How Do Consumers Respond Differently Toward
Targeted and Non-targeted Advertisement: from a
perspective of the moderating effect of involvement levels
Shilian Xia/ 11351543
Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science
Supervisor: Anne Kranzbuhler
Date of completion: 2018/2/01
How Do Consumers Respond Differently Toward
Targeted and Non-targeted Advertisement: from a
perspective of the moderating effect of involvement levels
Abstract
This research provides a close look at the influence of targeted advertisements on the consumers’ attitudes and what roles do perceived relevance and privacy concern play, with a moderating effect of product involvements. 180 participants are recruited online to participant in a 2×2 between-subjects factorial design experiment. The results show that the targeted advertising has a positive effect on participants’ attitudes towards the advertisement. There is no effect of targeted advertising influence on consumers’ attitudes through perceived relevance and privacy concern as we assumed. And no moderating effect of product involvement on participants’ attitudes is found either. According to the results, the advertisers and marketers should adjust their advertising strategies and pay more attention to targeted advertising. And the further study should look into how to improve the accuracy of targeting consumers.
Introduction
With the development of information and digital technology, the traditional forms of advertisements, such as broadcasting, newspapers, and magazines, are progressively being replaced by online advertisements (De Keyzer, Dens, & De Pelsmacker, 2015). In 2017, revenue in the social media advertising amounts to US$32,327 million, which currently corresponds to 0.05% of the U.S’s GDP. The amount is increasing and expected to show an annual growth rate of 10.9% resulting in a market volume of US$48,917 million in 2021("Social Media Advertising," 2017). In the meantime, some of the most important and popular internet companies, such as Amazon, Google, and Spotify, are targeting their consumers by collecting and analyzing their consumers’ online profiles, online activities and preferences (Zhu & Chang, 2016). They are using collaborative filtering to determine what music, books or movies that they should recommend to their consumers online. These phenomena indicate that the influence of online advertising is important and cannot be overlooked.
Due to the booming of online advertising, some studies have been carried out to test the effect of targeted advertising on consumers’ attitudes and behaviors. They find that targeting individual consumers helps the advertisers to minimize the wasted advertising resources (Johnson, 2013) and promoting the suitable products to the consumers who are interested (Choudhary, Ghose, Mukhopadhyay, & Rajan, 2005). On the other hand, the activities of collecting data without the permission of consumers raise the consumers’ concerns about their individual privacy (Awad & Krishnan, 2006; Chellappa & Sin, 2005). When the consumers perceive that the advertisers are overstepping the boundaries of their privacy, they develop a negative
attitude towards the advertisement (Chellappa & Sin, 2005). It is important for us to have a close look at the relationship between targeted advertising and consumers’ perceived relevance and privacy concern.
Current studies are mainly focusing on the influence of targeted advertisement as a whole, but overlooking the difference of high involvement products and low involvement products. Elaboration likelihood model shed light on peoples’ change of attitude format and decision-making processes. According to the ELM, people are making the decision through two different routes: central route and peripheral route (Clarke & Belk, 1979; Traylor, 1981). Two different routes lead to two different processes of decision making. When consumers are exposed to an online advertisement of low involvement products, they would be more likely to make their decision highly depended on the peripheral cues. And when the consumers are exposed to an advertisement of high involvement products, they would be more likely to make their decision after think profoundly and carefully (Belch & Belch, 2015; Morris & Boone, 1998). These two different decision-making processes bring out a different effects on consumers’ attitudes. It is important for us to included product involvements in our study and to compare the effect of high involvement products and low involvement products.
In summary, based on the knowledge of targeted advertising and elaboration likelihood model, I examine the effect of targeted advertising as compares to non-targeted ones on consumers’ attitudes and whether these effects differ on the level of product involvement. My research question is that: How do consumers respond
moderating effect of involvement levels. This research will help the advertisers and
marketers to better understand the relationship of targeted advertising, perceived relevance and privacy concerns, and products involvement, and to adjust their advertising and promotion strategies.
