Thesis Proposal

Journal of Social Marketing Moving social marketing beyond personal change to social change: Strategically using Twitter to mobilize supporters into vocal advocates Jeanine P.D. Guidry Richard D. Waters Gregory D. Saxton

Article information: To cite this document: Jeanine P.D. Guidry Richard D. Waters Gregory D. Saxton , (2014),"Moving social marketing beyond personal change to social change", Journal of Social Marketing, Vol. 4 Iss 3 pp. 240 - 260 Permanent link to this document: http://dx.doi.org/10.1108/JSOCM-02-2014-0014

Downloaded on: 04 November 2014, At: 16:40 (PT) References: this document contains references to 51 other documents. To copy this document: [email protected] The fulltext of this document has been downloaded 243 times since 2014*

Users who downloaded this article also downloaded: Lisa Chamberlin, Kay Lehmann, (2011),"Twitter in higher education", Cutting-edge Technologies in Higher Education, Vol. 1 pp. 375-391 Tay Pei Lyn Grace, (2009),"Wikis as a knowledge management tool", Journal of Knowledge Management, Vol. 13 Iss 4 pp. 64-74 Sylvia Tag, Miguel Ramos, Paul S. Piper, (2006),"Letting the grass grow: grassroots information on blogs and wikis", Reference Services Review, Vol. 34 Iss 4 pp. 570-574

Access to this document was granted through an Emerald subscription provided by All users group

For Authors If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.

About Emerald www.emeraldinsight.com Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services.

Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

Moving social marketing beyond personal change to social change

Strategically using Twitter to mobilize supporters into vocal advocates

Jeanine P.D. Guidry Richard T. Robertson School of Media and Culture, Virginia

Commonwealth University, Richmond, Virginia, USA

Richard D. Waters School of Management, University of San Francisco, San Francisco,

California, USA, and

Gregory D. Saxton Department of Communication, SUNY-Buffalo, New York, USA

Abstract Purpose – This paper aims to examine what type of messaging on Twitter is most effective for helping move social marketing beyond focusing on personal changes to find out what messages help turn members of the public into vocal advocates for these organizations’ social changes. Social marketing scholarship has regularly focused on how organizations can effectively influence changes in awareness and behaviors among their targeted audience. Communication scholarship, however, has repeatedly shown that the most influential form of persuasion happens interpersonally. As such, it is imperative that organizations learn how to engage audiences and facilitate the discussion about organizational messages between individuals. Social media provide platforms for such conversations, as organizational messaging can be shared and discussed by individuals with others in their networks. Design/methodology/approach – Through a content analysis of 3,415 Twitter updates from 50 nonprofit organizations, this study identifies specific types of messages that are more likely to get stakeholders retweeting, archiving and discussing the organizations’ messaging through regression analysis. Findings – Messages focusing on calls-to-action and community building generated the most retweets and Twitter conversation; however, they were also the least used strategies by nonprofit organizations. Originality/value – Research has regularly examined the types of messages sent out by nonprofit organizations on Twitter, but they have not tested those messages against measures of engagement. This study pushes the understanding of social media communication to the next level by analyzing those message categories against metrics provided by Twitter for each tweet in the sample.

Keywords Twitter, Social marketing, Situational theory of publics, Vocal advocacy

Paper type Research paper

Introduction Since its public launch in 2006, the microblogging service Twitter has grown to have more than 500 million active, daily users in less than eight years. Twitter has become the main social media platform used in social media marketing campaigns (Stelzner, 2009) primarily because of its possibilities for organizations to engage with and be responsive

The current issue and full text archive of this journal is available at www.emeraldinsight.com/2042-6763.htm

JSOCM 4,3

240

Journal of Social Marketing Vol. 4 No. 3, 2014 pp. 240-260 © Emerald Group Publishing Limited 2042-6763 DOI 10.1108/JSOCM-02-2014-0014

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

to their stakeholders. While the ways in which nonprofit organizations use Twitter have been relatively well-documented (Bortree and Seltzer, 2009; Lovejoy et al., 2012; Waters and Jamal, 2011), leading researchers in the field have failed to explore what types of Twitter practices are most effective for social marketing purposes. Existing studies essentially focus on what organizations do with Twitter, not on how target audiences respond. As a result, information is not known as to how publics respond to the messages sent or which types of messages elicit more favorable responses and higher levels of engagement.

This level of engagement is critical for moving social marketing beyond behavior change (Dooley et al., 2013). Traditionally, social marketing has focused on how organizations can use marketing strategies to influence knowledge, attitudes and ultimately behavior at the individual level (Lefebvre, 2011; Wood, 2012; Wymer, 2011). Social media, such as Twitter, provide an opportunity to move the field of social marketing beyond focusing on personal changes at the individual level. By tapping into individuals’ social networks, organizations can take advantage of interpersonal communication, which has widely been acknowledged as the most influential form of communication when it comes to changing personal attitudes and behavior (Forbus and Snyder, 2013; Pettigrew and Pescud, 2012). Organizations that provide updates on Twitter that are likely to be discussed, retweeted and archived for future reference increase their levels of stakeholder engagement (Burton and Soboleva, 2011; Castronovo and Huang, 2012), but also create an environment that transforms individuals from being active recipients of messages into a position of vocal advocacy, as they share the message and discuss it with others (Bernhardt et al., 2012).

The situational theory of publics from public relations scholarship outlines how individuals can transform into vocal advocates, or active publics, by increasing the recognition of an issue as being important, reducing obstacles to participation in conversation and events about the issue and, most importantly, becoming more involved with the issue (Grunig, 1997). Twitter facilitates all three of these criteria through the sharing of information about topics with personal networks; using the reply-function to engage in conversations about the topic; and becoming more involved with the topic through sharing it with personal networks, directly communicating with the sponsoring organization and archiving messages to return to them later for reference.

But, research has shown that all Twitter updates are not equal (Bortree and Seltzer, 2009; Lovejoy et al., 2012; Waters and Jamal, 2011). Consequently, this paper seeks to answer the question of which types of organizational messages elicit greater engagement from the public. Recent research has begun to outline the effectiveness of the messages organizations are posting on Facebook (Saxton and Waters, 2014), and this paper extends this approach to examine how stakeholders respond to different communication practices on Twitter. The purpose of this study is to examine the impact of different types of social media messages on stakeholder engagement. Specifically, through a content analysis of 3,415 tweets sent by 50 large American and international nonprofits over a two-week period in 2012, the research examines whether promotional/ mobilizational, informational or dialogic tweets elicit more stakeholder engagement with the organizations’ Twitter followers. This study seeks to develop an understanding of Twitter practices among nonprofit organizations as to how they more effectively

241

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

spread their messages and influence social change by focusing on moving beyond individual’s behavior change to the creation of vocal advocates for their social causes.

Literature review Situational theory of publics Organizations have many different stakeholders, which when organized into groups around common issues or characteristics, become important publics with which organizations communicate or otherwise interact. The situational theory of publics provides a framework for exploring the various factors involved in different publics’ attitudes and behaviors toward an organization based on their perceptions of an issue or situation (Grunig, 1989a; Hamilton, 1992) and outlines how organizations can approach these publics to boost engagement and involvement with an organization (Grunig, 1997). Much like the process advocated for behavioral change by social marketers (Andreasen, 2002; Lee and Kotler, 2011), the situational theory of publics proposes that persuasive communication can influence an individual’s behavior and ultimately become a supporter of an issue or idea as a result of well-planned and well-developed campaigns.

