If you reply to me, I will buy from you: A social influence examination

If you reply to me, I will buy from you: A social

influence examination of reciprocity on Twitter

⋆

Rosanna E. Guadagno

1,†

, Amanda Sardos

2,∗,†

and Amanda M. Kimbrough

3,†

University of Oulu, Oulu, Finland

University of Texas at Dallas, 800 W. Campbell Road, Richardson, TX 75080-3021, USA

Instagram, Dallas, TX, USA

Abstract

This research examines corporate communication practices on social media, focusing on potential

customers’ intentions to patronize a corporate brand after contacting the brand’s Twitter account.

Specifically, across two studies, participants reported their intentions to patronize one of two

restaurant chains after the corporate Twitter account responded in one of three ways to their

hypothetical message: direct reply, retweet, or did not respond. Based on Guadagno’s [11] model of

social influence online, we predicted that people would be more likely to patronize a restaurant that

responded to the tweet -- either via retweet or reply -- owing to the norm of reciprocity. Across both

studies, results indicated that participants reported stronger intentions to patronize the restaurant

after a Twitter interaction. Furthermore, women generally reported higher purchase intentions and

sensitivity to the different restaurant response conditions than did men. Thus, as predicted,

reciprocity in social media interactions between people and businesses is effective in influencing

people to patronize the business. Implications for user behavior change will be discussed.

Keywords

Reciprocity, social influence, social media, Twitter, gender, user engagement 1

1. Introduction

In recent years, social media (e.g., Facebook, Twitter/X, Instagram) has become the primary way

for people to communicate online. Social media sites, or social networking sites (SNS), as they

are often referred to, have been operationally defined by scholars as: “Web-based services that

allow individuals to (1) construct a public or semi-public profile within a bounded system, (2)

articulate a list of other users with whom they share a connection, and (3) view and traverse

their list of connections and those made by others within the system”[1]. Social media use has

been steadily growing globally over the past ten years. In 2005, only seven percent of American

adults used social media; ten years later, this number rose to sixty-five percent, an almost a ten-

fold increase. Among American adults under thirty years old, social media use is now almost

ubiquitous with ninety percent reporting active use on at least one platform. Furthermore,

BCSS 2024: The 12

International Workshop on Behavior Change Support Systems, April 10, 2024, Wollongong,

Australia.

∗

Corresponding author.

†

These authors contributed equally.

rosanna.guadagno@oulu.fi (R. E. Guadagno); [email protected] (A. Sardos); amkalbrigh[email protected]om (A. E.

Kimbrough)

0000-0001-8247-5154 (R. E. Guadagno)

CEUR

Workshop

Proceedings

ceur-ws.org

ISSN 1613-0073

social media use comprises about twenty percent of the total time Americans adults spend on

the Internet [20]. Thus, it is safe to state that social media use is extremely popular in

contemporary society.

Given the popularity of social media, scholars have begun to examine the ways in which

communication via social media affects social relationships e.g., [24]. While much of the extant

literature examines the implications of social media use for interpersonal relationships, there is

a paucity of research examining the relationships between consumers and businesses. Many

brands have a presence on social media that ties into their actual products. Indeed, during a trip

to the grocery store, most people encounter popular social media icons on product labels as well

as ads for tie ins between the physical product and the corporate social media presence. The

present investigation seeks to fill this gap by examining whether interacting with corporate

social media accounts via Twitter will increase people’s intentions to patronize the business.

1.1. Social influence online

The present investigation focuses on reciprocity as an influence tactic. Reciprocity has proven

successful in various offline contexts because returning favors, invitations, or gifts is ingrained

as a universal social norm [9]. Thus, the norm of reciprocity, which is also referred to as the

Golden Rule, indicates that people who give others favors have the reasonable expectation that

those favors will be returned. This is because when someone does another person a favor, the

person receiving the favor feels indebted to the favor-doer until they can return the favor.

Although the literature on compliance gaining in an online context is still somewhat limited,

results suggest that influence tactics that play on internal processes within the individual, rather

than those that rely on interpersonal processes will be more successful (see [13, 15, 16] for

reviews). Guadagno and colleagues argue that this is due to the primarily to the decreased

salience of the other person in many forms of online interaction.

