment-relevant rules.

Another group of attitude researchers posit two different judgment systems: an affect-based system and an information-based system (Koriat & Levy-Sadot, 1999). A third approach to dual processing of attitudes is proposed by Epstein and Pacini (1999). CEST (Cognitive-Experiential Self-Theory) proposed two different information-processing systems: a preconscious experiential system and a conscious rational system. All three models delineate two components of attitudes; affective and cognitive attitudes.

Based on the dual processing model of attitudes, visual information is posited to influence affective attitudes through imagery information processing, and verbal information is posited to influence cognitive attitudes through discursive information processing. Although it is possible that visual information may influence cognitive attitudes and vice versa, it is posited that the major influences on each component of attitudes are dependent on type of information (see Figure 1). Hypotheses Development Visual Information.

Visual information can vary by picture size. Both psychology and advertising research has shown that picture size is positively related to memory and attitudes (Kossyln, 1980; Mitchell & Olson, 1981; Rossiter & Percy, 1980, 1983). In an advertising context, when the same picture is used in different sizes, larger pictures engender significantly more favorable attitudes than the same picture in a smaller size. Rossiter and Percy (1978, 1983) found that a larger picture generated more positive effect on brand attitude than did a smaller picture.

Imagery research suggests that elaborated imagery processing affects behavioral intention (McMahon, 1973), and a large picture better facilitates imagery processing (MacInnis & Price, 1987). As compared to a small picture, a large picture is likely to have a more positive influence on elaborated imagery processing and subsequently affect behavioral intention (Rossiter & Percy, 1978; Smith et al. , 1984). Better elaborated imagery processing increases perceived likelihood of an event (MacInnis & Price, 154 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar 987), and people who imagined themselves performing a behavior showed a significant increase in their behavioral intentions (Gregory, Cialdini, & Carpenter, 1982). Mitchell and Olson (1981) found that positively evaluated visual stimuli increased attitude toward and purchase intention for a product. Thus, the following hypotheses were developed. In the present study, picture size is posited to determine the level of visual information. H1: As compared to people exposed to less visual information, those exposed to more visual information will have more positive attitudes toward the product. a: affective attitude, b: cognitive attitude) H2: As compared to people exposed to less visual information, those exposed to more visual information will have greater purchase intentions. Amount of Verbal Information. Prior research emphasizes the importance of verbal information in purchase decisions, especially in non-store retailing. Spiller and Lohse (1998) conjectured that product descriptions available on the Internet are equivalent to salespeople’s service at retail stores. Their analysis of 137 Internet retail stores revealed that good product descriptions influence ales in Internet shopping. Kim and Lennon (2000) posited that the perceived amount of verbal information moderates the level of perceived risk associated with television apparel shopping and subsequently increases purchase intentions. Other Internet shopping research further confirmed the positive role of product information on consumer behavior (Ballantine, 2005). Accordingly, the following hypotheses were developed. H3: As compared to people exposed to less verbal information, those exposed to more verbal information will have more positive attitudes toward the product. a: affective attitude, b: cognitive attitude) H4: As compared to people exposed to less verbal information, those exposed to more verbal information will have greater purchase intentions. In Internet retailing, the size of product pictures varies to a great extent and so does the amount of verbal product information. According to a recent content analysis of Internet apparel retailers (Kim et al. , 2006), picture sizes substantially varied across 111 apparel retail Web sites that were fairly good representations of Internet apparel retailers. Picture size ranged from 100 100 pixels to 800 600 pixels across apparel retail Web sites.

The researchers also found that the amount of verbal product information varied to a great extent across retail Web sites. When both visual and verbal information are available in Internet THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 155 retailing, it is further expected that visual and verbal information interact to influence consumer responses to the product. Therefore, the following hypotheses were developed. H5: Visual and verbal information will interact to affect attitudes toward the product. a: affective attitude, b: cognitive attitude) H6: Visual and verbal information will interact to affect purchase intentions. Prior research findings in both psychology and consumer research support the picture superiority effect in consumer memory and attitudes (Paivio & Foth, 1970; Peterson & McGee, 1974; Purnell & Solman, 1992). Therefore, the following hypotheses were developed. H7: Visual information will explain more variance in attitudes than verbal information. H8: Visual information will explain more variance in purchase intentions than verbal information.