According to the situational theory of publics, three independent variables – problem recognition, constraint recognition and level of involvement with the issue – predict two dependent variables – information seeking and information processing. Problem recognition is defined as the moment when people recognize that something should be done about an issue or situation, and stop and think about what to do. Constraint recognition happens when people perceive that there may be obstacles in the way of acting related to the problem, and level of involvement is the extent to which people connect with the issue or situation (Grunig, 1997).

The dependent variables, information seeking and processing, may reflect passive or active forms of communication, which influence how effective social marketing efforts are (Bernhardt et al. 2012). Passive or low levels of information seeking and processing may imply that an individual simply receives or consumes information that is presented to them. Active or higher levels of information seeking and processing, on the other hand, imply that individuals expend effort to locate or consume information about an issue or situation. As Grunig (1989b) stated, “people communicating actively develop more organized cognitions, are more likely to have attitudes about a situation, and more often engage in a behavior to do something about the situation” (p. 6). Social marketers have found that increased communication between individuals leads to more positive views toward an issue and a greater likelihood of both behavioral change and continued interest and vocal support for the issue (Dibb and Carrigan, 2013). Based on these variables, individuals can be considered latent, aware or active publics; a classification system that can help organizations determine information dissemination strategies and create communication campaigns (Aldoory and Sha, 2007). Additionally, those publics who are not impacted by an issue are labeled as nonpublics.

Subsequent studies since Grunig’s work have found support for and advanced the situational theory of publics in relation to social marketing causes. For instance, Aldoory (2001) focused on involvement related to health communication targeting women, while Sha (2006) highlighted the importance of cultural identity among various publics. These findings are important when considering the applicability of boosting stakeholder response to organizational messaging for social change.

JSOCM 4,3

242

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

A further dependent variable has been proposed by Weberling et al. (2012) in an application of the situational theory to individual reactions to charitable fundraising efforts following natural disasters. They proposed that information seeking and information processing were distinct from a third action carried out by stakeholders, behaviors that move beyond the need for acquiring and processing information. In their study, the involvement or behavioral component focused on making donations to relief efforts after the 2010 Haitian earthquake. Behavior, as they noted, could range significantly depending on the context of the situation, and the behavior could easily include retweeting and engaging in conversations when the construct is applied to Twitter and other social media platforms.

Though focusing on a very different topic, the results from the mobile giving study are similar to one that examines the adoption of a mobile phone application as part of a marine conservation social marketing campaign. Bowerman and DeLorme (2014) found that the perceptions an individual had about the efficacy of mobile phone applications and the social issue at hand had significant influence as to whether one would download and use the application. This parallels the finding that individual behaviors are driven by information seeking and information processing (Weberling et al., 2012). Users of the social marketing campaign’s mobile phone application expressed higher desires than non-adopters in regard to wanting to learn more about the marine conservation effort (e.g. information seeking) and actually exploring the campaign content (e.g. information processing).

The evolution of campaign messaging Scholars have long advocated for the incorporation of two-way channels in social marketing and public relations efforts (Andreasen, 2002; Bernhardt et al., 2012; Grunig and Hunt, 1984; Kent and Taylor, 1998). In social marketing, dialogue at the individual level (Watne and Brennan, 2012) and at the organizational level (Forbus and Snyder, 2013) helps create attitude and behavioral change. The push for two-way communications stands as a reminder that successful organizations consider the impact of on-going relationships with their stakeholders rather than focusing on one-way information. Social marketers have also been encouraged to explore the efficacy of a dialogic and relational approach in the social media environment as a way to become more effective in their overall goals (Dooley et al., 2013).

Grunig and Hunt (1984) describe four models of public relations that have evolved throughout the practice of communication campaigns. Press agentry is a one-sided model that relies on an emotional delivery of messaging to attract an audience; public information takes a similar one-way approach, but abandons emotional messaging in favor of researched, factual messages. Reflecting the two-way side of the continuum, two-way asymmetry reflects market research in that a faux dialogue is created with stakeholders for the sole reason of obtaining information that can be used for organizational benefit later, while two-way symmetry consists of legitimate conversations taking place between an organization and stakeholder with the end goal being mutual understanding. Two-way symmetry was argued as being the most ethical and idealistic form of communication, but Grunig and Hunt (1984) both argued that it was difficult to carry out due to its time-consuming and resource-intensive nature.

Following up on Grunig’s work, Kent and Taylor (1998) began publishing a series of studies that further argued that social marketers and those behind communication

243

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

campaigns should be working toward the facilitation of dialogue between organizations and their stakeholders. At the time of their studies, the increased use of traditional website technology and the subsequent growth of social media platforms have made online engagement and dialogue a reality that could not exist when the models were first discussed.

Kent and Taylor (2002) describe five crucial features of dialogue: facilitating participation by reducing obstacles to engagement, willingness to interact with individuals and publics that share differing ideas, providing an environment that enables trust be developed among parties, entering into a participative event with a commitment to conversation and genuineness, and mutual participation by multiple parties. Social media, when utilized to its full capacity, exemplifies these features, as it provides a venue for organizations and their publics to communicate in real time. Through social media, organizations create a climate where stakeholders’ participation is facilitated and not merely tolerated. Social media also opens a pathway for vulnerability and unscripted conversations, which is scary for many organizations. Yet this vulnerability provides an authentic voice that ultimately brings more genuine engagement from stakeholders.

Kent and Taylor (1998) also argued that successful incorporation of dialogue online was facilitated by basic online communication principles. These principles include designing an easily navigatable online platform, providing useful information for stakeholders, providing engaging content that encourages stakeholders to return to the platform in the future and finally engaging in conversations with stakeholders. Social media appears custom-designed to facilitate dialogue (Stelzner, 2009; Qualman, 2013), and literature confirms that Twitter and Facebook hold great potential in this area. However, Bortree and Seltzer (2009) found that dialogue with advocacy organizations did not unfold even when an interactive space was created.

Simply creating such a space does not mean communications will take place. Academic and popular literature agrees that creating and maintaining these give-and-take relationships takes consistent effort, time and energy (Baer and Naslund, 2011; Li and Bernoff, 2011; Li and Solis, 2013). For social marketing efforts to succeed online – particularly in the social media realm – communicators must be ready, willing and able to cultivate engagement with their fans and followers (Dibb and Carrigan, 2013). Immediate measureable results may be disappointing, but the payoff over time comes not only from increased awareness and behavior changes with those in an organization’s immediate circle but also from the mobilization of these active publics to help spread the organization’s messaging through peer-to-peer conversations (Bernhardt et al., 2012).

How nonprofits use Twitter Considering Twitter’s brief existence, existing academic research into its operational use is limited. However, there have been attempts to understand how Twitter can be used to increase levels of engagement with followers. These studies have centered on the characteristics of nonprofits’ Twitter use (e.g. message characteristics and profile content) rather than actual measures of engagement. These studies found that nonprofit organizations primarily use Twitter for one-way messages and fail to capitalize on the interactive nature of the platform (Lovejoy and Saxton, 2012; Waters and Jamal, 2011). This broadcast model of Twitter usage, termed shoveling tweets by Messner et al. (2011),

JSOCM 4,3

244

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

focuses on sending out as many tweets as possible without much consideration for if and how followers may respond. A recent study that analyzed Twitter engagement of leading nonprofit organizations in the US health sector found that, while there was a wide variety of results with regard to tweet frequency, the use of hashtags and retweeting, these organizations are still in need of best practice guidelines for Twitter to better engage with their publics (Messner et al., 2012).