Within the context of a brand-consumer exchange, reciprocity remains a relational or a give-

and-take process; the brand can give the consumer information or a special deal, and the

consumer can reciprocate that interaction by purchasing their product or spreading their

information to other consumers. Additionally, this can be seen in the ways consumers share

advice, information, and ideas from each other online and then reciprocating these shared

resources [3].

Ewell, Minney, and Guadagno [7] reviewed the literature on reciprocity online and found

that few studies have examined social influence with reciprocity in online contexts. For

instance, Eastwick and Gardner [6] examined whether people would comply with a requester

who used the reciprocity-based door-in-the-face (DITF) technique in a virtual world. With the

DITF, a requester initially asks their target for an exceedingly large request, expecting it to be

rejected, creating a situation in which retreating to a smaller request is perceived by the target

as a concession on the part of the requester. This invokes reciprocity and, compared to no large

request controls, generally increases compliance to the smaller, target request. Consistent with

the general research in this area, the author found a significant DITF effect, but only when the

avatar appeared to be White. When the avatar appeared to be Black, the DITF was unsuccessful

in the virtual world. Other scholarship suggests that reciprocity affects people’s social media

news consumption such that people with stronger beliefs in reciprocity consume more news

shared on social media [17] . Thus, while the evidence is limited, it appears that reciprocity in

communication can be effective in certain online contexts.

1.2. Gender differences in technology use

Human social behavior is often impacted by the expectations people have for others. This is

evident in the way reciprocity functions. However, societal expectations for behavior can also

differ between men and women in both on and offline contexts. Social role theory postulates

that men and women occupy different roles in society [5]. According to social role theory, the

different roles men and women fulfil in society produce gendered beliefs about the skills and

characteristics of men and women. Specifically, according to social role theory, perceivers

expect men to engage in more agentic behaviors such as assertiveness, competitiveness, and

independence. Conversely, perceivers expect women to be more communal in their conduct as

evidenced by a greater focus on relational goals and interpersonal processes. Not only are these

gender differences expected, research suggests that men and women conduct themselves in

accordance with these gender role expectations. These gendered differences in behavior have

been shown to translate to online environments as well, with women spending more time

maintaining interpersonal relationships online and men engaging in more independent and

competitive behaviors. For instance, Muscanell and Guadagno [22] found that women were

more likely to use social networking sites for maintaining existing relationships than were men.

Instead, men reported engaging in more agentic behavior in their social media use as evidenced

by more time spent establishing new romantic and platonic relationships and seeking job

opportunities. Other research has shown that women use mediated communication more

broadly defined (e.g., across social networking, texting, e-mail) more than do men [19].

Indeed, this tendency for women to adopt and use mediated channels of communication to

maintain social ties can be seen in the early days of communication on the Internet. For instance,

while both women and men report valuing email for its convenience and efficiency at equal

rates, women report greater satisfaction with email in their daily life and that have indicated

that communication via email is an important and meaningful aspect of their lives [8]. These

differences in communication satisfaction and importance can be traced to women’s greater use

of email for maintaining close social bonds with friends and family, whereas men were more

likely to use email as a source for seeking information [8, 19].

Across three studies, Guadagno and Cialdini [12, 14] investigated whether persuasion via

email differed significantly from face-to-face interactions. This line of research examined

people’s change in attitude towards undergraduate comprehensive exams – a popular topic of

persuasion, largely because it is so unpopular to students -- after hearing a series of persuasive

messages from a same-sex communicator via either an email or face-to-face interaction. This

research was framed by social role theory [5] and predicted that women, focused on establishing

connection would be more persuaded when the communicator was more salient (i.e., FtF) and

similar to them, while men would be focused on establishing agency and would not vary in

persuasion by communication mode. As expected, the results of their first study revealed that

women were significantly less persuaded when a same-sex confederate conveyed the persuasive

messages via email than via face-to-face. Men, on the other hand, did not differ in persuasion

as a function of the communication channel. Their second study replicated and expanded upon

this work by demonstrating that interacting with the confederate prior to the persuasive

exchange varied the outcome differentially for men versus women. Specifically, for women, any

prior interaction regardless of the nature (i.e., competitive vs. cooperative) attenuated the effect

of email on persuasion such that women reported the most negative attitudes toward the

comprehensive exams. This was not the case for male participants who instead reported the

most negative attitudes toward the exams when they had a competitive prior interaction with

the confederate before he attempted to persuade then in person. Study 3 replicated and

expanded on these findings by examining the impact of perceived similarity on persuasion.