METHODOLOGY This study employed a Web experiment using a mock retail Web site. According to Hantula (2005), Web experiments can be realistic and may be indistinguishable from real-life online interactions. For a mock Web site, a fictitious brand name was used to avoid any effects on attitudes and purchase intentions due to well-known brand names. A pretest was first conducted to develop visual and verbal stimuli for a mock Web site simulating Internet apparel shopping. The present study focused on apparel products consisting of multiple apparel categories such as tops, blouses, pants, skirts, and dresses.

As a key type of item sold online (Internet Retailer, 2003), apparel requires sensory evaluation to make a purchase decision and thus is deemed appropriate for the present study. Stimulus Development Visual Stimuli. First, visual stimuli (pictures of apparel items) were developed by downloading apparel pictures from commercial Web sites. Apparel items were selected from commercial Web sites because items sold on commercial Web sites are expected to be desirable by target customers, thus encouraging research participants to engage in simulated online shopping.

A total of 28 apparel items were initially selected, including woven shirts, knit tops, pants, skirts, dresses, sweaters, and jackets for women. To avoid extraneous factors, only pictures of garments without 156 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar models were selected. All pictures were shown on a body form. To assure a consistent size of garments, all 28 items were tried on and adjusted to fit the dummy model with full body, developed using Adobe Photoshop. The same image size of pants and shirts do not reflect the same garment size because pants are longer than shirts.

If their image sizes are the same, pants are probably smaller than shirts in terms of garment size. This problem was solved by fitting apparel items to the dummy body. Consistency in background, angle of photo shots, and the quality of pictures was achieved through a careful sampling process and touch-up using Photoshop. For the current research, the size of pictures was manipulated (small, large) to vary visual information. The small-size picture was one-fourth of the large-size picture. Verbal Stimuli.

Verbal product information to accompany apparel items was created using evaluative criteria for apparel purchases developed by Eckman, Damhorst, and Kadolph (1990). Eckman et al. categorized apparel evaluative criteria into intrinsic and extrinsic attributes. First, intrinsic criteria refer to product attributes that cannot be changed or manipulated without changing the physical characteristics of the product itself (e. g. , style, fiber content). Extrinsic criteria refer to product attributes that are not component parts of the physical products but are created by the manufacturer or retailer (e. g. , price, brand name).

Both intrinsic and extrinsic criteria are used in apparel purchase situations, but prior research shows that intrinsic criteria are more important to consumers than extrinsic criteria (Eckman et al. , 1990; Jacoby, Olson, & Haddock, 1971). In this study, extrinsic and intrinsic verbal stimuli were developed. The intrinsic criteria were (1) style (design features), (2) construction details, (3) fit, (4) fiber content (or fabric name), (5) care instruction, (6) color (also print information for printed fabrics), and (7) size; the one extrinsic criterion was (8) price. Amount of verbal information was manipulated as high and low.

The high amount of verbal information included all eight pieces of information; the low amount of verbal information included three pieces of intrinsic information (style, color, and size) and one piece of extrinsic information (price). This manipulation was based on prior research findings on consumer need for information in making apparel purchase decisions. When purchasing apparel, price, style, and color were the most frequently sought types of information, followed by fiber content (or fabric name), garment care instructions, brand name, and fit information (Davis, 1987; Martin, 1971).

