The Influence of Privacy Concern on Advertising Persuasion Effect

under Intelligent Recommendation

Junfeng Liao

1,2

, Yongyu Liao

, Chufang Huang

and Quzhen Zhaxi

Department of Electronic Business, South China University of Technology, Guangzhou, Guangdong, China

College of Economics and Management, Kashi University, Kashi, Xinjiang, China

School of International Relations and Public Affairs, Shanghai International Study University, Shanghai, China

Keywords: Privacy Concern, Intelligent Recommendation, Affective Response, Cognitive Response, Advertising

Persuasion Effect.

Abstract: This paper discusses the relationship and mechanism between consumers' privacy concerns, cognitive

responses and advertising persuasion effect. Based on the stimulus response (SOR) model, this paper uses

SPSS and AMOS to analyze the collected questionnaire, and concludes that consumers' privacy concerns are

negatively correlated with advertising persuasion effect under intelligent recommendation. Emotional

response (positive emotion and negative emotion) plays a mediating role between privacy concern and

advertising persuasion effect. Under intelligent recommendation, consumers' cognitive response is positively

correlated with advertising persuasion effect. There is a positive correlation between advertising persuasion

effect and attitude has a positive effect on the intention of continuous use by positively affecting the

willingness to accept.

1 INTRODUCTION

In recent years, with the rapid development of the

media driven by various high and new technologies,

the advertising industry has embraced the era of

intelligent advertising. As the future development

trend of online advertising, intelligent advertising has

received more and more attention. With its precision

and personalization of advertising, intelligent

advertising has attracted the attention of the majority

of advertising industry, which reduces the cost of

advertising and provides convenience and

personalized experience. However, the subsequent

tracking of users' network behavior data will lead to

the disclosure of users' privacy. With the

development of smart advertising and the awakening

of people's privacy awareness, an in-depth

understanding of the impact of privacy concern under

intelligent recommendation on consumer psychology

and behavior has played a vital role in the

development of smart advertising as a whole.

The research on information privacy in academia

mainly starts from three perspectives: concept

definition, privacy attitude and behavior, and

"privacy concern" is accepted by the public as an

indicator of privacy attitude. But privacy concerns of

this paper emphasizes the intelligent recommendation

of Internet privacy concerns, expanding the research

category of privacy concerns. In terms of advertising

persuasion effect, Xu (Xu, 2010) believes from the

perspective of communication that when consumers

have a certain attitude change, it is a general

persuasion. When advertising not only changes the

audience's attitude, but also leads to behavior change,

it is a meaningful persuasion. In addition, this

behavior and attitude change lasts for quite a long

time, which indicates that advertising persuasion

effect is successful. At present, the literature about

the influence of privacy concern and advertising

persuasion effect is almost blank, and the related

theories are scattered in the related research of

privacy concerns and consumers' attitudes or

behavioral willingness to behavior-oriented

advertising.

Therefore, in the context of intelligent

recommendation, this article regards consumers'

attitudes towards advertising, willingness to accept,

and willingness to continue to use advertisement as

the three progressive levels of advertisement

persuasion effect, and creatively introduces the SOR

theoretical model. Emotional response is used as an

intermediate variable to study the relationship

Liao, J., Liao, Y., Huang, C. and Zhaxi, Q.

The Inﬂuence of Privacy Concern on Advertising Persuasion Effect under Intelligent Recommendation.

DOI: 10.5220/0011189300003440

In Proceedings of the International Conference on Big Data Economy and Digital Management (BDEDM 2022), pages 515-520

ISBN: 978-989-758-593-7

515

between privacy concerns and the persuasive effect

of advertising. And targeted countermeasures are put

forward in the end according to the research results.

2 MATERIALS AND METHODS

2.1 Cognitive Response, Affective

Response and Advertising

Persuasion Effect

In this paper, positive emotion refers to the degree to

which an individual feels happy, or pleasurable when

watching the intelligent advertisement. Negative

emotion refers to the extent to which individuals feel

displeasurable or disgust watching intelligent

advertisement. As an important factor affecting

individual behavior, cognitive response has been

widely recognized, and emotional response also plays

a role that can not be ignored. SOR model believes

that consumers' emotional response will affect

consumer behavior. The cognition and emotion of

consumers constitute the basis and motivation of their

consumption intention. Pham et al. (Pham, et al,

2001) proposed that emotion, with its rapidity and

consistency, can directly guide people to think about

external stimuli, and has higher critical value

compared with cognitive evaluation of target things.