Results indicated that participants who perceived themselves to be highly similar to the

confederate were the most likely to be persuaded relative to those participants who did not

receive any similar feedback or were informed that they were highly dissimilar to the

communicator. Again, the results varied by participant gender such that male participants

evaluated the persuasive message on its merits when interacting via email, while women

participants were prone to reject entreaties from dissimilar others.

Given that this previous research demonstrated women’s greater focus on communal goals

in the context of online persuasion, we suggest that they may be more susceptible to reciprocity-

based social influence appeals are more likely to use, prefer, be more satisfied with, and report

greater importance of mediated communication channels in their lives, this could lead to them

being more persuaded by or compliant with requests presented via these channels, such as those

posed by brands online.

1.3. The present investigation

In two studies, we examined whether reciprocity affects purchase intentions after a

hypothetical interaction between corporate twitter accounts and consumers on Twitter.

Specifically, we examined how type of interaction with a business (reply, retweet, or no

response) affected people’s intentions to patronize a restaurant. We chose Twitter over other

social media platforms because it is a microblogging site and features more informational posts

and business to consumer messaging than social media sites focused on social networking such

as Facebook [10].

Based upon the relevant social influence on social media theoretical frameworks reviewed

above e.g.,[11], we predicted that the greater the response from a corporate twitter account to

a potential customer would increase intentions to patronize the brand. In addition, owing to

social role theory [5], we expected gender differences such that women would be more sensitive

to the different response types and therefore show greater intentions to patronize the brand

when the brand reciprocated the participant’s tweet with a personalized tweet in response. We

examined this first with an undergraduate sample (Study 1) and then again with an online

sample from Amazon’s Mturk (Study 2) in part to increase generalization of our results and in

part to increase the sample of men in our study. Other than this, there were no other differences

between the studies.

Specifically, our hypotheses are as follows:

H1: When brands interact with consumers through a reply or retweet, consumers will be

more likely to patronize the company.

H2: Women will be more likely than men to visit the company after seeing brands engage

with consumers on Twitter.

RQ1: Since the literature on the influence of reciprocity in interactions between businesses

and consumers is limited, we wanted to examine whether consumers perceived retweets and

replies similarly as they are both reciprocal forms of communication. Thus, this research

question is aimed at understanding whether retweets and responses are equally persuasive

when used in the same context.

2. Study 1

In this initial study, undergraduate students were recruited to test our hypothesis.

2.1. Participants

Participants were 139 undergraduate students (113 women, 26 men) from midsized

southwestern university who received partial course credit for their participation. Participants

ranged in age from 18-42 (M= 21.3, SD = 3.29). Self-reported ethnicity was as follows: Caucasian

54 (38.8%), African American 6 (4.3%), Asian 46 (33.1%), Pacific Islander 2 (1.4%), Hispanic 21

(15.1%), and Other 10 (7.2%). Additionally, almost half (47.5%) of our sample reported using

social media over 4 hours a week.

Table 1

Participants Social Media Use per Week in Study 1

Time Per Week Frequency Percent Cumulative Percent

0-1 hrs 11 7.9 7.9

1-2 hrs 24 17.3 25.2

3-4 hrs 38 27.3 52.5

4 or more hrs 66 47.5 100

Total 139 100

2.2. Design and measures

Design. The experimental design was a 3 (restaurant’s social media response: retweet vs. reply

vs. none) X 2 (participant gender: men vs. women) between subjects factorial design. The

restaurant’s social media response options were as follows: replying to the consumer’s tweet,

retweeting the consumer’s tweet, or not responding to the consumer at all. Participants were

randomly assigned to one of these three conditions by Qualtrics, the survey software used to

collect the data. To increase generalizability of our findings, stimulus materials were created for

two different restaurants with participants randomly assigned to view one: Chili’s or Gloria’s.

See Figure 1 for example images of the stimulus materials for each restaurant.

Measures: To assess users’ behavioral intentions with a company after interacting with

them on Twitter, a scale was created using three author-generated items, the reliability of which

was quite high, Chronbach’s α = .89: (1) “How likely are you to visit this company in person?”