Thus, the low verbal condition included the most needed information (style, color, price) in addition to information about size. Size information was added because a shopper must choose a size to proceed with a purchase. For the high verbal condition, further information was added that consumers seek when purchasing apparel, such as fiber content (or fabric name), care instructions, and fit (Davis, 1987; Martin, 1971). Detailed style information THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 157 nd construction details were also added to the high verbal condition, based on the suggestion that more sensory-oriented, tactile descriptions of a product are desired in Internet shopping (Park & Stoel, 2005). Therefore, the low verbal condition was designed to include information necessary to make an apparel purchase, whereas the high information condition was intended to include additional information that consumers are likely to desire when shopping for apparel online. Pretests A pretest of apparel pictures was first conducted to select visual stimuli for the main study and also to perform a manipulation check on verbal stimuli.

The goal was to select apparel items that were neutral in terms of attractiveness, fashionableness, and likableness, to minimize the potential effect of apparel items per se on attitudes and purchase intentions (e. g. , a very attractive apparel item will be desired by many people regardless of presentation format). College women (n 44) participated in the pretest, using a mock Web site in exchange for course credit. During the pretest, all participants evaluated 28 apparel items in the same size (picture only) on attractiveness, fashionableness, and likableness measures, one apparel item at a time.

Three evaluative measures used a 7-point rating scale from 1 (highly unattractive; highly unfashionable; highly unlikable) to 7 (highly attractive; highly fashionable; highly likable). To select neutral apparel stimuli, scores from the three evaluative measures for each apparel item were collapsed, based on the consistency of three measures (all s 0. 90). The possible summed scores per item ranged from 3 to 21. The 10 apparel items with the most neutral ratings on the three measures (summed scores ranged from 11 to 13; midpoint 12) were selected for the main study.

To assess order effects, three different presentation orders of the 28 apparel items were used in the pretest. MANOVA revealed no effect for presentation order [Wilks’s l 0. 94, F(6, 70) 0. 37, p 0. 90] on the three dependent variables (attractiveness, fashionableness, and likableness). During the pretest, a manipulation check of verbal stimuli was also conducted. After evaluating apparel items, participants were randomly assigned to one verbal condition (high vs. low) to evaluate their perceptions of the amount of verbal product information.

For stimulus sampling purposes, two apparel items were evaluated in each verbal condition. Pretest participants were randomly assigned to one of the verbal conditions (high vs. low) and viewed both apparel items under their assigned condition. After viewing the items, participants rated the perceived amount of verbal information in the product description using a 7-point rating scale from 1 (very little) to 7 (very much). Responses evoked by both outfits were summed ( 0. 85) and used as a measure of the perceived amount of information.

One-way ANOVA was performed to examine the effect of the verbal manipulation on perceived amount of information 158 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar and found a main effect for verbal condition on perceived amount of information [F(1, 42) 6. 63, p . 05] Pretest participants exposed to the high amount of verbal information (M 11. 45, SD 2. 04) perceived more information than those exposed to the low amount of information (M 10. 00, SD 1. 69). Thus, the manipulation of the amount of verbal information was appropriately perceived by the pretest participants. Instrument Development Attitude Measures.

Attitude items were adopted from Hirschman (1986). Affective attitudes were measured by attractiveness and likableness, and cognitive attitudes were measured by perceived amount of information and perceived usefulness of information. All attitude items used 7-point Likert scales with endpoints of 1 (strongly disagree) and 7 (strongly agree). Purchase Intention Measure. One item was used to measure Internet purchase intention. Adopted from Taylor and Baker (1994), this item addressed the intention to purchase an apparel item viewed during the Web experiment in a certain time frame (i. . , in the upcoming year) using a 7-point Likert rating scale with endpoints of 1 (strongly disagree) and 7 (strongly agree). Other Measures. Two items were developed to measure perceptions of picture size and amount of verbal information, respectively. Other items assessing prior experience with the Internet and Internet shopping and demographic information were also included. Except for demographic items, all items used 7-point rating scales. To enhance the realism of the experiment, this study used a Web experiment so that participants could participate when and where convenient.

Unlike lab experiments in which participants use the same types of computers, the participants in this study could use various types of computers and monitors. Though improving realism, this method posed a concern due to additional variability with regard to picture size as a function of types of computers used to participate in the experiment. Therefore, information about types of computer, monitor size, and monitor resolution was collected to better interpret the results. Instructions were provided to participants about how to find the information about resolution of their monitors.