When the subject thinks that the external stimulus is

related to its needs, it will produce subjective

feelings, and then generate the motivation of behavior

intention. In online consumption, consumers' strong

emotional and cognitive responses can enhance their

consumption intentions. Therefore, it is proposed

that: H5: The positive emotion of consumers is

positively correlated with the attitude of intelligent

advertising. H6: Consumers' negative emotions are

negatively correlated with their attitudes towards

intelligent advertising. H11: The cognitive response

of consumers is positively correlated with their

attitude towards intelligent advertising.

The cognitive process of the subject to the

external stimulus should precede the emotional

response, that is the subject generates the emotional

response based on the cognitive evaluation of the

attribute and function of the object. In the online

consumption, consumers will first generate cognitive

evaluation and then form emotion in the face of all

kinds of intelligent advertisements, and the cognitive

response will enhance the emotional response. Thus,

we propose: H1: The cognitive response of

consumers is positively correlated with positive

emotions. H2: The cognitive response of consumers

is negatively correlated with negative emotions.

2.2 The Relationship between

Advertising Persuasion Effect

This paper studies the persuasion effect of intelligent

advertising, emphasizing the psychological change

process of consumers to accept intelligent

advertising. This paper draws lessons from three

levels of persuasion effect, using attitude, accepting

willingness and continuing willingness to use to

measure the persuasion effect of intelligent

advertising. For intelligent advertising, consumers'

attitudes affect their willingness to continue using

through their willingness to accept. Based on these

arguments, we propose: H7: Consumers' attitude

towards intelligent advertising is positively correlated

with their willingness to accept it. H8: Consumers'

willingness to accept intelligent advertisements is

positively correlated with their willingness to

continue using them.

2.3 Privacy Concern, Emotional

Response and Advertising

Persuasion Effect

Intelligent recommendations can capture the needs of

users more accurately, but it also brings some privacy

issues. Cockcroft and Heales (Cockcroft, Heales,

2005) proposed that consumers' privacy concerns

were negatively correlated with their transaction

intentions, with risk as a mediating variable. Eastlick

and Lotz (Eastlick, Lotz, 2006) proposed that

consumers' privacy concern was negatively

correlated with their purchase intention through trust

and loyalty, and trust and loyalty were regarded as

intermediary variables. In this study, consumers'

concern about privacy affects their thoughts and

behaviors, which indirectly affects the effectiveness

of intelligent advertising. Under any circumstances,

emotional response plays a crucial role. Therefore,

we can put forward the hypothesis: H3: Consumer

privacy concerns are negatively correlated with

positive emotions. H4: Consumer privacy concerns

are positively correlated with negative emotions. H9:

Emotional response plays a mediating role between

consumers' privacy concerns and their attitudes

towards intelligent advertising. H10: Consumer

privacy concerns are negatively correlated with

attitudes towards intelligent advertising.

To sum up, this paper proposes the following

research models:

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

516

Figure 1: Theoretical model diagram.

3 RESEARCH DESIGN

This paper adopts the Likert five-point scale, each

variable is measured with reference to the mature

scale related to it at home and abroad, and combined

with the characteristics of intelligent advertising in

this paper and the results of small-scale interviews.

The measurement of privacy concern was based on

the scales of Malhotra (Malhotra 2004) and Hong

(Hong 2013). Assessment of cognitive response was

made by reference to the scale of Shih (Shih 2013).

The evaluation of emotional responses refers to the

scale of Dholakia (Dholakia 2000). The scales of

Taylor & Todd (Taylor, Todd 1995), and Xu (Xu

2007) are the main references for the attitude of

advertising effect, and the scale of Merisavo

(Merisavo 2007) is the reference for the measurement

of acceptance intention. Intention to continue use was

measured by the scale of Bhattacherjee A

(Bhattacherjee 2001).

In this paper, a total of 500 questionnaires were

collected online, of which 419 were valid. The

effective rate of the questionnaire was 83.9%.

4 RESULTS & DISCUSSION

The Cronbach's α value of each variable in the

questionnaire scale is all greater than the standard of

0.7, which has good internal consistency and good

reliability. The KOM coefficients of all variables are

above 0.7, and the significance probability of

Bartlett's sphere test is 0.000, indicating that this

questionnaire scale has high reliability and validity

and is suitable for subsequent analysis.