(2) “How likely are you to buy this company’s product?” and (3) “How likely are you to

recommend this company?” Responses were measured using a 7-point Likert scale ranging from

(1= not important at all; 7=very important).

Additionally, before seeing the stimulus materials or rating their behavioral intentions,

participants were asked questions about their overall social media use in order to get a sense of

how important brand interaction over social media would be. Participants were asked how

much they used social media in a given week and were asked to choose the answer that best

reflected their general habits from 0-1 hours, 1-2 hours, 3-4 hours, or 4 or more hours. Finally,

after seeing the stimulus materials and rating their behavioral intentions, participants were

specifically asked to rate how important it is that a brand engages with them on twitter on a 1-

7 Likert scale from (1) not at all important to (7) extremely important.

2.3. Procedure

Participants in this study first answered survey questions pertaining to their overall social

media use to get a sense of the importance social media plays in their everyday lives. Then

participants were asked to imagine that they sent a tweet to a restaurant while planning an

evening out with friends. They were then presented with one of the three tweet conditions

(response, retweet, or no response) from either of the two restaurants (Chili’s or Gloria’s). After

reading the stimulus material, the participant was asked to rate, given their imagined

interaction with the company, their likelihood to visit the company, buy the company’s

products, and recommend this company to others. Finally, participants were asked to rate the

importance of brand engagement on Twitter as well as complete a series of demographic

measures.

2.4. Results and discussion

Data were analyzed using IBM’s SPSS. To ensure that there were no discernable differences

between the two restaurants, we first conducted an ANOVA to compare behavioral intentions

across the type of restaurant. As anticipated, this analysis failed to reveal a significant difference

between Chili’s (n = 70; M = 4.94, SD = 1.5) and Gloria’s (n = 69; M = 4.65, SD = 1.5);

F(1,137)=1.23, p=.269, ηp2 = .009, so for the remaining analyses, participants’ responses were

collapsed across the type of restaurant.

When asked how much they engaged with social media in a given week almost half (47.5%)

of our participants indicated that they use social media more four hours per week (3-4 hours –

27.3%, 1-2 hours – 17.3%, 0-1 hours – 7.9%) indicating that social media played a prominent role

in their weekly activities. However, on a 1-7 scale of how important was it that a brand engaged

with them on Twitter, participants rated it as low importance (M = 2.95; SD = 1.79).

Next, to test H1 and H2, we conducted a 3 (restaurant’s social media response: retweet vs.

reply vs. none) X 2 (participant gender: men vs. women) ANOVA on individual’s behavioral

intentions to patronize the restaurant. Failing to support H1, there was no significant difference

for restaurant response [F(2, 139) = .433, p = .649, ηp2 = .017]. Nor was there a significant main

effect for participant gender [F(1, 139) = 2.34, p = .128, ηp2 = .006]. However, in support of H2,

the predicted interaction between the restaurant’s response type and participant gender was

marginally significant, F(2, 139) = 2.61, p = .077, ηp2 = .038. Planned pairwise comparisons

revealed that women had stronger behavioral intentions to patronize the restaurant after a reply

(M = 5.57, SD = 1.31) than after a retweet (M = 4.79, SD = 1.54, p = .024) or no response (M =

4.36, SD = 1.57, p < .001). There were no significant differences for behavioral intentions for

men across response type (retweet M = 4.33, SD = .49, reply M = 4.12, SD = .46, no response M

= 4.76, SD = .55). This analysis also addressed research question 1 by demonstrating that men

and women differed in how they responded to a reply versus a retweet.

Table 2

Study 1 Results by Condition

Overall No Response Response Retweet

N M(SD)

M(SD)

Women 113 4.89(1.55) 40 4.37(1.57) 36 5.57(1.31) 37 4.79(1.54)

Men 26 4.38(1.33) 7 4.76 (.92) 10 4.17(1.49) 9 4.33(1.48)

Total 139 4.79(1.51) 47 4.43(1.48) 46 5.27(1.46) 46 4.70(1.52)

Consistent with predictions, we found that women differentiated more between the

different restaurant response options than did men. Specifically, women participants reported

that they would be more likely to patronize a restaurant after the restaurant directly engaged

with them by replying to their hypothetical tweet personally. Men did not differentiate between

the different response options. However, given the low sample size for men (n = 26 across the

design), it remains unclear whether our predications for men were supported. Study 2 was

deigned to address this issue by increasing the size of the male sample.