Theoretical background
Targeted advertising
The concept of targeted advertising has extended from targeting by specific characteristics of a group of consumers to a particular individual based personalization, from targeting a group of people to one-to-one specific individuals (De Keyzer et al., 2015). The advertisers not only gather the geographic and socio-psychographics information of the consumers but also the information that individual consumers provided on each social network website and their online activities (Jansen, Moore and Carman, 2013). Doing so, the advertisers can precisely target one specific individual and provide s/he with specialized advertisements. In this study, I define targeted advertising as an advertising strategy that companies use to adapt and shape their advertising content based on the individual consumers’ preferences. Thus, they provide specialized, customized and individualized advertising messages to their targeted consumers.
Some studies have examined the effectiveness of targeted advertising. They found that targeted advertisements improve the effectiveness of advertising as compared to non-targeted advertising (Arora et al., 2008; De Keyzer et al., 2015; Kalyanaraman & Sundar, 2006; Plummer, Rappaport, Hall, & Barocci, 2007).
First of all, targeted advertisements can improve the consumers’ satisfaction (Arora et
al., 2008; Chen & Zhang, 2009). Targeted advertising uses consumers’ online
information and online behaviors to estimate consumers’ valuations and preferences. Based on consumers’ valuations and preference, the companies target the consumers with the advertisement that focuses on the special characteristics the consumers valued most and a finer price the consumers can afford (Choudhary, et. al., 2005). Doing so, the consumers are more likely to be satisfied by the advertisements. The consumers who are satisfied by the advertisements are more likely to have a positive attitude towards the advertisement.
Second, targeted advertisement are more likely to successfully persuade the consumers. By collecting and analyzing the consumers’ preferences and previous online behaviors, the targeted advertising use the messages that closely related to consumers’ expectations and perceived values in their advertisement. Palmgreen (1984) and Herr (1986) indicate that the consumers communicate the world base on their expectations and their perceived values, and the consumers’ expectations have an impact on their attitudes and behaviors. The messages which is related to the consumers’ expectations and values are more likely to successfully persuade the consumers.
Alternatively, targeted advertisements are closely associated with consumers’ memories. By analyzing consumers’ previous online behaviors, the advertisers target the consumers with the products that they have searched before and browsing history (Jai, Burns, & King, 2013). According to the priming theory (Tulving & Schacter, 1990), when people are exposed to a stimulus, their long term memory will influence
on their response to this stimulus. Thus, the targeted advertisements which are closely associated with consumers’ memories will bring a positive influence on the consumers’ responses to the stimulus and generate a more positive attitudes towards the advertisements.
The first hypothesis is formulated as:
H 1: Consumers will be more likely to have a positive attitude towards the
advertisement being exposed to a targeted advertisement compared to a non-targeted advertisement.
Perceived relevance
Celsi and Olson (1988) define the personal relevance of an advertisement as perceiving the individual’s needs, goals, self-knowledge and their product knowledge from the products. In other words, a message is likely to be perceived as relevant by an individual if it is seen as responding to his or her particular circumstance, life experience, or predisposition and other such characteristics (Kreuter & Wray, 2003). Characteristics of the product are associated with their dispositions and personal goals. I define perceived relevance as the level of self-related feelings of achieving goals, needs and values that the consumers perceived.
First of all, targeted advertisements has a positive effect on consumers’ perceived relevance because of consumers’ higher level of self involvement as compares to non-targeted. The targeted advertisements usually address their consumers by the consumers’ name or a second personal pronoun (Smith, 2004). According to the self-referencing theory, people tend to decode information differently depending on the
level of involvement of self in the information, and people are more likely to remember the information that is relevant to themselves (Rogers, Kuiper, & Kirker, 1977). When the consumers’ levels of involvement of self are increasing, the consumers are more likely to process and remember the information contained in the advertisements. Thus, when the consumers are exposed to a targeted advertisement that contains the information that they are familiar and related with, their perceived relevance increases (Kreuter & Wray, 2003).