4.1 Model Fitting Index

When testing the model, multiple fitting indexes

provided by SEM should be considered

comprehensively instead of relying solely on a single

index. As shown in the table below, this paper

comprehensively used a variety of fitting indicators,

and the results were as shown in the table. All

indicators reached the standard level, indicating that

the model fitted well.

Table 1: Integration of fitting index results.

4.2 Hypothesis Testing and Path

Analysis

Figure 2: Model path diagram.

We can conclude from the path coefficients in the

model path analysis that all relevant assumptions are

valid. Consumer privacy concern has a negative

impact on advertising persuasion effect. The higher

the degree of privacy concern, the worse the

advertising persuasion effect. Advertising to

convince the audience is a progressive process, under

the background of intelligent recommendation. The

persuasion process of smart advertising is to

influence the audience’s attitude towards advertising

on the first level, influence the audience’s willingness

to accept advertising on the second level, and

influence the audience’s willingness to continue

using smart advertising on the third level. Attitude

has a positive influence on willingness to continue to

use through positive influence on acceptance.

Figure 3: Relationship between cognitive response and

privacy concern and advertising persuasion effect.

The influence of privacy concerns and cognitive

responses on the persuasive effect of advertising

through emotional responses (positive emotions &

The Inﬂuence of Privacy Concern on Advertising Persuasion Effect under Intelligent Recommendation

517

negative emotions) is shown in Figure 3. It can be

found that cognitive response positively (negative)

affects positive (negative) emotions, and positively

affects the persuasive effect of advertising, while

privacy concerns negatively (positively) affect

positive (negative) emotions and negatively affect the

persuasive effect of advertising. That is, not only the

role of privacy concern and cognitive response on

advertising persuasion effect is opposite, but also the

mechanism of both through emotional response

(positive emotion & negative emotion) on advertising

persuasion effect is opposite.

4.3 Mediating Effect of Affective

Response

This article refers to the comparison method proposed

by Mayer & Davis (1999), constructs nested models

as shown in Table 2, and then conducts path analysis

for each model separately.

Table 2: Nested model settings.

Table 3: Test of main path relationship of each competition

model.

From the above results, it can be seen that the path

coefficient from privacy attention to attitude on the

basic model (M) is smaller than the non-mediation

model (MC0, MC1, MC3) and the complete

mediation model (MC2). That is, after the emotional

response is added, the relationship between privacy

attention and attitude is still significant but weakened,

indicating that emotional response plays a part of the

mediating role in the relationship between privacy

attention and attitude. Therefore, emotional response

has a mediating role between privacy concerns and

attitudes towards smart advertising. That is, the

higher the degree of privacy concern, the stronger the

negative emotions, and the worse the advertising

persuasion effect; the lower the degree of privacy

attention, the stronger the positive emotions, and the

better the advertising persuasion effect. The

hypothesis H9 is verified.

5 CONCLUSIONS AND

IMPLICATIONS

5.1 Research Conclusion

In the context of intelligent recommendation, this

paper builds a theoretical model based on the

stimulus-response theory, adds cognitive response as

another independent variable, and introduces

consumer emotional response as an intermediary

variable. After analysis of the results, the following

conclusions can be drawn for the research problems

in this paper:

First, the privacy concern has a significant

negative impact on the advertising persuasion effect,

the higher the consumer privacy concern, the worse

the advertising persuasion effect. Second, the higher

the consumer cognitive response, the better the

advertising persuasion effect and the cognitive

response has a significant positive impact on the

advertising persuasion effect. Third, positive and

negative emotions play a partial mediating role

between consumer privacy concern and advertising

persuasion effect. The higher the negative emotion of

users with high privacy concern, the worse the

advertising persuasion effect; the higher the positive

emotion of users with low privacy concern, the better

the advertising persuasion effect. Fourthly, the

influence mechanism of privacy concern and

cognition on advertising persuasion effect through

emotional response is completely opposite. Privacy

concerns affect the positive (negative) emotion in the

negative (positive) direction, and negatively affect

the persuasive effect of advertising; while the

cognitive response affects the positive (negative)

emotion in the positive (negative) direction, and

positively affect the persuasive effect of advertising.