3. Study 2

Due to the small sample of men in Study 1, we were unable to thoroughly investigate whether

men and women evaluated the corporate Twitter response (or lack thereof) differentially. Thus,

Study 2 was conducted to recruit a larger sample of men from among a sample of US adults

using the Amazon.com data collection website Mechanical Turk (www.MTurk.com).

Contemporary research indicates that data collected using this online service yields results

comparable to Psychology subject pools [2]. Thus, our predictions and procedure for Study 2

were identical to Study 1. The sample changed and the data were collected a semester later than

Study 1.

3.1. Participants

Participants were 306 (115 female, 191 male) adults from the USA recruited from Amazon’s

Mechanical Turk. Participants ranged in age from 18-68 (one participant did not provide data;

(M= 30.2, SD = 8.95). The ethnicity breakdown was as follows: Caucasian 215 (70.3%), African

American 24 (7.8%), Asian 5 (1.6%), Pacific Islander 1 (0.3%), Hispanic 21 (6.9%), and Other 5

(1.6%). Participants in this study also reported being heavy social media users with 42.8 percent

reporting using social media over four hours per week.

Table 3

Participants Social Media Use per Week in Study 2

Time Per Week Frequency Percent Cumulative Percent

0-1 hrs 31 10.1 10.1

1-2 hrs 77 25.2 35.3

3-4 hrs 67 21.9 57.2

4 or more hrs 131 42.8 100

Total 306 100

3.2. Measures

To assess users’ behavioral intentions to patronize the restaurant after interacting with them

on Twitter, we used the same three items to make the scale as in Study 1. Again, the reliability

was quite high, Cronbach’s α = .94. The same measures of social media use and importance of

brand engagement on Twitter were taken before and after the stimulus material and behavioral

intentions questions.

3.3. Procedure

The procedure was identical to Study 1.

3.4. Results and discussion

As with Study 1, an ANOVA was conducted to compare behavioral intentions across the type

of restaurant. Again, as anticipated, there was no difference between Chili’s (n =154; M = 4.65,

SD = 1.65) and Gloria’s (n = 152; M = 4.83, SD = 1.29); F(1,306)=1.18, p=.278, ηp2 = .004. For the

remaining analyses, responses were collapsed across the type of restaurant.

Much like our first sample, participants in Study 2 also indicated that social media was a

prominent part of their weekly activities with over half indicating they spent more than 3 hours

per week on social media (4 hours or more – 42.8%, 3-4 hours – 21.9%, 1-2 hours – 25.2%, 0-1

hour – 10.1%). However, in this second sample, participants rated brand engagement on Twitter

as slightly higher in importance than in the first sample (M = 3.05, SD = 1.67).

To examine H1, we examined how the different responses (no response, reply, or retweet)

affected participants’ intentions to patronize the restaurant. In support of H1, this analysis

revealed a significant main effect for response F(2, 306) = 3.28, p = .039, ηp2 = .021. Specifically,

replies (M = 4.99, SD = 1.58) and retweets (M = 4.87, SD = 1.34) were more likely to lead to

higher behavioral intentions with a company than no response (M = 4.36, SD = 1.46).

H2 was also supported. Our analyses revealed a significant main effect for gender, F(1, 306)

= 6.29, p = 0.013, ηp2 = .021. Specifically, women (M = 5.05, SD = 1.49) were more likely to

visit/recommend/purchase from a company engaged with them on Twitter than were men (M

= 4.55, SD = 1.45). While there was no significant interaction between gender and response type

F(2, 306) = .195, p = .82, ηp2 = .001. However, because we had predictions based on gender, we

examined pairwise comparisons, which revealed that female respondents had significantly

greater intentions to patronize the restaurant after a personal reply (M = 4.86, SD = 1.57, p =

.01) or a retweet (M = 4.71, SD = 1.26, p = .04) than after no response (M = 4.19, SD = 1.43), but

there was no difference for men across response type. This analysis also answered RQ1 and

indicated that, as with Study 1, the was a difference between responses and retweets on this

measure for women, but not men.