Secondly, perceived relevance will lead to greater attention, greater elaboration message processing for the consumers (Bright & Daugherty, 2012; De Keyzer et al., 2015). When the consumers are exposing to the targeted advertisements, they perceived relevance and their attention and elaboration are increasing. A higher attention and elaboration help the consumers to remember the advertisement better and generate a more positive attitude towards the advertisement compared to non-targeted one. Thus, non-targeted advertising based on individual characteristics and tailored to individual preferences is an efficient way for the marketers to promoting (Kreuter & Wray, 2003; Pavlou & Stewart, 2000; Xia & Bechwati, 2008).
On the top of that, when the consumers perceive the advertisement as relevant, they are more likely to have an intention to process the information on the advertisement. According to attention and comprehension processes theory, when the consumers’ memory is activated by exposing to information containing their relevant knowledge, they will have a motivation to process the information (Celsi and Olson, 1988). A strong intention helps the consumers to process and remember the advertisements better, which leads to a positive attitude. Thus, perceived relevance plays an important
role in generating a positive effect on advertising effectiveness. It has a positive mediation effect on consumers’ attitudes.
I formulate my second hypothesis:
H 2: A targeted advertisement will increase consumers’ perceived relevance, which
leads to a positive attitude towards the product compared to the non-targeted advertisement.
Privacy concern
Smith, Milberg, and Burken (1996) developed a measurement of information privacy concerns on four dimensions: improper access to personal data, collection, errors and unauthorized secondary use. The privacy associated with the rights of people whose information is shared (Okazaki, Li, & Hirose, 2009). I define privacy concern as the
consumers’ control ability of protecting their personal information online.
In order to gain an economical access to the unlimited online resources, the online consumers are assumed willingly to give up an amount of privacy (De Corniere & De Nijs, 2016). The problem is how much efficient amount of privacy does the consumers willing to give up. Thus, the consumers are continually negotiating and managing the tension between perceived privacy concern and expected benefits (Debatin et al., 2009). Some studies found that targeted advertisements do not always have a positive effect on the consumers’ attitudes towards the advertisement. On the contrary, it has a negative effect on the consumers ’attitude towards the advertisement due to the activated privacy concerns (Awad & Krishnan, 2006; Chellappa & Sin, 2005; Jung, 2017).
Targeted advertising leads to an increasing of privacy concern(Awad & Krishnan, 2006; Chellappa & Sin, 2005; Jung, 2017). When the consumers receive targeted advertisements, they realize their information are collected without their awareness by a third party. They perceive a violation of privacy by unauthorized people, their privacy concern is activated (Jung, 2017).
Increasing of privacy concern leads to a negative effect on consumers’ attitudes. When the consumers’ privacy concern arises, the consumers tend to experience a feeling of losing control of their personal information (Jung, 2017). They are experiencing the feeling that their information is vulnerable and that they do not have the ability to control their personal information (Sheng, Nah, & Siau, 2008), which leads to a negative effect on their attitude towards the advertisements. Thus, targeted advertising leads to privacy concern, and privacy concern has a negative effect on consumers’ attitude.
I formulate my third hypothesis:
H 3: A targeted advertisement will increase consumers’ privacy concerns, which leads
to a negative attitude towards the product compared to the non-targeted advertisement.