Several studies have focused on how nonprofits use the specific Twitter communication tools. A 2012 study of the Nonprofit Times 100 organizations with Twitter accounts found that 68 per cent of the tweets in their sample included hyperlinks – a considerably higher percentage than the average Twitter handle (Lovejoy et al. 2012). This study also found that the sampled nonprofits used public messages addressed directly to specific Twitter users (i.e. any message that started with the @ symbol) more frequently than found in previous studies. The retweet function was used less by the nonprofits than among Twitter users in general – only 16.2 per cent of the total number of tweets studied by Lovejoy shared other users’ tweets.

A more recent study among 188 advocacy organizations found that more than 60 per cent of all tweets in their study contained one or more hashtags, more than double than the Lovejoy study (Guo and Saxton, 2014). This study also found that 22.4 per cent of the tweets in their sample were retweets – again higher than the 16.2 per cent found in the study of Nonprofit Times 100 organizations (Lovejoy et al., 2012).

Shifting from the communication tools used on tweeting to communication style, Waters and Jamal (2011) compared nonprofits’ tweets to Grunig and Hunt’s traditional public relations’ models and found that one-way public information messaging was the most widely used model. When it came to two-way communication, nonprofits were more likely to use asymmetrical rather than symmetrical dialogue. Even when using a level of dialogue on Twitter, the emphasis tended to be on what the stakeholder could accomplish for the organization instead of a truly dialogic model seeking stakeholder input.

Several studies have used a coding scheme based on Kent and Taylor’s (1998) previously mentioned principles for dialogue in Internet communications to measure Twitter communication. In a study of Fortune 500 companies (Rybalko and Seltzer, 2010), the most commonly enacted principle was the incorporation of a dialogic loop (i.e. responding directly to stakeholders’ tweets as well as asking questions on Twitter). However, the authors concluded that the majority of communication on Twitter was still one-way. Echoing this conclusion, a recent study found that among the largest US nonprofits, fewer than 20 per cent of the tweets are of a conversational nature (Lovejoy et al., 2012).

Lovejoy and Saxton (2012) developed the most recent typology of organizational microblogging functions in an attempt to unify the disparate approaches to understand organizational Twitter usage. Their approach centered on classifying tweets as informational, which involve spreading information about the organization, its activities or anything of potential interest to followers; community building, which attempt to build relationships, networks and communities through tweets that promote interactivity; and action-based promotional messages, which endeavor to get followers to do something for the organization.

Lovejoy and Saxton (2012) suggest that the information– community–action triad may represent a “ladder” of organizational communication functions: information as a

245

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

core activity to attract followers; community-focused messages to engage a following of users; and action-oriented messages to mobilize the audience that has been developed through the first two categories. Using this typology, they found that 58.6 per cent of all tweets in their study fell in the information category, 25.8 per cent in the community category and the remaining 15.6 per cent in the action category. While this study helped create a measurement scheme that advanced the development of a messaging strategy for Twitter, it failed to measure how the stakeholders responded to those messages.

How do publics respond? Most of the early studies on organizational Twitter usage mention that future studies should focus on how followers respond to organizational tweets and otherwise engage with nonprofits on Twitter. Social marketing scholars have also encouraged social media research to move beyond organizational behavior to focus on the outcome of socially mediated communication (Dooley et al., 2013). To date, response-oriented research has primarily examined the impact of Facebook. An early study of stakeholders of three Texas-based nonprofits provided promising data indicating that fans and followers are likely to become more involved in the nonprofit’s cause when nonprofits send relevant updates and actively promote conversations on Facebook (Menezes, 2010). Saxton and Waters (2014) recently presented the first in-depth study into stakeholders’ responses to Facebook posts; in other words, the research examined what information stakeholders liked, commented on and shared on Facebook. Prior to this research, the focus of exploration had been entirely on the organization and its activity through Web 2.0 and social media, not on the dynamic response of the public to that activity.

Social media allows near-real-time observation of the relationships between organizations’ actions and the public’s reactions, and this research was the first to attempt to measure and quantify this dynamic. Their findings showed that nonprofits’ Facebook publics are most receptive to community building and certain types of call-to-action messages (advocacy, lobbying or volunteering efforts), and they were least responsive to event promotion and fundraising messages, with informational messages falling in the middle. For the first time from an academic perspective, their study found that community-building messages sent through social media do produce more dialogic outcomes. This reflects social marketing practice in that campaigns that take advantage of established relationships are more effective in changing attitudes and behavior (Watne and Brennan, 2012). Community building messages are perceived more favorably and attract considerably more comments than informational messages because they facilitate back-and-forth exchanges. Dialogic messages, however, are not the only type to yield a higher level of engagement; call-to-action messages, which ask stakeholders to volunteer for or advocate on behalf of the organization, prompted the highest level of engagement from the public (Saxton and Waters, 2014).

This research on Facebook provides the foundational framework for the current study. Given that Twitter has become the social media platform used most often in social marketing efforts (Stelzner, 2009), it is important to see how stakeholders respond to these messages as well. To understand how social marketing can move beyond creating awareness and changing behaviors, it is critical to understand what organizations can do to mobilize their active publics, according to situational theory of publics labeling. Calls have been put forth by social marketing scholars to understand

JSOCM 4,3

246

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

how social media can best be used to mobilize message recipients (Dooley et al., 2012; Kotler, 2011). If organizations can provide specific content that promotes retweeting and conversations among active audiences, then these stakeholders can transition into vocal advocates within their own social networks to further the causes and pro-social changes desired by nonprofit organizations. Taking into consideration the lack of research concentrating on publics’ response to the organizational tweets they are sent, this study is the first to focus on the types of tweets that will elicit engagement from stakeholders.

Previous studies have shown that most tweets sent by nonprofit organizations are information-focused (Guo and Saxton, 2014; Lovejoy and Saxton, 2012, Waters and Jamal, 2011). Social marketing literature regularly separates information-messages into either awareness advertising, which occurs when organizations advise the public about a greater social issue, or direct-to-consumer advertising, which promotes specific brands or solutions to an issue rather than simply promoting awareness of the issue (Hoffman and Novak, 1996). Based on this dichotomization, informational messages can be delineated into two discrete types: public education messages and marketing messages.

It should be noted that marketing messages do not have to be focused on a specific brand, as many social marketing campaigns are not done for a particular organization. Instead, the true differences between public education and marketing messaging as discussed in the current study stem from the nature of communication as discussed by Grunig and Hunt (1984). Public education messages are fact-based and driven by statistics, evidence and testimonials rather than emotionally charged language used by marketing messages. Attempts at educating through persuasive argumentation that is more neutral in tone have been shown to be more effective than publicity-oriented messages using catchy phrases and emotional word choices (Saxton and Waters, 2014). To determine whether these results hold true in the nonprofit setting, the study’s first research question is posed:

RQ1. Do public education tweets elicit more engagement from publics than marketing-focused tweets?

Saxton and Waters (2014) found that call-to-action posts produced the largest number of “likes” and comments among Facebook stakeholders. While there are significant platform differences between Facebook and Twitter, both are popular social media platforms that allow text posts and multimedia posts. Therefore, the study’s second research question asks whether a similar pattern found on Facebook by Saxton and Waters (2014) is present on Twitter:

RQ2. Are messages focusing on calls-to-action more likely to elicit engagement from publics than event promotion or fundraising tweets?