EXPERIMENT 1 Procedure Experiment 1 was a 2 (Visual: Large vs. Small) 2 (Verbal: High vs. Low) between-subjects design. When participants logged onto the mock THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 159 Web site, they were randomly assigned to one of the four treatment conditions and evaluated all 10 apparel items for stimulus sampling purposes (Fontenelle, Phillips, & Lane, 1985). Stimulus sampling is used so that results can be generalized over more than one stimulus (i. e. , to increase external validity).

In this research, using 10 apparel items ensures that any significant effects are not due to idiosyncratic characteristics of a single stimulus. The order of presentation of the apparel items was completely randomized to distribute any order effects randomly over the four treatment conditions. Participants were instructed to assume that they had enough money to purchase any items they wished to buy, to minimize the effect of monetary constraints on purchase intentions. Participants College women (n 159) enrolled at a large Midwestern university participated in a Web experiment in exchange for course credit.

College women were recruited for the study because young women comprise a significant portion of Internet shoppers. According to Internet Retailer (2004c), they make up 63% of shoppers at online apparel and beauty sites. Additionally, research evidence supports that college students do not differ from typical consumers in terms of beliefs and attitudes (Duvasula et al. , 1997). After eliminating unusable responses due to incomplete questionnaires, there were 145 usable questionnaires. The mean age of participants was 22 (see Table 1).

More than 80% of participants were juniors or seniors. Over 88% of participants owned PCs and almost 95% of them had Internet access at home. Results Manipulation Checks. After completing the dependent measures, participants rated the perceived picture size and perceived amount of verbal information. As anticipated, ANOVA results indicated that actual picture size had a significant effect on perceived size of picture [F (1, 143) 184. 02, p . 001] and actual amount of verbal information had a significant effect on perceived amount of verbal information [F(1, 143) 56. 9, p . 001]. Participants who viewed large pictures perceived pictures to be larger (M 5. 07, SD 1. 06) than those who viewed small pictures (M 2. 43, SD 1. 28). Participants exposed to the high amount of verbal information perceived more verbal information (M 5. 23, SD 1. 18) than those exposed to the low amount of information (M 3. 55, SD 1. 48). Thus, experimental manipulations were successful. Preliminary Analysis. Participants evaluated all 10 apparel items on the four attitudinal items and purchase intention. After checking reliabilities (all s 0. 0), scores for each item were collapsed for all 10 apparel 160 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar Table 1. Demographic Characteristics of Participants. Experiment 1 (n 145) Characteristics Age 20 20–25 26–30 30 Academic standing Freshman Sophomore Junior Senior Graduate student Own PC Yes No Access to the Internet Yes No Monitor size Smaller than 15? 15? –17? 19? –21? Bigger than 21? Monitor resolution 640 480 800 600 1024 768 Others f 12 121 8 4 3 12 62 58 10 128 17 137 8 12 93 34 6 4 93 45 3 % 8. 3 83. 4 5. 5 2. 8 2. 1 8. 3 42. 8 40. 0 6. 9 88. 3 11. 7 94. 5 5. 8. 3 64. 1 23. 5 4. 1 2. 8 64. 1 31. 0 2. 1 f 15 133 1 1 1 18 74 55 2 122 28 144 6 15 102 30 3 11 84 48 7 Experiment 2 (n 150) % 10. 0 88. 7 0. 7 0. 7 0. 7 12. 0 49. 3 36. 7 1. 3 81. 3 18. 7 96. 0 4. 0 10. 0 68. 0 20. 0 2. 0 7. 3 56. 0 32. 0 4. 7 stimuli. Scores for each stimulus ranged from 10 to 70 (10 stimuli with a 7-point rating scale). Then the two items tapping affective attitudes were summed ( 0. 97); likewise the two items tapping cognitive attitudes were summed ( 0. 96). Affective attitude scores ranged from 20 to 140 and cognitive attitude scores ranged from 20 to 138.