The moderating effect of involvement levels
The elaboration likelihood model is a dual process theory that has been widely used as a framework for the study of persuasion in the field of social psychology (Petty and Cacioppo, 1977). It consists of the central route and the peripheral route and suggests
that there are different consequences resulting from the two processing routes (Bitner & Obermiller, 1985; Petty & Cacioppo, 1986). This theory has been used to examine the effectiveness of advertisements. It predicts the change of attitude forms and also has been successfully used to predict the consumers’ attitude towards the products (Trampe et al., 2010)
According to the model, when the elaboration likelihood is low, the consumers will be more likely to be persuaded by peripheral cues. Whereas the elaboration likelihood is high, the consumers will be more likely to be persuaded by reasonable argumentations. Exposing to a highly involved product generates more elaboration and cognitive processing (Petty and Cacioppo, 1977). When the consumers are exposed to high involvement products, they tend to elaborate the messages conveyed by the advertisement more carefully and thoughtfully. The targeted advertisement which contains the information that associated with the consumers will be interpreted as a persuasive and reasonable argumentation, and the consumers will be more likely to accept it as well (Trampe et al., 2010). And when the consumers are exposed to low involvement products, a less elaboration cognitive processing is generated. The consumers make their decision merely relying on the peripheral cues, the effect of targeting is weakened.
Thus, I formulate my third hypothesis:
H 4: The positive effect of targeted advertisement on consumers’ attitude towards the
advertisement of high involvement products are stronger, compared to the effect of targeted advertisement of low involvement products.
Diagram 1: Conceptual model
Method
Experiment
An experiment with a 2×2 between-subjects factorial design: 2 (between-subjects factor targeted advertisement vs. non targeted advertisement) × 2 (between-subjects factor ‘product involvements’: high involvement vs. Low involvement) was conducted. Doing so, the effect of targeted and non-targeted and high involvement and low involvement on participants’ attitude towards the advertisement can be isolated and analyzed separately.
To operationalize the “targeted and non-targeted”, I use participants’ educational backgrounds as the targeting manipulation, because educational background information is that the information of educational background is objective and it is convenient to obtain. The participants who are in one of the five most popular majors: computer science, mathematics, law, social science, and accountancy, are randomly assigned to four conditions (As in table 2.). The participants who are in other majors are randomly assigned to two conditions (condition 3 or condition 4). The targeted
Perceived relevance Targeted/ Non-targeted Attitude s Privacy concerns Low/high involvement
group received an advertisement contains the message that related to their major. And the non-targeted group received an advertisement not targeted specific major.
Table 2: 2*2 between-subjects factorial design factorial design
Between-subjects
Targeted advertisement
Non-targeted advertisement
High involvement product 1 3
Low involvement product 2 4
To operationalize the “high involvement and low involvement”, I use the laptop as the high involvement product and notepad as the low involvement product. They are typical high involvement and low involvement products. High involvement products have the characteristics of high prices and complex technical features, and the purchase behaviors of high involvement products require a long and careful consideration (Jiang et al., 2015). The laptop conforms to those characteristics. Whereas low involvement product has the characteristics of low prices and simple features, and the purchase behaviors of low involvement products are lack of the motivation to elaborate (Furbeck & Sjödin, 2017). The notepad conforms to those characteristics.
The experiment consists of 3 sections: (1) First, the participants were reached out online. They were asked to fill out a questionnaire. The questionnaire begins with a short introduction and a set of questions about their demographic information. Their educational background information are used as the targeting manipulation; (2) Second, the participants were randomly assigned to one of four conditions. There are
total 12 experimental material. The experimental material for each major are specially designed (see the Appendix). (3) Last, participants were asked to fill out the rest questions which measure their attitudes, perceived relevance, and privacy concerns. The experiment took about 5 minutes.
Measurement
The dependent variable is the participants’ attitude towards the advertisement. It is measured on a 7-point Likert scale using item 1: How do you like this product; item 2: Are you satisfied with the picture; and item 3: How likely are you going to purchase the product in the picture, adapted from attitude component models and Fort-Rioche and Ackermann (2013). The mediator variable is perceived relevance. It is measured on a 7-point Likert scale using item 4: I thought these personal ads fit my interest/preference/tastes, item 5: I was connected to the products, item 6: I thought these ads were made for me, from Clarke and Belk (1979) and Zhu and Chang (2016). And the other mediator variable is privacy concern. It is measured on a 7-point Likert scale using item 7: It would bother me if my information could be collected by a third-party company via the internet without my permission, item 8: It would bother me if a third-party company knew too much information about me, and item 9: It would bother me if my information could be used by a third-party company in ways I could not foresee, adapted from Sheng, Nah, and Siau (2008).
