Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
The concept of Informational Social Influence (ISI) and its impact on perception, specifically in the context of the Internet and consumer products. The research presented demonstrates how individuals' opinions can be shaped by others, particularly when exposed to expert opinions or online recommendations. The document also discusses the limitations and future directions of ISI research.
What you will learn
Typology: Study Guides, Projects, Research
1 / 58
This page cannot be seen from the preview
Don't miss anything!
by Christopher P. Golda A thesis Presented to the faculty of Towson University in partial fulfillment of the requirements for the degree Master of Arts Department of Psychology Towson University Towson, Maryland 21252 (December, 2014)
ii
iv
Abstract Information Social Influence and the Internet: Manipulation in a Consumptive Society Christopher P. Golda Research on Informational Social Influence (ISI) has demonstrated that an individual’s opinion can be shaped by others. Subsequent research has shown that an individual’s sensory perceptions can be influenced in the same manner, with significant implications for consumer advertising. The present study extends the current body of literature on ISI by applying this premise to artwork evaluated over the Internet. In this study, 60 participants were randomly assigned to one of four groups wherein they were asked to evaluate a piece of artwork over the Internet while unsuspectingly utilizing confederate ratings that were manipulated by opinion and credibility. The results demonstrated that confederate ratings influenced a participant’s evaluation of artwork; however, confederate expertise and the interaction were not significant. Ultimately, these data demonstrate that ISI can generalize to aesthetic qualities. That is, perceived beauty of a piece of artwork in an online context varies with external influence. Keywords : Informational Social Influence, Internet, artwork, beauty, perception
v
Table of Contents
Introduction Topical Significance 1 Research Road Map 3 Literature Review Conventional ISI 5 Expertise Moderating ISI 6 ISI and Consumer Products 8 ISI, Products, and Expertise 11 ISI and Online Products 15 Present Study 20 Method Design 22 Participants 22 Apparatus 22 Procedures 26 Results 29 Discussion 31
Introduction Topical Significance A highly exploitable characteristic of human nature is our susceptibility to social influence strategies. This is a phenomenon that covers a wide variety of situations ranging from influence on individual perception of movement to manipulation of one’s acceptance of certain social and political viewpoints (Fein, Goethals, & Kugler, 2007; Reingen, 1982; Sherif, 1937). Somewhere in the middle of these two extremes, we find social influence as it relates to consumer products (Burnkrant & Cousineau, 1975; Cohen & Golden, 1972; Fein, Goethals, & Kugler, 2007; Pincus & Waters, 1977; Sherif, 1937). The ability to influence an individual’s attitude towards a particular consumer product can be a powerful and financially prosperous tool, especially for those who would strategically use it to take advantage of others (Burnkrant & Cousineau, 1975; Cohen & Golden, 1972; Pincus & Waters, 1977). For example, a department store using social influence strategies would likely be more successful in convincing a customer to buy a particular vase or furnishing than one that did not use such a ploy. Furthermore, high end boutiques could use the same strategies to convince a customer that a particular line of clothing was the most attractive and fashionable brand on the market, thus worth the extreme markup in price. Indeed, research has demonstrated that subtle influence strategies are effective at swaying opinion in many different circumstances, including an individual’s perception of the quality or craftsmanship of a particular product (Burnkrant & Cousineau, 1975; Cohen & Golden, 1972; Pincus & Waters, 1977). Moreover, the effectiveness of such psychological strategies can be further enhanced by presenting credible “expert” opinions
that solidify the evaluations presented for particular products (Lafferty, Goldsmith, & Flynn, 2005; Priilaid, Feinberg, Carter, & Ross, 2009). Certainly, experimental research does support the view that the expertise of the influencer can increase an individual’s susceptibility to social influence in a variety of situations (Lafferty, Goldsmith, & Flynn, 2005; Priilaid, Feinberg, Carter, & Ross, 2009). Unfortunately, the bulk of the research in this area has primarily been accomplished in manufactured environments where the use of confederates was too contrived to be convincing, making the process of extrapolating from existing data to real life scenarios challenging (Lafferty, Goldsmith, & Flynn, 2005; Priilaid, Feinberg, Carter, & Ross, 2009). With the advent of the Internet, bridging this gap between experimentation and real-world application has become an easier task (Chen, 2008). Researchers have not been remiss in recognizing the potential of the Internet as a venue where conformity as it relates to consumerism can thrive (Chen, 2008; Harmon, 2004; Huang & Min, 2007). Such effects are not limited to lab environments, however. It would be a simple feat for one to manipulate product information on a legitimate e- commerce website and would only be limited by the business’ ethics and morals. Harmon (2004) has taken this idea from hypothetical supposition to proven application by revealing the rampant falsification of real-world product evaluations that became evident when Amazon’s Canadian site malfunctioned. Despite the relative ease with which businesses have employed social influence strategies over the Internet, it may come as a surprise to learn that very little research has been conducted regarding social influence over the Internet in relation to perception change (Chen, 2008; Harmon, 2004; Huang & Min, 2007). Generally, the focus has been