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

518

5.2 Practical Inspiration

In order to better improve the shopping experience of

consumers, to achieve good advertising persuasion

effect, and then enhance the attraction of the website

to consumers, based on the above research results,

this paper tries to put forward the following

suggestions for shopping websites:

5.2.1 Subdivide Users with Composite

Labels

Privacy concern and cognitive response, as two

important factors affecting the persuasive effect of

intelligent advertising, can be considered as a new

label for user segmentation. Among them, users in the

(high cognitive response × low privacy concern)

group should be the most suitable users to be pushed

smart ads, while (low cognitive response × high

privacy concern) group should be the least suitable

users. Different marketing strategies should be

adopted for different groups of users. For users in the

(high cognitive response × low privacy concern)

group, they should focus on analyzing user data,

mining user preferences, and continuously

optimizing intelligent advertising. It is obviously not

a good idea to push smart ads to users in the (low-

cognition reaction × high-privacy concern) group.

Other methods should be used to help users improve

the shopping experience, such as recommending

products on the list.

5.2.2 Develop Two Parallel Lines to

Cultivate Users

Intelligent advertising is in the development stage,

and many users have no clear understanding of the

difference between it and the ordinary advertising.

Considering marketing means, word of mouth

communication, opinion leader recommendation and

friend recommendation are all helpful to form users'

cognition of intelligent advertising. Of course, in the

long run, we need to work together to create a good

industry atmosphere for the development of

intelligent advertising and e-commerce, including

enterprises' respect and practical protection for user

privacy, government and relevant departments'

supervision and safety of online privacy legislative

protection and consumers' general increased

awareness of privacy and rights protection. After all,

what really disgusts and cares about users of high

privacy concerns is not the smart advertisement itself,

but the risk of personal privacy leakage hidden

behind the smart advertisement. The control and

prevention of such risks cannot be achieved

overnight.

5.2.3 Pay Attention to the Emotional

Response of Users and Optimize the

Service in Time

At present, the after-sales evaluation of products on

online shopping websites is mainly centered on

product quality, store services and logistics. We can

consider increasing the user's evaluation of the use of

intelligent advertising (for users through intelligent

advertising transactions), analyze whether the users'

emotional response to smart advertising is positive or

negative through user evaluation, and then adjust the

service policy to find out where users are dissatisfied

with smart advertising as much as possible and make

timely improvements. This will help keep users in the

(high cognitive response × low privacy concerns)

group, which will have a positive impact on users in

other groups.

ACKNOWLEDGEMENTS

Thanks for the support from The National Social

Science Fund of China (18BGL110).

REFERENCES

Bhattacherjee A. (2001). An empirical analysis of the

antecedents of electronic commerce service

continuance. Decision Support Systems, 32(2):201-

214.

Crossler R. E. (2011). Privacy in the digital age: a review

of information privacy research in information systems.

Society for Information Management and The

Management Information Systems Research Center.

Dholakia U. M. (2000). Temptation and resistance: An

integrated model of consumption impulse formation

and enactment. Psychology & Marketing, 17(11):955-

982.

Gao Lin, Li Wenli, Ke Yulong. (2017). The Influence of

Internet Word of Mouth on Consumers' Purchasing

Intentions in Social Commerce: The Mediating Effect

of Emotional Response and the Moderating Effect of

Curiosity. Chinese Journal of Management

Engineering, 31(4): 15-25.

Pham M. T., Cohen J. B., Pracejus J. W. (2001). Affect

Monitoring and the Primacy of Feelings in Judgment.

Journal of Consumer Research, 28(2):167-188.

Xiaojuan Xu. (2010). On the Persuasion Effect of

Advertising. Journal of Beijing Business University

(Social Sciences Edition), (02):79-83.

The Inﬂuence of Privacy Concern on Advertising Persuasion Effect under Intelligent Recommendation

519

Zhang Jianqiang, Liu Juan, Zhong Weijun. (2019).

Advertising accuracy and advertising effects: field

experiments based on privacy concerns. Management

Science, 32(06): 123-132.

Zhu Qiang, Wang Xingyuan, Xin Luqi. (2018). Research

on the Influence Mechanism of Privacy Concerns on

Click Intention of Online Precise Ads——The Role of

Consumers' Risk Perception and Self-Efficacy. Soft

Science, 32(4): 115-119.

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

520