Purchase intention scores ranged from 10 to 67. Analyses. All hypothesized relationships were initially examined in a multivariate analysis of variance (MANOVA) with affective and cognitive attitudinal responses to apparel stimuli and purchase intention as dependent variables; visual information varied by picture size and verbal information varied by the amount of product information were the independent variables. Results indicated that amount of verbal information was significantly related to the set of dependent variables [Wilks’s 0. 82, F(3, 139) 10. 17, p . 0001].

Follow-up univariate analyses of THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 161 variance indicated that the amount of verbal information was related to both affective [F(1, 141) 6. 9, p . 05, 2 0. 04] and cognitive [F(1, 141) 30. 74, p . 0001, 2 0. 17] attitudes. Participants exposed to more verbal information about the product expressed stronger affective attitudes (M 95. 07, SD 21. 53) than those exposed to less verbal information (M 85. 25, SD 24. 72). Also, participants exposed to more verbal information exhibited stronger cognitive attitudes (M 112. 6, SD 19. 40) than those exposed to less verbal information (M 89. 34, SD 29. 38). Therefore, H3a and H3b positing the effects of verbal information on affective and cognitive attitudes were supported. No main effect for verbal information on purchase intention was found, and no significant multivariate effect was found for visual information or the interaction. Thus, the remaining hypotheses were not supported. Post-hoc Analysis. Results from Experiment 1 were surprising in that no effects for visual information were found, despite evidence from previous literature supporting picture superiority.

Picture size had no effect on affective attitudes, whereas the amount of verbal information had a significant effect on affective attitudes. Additional analyses were performed to see whether participant perceptions of visual and verbal information have different effects on attitudes and purchase intentions. Simple regression analyses were performed using perceived picture size and perceived amount of information as independent variables and both components of attitudes and purchase intention as dependent variables. Simple regression analyses revealed that perceived picture size was positively related to affective attitudes [F(1, 143) 19. 0, p . 0001] and also to cognitive attitudes [ F(1, 143) 5. 40, p . 05]. Perceived amount of verbal information was also a significant predictor of affective attitudes [F(1, 143) 23. 32, p . 0001] and cognitive attitudes [F(1, 143) 155. 12, p . 0001]. Further, both perceived picture size and perceived amount of verbal information were significant predictors of purchase intention [F(1, 143) 14. 34, p . 0001; F(1, 143) 17. 52, p . 0001, respectively]. Whereas objectively manipulated picture size had no effect on either affective or cognitive attitudes, perceived picture size was a significant predictor of both components of attitudes.

In addition, both visual and verbal information had a significant influence on purchase intention when participant perceptions of picture size and amount of verbal information were used instead of objectively manipulated visual and verbal information. Examination of monitor size and monitor resolution provided useful insights to explain why there were no effects for visual information as hypothesized. Monitor size used for the experiment greatly varied from 15? to 21? (see Table 1). For resolution, about 64% of participants used 800 600 pixels and 31% used 1024 768 pixels.

Thus, although picture size was objectively manipulated in the experiment, the actual size of pictures that participants saw during the experiment varied depending on both the size and resolution of monitors used to participate in the 162 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar study. Although participants were instructed to participate in this study by logging onto the Web site when and where convenient in order to enhance the realism of the Internet shopping context, variations in monitor size and resolution may have confounded true effects of visual information.

To avoid such problems, Experiment 2 was conducted to expose participants to all four treatment conditions. In this way, although actual picture size viewed by participants might vary depending on monitor size and resolution, participants could see relative differences between large and small pictures. This context is also more realistic, given that in actual Internet shopping situations picture sizes vary greatly across different retail Web sites as do shoppers’ computer monitors. EXPERIMENT 2 Procedure Experiment 2 was a 2 (Visual: Large vs.