Overall, Study 2 replicated the results of Study 1 in that women responded with stronger

behavioral intentions when the brand directly engaged the consumer by responding to a tweet.

With a larger sample of men, we found that they responded similarly but the magnitude of the

difference was smaller than with women. Furthermore, with a sample that was better balanced

with respect to gender, we also found support for H1 revealing that participants’ intentions to

patronize were higher when the corporate Twitter account responded to their hypothetical

tweet.

Table 4

Study 2 Results by Condition

Overall No Response Response Retweet

N M(SD) N M(SD) N M(SD) N M(SD)

Women 115 5.04(1.49) 29 4.78(1.49) 43 5.18(1.58) 43 5.09(1.49)

Men 191 4.55(1.45) 73 4.19(1.43) 58 4.86(1.57) 60 4.70(1.26)

Total 306 4.74(1.48) 102 4.36(1.46) 101 4.99(1.58) 103 4.86(1.34)

4. General discussion

We began this paper by questioning whether reciprocal communication – operationalized as

reciprocity in responses between a consumer and a corporate Twitter account could affect

intentions to patronize said company. Our predictions were that responses would be more

effective than no response and that women would be more sensitive to the nuances of different

response options than would men. Across two studies, one an undergraduate sample and a

second an online sample of American adults, the results supported our hypotheses. Thus, these

results provide evidence that directly engaging with consumers via Twitter increases the

likelihood these consumers will give their business to the company. Our results have direct

implications for the ways corporations can utilize Twitter for their benefit to connect with their

current and potential customers. Communication via Twitter can also involve a one-on-one

interaction, so when brands respond directly, the consumer feels connected, and this produces

positive feelings toward the brand. This paper also fills a gap in the literature as there are few

studies that examine business to consumer social influences process on social media

applications such as Twitter.

We expected that women would be more influenced than men to visit the company after

receiving a reciprocal communication from the brand. As social role theory [5] predicts, these

results indicate that compared to men, women are more likely to visit a company when the

company has interacted with them on Twitter. Illustrative of this, research indicates that

women are more likely to engage in Internet use to fulfil relational goals and men are more

likely to use the Internet to seek information [22]. Our results support this previous research

and suggest that the reason reciprocity was more effective for women than men could be that

women may be using Twitter to form and maintain relationships with brands as well as people.

Indeed, some evidence suggests that women may be more apt to be persuaded or influenced by

relationship building online than men [12, 14]. Men, who are generally less relationally focused,

may not show the same pattern of results is that given Twitter’s limited capacity for text, men

may not see it as an efficient way to gain knowledge and information. Social networking’s focus

on building and maintaining relationships is one reason Kimbrough et al. [19] assert that

women use and prefer mediated technology more than men.

4.1. Limitations and future research

The present study reveals implications for both brands and consumers. For corporations, these

results demonstrate an easy and effective means of obtaining new customers. For consumers,

these results suggest that it may be beneficial to engage with a corporate Twitter account as an

initial means of gauging likely levels of customer attentiveness. Furthermore, it is an open

question as to whether these results generalize beyond people already likely to patronize a

specific restaurant. Future research should examine whether these results are equally applicable

to new versus existing patrons. Also, given that this is the first study that we are aware of that

has revealed that in mediated interactions, men and women respond differently to reciprocity-

based social influence techniques, we suggest that future research continue to examine this

finding through replication and expansion so that scholars and practitioners alike can gain an

understanding of the boundary conditions of this finding.

This study focused on one type of consumer product: a restaurant. It may be that women

and men may respond more similarly (or even more differently) as a function of the type of

product or company interacting with them via Twitter. Future research should focus on more

than one category of product as a means of exploring the generalizability of the results. Finally,

these results may not generalize beyond Twitter, a microblogging platform. Future research

should examine whether similar dynamics between customer and business exists on other social

media platforms, especially those with different functions.

4.2. Implications for user behavior change

Overall, the results of these two studies highlight the importance of communication between a

corporation and a user. Specifically, these results suggest that responding directly to a

prospective customer will likely increase the likelihood of that customer patronizing the

business.

Acknowledgements

The research reported in this manuscript was part of Amanda Sardos' master’s project at the

University of Texas, Dallas.

Disclosure of interests. The authors have no competing interests.