The study’s first two research questions address the differences that exist within two of the three Twitter communication strategies identified by Lovejoy and Saxton (2012); however, they omit community building and fail to collectively compare which of the Twitter message strategies are most efficient at securing stakeholder engagement. With the goal of moving social marketing beyond behavior change to focusing on the creation of vocal advocates, it is important for social marketers to know which communication style is most likely to generate retweets, archiving and conversation (Bernhardt et al., 2012; Lee and Kotler, 2011). As discussed by Dooley et al. (2013), social marketing literature needs more detailed examinations of social media content and outcomes.

247

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

While the two previous research questions are based on comparisons of one-way message strategies in RQ1 and comparisons of one-way and two-way strategies in RQ2, there has not been an analysis of the specific types of strategies that are most effective in securing particular types of stakeholder engagement and responses. Specifically, to advance social marketing beyond behavioral change to becoming an advocate for an issue requires a more complete examination of what strategies individuals are most likely to use in their own online advocacy efforts. Given that Twitter messaging strategies’ impact on an individual’s advocacy efforts is an emerging area of study, the study’s final research question was used to determine which strategies were most effective in creating interpersonal advocacy. The study’s third research question helps clarify the understanding of Twitter message strategies in producing outcomes that move beyond the individual to an individual’s entire social network:

RQ3. Which of the Twitter communication message strategies is most likely to impact stakeholder advocacy and engagement positively?

Method To test the study’s three research questions, a content analysis of nonprofit organizations’ tweets during a two-week period was conducted. The structured sample consisted of 50 nonprofit organizations from the Nonprofit Times 100 and Top Nonprofits on Twitter lists. The Nonprofit Times 100 is published annually to highlight the top 100 legally recognized nonprofits in the USA that must have at least 10 per cent of revenue coming directly from public support and must not be a university or a donor-advised fund. As such, the Nonprofit Times 100 organizations are those working toward social change with active programs and services for the communities they serve. The Top Nonprofits on Twitter list (http://topnonprofits.com/), in turn, contains a regularly updated list of US and international nonprofits (excluding churches, universities and colleges) with the most followers on Twitter. The sample was thus structured in a sense that two groupings were examined. First, 25 organizations were chosen from a random sample of the top 50 nonprofits on the Nonprofit Times 100 list; this ensured that the largest nonprofit organizations were being examined. Second, 25 additional random organizations were chosen from the 50 most followed Twitter accounts on the Top Nonprofits on Twitter list. Random sampling was done to ensure greater variety of organizations that comprised the list, as relying solely on the top 25 organizations on the list resulted in a more homogenous sample than the random sample produced with the assistance of a random sample generator. Random samples were pulled from these two lists intentionally to ensure that nonprofits with large revenues were included in the overall sample, as they are the one most likely to have the resource to develop a strategic social media team and to make sure that nonprofits with the largest social media presence are included. As the random sampling indicated, these two lists were quite different in makeup, as only eight organizations were present on both lists.

Given the presence of eight organizations in both random samples, an additional eight organizations were randomly drawn from those lists. The eight organizations that were drawn twice were kept on the list that they were drawn first and replaced on the second list. For example, if the YMCA appeared on both lists and was randomly drawn second on the Nonprofit Times 100 list and fourth on the Top Nonprofits on Twitter list,

JSOCM 4,3

248

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

it would be kept on the Nonprofit Times 100 list because it was drawn before the second list. A replacement organization would have been randomly drawn to replace it on the Top Nonprofits on Twitter list.

During the eight-week period between June 14 and August 9, 2012, content from the Twitter accounts of these 50 nonprofit organizations was sampled during a constructed two-week period. Each day of the week was randomly selected twice within the eight-week period. This produced a sample of the following 14 non-consecutive days: June 14, 20, 23, 26 and 29; July 1, 3, 5, 19, 16, 23 and 27; and August 1 and 4. On each of the days in the sample, all tweets and retweets by each of the 50 organizations were archived and analyzed.

During these constructed two weeks, 3,415 tweets were produced by the 50 nonprofit organizations. Using a Twitter classification scheme based on the one developed by Lovejoy and Saxton (2012), each of the 3,415 tweets was assigned one of six codes according to what was considered to be its primary purpose: informational tweets (i.e. those spreading information about the organization, its activities or anything of potential interest to followers), community-building messages (i.e. those attempting to build relationships, networks and communities by promoting interactivity and dialogue) and promotion and mobilization (i.e. those seeking to get followers to act on behalf of the organization). The researchers decided that it would be helpful to further break down the informational and promotion and mobilization categories to more accurately reflect distinct dimensions of these messages. As shown in Table I, informational tweets can take on the nature of a public education post by making an effort to inform and educate the public about social marketing issues broadly or they could take on the form of distributing information about the work the nonprofit does directly. Likewise, the promotion and mobilization messages represent three different categories. Events and promotions are tweets that promote an organization’s upcoming events, fundraising tweets focus on revenue-generation activities whether that be through donation solicitation or sales, and those that have a clear goal of soliciting the public’s help in specific lobbying, advocacy or volunteering efforts were classified as call-to-action messages.

All 3,415 tweets in the sample were coded for message type and the level of engagement created by each tweet. Engagement was measured by metrics reported by Twitter and included identifying the number of times a tweet was retweeted, the number of times a tweet was favorited, the number of times a bit.ly hyperlink (if present) was clicked on and was shared and whether a tweet resulted in a conversation. Bit.ly hyperlinks allow organizations to link to websites outside of Twitter using a URL-shortening service, which is important for Twitter given its 140-character limit. Though many services exist to provide shortened URLs, Bit.ly still continues to be a dominant service used because of its provision of click-through metrics for organizations.

After a 90-minute training session where the content analysis framework and codebook were discussed, two coders were used to test intercoder reliability. Following suggestions by Neuendorf (2002), 10 per cent of the sample was coded to test for intercoder reliability. Using Scott’s �, the intercoder reliability coefficient varied between 0.89 and 0.96 for tweet content category and the four methods of engagement.

249

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

Results The 50 randomly sampled nonprofit organizations represented a cross-section of the nonprofit sector, ranging from the American Heart Association to the New York Public Library, and from Habitat for Humanity to the World Wildlife Fund. Using the National Taxonomy of Exempt Entities classification system, 42 per cent of the organizations operate in the field of international and foreign affairs (n � 21), 18 per cent in human services (n � 9), 16 per cent in health (n � 8) and the remainder in arts, culture and humanities; environment; education; housing and shelter; youth development; and public and societal benefit. At the time of coding, these nonprofits had an average of 312,329.4 people following their accounts (SD � 372,414.5), which ranged from a minimum of 411 followers to a high of 1,373,979 followers. The nonprofits followed 23,320.1 other Twitter users (SD � 57,186.8), which ranged from only following five other accounts to a high of following 276,214. During the constructed two-week period, the sampled nonprofits had an average of 68.3 tweets (SD � 50.92), which ranged from

Table I. Descriptions and examples of Twitter content and categories

Content category Description Example tweets

Informational Marketing Tweets focusing on direct marketing

the work carried out by the organization

JUST IN: Greenpeace activists shut down 45 Shell petrol stations www.guardian.co.uk/environment/ 2012/jul/16/greenpeace-activists- shell-petrol. .. #tellshell #savethearctic (tweet by @greenpeace)

Public education Tweets focusing on informing and educating the public about policy issues and the cause advocated by the organization

Power out tip: Never use a generator, grill, or other gasoline, propane, natural gas, or charcoal- burning devices inside an enclosed area (tweet by @redcross)