Small) 2 (Verbal: High vs. Low) within-subjects design. Eight apparel items were selected from the 10 items used in Experiment 1 by eliminating two items deemed inappropriate due to seasonal change. Participants were exposed to all four visual by verbal treatment conditions and evaluated two apparel items in each condition, for a total of eight apparel items rated. Sixteen different presentation orders of the experimental conditions were used to balance out order effects. The presentation order of eight apparel items was fully randomized.

The questionnaire used in Experiment 2 was modified from the questionnaire used in Experiment 1 by eliminating the questions about perceptions of picture size and amount of information. Aside from these differences, the stimulus materials, procedures, independent variables, and dependent variables were identical to those of Experiment 1. Scores on the dependent variables were collapsed for two apparel items within each treatment condition. Participants College women (n 160) enrolled at a large Midwestern university participated in a Web experiment in exchange for course credit.

Participants in Experiment 2 did not overlap with participants in Experiment 1. After eliminating unusable responses due to incomplete questionnaires, there were 150 useable questionnaires. The mean age of the participants was 21 (see Table 1). A majority of participants were juniors or seniors. More than 81% of participants owned PCs and 96% of them had Internet access at home. Overall, demographic characteristics of participants in Experiment 2 were similar to participants in Experiment 1. THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 002/mar 163 Results Repeated measures MANOVA involving all dependent measures indicated that one or more dependent variables differed by visual information [Wilks’s 0. 91, F (3, 147) 5. 06, p . 01], by verbal information [Wilks’s 0. 31, F (3, 147) 108. 00, p . 0001], and by an interaction between visual and verbal information [Wilks’s 0. 92, F (3, 147) 4. 15, p . 01]. Follow-up repeated measures ANOVAs for visual information found that visual information had significant main effects on affective [F(1, 149) 7. 73, p . 01, 2 0. 04] and cognitive attitudes [F(1, 149) 11. 63, p . 1, 2 0. 07], thus supporting H1a and H1b. Purchase intention was not related to visual information, thus failing to support H2. Subsequent ANOVAs for verbal information indicated that amount of verbal information had significant main effects on affective [F(1, 149) 23. 50, p . 0001, 2 0. 13] and cognitive attitudes [F(1, 149) 289. 57, p . 0001, 2 0. 70]. Purchase intention was also significantly related to the amount of verbal information [F(1, 149) 7. 64, p . 01, 2 0. 04]. H3a, H3b, and H4 positing the main effects of verbal information on dependent variables were all supported.

Follow up ANOVAs were further conducted for interactions (see Figure 2). Results indicated that visual by verbal interaction effects were significant for cognitive attitudes [F(1, 149) 9. 68, p . 01, 2 0. 05] and purchase intentions [F(1, 149) 3. 95, p . 05, 2 0. 02]. Simple effects tests revealed that the effect of visual information on cognitive attitudes was significant [F(1, 149) 17. 30, p . 0001] only when the amount of verbal information was low. When the amount of verbal information was low, participants who viewed the large picture exhibited more positive cognitive attitudes (M 18. 6, SD 4. 68) than those who viewed the small picture (M 16. 92, SD 4. 70). This difference was larger when the amount of verbal information was low rather than when the amount of verbal information was high. The effect of verbal information on cognitive attitudes was significant both when picture size was large [F(1, 149) 182. 81, p . 0001] and small [F(1, 149) 268. 20, p . 0001]. Data supported H5b, but not H5a. Simple effects tests further indicated that the effect of visual information on purchase intention was significant when the amount of verbal information was low [F(1, 149) 5. 9, p . 05] (see Figure 2). When the amount of verbal information was low, participants who viewed large pictures (M 7. 05, SD 3. 08) exhibited stronger purchase intentions than those who viewed small pictures (M 6. 50, SD 2. 81). Simple effects tests also indicated that the effect of verbal information on purchase intention was significant when small pictures were used [ F(1, 149) 10. 52, p . 01]. When exposed to small pictures, participants who received more verbal information (M 7. 30, SD 3. 08) expressed stronger purchase intentions than those who received less verbal information (M 6. 0, SD 2. 81). H6 was thus supported. 164 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar 24 23. 32 22. 97 Purchase Intentions 7. 3 7. 2 7. 30 7. 23 7. 05 Verbal Info High Low Cognitive Attitudes 22 Verbal Info 20 18. 36 18 16. 92 High Low 7. 1 7. 0 6. 9 6. 8 6. 7 6. 6 16 Large Small 6. 5 Large 6. 50 Small Picture Size Picture Size Figure 2. Visual by verbal interaction. H7 and H8 posited that visual information varied by picture size would explain more variance in attitudes and purchase intentions than verbal information.