References

[1] D. Boyd, N.B Ellison, Social network sites: Definition, history, and scholarship. Journal of

Computer-Mediated Communication. 13(1):11, (2007).

[2] M. Buhrmester, T. Kwang, S.D. Gosling, Amazon's Mechanical Turk a new source of

inexpensive, yet high-quality, data? Perspectives on Psychological Science. 6:3-5, (2011).

[3] K. Chan, S. Li, Understanding consumer-to-consumer interactions in virtual communities:

The salience of reciprocity. Journal of Business Research. 63:1033-1040, (2010).

[4] R.B. Cialdini R, Influence: Science and practice (Vol. 4). Boston, MA: Pearson education;

(2009).

[5] A.H. Eagly, Sex differences in social behavior: A social-role interpretation. Hillsdale, NJ:

Erlbaum; (1987).

[6] P.W. Eastwick, W.L. Gardner, Is it a game? Evidence for social influence in the virtual

world. Social Influence, 4(1):18-32, (2009).

[7] P.J. Ewell, J.A. Minney, R.E. Guadagno, Social Influence Online: An Updated View. In:

Khosrow-Pour M, ed. Encyclopedia of Information Science & Technology. 3rd ed. Hershey,

PA: IGI Global; 6762-6772(2014).

[8] D. Fallows, How women and men use the Internet. Pew Internet & American Life Project.,

28:1-45, (2005).

[9] A.W. Gouldner, The norm of reciprocity: A preliminary statement. American sociological

review. 161-178, (1960).

[10] R. E. Guadagno, Psychological processes in social media: Why we click. Academic Press.

(2024).

[11] R.E. Guadagno, Compliance: a classic and contemporary review. The Oxford handbook of

social influence, 1:107-28, (2017).

[12] R.E. Guadagno, R.B. Cialdini, Online persuasion: An examination of gender differences in

computer-mediated interpersonal influence. Group Dynamics: Theory, Research, and

Practice. 6(1):38, (2002).

[13] R.E. Guadagno, R.B. Cialdini, Online persuasion and compliance: Social influence on the

Internet and beyond. In: Amichai-Hamburger Y, ed. The social net: The social psychology

of the Internet. New York: Oxford University Press; 91-113, (2005).

[14] R.E. Guadagno, R.B. Cialdini, Persuade him by email but see her in person: Online

persuasion revisited. Computers in Human Behavior, 23(2):999-1015, (2007).

[15] R.E. Guadagno, Social influence online: The six principles in action. Casing persuasive

communication, 319-44 (2013).

[16] R.E. Guadagno, Compliance: A Contemporary Review. The Oxford Handbook of Social

Influence, 1:107-128 (2017).

[17] A.E.Holton, M. Coddington, S.C. Lewis, H.G. De Zuniga, Reciprocity and the news: The

role of personal and social media reciprocity in news creation and consumption.

International Journal of Communication, 9:22 (2015).

[18] E. Kim, Y. Sung, H. Kang, Brand followers’ retweeting behavior on Twitter: How brand

relationships influence brand electronic word-of-mouth. Computers in Human Behavior,

37:18-25 (2014).

[19] A. M. Kimbrough, R.E. Guadagno, N.L. Muscanell, J.Dill, Gender differences in mediated

communication: Women connect more than do men. Computers in Human

Behavior, 29(3), 896-900 (2013).

[20] A. Lella, A. Libsman, U.S. Cross Platform Report 2016: Future in Focus [White paper].

Retrieved August 3, 2016 from ComScore:

https://www.comscore.com/Insights/Presentations-and-Whitepapers/2016/2016-US-

Cross-Platform-Future-in-Focus

[21] K. Lovejoy, R.D. Waters, G.D. Saxton, Engaging stakeholders through Twitter: How

nonprofit organizations are getting more out of 140 characters or less. Public relations

review, 38(2):313-8 (2012).

[22] N.L. Muscanell, R.E. Guadagno, Make new friends or keep the old: Gender and personality

differences in social networking use. Computers in Human Behavior, 28(1), 107-112 (2012).

[23] R.E. Wilson, S.D. Gosling, L.T. Graham, A review of Facebook research in the social

sciences. Perspectives on Psychological Science, 7(3), 203-220, (2012).

A. Conditions by brand