Community building Tweets focusing on building relationships, networks and communities through tweets that promote interactivity and dialogue

@citlau She’s in our thoughts. We’re hoping for the best. (tweet by @American_Heart)

Promotional/Mobilization Call-to-action Tweets soliciting the public’s help in

specific lobbying, advocacy or volunteering efforts

Take Action: Drop charges Against 11-Year-Old “Protester” in Bahrain http://owl.li/bq250 #Bahrain (tweet by @Amnesty)

Events and promotion Tweets promoting the organization’s upcoming events

Who doesn’t love an @ResourceMom Twitter Party? Especially when it benefit babies & moms! Please RSVP: http://cintl. us/10h (tweet by @compassion)

Fundraising Tweets aiming at soliciting donations or sales

Woot! We’re 15 #text4tigers away from 1000! Pls help WWF save tigers—text TIGERS to 20222 to donate $10 http://bit.ly/tftTu (tweet by @World_Wildlife

JSOCM 4,3

250

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

a low of 7 to a high of 226. The sampled nonprofits had an average of 421.9 publicly shared updates (SD � 721.7), which ranged from a minimum of 7 to a maximum of 226.

Of the 3,415 tweets, the largest group of tweets represented informational messaging, as 67.9 per cent (n � 2,321) represented this type of communication style. Dividing informational messages into public education and marketing revealed that marketing was used slightly more often (39.8 per cent, n � 1,358) than public education (28.2 per cent, n � 963).

Overall, 661 tweets (19.4 per cent) made up the promotion and mobilization grouping. Looking at the three sub-groupings of mobilization, the largest category was call-to-action with 8.1 per cent (n � 278) of the tweets, followed by event promotion (6.9 per cent, n � 234), and finally fundraising (4.4 per cent, n � 149). Reflecting previous social media studies, community building represented the smallest portion of tweets, as only 12.6 per cent (n � 248) of the total sample focused on attempts to engage and interact with the audience.

Though attempts to engage audiences on Twitter were only minimally used, that does not mean that nonprofit organizations were unable to solicit engagement with other types of messaging. RQ1 sought to understand what types of information sharing elicited more engagement from stakeholders by comparing public education and marketing messages against the public response they generated through retweets, favoring the organization’s tweets and creating conversations. To test the engagement impact of different informational messages, negative binomial regression analysis tests were carried out to determine the effect of the use of education and marketing information content on the levels of engagement. For all measures, marketing content was unlikely to generate retweets (r � �0.33, p � 0.001), be archived (r � �0.22, p � 0.01) or generate conversations (r � �0.54, p � 0.001). Public education informational messages, on the other hand, were more likely to generate retweets (r � 0.15, p � 0.05), be archived (r � 0.22, p � 0.01) and generate conversations with the organization (r � 0.28, p � 0.001). It can be concluded that there is evidence to support that public education tweets elicit more engagement than marketing-oriented messages.

RQ2 carried out a similar test as RQ1; however, the focus transitioned from information-sharing messages to those focusing on promotion and mobilization. The research question asked whether call-to-action messages were more likely to result in stakeholder engagement than messages that promoted events or focused on fundraising efforts. To test RQ2, negative binomial regression analyses were also carried out to determine the effect of the three different mobilization and promotion content categories on stakeholder engagement. For the generation of retweets (r � 0.57, p � 0.001) and conversations (r � 0.19, p � 0.05), call-to-action tweets were more likely to create this form of engagement. Both fundraising (r � �0.36, p � 0.001) and event promotion (r � �0.77, p � 0.001) demonstrated that Twitter followers were highly unlikely to retweet these messages to their own networks. Similarly, fundraising (r � �0.73, p � 0.001) and event promotion (r � �0.13, p � 0.05) had similar adverse reactions for archiving or favoriting the messages for future usage, but it should be noted that the regression was not statistically significant for event promotion though it had statistical leanings toward significance. Call-to-action messages showed signs of positive reactions by stakeholders for archiving them (r � 0.10, p � 0.05); however, they too failed to reach statistical significance. RQ2 has inconclusive findings, as call-to-action messages were more likely to generate retweets and conversation, but none of the

251

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

promotional and mobilization messages had a positive impact on whether someone would favorite or archive the messages for access at a later date.

RQ3 sought to broaden the discussion of stakeholder engagement from looking at the two most commonly used messaging strategies to the entire spectrum of Twitter message styles. Five regression models, presented in Table II, were run using public education informational messages as the baseline for comparison. The decision was made to use this category as the baseline for comparison given past studies that have documented the dominance of awareness and information messages across social media platforms and the results of the first hypothesis that showed that public education informational messaging was more likely to create engagement than marketing messages.

The first model sought to test which message strategy was most likely to generate retweets using logit regression. Compared with public education information messaging, both marketing information (r � �0.29, p � 0.10) and community building (r � �2.91, p � 0.01) were less likely to be retweeted. There was no statistical difference for public education messages and event promotion (r � 0.39), fundraising (r � 0.25) and call-to-action (r � 0.19) messages. The second model used logit regression to determine whether Twitter message strategies were likely to be favorited or archived. Again using public education messages as a baseline for comparison, the regression model revealed that community building (r � �2.11, p � 0.01) and fundraising (r � �0.32, p � 0.10) were less likely to be archived or favorited than public education messages. There were no statistical differences for marketing information (r � �0.13), event promotion (r � 0.08) and call-to-action (r � 0.04) messages. The third logit model sought to determine which message strategies were most likely to generate conversation. Contrary to the first two logit models, community building was significantly more likely to generate conversation (r � 7.11, p � 0.01). Specific calls-to-action also were more likely to generate conversation surrounding the message than public information messages (r � 1.24, p � 0.10). Marketing information (r � 0.06), event promotion (r � �0.09) and fundraising (r � 0.48) messages did not generate conversations in different proportion to public education messages.

The fourth and fifth models presented in Table II illustrate the results of negative binomial regressions that continued the use of public education messages as a baseline for comparison. Rather than evaluating the strategies against whether or not the messages were likely to generate retweets, archiving or conversation using a dichotomous yes/no outcome, these models used the overall number of retweets and instances of archiving/favoriting as a continuous variable. The negative binomial regression in the fourth model showed that marketing information (r � �0.48, p � 0.01), event promotion (r � �1.00, p � 0.01) and fundraising (r � �0.59, p � 0.01) generated significantly fewer retweets than public education messages. Community building messages generated similar numbers of retweets as public education messages (r � �0.12), as there was no statistically significant difference. Call-to-action tweets, on the other hand, are associated with significantly more retweets than public education informational messages (r � 0.34, p � 0.01).

The fifth regression model sought to determine which type of Twitter message strategy was most likely to generate favoriting or archiving by Twitter users. None of the Twitter message types generated more favoriting than public education information tweets. Marketing information (r � �0.44, p � 0.01), fundraising (r � �0.93, p � 0.01)

JSOCM 4,3

252

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

Table II. Regression models for overall comparison of

engagement actions by different message types†

M od

el 1:

re tw

ee te

d (Y

/N )

M od

el 2:

fa vo

ri te

d (Y

/N )

M od

el 3:

co nv

er sa

ti on

(Y /N

) M

od el

4: nu

m be

r of

re tw

ee ts

M od

el 5:

nu m

be r

of fa

vo ri

te s

M ar

ke ti

ng in

fo rm

at io

n �

0. 29

� (0

.1 6)

� 0.