Information Social Influence (ISI), wherein an individual will look to others to formulate an opinion when faced with contextual ambiguity (Deutsch & Gerard, 1955; Sherif, 1937). Further research in this area led to the discovery of another powerful phenomenon, called Normative Social Influence (NSI), which occurs as a consequence of peer pressure (Asch, 1955; Deutsch & Gerard, 1955). In contrast to Sherif’s (1937) research, Asch (1955) devised a social influence experiment deliberately designed to provide a clear, correct answer. In this study, participants were asked to compare two cards, one with a line of fixed length and one with three variable length line segments, to find the matching pair (Asch, 1955). Although the answer was apparent, Asch (1955) found that participants incorrectly responded in line with the majority opinion of the group comprised of confederates because of their inherent desire to fit in (Asch, 1955). This form of social influence, where peer pressure becomes the driving force, has become known as NSI in the literature (Asch, 1955; Deutsch & Gerard, 1955; Sherif, 1937). Although these two types of social influence, ISI and NSI, can coexist under the right conditions, they are in fact two different phenomena that thrive in different social contexts (Asch, 1955; Deutsch & Gerard, 1955; Keasy, Walsh, & Moran, 1969; Sherif, 1937). For instance, ISI is most effective in vague or ambiguous situations where a clear answer is either subjective or otherwise unavailable. NSI, on the other hand, does not concern itself with achieving the right answer, but rather, the desire to fit in and or avoid rejection from a group because of a contrasting opinion (Asch, 1955; Deutsch & Gerard, 1955; Keasy, Walsh, & Moran, 1969; Sherif, 1937). Perhaps more important than the
mechanical differences between the two is their influential sustainability (Asch, 1955; Deutsch & Gerard, 1955; Sherif, 1937). In regards to their long-term effects, NSI typically gains a foothold in situations where peer pressure or group identification is the primary catalyst and persists as long as that pressure is exerted, as demonstrated in Asch’s experiment (Asch, 1955; Deutsch & Gerard, 1955; Keasy, Walsh, & Moran, 1969; Sherif, 1937). In contrast, ISI can lead to a perceptual change of opinion, and for that reason, can affect the individual’s judgment indefinitely. Because ISI can be internalized, it may play a major role in shaping society’s opinions and actions in seemingly endless ways (Asch, 1955; Deutsch & Gerard, 1955; Keasy, Walsh, & Moran, 1969; Sherif, 1937). As such, it has significant implications for consumerism and the ability to generate a sustained customer base. For this reason, ISI is the focus of this paper, and hereinafter, will be discussed in the various forms and contexts in which it operates (Deutsch & Gerard, 1955; Keasy, Walsh, & Moran, 1969; Reingen, 1982). Conventional ISI. Following the discovery and distinction between ISI and NSI, efforts were made to untangle the two phenomena in subsequent studies (Deutsch & Gerard, 1955; Keasy, Walsh, & Moran, 1969). Indeed, researchers made attempts to isolate the effects of ISI in various contexts using a variety of creative constructs. For example, Keasy, Walsh and Moran (1969) developed an experiment where confederates’ responses regarding the color of a specific wavelength presented on a color wheel were made available to participants. The results conformed to the ISI model in that participants, faced with ambiguity, perceived the color on the card to be closer to the confederates’ answers. More importantly, the absence of physically present confederates
from the blocks, only 13% picked up an order card. This is in contrast to the percentages calculated with a referent individual in the waiting room. Here, 61% of participants decided to pick up an order card when the referent comments were positive and 40% when neutral (Ross, 1973). The difference between these percentages indicates that an expert’s opinion holds more influential weight than a referent individual’s with regards to child rearing advice (Ross, 1973). Likewise, Strong and Dixon’s (1971) experiment on expertise and influence revealed that an individual with expertise still had the power to persuade someone even if they were discourteous or incapable of effective communication. Participants in the experiment were engaged by one of two types of counselors regarding self-motivation perceptions: a personable counselor or a socially ineffective one. The results showed both had a substantial influence provided they were experts; however, if the counselor was thought to be a first year student filling in for the lead counselor, they were only influential if they were friendly and relatable. In contrast, the inexperienced, unpleasant counselor was considered the least influential or liked according to the post assessments. As pointed out by Strong and Dixon (1971), these results indicate that expertise can compensate for other negative characteristics when it comes to influence. With some minor caveats, the data from Strong and Dixon (1971) and Ross (1973) have demonstrated that the influential effects of ISI can be amplified by the perceived expertise of the advice giver. Although these studies are analogous to earlier conformity research performed in a general context, ISI has applications in the area of perception of certain consumer products as well (Cohen & Golden, 1972).