Table 3: the results of PCA
Source Items Cronbach’s
alpha
Loading
1 2 3
Attitude
component models and Fort-Rioche and Ackermann (2013)
1. How do you like this product?
1.00
2. Are you satisfied with the picture? (deleted)
0.87
3. How likely are you going to purchase the product in the picture? (deleted)
0.87
Perceived relevance Clarke and Belk (1979) and Zhu and Chang (2016)
1. I thought these personal ads fit my interest/preference/tastes.
0.86 0.87
2. I was connected to the products?
0.85
3. I thought these ads were made for me.
0.75
Privacy concern Sheng, Nah, and Siau (2008).
1. It would bother me if my information could be collected by a third-party company via the internet without my permission.
0.89 0.95
2. It would bother me if a third-party company knew too much information about me.
0.91
3. It would bother me if my information could be used by a third-party company in ways I could not foresee.
0.85
An orthogonal rotation principle component analysis (PCA) is conducted. As we can see from the table 1. There are three component are extracted: attitude, perceived relevance and privacy concern. Item 4, item 5 and item 6 correlate positively with the component 2. Item 4 “I thought these personal ads fit my interest/preference/tastes”
has the strongest association (factor loading is .87 ). And the reliability of the scales is good and reliable (Cronbach’s alpha = .86). Item 7, item 8 and item 9 correlate positively with the component 3. Item 7 “It would bother me if my information could be collected by a third-party company via the internet without my permission” has the strongest association (factor loading is .95). Reliability of the scales is good and reliable (Cronbach’s alpha = .89) as well. Together, they explain 83.38% of the variance in the original items. The participants’ attitude was measured by item 1 (M = 4.10, SD = 2.12). Participants’ perceived relevance was measured by item 4, item 5 and item 6 together (M= 3.62, SD = 1.75), and participants’ privacy concern was measured by item 7, item 8 and item 9 together (M = 5.21, SD = 1.91).
Result
Sampling
180 participants were recruited via personal approach online. 33 participants took English version questionnaire, and 147 took Chinese version. 52 of them are males and 128 are females, with a mean age of 30.27 (SD = 10.37). Participants’ educational backgrounds range from less than high school degree to doctoral degree. 2.2% of the participants have less than high school degree. 9.4% of them have a high school graduate, 49.2% of them have a bachelor degree, and 12.2% of them have a master degree. Most participants are major in social science (29.8%) and accountancy (11%). 3.9% of them major in computer science, 2.8% of them major in law, and 50.8% of them major in others.
Effect of targeted and non-targeted advertisement on participants’ attitude
To test H1, I use participants’ attitude as the dependent variable. And targeted and non-targeted is the independent variable. An independent sample T-test was conducted to compare the effect of targeted advertisement (N = 59, M = 4.56, SD = 1.94) to non-targeted advertisement (N = 117, M = 3.87, SD = 2.18) on the participants’ attitude towards the product. The result shows that t (174) = 2.05, p = .042, 95% CI [.25, 1.35]. The result shows that the targeted group (M = 4.56, SD = 1.94) has a more positive attitude towards the advertisements than the non-targeted (M = 3.87, SD = 2.18), and the difference between them is statistically significant. Thus, people are more likely to have a positive attitude towards the advertisement after being exposed to a targeted advertisement compared to a non-targeted advertisement. H1 was accepted.
Mediation effect of perceived relevance
According to Baron and Kenny (1986) and Hayes Andrew (2013), when the effect of the independent variable transit through a third variable, the third variable is the mediator variable. The mediation can be tested by regression analysis. In the experiment, perceived relevance and privacy concern are the parallel mediation variables.