Omega squared ( A2) was used to assess the relative importance of each of the independent variables. According to Cohen’s (1977) guidelines, 2 0. 15 is a large effect, 2 0. 06 is a medium effect, and 2 0. 01 is a small effect. As shown in Table 2, verbal information had a larger effect on attitudes than visual information. Contrary to prediction of H7, verbal information accounted for 13% of the total variance in affective attitudes, whereas 4% was accounted for by visual information. For cognitive attitudes, verbal information accounted for 10 times more variance than visual information ( 2 0. 0 vs. 2 0. 07) in cognitive attitudes. In regards to purchase intention, verbal information had a moderate effect on purchase intention ( 2 0. 04), whereas visual information did not have a significant effect. Discussion The present study examined how different presentation formats in Internet shopping influence consumer attitudes toward the product and subsequent purchase intention. Extending visual and verbal research in advertising into the Internet shopping context, this study investigated the effects of visual and verbal information on consumer responses.

Table 2. Comparisons of Effects ( Experiment 2. 2 A ) of Visual and Verbal Information in Visual by verbal interaction ns 0. 05 0. 02 Visual information Affective attitude Cognitive attitude Purchase intention 0. 04 0. 07 ns Verbal information 0. 13 0. 70 0. 04 THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 165 Table 3. Summary of Experiment 1 and 2. Experiment 2 (within-subjects design) MANOVA Visual Sig. ** Sig. ** ns Verbal Sig. **** Sig. **** Sig. **

Experiment 1 (between-subjects design) MANOVA Visual Affective attitudes Cognitive attitudes Purchase intentions *p . 05. **p . 01. ***p . 001. ****p . 0001. Regression Perceived visual Sig. **** Sig. * Sig. **** Perceived verbal Sig. **** Sig. **** Sig. **** Verbal Sig. * Sig. **** ns ns ns ns As summarized in Table 3, the findings showed that both visual and verbal information have significant effects on consumers’ affective and cognitive attitudes toward apparel products. However, only verbal information had a significant effect on purchase intention.

One noteworthy finding is that when consumer perceptions of picture size and the amount of verbal information were used instead of actual picture size and amount of verbal information, both visual and verbal information significantly influenced both affective and cognitive attitudes and also affected purchase intentions. The findings further support verbal superiority in product presentation in Internet shopping. This is contrary to the predicted visual superiority based on previous literature supporting picture superiority and need for sensory evaluation for apparel products.

Although it was anticipated that visual information would have stronger effects on consumer attitudes toward apparel products, especially affective responses, and purchase intentions, the results of the study suggest verbal superiority for both attitudes (affective and cognitive) and purchase intention. Based on the guidelines for effect sizes ( 2) of Cohen (1977), it was observed that verbal information had large effects for both affective and cognitive attitudes and had a moderate effect on purchase intention.

Although both visual and verbal information had significant effects on affective attitudes, visual information had a weaker impact on affective attitudes than verbal information. In regards to cognitive attitudes, both visual and verbal information influenced cognitive attitudes, and verbal information had a stronger impact on cognitive attitudes as expected. The effect of visual information on cognitive attitudes was significant only when the amount of verbal information was low and it was a medium effect. Regarding purchase intentions, verbal information had a significant effect only when small pictures were used, and it was a small effect.