ISI and Consumer Products. In an attempt to determine ISI’s applicability to consumer products, Cohen and Golden (1972) designed an experiment regarding coffee brand taste testing wherein the ratings of a fabricated reference group were manipulated to see the influential effect on a participant’s rating of taste. In their experiment, participants were assigned to groups receiving one of three rating set-ups: a fabricated reference rating that was fairly uniform and skewed to the positive end of the rating scale, one with variable fabricated ratings calculated to the same mean as the previous group, or a control group in which no reference rating was provided. Additionally, groups varied on whether or not they were led to believe that their evaluations were going to be made public or kept confidential. The participants were then instructed to taste and rate the coffee product. The results demonstrated that the participant ratings were higher in the groups that received reference ratings than the control group in which no reference was provided. Additionally, their experiment detected differences between the group where the individual fabricated reference ratings were variable and the one where ratings converged on a similar point on the scale; however, these results were not statistically significant. Lastly, the data showed no significant differences between the low and high anonymity groups, indicating that potential confounds with NSI were minimal. The data from their experiment provided evidence that ISI can operate under conditions where an individual can use other opinions as a reference, but the efficacy of confederate rating unanimity is still debatable (Cohen & Golden, 1972). A similar experiment was later conducted by Burnkrant and Cousineau (1975), again within the context of coffee brand taste testing preferences. The most notable inclusion in this study was the experimenter’s focus on the effects of confederate rating
in this experiment, but this may be due to the fact that the title of “food expert,” as Burnkrant and Cousineau (1975) suggested, did not hold much weight in the eyes of the participants. Subsequent experiments on ISI and product evaluation have found similar main effects as previous studies, but more importantly, they have also started to look at how negative ratings affect an individual’s judgment (Burnkrant & Cousineau, 1975; Cohen & Golden, 1972; Pincus & Waters, 1977). In an expansion of Cohen and Golden’s (1972) earlier work, Pincus and Waters (1977) devised an experiment to determine if ISI could be generalized to include other consumer products, such as paper plates. In addition to previous ISI research, they also included a negative reference rating to their design (Pincus & Waters, 1977). In their experiment, Pincus and Waters (1977) hypothesized that participant ratings of paper plate quality would converge with group opinion, whether in a negative or positive direction, while the ratings would correlate higher when the reference opinions tended to cluster together rather than when they were dispersed. In order to test their hypothesis, they devised a between groups experiment using 170 female college students as participants (Pincus & Waters, 1977). Participants were assigned to groups in which rating type (i.e., positive and negative) and rating uniformity (i.e., variable and uniform) was manipulated. Each group was required to rate the quality of the paper plate on a scale of 1 to 15 with 1 being the lowest of quality and 15 being the highest. The results of the experiment detected that participant ratings were significantly higher when they were shown a positive reference rating than when shown a negative one, particularly when the visible confederate responses were highly uniform. The work of Pincus and Waters (1977)
provided converging evidence supporting the hypothesis that ISI affects one’s perception of product quality (Pincus & Waters, 1977). ISI, Products, and Expertise. In addition to the successful extrapolation of ISI effects on consumer products, there has also been some evidence to demonstrate that expertise enhances the ISI effect on perception of consumer products in a similar fashion as shown in earlier experiments. More recently, Lafferty, Goldsmith, and Flynn (2005) conducted an experiment to determine if an endorser’s perceived technical expertise would influence an individual’s perception of the innovativeness of technological products. More specifically, they hypothesized that those less experienced, or non- innovative, product users would be more susceptible to influence than those considered innovators, which was operationally defined as an early adopter of new technological products. To test this hypothesis, the researchers first evaluated