To test H2 and H3, I first conducted a simple regression analysis to test the relationship between targeted and non-targeted and participants’ attitude. The result of simple regression shows that the model as a whole is significant, F (1, 174) = 4.19, p = .042. The regression model can be used to predict the participants’ attitude. And 2 percent of the variance in participants’ attitude can be predicted by the variable
targeted and non-targeted. A statistically significant difference is found between targeted and non-targeted. (R2 = .02). b* = .15, t = 2.05, p < .042, 95% CI [1.35, -.03]. The participants’ attitude of non-targeted was .69 more negative than targeted group.
Secondly, I conducted a simple regression analysis to test the relationship between targeted and non-targeted and participants’ perceived relevance. The result shows that the model as a whole is not significant, F (1, 174) = .29, p = .591. The regression model cannot be used to predict the participants’ perceived relevance. Zero percent of the variance in participants’ perceived relevance can be predicted by the targeted and non-targeted (R2 = .00).
Next, a simple regression analysis was conducted to test the relationship between perceived relevance and participants’ attitude. The result shows that the model as a whole is not significant, F (1, 174) = .06, p = .804. The regression model cannot be used to predict the participants’ attitude. Zero percent of the variation in participants’ attitude can be predicted by participants’ perceived relevance.
Lastly, a multiple regression was conducted to test the relationship between targeted and non-targeted, perceived relevance and participants’ attitude. The results of multiple regression show that the model as a whole is not significant, F (1,173) = 2.14, p = .121. Perceived relevance does not mediate the effect of targeted and non-targeted advertisements on participants attitude. Only 2 percent of variation in participants’ attitude are explained by targeted and non-targeted. And the effect do not transit through perceived relevance. People exposed to a targeted advertisement do
not increase their perceived relevance and do not leads to a positive attitude towards the product compared to the non-targeted advertisement. H2 was rejected.
Mediation effect of privacy concern
Similarly, a simple regression was conducted to test whether the targeted and non-targeted and privacy concern predicting participants’ attitude. The regression model shows that the privacy concern do not have an effect on participants’ attitudes. F (1,174) = 1.60, p = .207. Only one percent of the variance in participants’ privacy concern can be explained by the targeted and non-targeted (R2 = .01).
Secondly, a simple regression was conducted to test the privacy concern predicting participants’ attitude. The result shows that F (1,174) = .540, p = .462. Zero percent of the variance in participants’ attitude can be predicted on privacy concern (R2 = .00). A multiple regression was conducted to test the relationship between participants’ attitude, targeted and non-targeted and privacy concern. The results show that F (2,173) = 2.23, p = .110. Only 3 percent of the variance in participants’ attitude is predicted by the targeted and non-targeted (R2 = .03), not by privacy concern. Thus, privacy concern cannot be used to predict participants attitude. And privacy concern and targeted and non-targeted together cannot be used to predict participants attitude. People exposed to a targeted advertisement do not increase their privacy concern and do not leads to a negative attitude towards the product compared to the non-targeted advertisement. H3 was rejected.
Moderation of products involvement
independent variable on the dependent variable differ according to the level of the third variable, moderation occurs. Moderation is usually tested with analysis of variance. Therefore, to test whether the effect of an exposure to a targeted or non-targeted ad depends on the level of product involvement, a two-way ANOVA test was conducted with targeted and non-targeted and product involvements as the independent variables and attitude as the dependent variable.
The result shows that a significant effect was found among the participants’ exposure to the targeted and non-targeted advertisement on the participants’ attitudes towards the advertisement, F (1, 172) = 4.26, p = .040, η2 = .02. But the effect size is small. The participants who are exposed to the targeted advertisement (M = 4.57, SD = 1.94) appear to have a more positive attitude towards the advertisement than the participants who are exposed to non-targeted advertisement (M = 3.87, SD = 2.18). There is no significant effect of products involvement, F (1, 172) = 1.12, p = .291, η2 =.01. The participants who are exposed to the high involvement (M = 4.04, SD = 2.03) and the participants who are exposed to low involvement (M = 4.39, SD = 1.66) appears to have the same attitude towards the advertisement. No significant interaction effect between products involvement and exposure to targeted and non-targeted advertisement was found either, F (1, 172) = 1.25, p =.266, η2 = .01. Product involvements do not have an influence on the strength of the relationship between targeted and non-targeted advertisement and participants’ attitude towards the advertisement. People are more likely to have a positive attitude towards the advertisement after being exposed to a targeted advertisement compared to a non-targeted advertisement, but the effect is not more prominent for low involvement product than high involvement products. H4 was rejected.