Although the findings of verbal superiority were unexpected, the findings of the study are generally consistent with findings in Smith (1991). In an 166 KIM AND LENNON Psychology & Marketing DOI: 10. 1002/mar advertising context, Smith found that the effect of visual information was dominant only when visual information conveys different messages from verbal claims. When both visual and verbal information conveyed the same message, Smith found that inferences based on visual stimuli were weaker than inferences based on verbal claims due to higher uncertainty associated with visual stimuli.

Verbal information in an ad makes explicit, specific claims about product attributes or performance, which facilitate inferences about unknown information about a product. On the contrary, claims made using visual information tend to be less explicit and less specific, which is likely to result in a heightened uncertainty of inferences. Thus, in the present study, it is possible that the effect of visual information was lessened compared to verbal information because both visual and verbal stimuli conveyed the same messages about the product to some extent, especially for style information and construction details.

In addition, the way visual and verbal stimuli were manipulated in this study may have contributed to the stronger effects of verbal information than visual information. For visual stimuli, picture size was manipulated such that a small picture was one-fourth of a large picture. Despite the size difference, the same pictures were used. However, for verbal stimuli, the amount of verbal information was manipulated such that a low verbal condition did not include four pieces of intrinsic information that were provided in the high verbal condition.

Therefore, the difference between high and low verbal conditions (i. e. , absence of information) may be larger than the difference between visual condition groups, resulting in larger effects of verbal information. Indeed, visual information had a significant impact on both affective and cognitive attitudes, albeit weaker effects than verbal information. Another plausible explanation of the findings of the study is that perhaps verbal product information used in this study evoked imagery information processing in addition to discursive information processing because of concrete verbal stimuli (e. . , construction details of apparel). Imagery processing can be induced by a number of external sources. Pictures are the most well-known predictor of imagery (Paivio, 1971; Shepard, 1967). The superiority of visual information has been attributed to the imagery induced by visual information as compared to discursive information processing by verbal information (Childers & Houston, 1984; Lutz & Lutz, 1977; Paivio, 1971). In addition to pictures, concrete verbal stimuli can stimulate imagery processing (Paivio & Csapo, 1973; Paivio & Foth, 1970; Richardson, 1980).

The level of the concreteness of words was found to be significantly related to the level of imagery value (Pavio, Yuille, & Madigan, 1968). Paivio (1971) also posited that the verbal superiority of high imagery values can occur. In this study, the difference between high and low verbal conditions was the amount of product information, especially intrinsic product information including construction/style details, fit, fiber/fabric information, and care instructions. THE EFFECTS OF VISUAL AND VERBAL INFORMATION ON ATTITUDES Psychology & Marketing DOI: 10. 1002/mar 167

In particular, construction details and style information provide concrete information about apparel products (e. g. , the pointed collar and barrel cuffs, pearl buttons for front closure, rounded shirt bottom, and single chest pocket; two layers of silk with a sheer top, a pattern of slender roses with delicate, thorny stems in deep brown and green, transparent seed and bugle beads across the upper layer). Such verbal information may have led participants to engage in imagery information processing as well as discursive information processing evoked by other verbal messages.

Additionally, previous research supports that the effects of visual and verbal stimuli are interactive in such a way that the addition of verbal stimuli that explains the message conveyed by visual stimuli enhances the use of imagery processing (Bower, Karlin, & Dueck, 1975; Childers & Houston, 1984). Concrete verbal descriptions of style information and construction details of apparel items may have helped participants interpret the picture of the item and thus may have stimulated imagery processing. These interpretations may explain why verbal information had stronger effects on both affective and cognitive attitudes.

The concept of perceptual fluency provides useful insights to interpret the findings of the study. Concrete verbal descriptions used in this study are likely to improve perceptual fluency (e. g. , the ease of identifying the physical identity of the stimulus). The availability of concrete verbal infor

How to cite this page

Choose cite format:
The Effect of Visual and Verbal Information on Attitudes. (2017, Mar 21). Retrieved June 27, 2019, from https://phdessay.com/the-effect-of-visual-and-verbal-information-on-attitudes-and-purchase-intentions-in-internet-shopping/.