the participant’s innovativeness using a series of questions on a 6 item assessment. Participants were then assigned to either a novice or expert endorser group where they were asked to evaluate a new cell phone advertised as the next generation mobile phone technology. In this 2 x 2 factorial experiment, both groups received the same advertisement; however, the written testimonials were identified as first time phone users for the novice group and experienced phone users for the expert group. After viewing the advertisement, the participants were asked to fill out a questionnaire which assessed their attitudes about the brand, purchasing intentions, and willingness to purchase the new phone (Lafferty, Goldsmith, & Flynn, 2005). The data provided some insightful discoveries about how expert influence works with varying levels of participant expertise on the topic at hand (Lafferty, Goldsmith, &
The interaction between ISI and expertise was further supported in Aqueveque’s research (2006) wherein a 2 X 2 X 2 factorial experiment was conducted to investigate the effects of manipulating social setting, price, and expert opinion in relation to an individual’s purchasing intentions for a particular type of wine. In the social settings manipulation, the participants were asked to imagine various settings where they might find themselves purchasing wine. For example, in one group, participants were asked to imagine that they were looking for wine that they could bring to a party. In the other group, they were asked to imagine that they were simply purchasing wine to drink by themselves at home. In the price manipulation, participants were asked to imagine that they went to the store and happened upon a cheap $3 or $9 dollar bottle of wine. Lastly, the expert manipulation consisted of having participants imagine that the particular wine was either poorly or highly recommended by wine experts. After the imaginative story, the participants were asked to fill out a questionnaire that gauged their purchasing intentions (Aqueveque, 2006). The results of this study showed that the expert opinion was the most salient and influential factor when it came to the participant’s inclination to purchase a particular wine (Aqueveque, 2006). Although this study demonstrates the efficacy of the expert influence in relation to other significant variables, the research was significantly limited in a couple of ways. First, the level of ecological validity was questionable because of the dissociative nature of the imaginative storylines. Simply put, there is too large of a difference between what actually happens and imagining what might happen in regards to consumer behavior. Second, while the expert influence is significant, it only touches on purchasing intentions and does not extend to perception change. Despite these
limitations, the research supports the body of literature regarding ISI and the enhancing effects of expert influence. However, it is prudent to look further into similar research experiments to determine if expert influence can truly manipulate an individual’s perception of a product (Aqueveque, 2006). In a related study, researchers set out to determine if a participant’s taste of wine could be influenced by the ratings of wine experts using a similar but more realistic test methodology than the previous experiment. Here, Priilaid, Feinberg, Carter, and Ross (2009) conducted an experiment in which 32 participants were asked to taste test different wines that were provided in random order. In the first trial block, the participants were asked to rate the wines blindly without any cues given except for the type of wine, which was Merlot. In the second trial block, the participants were asked to taste the wines “sighted,” meaning they were able to assess the value of the wine and were given an idea of its desirability using a wine expert’s rating guide. The 224 paired data points of blind and sighted taste test results revealed that the strength of the rater’s expertise resulted in a five-fold increase in influential strength. Additionally, demographic data collected prior to the experiment revealed that the influence of the expert became more pronounced with inexperienced wine drinkers. Conversely, there was a negative correlation between years of wine consumption and the influential effects of the expert. The results of this experiment clearly demonstrate that the effects of social influence can be amplified by expert credibility; however, the susceptibility to influence varies with wine drinking experience (Priilaid, Feinberg, Carter, & Ross, 2009). Although evidence has aptly demonstrated that ISI, along with expert influence, affects an individual’s perception of consumer products, research pertaining to the