Discussion
The results offer several theoretical and practice implications on targeted advertising. First, it provides first-hand evidence of the effectiveness of targeted advertising. According to significant result of hypothesis 1, targeted advertising play an important role in influencing people’s attitude towards the advertisement. The participants on average hold a more positive attitude towards the targeted advertisement than non-targeted, and the difference between them is statistically significant. Second, it looks into the relationship between targeted advertising and perceived relevance and privacy concern. Targeted advertising does bring a positive influence on consumers’ attitude towards the advertisement, but the effects cannot be explained by perceived relevance, privacy concern. Targeted advertising does not increase the participants’ perceived relevance, and participants’ perceived relevance do not increase their positive attitude towards the advertisement. And targeted advertising does not always lead to a higher privacy concern, while privacy concern does not lead to a more negative consumers’ attitudes. Third, it attempts to apply elaboration likelihood model to targeted advertising. But the product involvements do not have the effect on consumers’ attitudes as we assumed. And the effect of targeted versus non-targeted advertisements does not differ for low as compared to high involvement products. The effect of product involvements remains in doubt.
There are three limitations in my study. First, only one targeting manipulation is used. In the experiment, I use the educational background information as the targeting manipulation. Some participants who are in targeted condition may be not interested
in or perceive relevant from their major. Thus, we should take other traits, such as the gender and the age generation, into considerations.
Second, the sampling is small. A larger sampling size can decrease the margin of error. In the experiments, there are 29 participants are targeted by high involvement products and 30 participants are targeted by low involvement products. The sampling size is tolerated to conduct variance test, but a small sampling decreases the accuracy of the results. Thus, with more adequate resources, a large size of sampling can increase the accuracy of the experiment results.
And last, the targeted experiment material is visually more attractive than the non-targeted one. The product presented in the non-non-targeted advertisement is not designed, while the product presented in the targeted advertisement is designed with targeting message in stylish letters or characters. The visual difference rather than being targeted also may lead to a more positive attitude.
For future study, we could look into how to improve the effectiveness of targeting the consumers. By using more targeting manipulations, and combining two or more manipulation together, we can see whether it improves the accuracy of targeting. For example, targeting the consumers by geographic locations is a good way to improve the accuracy of targeting. Doing so, the advertisers and marketers could recommend the closest stores and restaurants to their consumers. It helps the advertisers and marketers to save resources and increase profits. Additionally, if there are adequate resources the sampling size should be enlarged.
Reference
Arora, N., Dreze, X., Ghose, A., Hess, J. D., Iyengar, R., Jing, B., . . . Neslin, S. (2008). Putting one-to-one marketing to work: Personalization, customization, and choice. Marketing Letters, 19(3-4), 305.
Awad, N. F., & Krishnan, M. S. (2006). The personalization privacy paradox: an empirical evaluation of information transparency and the willingness to be profiled online for personalization. MIS quarterly, 13-28.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of personality and social psychology, 51(6), 1173. Belch, G. and Belch, M. (2015). Advertising and promotion. McGraw-Hill Education. Bright, L. F., & Daugherty, T. (2012). Does customization impact advertising
effectiveness? An exploratory study of consumer perceptions of advertising in customized online environments. Journal of Marketing Communications,
18(1), 19-37.
Celsi, R. L., & Olson, J. C. (1988). The role of involvement in attention and comprehension processes. Journal of consumer research, 15(2), 210-224. Chellappa, R. K., & Sin, R. G. (2005). Personalization versus privacy: An empirical
examination of the online consumer’s dilemma. Information technology and
management, 6(2), 181-202.