13 (0

.0 9)

0. 06

(0 .6

5) �

0. 48

** (0

.0 7)

� 0.

44 **

(0 .0

8) C

om m

un it

y bu

ild in

g �

2. 91

** (0

.1 6)

� 2.

11 **

(0 .1

4) 7.

11 **

(0 .5

2) �

0. 12

(0 .0

9) �

0. 07

(0 .1

0) E

ve nt

pr om

ot io

n 0.

39 (0

.3 3)

0. 08

(0 .1

6) �

0. 09

(0 .1

4) �

1. 00

** (0

.1 2)

� 0.

33 *

(0 .1

3) F

un dr

ai si

ng 0.

25 (0

.3 9)

� 0.

32 �

(0 .1

8) 0.

48 (1

.1 2)

� 0.

59 **

(0 .1

4) �

0. 93

** (0

.1 6)

C al

l-t o-

ac ti

on 0.

19 (0

.2 9)

0. 04

(0 .1

5) 1.

24 (0

.7 1)

0. 34

** (0

.1 1)

� 0.

11 (0

.1 2)

C on

st an

t 2.

62 **

(0 .1

3) 0.

83 **

(0 .0

7) �

5. 48

** (0

.5 0)

3. 57

** (0

.0 5)

1. 76

** (0

.0 6)

N 3,

41 5

3, 41

5 3,

18 1

3, 41

5 3,

41 5

P se

ud o

R 2

0. 15

7 0.

09 8

0. 42

6 0.

04 1

0. 01

6 M

od el

si gn

ifi ca

nc e

(� 2 )

58 2.

01 **

35 3.

43 **

17 68

.1 6*

* 14

2. 79

** 56

.5 9*

* L

og lik

el ih

oo d

� 10

80 .8

3 �

20 87

.2 6

� 28

7. 64

� 13

72 2.

06 �

80 52

.6 6

N o te

s: †

P ub

lic ed

uc at

io n

tw ee

ts ar

e us

ed as

a ba

se lin

e fo

r co

m pa

ri so

n in

th es

e m

od el

s; th

us ,t

he ca

te go

ry is

re m

ov ed

fr om

th e

ta bl

e lis

ti ng

.S ta

nd ar

d er

ro rs

in pa

re nt

he se

s. M

od el

s 1

th ro

ug h

3 us

e lo

gi tr

eg re

ss io

n; m

od el

s 4

an d

5 us

e ne

ga ti

ve bi

no m

ia lr

eg re

ss io

n. W

it h

m ax

im um

-li ke

lih oo

d m

od el

s su

ch as

in ne

ga ti

ve bi

no m

ia lr

eg re

ss io

n, th

er e

is no

tr ad

it io

na lR

2 ;f

or th

is re

as on

,a n

an al

og ou

s “p

se ud

o- R

2 ”

is ty

pi ca

lly re

po rt

ed .T

he R

2 sh

ow n

he re

is th

e M

L (C

ox –S

ne ll)

R 2 ;

� p

� 0.

10 ;

* p

� 0.

05 ;

** p

� 0.

01

253

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

and event promotion (r � �0.33, p � 0.05) were statistically less likely to generate archiving by Twitter users. Community building (r � �0.07) and calls-to-action (r � �0.11) were not statistically different from the number of public education messages that were favorite, though the negative relationship indicates that fewer messages in these categories were actually archived by users.

Overall, the answer to RQ3 paints a mixed picture. Depending on the type of engagement that an organization hopes to achieve, different message types may be most helpful. In terms of generating dialogue and actual engagement with a stakeholder, community building and call-to-action messages are the most successful. However, one-way public information messages that focus on education and awareness appear to generate more retweets than community building. Public education tweets also are archived or favorited more than other types of tweets.

Discussion This study found that the types of tweets nonprofit organizations disseminate as part of their social marketing efforts have an impact on how their stakeholders reach out and interact with the organizations. During the constructed two-week period, nearly 68 per cent of tweets sent out by the nonprofits were informational sharing in nature, while 19 per cent represented the promotion and mobilization category. Only 13 per cent were seeking to engage and interact with community-building messages. Comparing this breakdown with what messages trigger stakeholder engagement shows that the two categories that trigger conversations with stakeholders are rarely being used. Under the promotional and mobilization category, call-to-action messages elicited large numbers of messages retweeted, or shared, among the individuals’ social networks. Community-building messages, on the other hand, were the most successful at creating dialogue with audiences. Marketing and public education informational messages were generally outperformed on all engagement metrics. These results produced similar results to Saxton and Waters’ (2014) study on Facebook engagement, and they reflect similar suggestions by other social marketing scholars about using social media to affect social change (Thackeray and Hunter, 2010).

For social marketers, these results indicate that time spent on social media efforts are mostly off target with their stakeholders (Hestres, 2014). Studies from marketing and public relations highlight that organizations routinely use social media for information-sharing purposes (Bortree and Seltzer 2009; Obar et al., 2012). However, these messages were not the ones that generated stakeholder engagement. Castronovo and Huang (2012) argue that organizations have to decide how to best use social media to their advantages. Rather than dedicating significant time and resources to maintaining profiles that are bastions of dialogue and engagement, understanding how an organization’s stakeholders use social media can help make time spent online more effective.

Burton and Soboleva (2011) discuss a strategy that may transform the practice of social marketing. Instead of spending the bulk of marketing resources focusing on educating, attitude change and behavior change, organizations would be better served to provide content that organizations are more likely to share and discuss within their own social networks (Hestres, 2014); in essence, organizations should work to determine how to transform their supporters into vocal advocates for their causes (Bernhardt et al., 2012). The current study found that call-to-action tweets generated the most retweeting,

JSOCM 4,3

254

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

so organizations should provide more messages of this nature to spread the word about their organizational efforts organically. Tapping into the interpersonal networks social media platforms provide access to for messaging is considered to be more trustworthy than direct organizational messaging because the individual sharing or retweeting the message has already vetted its credibility (Wood, 2012; Wymer, 2011).

Increasing an individual’s level of involvement with an organization positions them further into the situational theory of publics classification, and Bowerman and DeLorme (2014) concluded that involvement in an issue was a key component of becoming more active in a social marketing campaign. Active publics, those who recognize an issue as being important, have few obstacles preventing their involvement, and who have high levels of involvement, are quick to mobilize and reach out to others. Twitter facilitates turning these active publics into vocal advocates on behalf of the organization, thus allowing the organization to spread its message about an issue throughout an interpersonal network while tapping into perceptions of vetted message credibility (Neiger et al., 2013). Coombs (2007) argued that organizations facing crises can reduce the damage to their reputation if they mobilized their supporters online to share organizational messaging on blogs and other social media platforms rather than solely focusing on organizational spokespeople and message point. The credibility and trust gained through the interpersonal networks of the vocal advocates enhances the organizational messaging and ultimately creates more engagement for the organization.

Grunig (2009) echoed these thoughts in a review of strategic communications in the digital era. Organizations that use their social media platforms as a megaphone are missing the point of the technologies. Whether Facebook or Twitter, messages are meant to be created, shared and discussed. Using situational theory of publics terminology, active publics who can turn into vocal advocates for the organization can help influence latent and aware publics to be more receptive of organizational messages. Mobilizing stakeholders to become vocal advocates on behalf of the nonprofits’ issues online frees up social marketers to focus more on taking their communication efforts offline, which generally requires more time and resources (Naik and Peters, 2009).

The question is no longer whether nonprofits use social media, but instead how to use it effectively. Qualman (2013) argues that if organizations facilitate engagement and conversation rather than grandstanding and using social media as a publicity channel, then their overall goals would be achieved more easily, as individuals start responding to what they see and hear within their own social networks and not just from organizations. Although this study found that certain types of Twitter updates produced engagement outcomes for large and highly followed nonprofit organizations, it is crucial for social marketers to understand how their own audiences respond to these social media efforts. Reviewing their own social media strategies to see what has worked best in past social marketing campaigns can lead to an optimized approach with future efforts and resource allocation to reach positive outcomes not just for education, attitude change and behavioral change but also for mobilizing a community around the social issue so that their voices are amplified to truly create systematic social change.

Conclusion As social marketers, significant amounts of time are spent on trying to change the public’s audiences and behaviors. Practitioner and academic literature both echo the importance of a combined online and offline strategy to influence social change;

255

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

however, nonprofits face limited resources and time for their social marketing efforts. Recognizing that social marketing can become more effective by tapping into the personal networks of vocal advocates, communication practitioners need to understand how to use Twitter to mobilize their supporters and tap into the perceived credibility of shared organizational messaging.

Though not the original intention of this study, the results suggested three best practices for Twitter usage in social marketing efforts. First, the most successful engagement stems from communication that asks questions of social media followers and encourages them to respond to online and offline calls-to-action rather than simply focusing on publicity and information sharing. Second, social marketers must dedicate portions of their day to engaging in conversations and responding to individuals’ social media comments and questions. Finally, even though fundraising, event promotion and publicity tweets did not result in significant engagement, they have their place in social marketing. Organizations should strive for retweets, conversation and having their tweets favorited, but they should not avoid spreading messages about their own initiatives concerning fundraising, events and marketing. These messages have their time and place, and they may also attract audiences when creativity and multimedia are used to help relay the messages even though they may not fully engage using Twitter metrics.

Limitations of the study Though the current study found statistical significance in relation to which message types elicit certain types of stakeholder engagement on Twitter, certain limitations of the study must be taken into consideration by social marketers reconsidering their approaches to their campaigns. First, it is necessary to stress that not all nonprofit organizations engage in social marketing. The sector is quite broad with diverse missions, and not all nonprofits are seeking to create widespread social change. However, the organizations in the current study represent two diverse groups – those organizations listed in the Nonprofit Times 100 and Top Nonprofits on Twitter. While these organizations represent the largest and most successful US and international nonprofits in terms of fundraising and generating social media followers, respectively, an argument could be made that smaller community-based nonprofits may use social media more effectively, given the closer ties to the community and the lack of a bureaucracy dictating how social media must be used for stakeholder communications. Second, the study looks at metrics provided by the Twitter platform (e.g. number of retweets, favorites and public @-replies); it does not examine private messaging that happens on Twitter. As such, fundraising or event promotion messages may actually trigger direct messages being sent to the nonprofit, though it would not be reflected in Twitter measures of engagement. Third, the tweets in the sample represent a constructed two-week time frame that happened over three months in the summer of 2012. Though there is no evidence that the time of year influences content of social media messages, it would be wise to expand the analysis over a longer period to be more reflective of the nonprofits’ social media actions.

Future research This study makes an argument that social marketing can move beyond behavioral change by enabling audiences to become vocal advocates for causes they are highly connected to, but is this an outcome that social marketers want? Kent and Taylor (2002)

JSOCM 4,3

256

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

note that unscripted conversations are often uncomfortable for strategic communicators, who give up control when engaging in social media conversations. Likewise, enabling active publics to speak on behalf of organizational issues may introduce a level of discomfort for social marketers. It would be helpful to survey social marketers for their interest tapping into the power of the social web. Perhaps they do not wish to use Twitter, Facebook and other social platforms to engage in meaningful interactions, but prefer to use social media as a one-way message-sharing platform. Understanding the messaging preferences of social marketers would also identify whether they use different strategies for different audiences on social, shareable platforms. Though this study is one of the first that connects message type with actual measures of public engagement, it would also be helpful to ask Twitter users about their motivations for retweeting, favoriting and replying to organizational messages. Qualitative research could explore why some messages are more successful at engaging audiences and producing retweets and replies. If social marketers have a better understanding of why and how people use Twitter, then they would be in a better position to use the platform to move beyond behavioral change to create large-scale social change.

References Aldoory, L. (2001), “Making health communications meaningful for women: factors that

influence involvement”, Journal of Public Relations Research, Vol. 13 No. 2, pp. 163-185.

Aldoory, L. and Sha, B. (2007), “The situational theory of publics: practical applications, methodological challenges, and theoretical horizons”, in Toth, E. L. (Ed), The Future of Excellence in Public Relations and Communication Management, (pp. 339-355). Lawrence Erlbaum, Mahwah, NJ.

Andreasen, A.R. (2002), “Marketing social marketing in the social change marketplace” Journal of Public Policy & Marketing, Vol. 21 No. 1, pp. 3-13.

Baer, J. and Naslund, A. (2011), The Now Revolution: 7 Shifts to Make Your Business Faster, Smarter, and More Social, John Wiley & Sons, Inc, New York, NY.

Bernhardt, J.M., Mays, D. and Hall, A.K. (2012), “Social marketing at the right place and right time with new media”, Journal of Social Marketing, Vol. 2 No. 2, pp. 130-137.

Bortree, D.S. and Seltzer, T. (2009), “Dialogic strategies and outcomes: an analysis of environmental advocacy groupsõ Facebook profiles”, Public Relations Review, Vol. 35 No. 3, pp. 317-319.

Bowerman, K. and DeLorme, D.E. (2014), “Boaters’ perceptions of a mobile app for a marine conservation social marketing campaign”, Social Marketing Quarterly, Vol. 20 No. 1, pp. 47-65.

Burton, S. and Soboleva, A. (2011), “Interactive or reactive? Marketing with Twitter”, Journal of Consumer Marketing, Vol. 28 No. 7, pp. 491-499.

Castronovo, C. and Huang, L. (2012), “Social media in an alternative marketing communication model”, Journal of Marketing Development & Competitiveness, Vol. 6 No. 1, pp. 117-136.

Coombs, W.T. (2007), “Protecting organization reputations during a crisis: the development and application of situational crisis communication theory”, Corporate Reputation Review, Vol. 10, 163-176.

257

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

Dibb, S. and Carrigan, M. (2013), “Social marketing transformed: Kotler, Polonsky and Hastings reflect on social marketing in a period of social change”, European Journal of Marketing, Vol. 47 No. 9, pp. 1376-1398.

Dooley, J.A., Jones, S.C. and Iverson, D. (2012), “Web 2.0: an assessment of social marketing principles”, Journal of Social Marketing, Vol. 2 No. 3, pp. 207-221.

Dooley, J., Jones, S.C. and Iverson, D. (2013), “Web 2.0 marketing: the need for a new research agenda for social marketing”, Journal of Digital & Social Media Marketing, Vol. 1 No. 2, pp. 201-215.

Forbus, R. and Snyder, J.L. (2013), “Use of comforting to enhance social marketing success: a case study”, Social Marketing Quarterly, Vol. 19 No. 2, pp. 97-109.

Grunig, J.E. (1997), “A situational theory of publics: conceptual history, recent challenges and new research”, in Moss, D. MacManus, T. and Vercic, D. (Eds.), Public Relations Research: An International Perspective, (pp. 3-38). International Thomson Business Press, London, England.

Grunig, J.E. (2009), “Paradigms of global public relations in an age of digitalization”, PRism, Vol. 6 No. 2. www.prismjournal.org/fileadmin/Praxis/Files/globalPR/GRUNIG.pdf. (accessed 20 February 2014).

Grunig, J.E. (1989a), “Publics, audiences and market segments: segmentation principles for campaigns”, in Salmon, C.T. (Ed), Information Campaigns: Balancing Social Values and Social Change (pp. 199-223). Sage, Beverly Hills, CA.

Grunig, J.E. (1989b), “Sierra Club study shows who becomes activists”, Public Relations Review, Vol. 15 No. 3, pp. 3-24.

Grunig, J.E. and Hunt, T. (1984), Managing Public Relations. Holt, Rinehart, and Winston, New York, NY.

Guo, C. and Saxton, G.D. (2014), “Tweeting social change: how social media are changing nonprofit advocacy”, Nonprofit and Voluntary Sector Quarterly, Vol. 43 No. 1, pp. 57-79.

Hamilton, P.K. (1992), “Grunig’s situational theory: a replication, application, and extension”, Journal of Public Relations Research, Vol. 4 No. 3, pp. 123-149.

Hestres, L.E. (2014), “Preaching to the choir: internet-mediated advocacy, issue public mobilization, and climate change”, New Media & Society, Vol. 16 No. 2, pp. 323-339.

Hoffman, D.L. and Novak, T.P. (1996), “Marketing in hypermedia computer-mediated environments: conceptual foundations”, Journal of Marketing, Vol. 60 No. 3, pp. 50-69.

Kent, M.L. and Taylor, M. (1998), “Building dialogic relationships through the World Wide Web”, Public Relations Review, Vol. 24 No. 3, pp. 321-324.

Kent, M.L. and Taylor, M. (2002), “Toward a dialogic theory of public relations”, Public Relations Review, Vol. 28 No. 1, pp. 21-37.

Kotler, P. (2011), “Reinventing marketing to manage the environmental imperative”, Journal of Marketing, Vol. 75 No. 4, pp. 132-135.

Lee, N.R. and Kotler, P. (2011), Social Marketing: Influencing Behaviors for Good. Sage, Thousand Oaks, CA.

Lefebvre, R.C. (2011), “An integrative model for social marketing”, Journal o Social Marketing, Vol. 1 No. 1, pp. 54-72.

Li, C. and Bernoff, J. (2011), Groundswell, Expanded and Revised Edition: Winning in a World Transformed by Social Technologies. Harvard Business School Publishing, Boston.

Li, C. and Solis, B. (2013), The Seven Success Factors of Social Business Strategy. Jossey-Bass, San Francisco.

JSOCM 4,3

258

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

Lovejoy, K. and Saxton, G.D. (2012), “Information, community, and action: how nonprofit organizations use social media”, Journal of Computer-Mediated Communication, Vol. 17 No. 3, pp. 337-353.

Lovejoy, K., Waters, R.D. and Saxton, G.D. (2012), “Engaging stakeholders through Twitter: how nonprofit organizations are getting more out of 140 characters or less”, Public Relations Review, Vol. 38 No. 2, pp. 313-318.

Menezes, Eva L.T.de. (2010), Nonprofits and Social Media: Can Online Actions Translate into Social Good? Master’s thesis presented at the University of Texas at Austin.

Messner, M., Linke, M. and Eford, A. (2011), “Shoveling tweets: an analysis of the microblogging engagement of traditional news organizations”, Paper presented at the International Symposium on Online Journalism, Austin, TX.

Messner, M., Jin, Y., Medina-Messner, V., Meganck, S., Quarforth, S. and Norton, S. (2012), “140 characters for better health: an exploration of the Twitter engagement of leading nonprofit organizations”, in Al-Deen, H.S.N. and Hendricks, J.A. (Eds.), Social Media and Strategic Communications (pp. 119-136). Palgrave Macmillan, New York.

Naik, P.A. and Peters, K. (2009), “A hierarchical marketing communications model of online and offline media synergies”, Journal of Interactive Marketing, Vol. 23 No. 4, pp. 288-299.

Neiger, B.L., Thackeray, R., Burton, S.H., Giraud-Carrier, C.G. and Fagen, M.C. (2013), “Evaluating social media’s capacity to develop engaged audiences in health promotion settings: use of Twitter metrics as a case study”, Health Promotion Practice, Vol. 14 No. 2, pp. 157-162.

Neuendorf, K. A. (2002), “The content analysis guidebook.”

Obar, J.A., Zube, P. and Lampe, C. (2012), “Advocacy 2.0: an analysis of how advocacy groups in the United States perceive and use social media as tools for facilitating CIIC engagement and collective action”, Journal of Information Policy, Vol. 2. http://jip.vmhost.psu.edu/ojs/ index.php/jip/article/viewArticle/80 (accessed 10 June 2014).

Pettigrew, S. and Pescud, M. (2012), “Improving parents’ child-feeding practices: a social marketing challenge”, Journal of Social Marketing, Vol. 2 No. 1, pp. 8-22.

Qualman, E. (2013), Socialnomics: How Social Media Transforms The Way We Live And Do Business. John Wiley & Sons, Inc, New York.

Rybalko, S. and Seltzer, T. (2010), “Dialogic communication in 140 characters or less: How Fortune 500 companies engage stakeholders using Twitter”, Public Relations Review, Vol. 36 No. 4, pp. 336-341.

Saxton, G.D. and Waters, R.D. (2014), “What do stakeholders like on Facebook? Examining public reactions to nonprofit organizations’ status updates”, Journal of Public Relations Research, Vol. 26 No. 3, pp. 280-299.

Sha, B.-L. (2006), “Cultural identity in the segmentation of publics: an emerging theory of intercultural public relations”, Journal of Public Relations Research, Vol. 18 No. 1, pp. 45-65.

Stelzner, M.A. (2009), “Social media marketing industry report”, www.socialmediasummit09. com/ (accessed 26 February 2010).

Thackeray, R. and Hunter, M. (2010), “Empowering youth: use of technology in advocacy to affect social change”, Journal of Computer-Mediated Communication, Vol. 15 No. 4, pp. 575-591.

Waters, R.D. and Jamal, J.Y. (2011), “Tweet, tweet, tweet: a content analysis of nonprofit organizations’ Twitter updates”, Public Relations Review, Vol. 37 No. 3, pp. 321-324.

259

Moving social marketing

beyond personal change

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)

Watne, T. and Brennan, L. (2012), “Behavioural change starts in the family: the role of family communication and implications for social marketing”, Journal of Nonprofit & Public Sector Marketing, Vol. 23 No. 4, pp. 367-386.

Weberling, B., Waters, R.D. and Tindall, N.T.J. (2012), “The role of text messaging in public relations: testing the situational theory of publics for mobile-giving campaigns”, in Duhe S. (Ed), New Media and Public Relations (second edition, pp. 189-197). Peter Lang, New York.

Wood, M. (2012), “Marketing social marketing”, Journal of Social Marketing, Vol. 2 No. 2, pp. 94-102.

Wymer, W. (2011), “Developing more effective social marketing strategies”, Journal of Social Marketing, Vol. 1 No. 1, pp. 17-31.

Corresponding author Richard D. Waters is the corresponding author and can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

JSOCM 4,3

260

D ow

nl oa

de d

by N

ew Y

or k

U ni

ve rs

it y

A t

16 :4

0 04

N ov

em be

r 20

14 (

P T

)