Clarke, K., & Belk, R. W. (1979). The effects of product involvement and task definition on anticipated consumer effort. NA-Advances in Consumer
De Corniere, A., & De Nijs, R. (2016). Online advertising and privacy. The RAND
Journal of Economics, 47(1), 48-72.
De Keyzer, F., Dens, N., & De Pelsmacker, P. (2015). Is this for me? How consumers respond to personalized advertising on social network sites. Journal of
Interactive Advertising, 15(2), 124-134.
Edwards, J. R., & Lambert, L. S. (2007). Methods for integrating moderation and mediation: a general analytical framework using moderated path analysis.
Psychological methods, 12(1), 1.
Furbeck, J., & Sjödin, S. (2017). Is frozen the new fresh?: An observational study of low-involvement product choices.
Jai, T. M. C., Burns, L. D., & King, N. J. (2013). The effect of behavioral tracking practices on consumers' shopping evaluations and repurchase intention toward trusted online retailers. Computers in Human Behavior, 29(3), 901-909.
Jansen, B. J., Moore, K., & Carman, S. (2013). Evaluating the performance of demographic targeting using gender in sponsored search. Information Processing & Management, 49(1), 286-302.
Jiang, C., Duan, R., Jain, H. K., Liu, S., & Liang, K. (2015). Hybrid collaborative filtering for high-involvement products: a solution to opinion sparsity and. dynamics. Decision Support Systems, 79, 195-208.
Johnson, J. P. (2013). Targeted advertising and advertising avoidance. The RAND
Journal of Economics, 44(1), 128-144.
Jung, A.-R. (2017). The influence of perceived ad relevance on social media advertising: An empirical examination of a mediating role of privacy concern.
Hayes, A. F. (2013). Mediation, moderation, and conditional process analysis. Introduction to Mediation, Moderation, and Conditional Process
Analysis: A Regression-Based Approach edn. New York: Guilford
Publications, 1-20.
Kalyanaraman, S., & Sundar, S. S. (2006). The psychological appeal of personalized content in web portals: does customization affect attitudes and behavior?
Journal of Communication, 56(1), 110-132.
Morris, J. D., & Boone, M. A. (1998). The effects of music on emotional response, brand attitude, and purchase intent in an emotional advertising condition. ACR
North American Advances.
Pavlou, P. A., & Stewart, D. W. (2000). Measuring the effects and effectiveness of interactive advertising: A research agenda. Journal of Interactive Advertising, 61-77.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. experimental social psychology, 19, 123-205.
Petty, R. and Cacioppo, J. (1977). Forewarning, cognitive responding, and resistance to persuasion. Journal of Personality and Social Psychology, 35(9), pp.645-655.
Plummer, J., Rappaport, S. D., Hall, T., & Barocci, R. (2007). The online advertising
playbook: proven strategies and tested tactics from the advertising research foundation: John Wiley & Sons.
Sheng, H., Nah, F. F.-H., & Siau, K. (2008). An experimental study on ubiquitous commerce adoption: Impact of personalization and privacy concerns. Journal
Smith, K. (2004). I am me, but who are you and what are we?': The translation of personal pronouns and possessive determiners in advertising texts.
Multilingua, 23(3), 283-304.
Social Media Advertising. (2017).
Trampe, D., Stapel, D. A., Siero, F. W., & Mulder, H. (2010). Beauty as a tool: The effect of model attractiveness, product relevance, and elaboration likelihood on advertising effectiveness. Psychology & Marketing, 27(12), 1101-1121. Tulving, E., & Schacter, D. L. (1990). Priming and human memory systems. Science,
Appendix
1. Non-targeted high/low involvement product advertisement:
1. Targeted high/low involvement product advertisement for math major:
2. Targeted high/low involvement product advertisement for accountancy major:
3. Targeted high/low involvement product advertisement for social science major:
4. Targeted high/low involvement product advertisement for law major:
