Understanding Destination Management Organizations Use of Twitter: A Content Analysis of Tweets

UNDERSTANDING DESTINATION MANAGEMENT ORGANIZATIONS USE OF TWITTER: A CONTENT ANALSYIS OF TWEETS

Chris Gibbs

Assistant Professor Ryerson University

Ted Rogers Institute for Tourism and Hospitality Research [email protected]

Anita Dancs

Undergraduate Student Ryerson University

Ted Rogers Institute for Tourism and Hospitality Research [email protected]

This study contributed to the theme of “doing more with less” by using innovative technology such as Ncapture and Nvivo to uncover the use of the social media platform Twitter in the tourism industry.

Refereed Paper Submission to the 2013 TTRA Canada

Enquiries should be addressed to Chris Gibbs

UNDERSTANDING DESTINATION MANAGEMENT ORGANIZATIONS USE OF TWITTER: A CONTENT ANALSYIS OF TWEETS ABSTRACT Social media platforms like Twitter represent a new channel of communication for attracting visitors.

Research about Twitter in a tourism capacity at this time is limited. This exploratory study attempts to use content analysis to understand how Canadian destination marketing organizations use Twitter. Initial findings suggest that Twitter is used to share links with information and be conversational with Twitter users. The usage of Twitter by different organizations varies significantly, suggesting that this mode of communication is still evolving and could benefit from research. In an increasingly competitive market to attract visitors, this study will provide insight for the use of Twitter by destination marketing organizations. To the best of our knowledge this is the first such content analysis study conducted about destination marketing organizations’ use of Twitter.

INTRODUCTION

The global tourism market is highly competitive and saturated with more destinations

using more sophisticated channels of marketing than ever before. Social media has become an

important element of marketing and communications strategies for destination marketing

organizations (DMO’s). Along with information search, social media has been identified as one

of two “mega trends” that will significantly impact the tourism industry (Xiang & Gretzel, 2010).

Within a destination marketing organization context the Canadian Tourism Commission (CTC)

has identified social media as an important platform to their overall marketing and

communication strategy (2013). Hotels and other tourism businesses have embraced social

media platforms like Facebook, Twitter, YouTube and Flickr into their marketing channels,

featuring links prominently on website home pages (Leung, Law, van Hoof, & Buhalis, 2013).

Despite the adoption and usage of social media within the tourism industry the practices of

managing social media are largely unknown to both scholars and practitioners.

From a practitioner perspective, thousands of how-to books and guides are available to

learn from. From a scholar perspective, recent research has started to emerge that links the use of

social media to national tourism organizations (Dwivedi, Yadav, & Venkatesh, 2012) and

identifies a significant increase in tourism and hospitality peer reviewed published articles

related to social media (Leung et al., 2013). Both of these recent studies confirm the increased

interest for social media research amongst industry practitioners and academic researchers.

While social media is starting to attract attention amongst academic researchers, it could still be

considered a research topic in its infancy (Leung et al., 2013). It has been suggested that

quantitative content analysis be used to focus on specific sectors in order to provide a clearer

picture of social media use in hospitality and tourism (Leung et al., 2013).

Following the adoption (Dwivedi et al., 2012), the increased use by industry, and the

infancy stage for Twitter research, this inquiry will attempt to provide further insight into social

media use by destination marketing organizations. Rather than focus on multiple different social

media platforms like previous research has, this research will focus exclusively on the use of

Twitter by Canadian DMO’s.

LITERATURE REVIEW

Twitter is an increasingly popular form of communication. It started in 2006 with the idea

to share “What are you doing?”. Initially Twitter was designed as a form of micro blogging

where users can share their current status through instant messages, mobile phones, e-mail and

the Web. Compared to weblogs which feature short essays, a tweet is only 140 characters and is

designed to be a much faster form of communication. Since the introduction of Twitter in 2006,

it has grown to be one of the most popular social media platforms with over 200 million active

users (Twitter, 2013).

What has made Twitter an influential social media has been its initial design as a mobile

form of communication and its ability to quickly transmit information. The Pew Internet and

American Life Project estimated that 15% of American online adults used Twitter in 2012,

compared to only 8% in 2010 (Smith & Brenner, 2012). A rise in smartphone adoption is

suspected to be one of the driving forces for increased usage of Twitter (Smith & Brenner, 2012).

Unlike other social media platforms, Twitter has become a source for news by mainstream media

like newspaper and television stations (Mitchell, Rosenstiel, & Christian, 2012; Schultz &

Sheffer, 2010). The mainstream media’s use of Twitter as a source for news amplifies the

importance of Twitter as a marketing tool.

While DMO’s like the Canadian Tourism Commission (CTC) have stated that social

media channels stimulate people to share stories about Canadian travel experiences, limited

research has been done to identify how this goal is achieved. A recent literature review identified

44 published articles about social media in a tourism or hospitality context that were either from

the consumer or supplier perspective (Leung et al., 2013). Amongst the supplier perspective no

studies exist where tweets are a unit of analysis. From a consumer perspective the only Twitter

study identified a need for trained persons to monitor and respond to reviews across multiple

platforms plus the need for communicating directly with visitors to stimulate dialogue (Sotiriadis

& van Zyl, 2013).

Although Twitter messages, known as tweets, are limited to 140 characters, Twitter is

able to influence news, opinion, search, and advertising through end-user innovation (Johnson,

2009). Users of Twitter have created conventions for use such as RT for ReTweet, @ for reply,

short URL’s and # for hashtag trend finding (Boyd, Golder, & Lotan, 2010; Honey & Herring,

2009). In one of the earlier investigations about Twitter, a content analysis of 36,987 tweets

identified that the @ symbol expanded the types of use for Twitter beyond the “What are you

doing?”(Honey & Herring, 2009). In a dataset of 720,000 tweets from January to June 2009,

22% of tweets included a URL, 36% identified a user using the @ symbol, and 5% of tweets

used a hashtag (Boyd et al., 2010). With the increased popularity of Twitter and the ability for

Twitter to serve multiple purposes, applications have evolved to help suppliers manage their

content to meet the needs of their audience. Services such as HootSuite.com and Tweetdeck.com

help users manage profiles across different social media platforms, Bit.ly shortens long web

address for easy tweeting, and Buffer schedules tweets at the optimal time of day for Twitter

followers.

Based on the increased usage of Twitter and growing interest for researching social

media there is a gap in knowledge about Twitter use. Increasingly DMO’s and other Twitter

users are using tools to help them manage their social media applications. However, due to a lack

of understanding about how DMO’s are using Twitter, limited information exists to improve the

use of Twitter as a tool for sharing destination travel experiences.

METHOD

The methodology adopted was exploratory and interpretive in nature. The research

method used a two stage approach to collecting data to understand the use of Twitter by

Canadian DMO’s. With 13 provinces and territories and 709 communities with more than 5,000

people (Statistics Canada, 2011), a sampling framework was established that focused on all

provincial or national DMO’s plus cities with more than 500,000 people. Including the CTC, the

study reviewed Twitter use for 21 Canadian DMO’s.

Stage One. Stage one of the data collection attempted to understand a user profile for

each of the DMO’s. Data was collected on June 25, 2013 for different DMO’s Twitter accounts

using www.twitter.com and www.followerwonk.com. This initial collection was designed to

understand the DMO’s style and use of Twitter.

Stage Two. Stage two of the data collection used NCapture, a service of the qualitative

analysis software Nvivo to collect the tweets from the 21 Canadian DMO’s in the sample.

Initially, the sample size consisted of 58,949 tweets collected from May 6, 2009 to July 17, 2013.

The most current month of tweets, June 2013 was selected for analysis. This reduced the sample

size to 4,663. The new sample size was still deemed too large and time consuming; therefore a

systematic approach was undertaken, considering every fourth tweet. In total 1,166 tweets were

collected for quantitative content analysis. This collection of tweets was designed to understand

Canadian DMO’s use of Twitter.

From the collection of tweets, a content analysis was chosen to categorize the use.

Content analysis is commonly used to analyze tweets to understand usage. Early seminal

research using content analysis of tweets gathered large datasets with 150,000 or more tweets to

gain insight into the application of Twitter (Jansen, Zhang, Sobel, & Chowdury, 2009; Java,

Song, Finin, & Tseng, 2007; Naaman, Boase, & Lai, 2009). Later research about Twitter using

content analysis attempted to understand Twitter use within a sample of user types. Compared to

the early seminal research, the datasets were a much smaller sample size of tweets and focused

on a coding structure specifically related to the type of user. Content analysis has been used to

investigate Twitter use by professional athletes (Hambrick, Simmons, Greenhalgh, & Greenwell,

2010), libraries (Aharony, 2010), United States congress people (Golbeck, Grimes, & Rogers,

2010), and city police (Heverin & Zach, 2010).

From the Twitter content analysis research papers cited, different coding schedules and

formats were used for each, suggesting that there is not one standard coding schedule available

for understanding Twitter use. For this research a six step content analysis approach was used to

identify how Canadian DMO’s use Twitter (Hansen, 1998). To assist with defining the analytical

categories for coding, the Twitter content classification framework developed by Dann (2010)

was used as a guiding framework. Dann considered the following categories: conversational,

status, pass along, news and phatic. Dann used 2,841 tweets from March 2007 to August 2009

and found that the most common usages were conversational, consisting of responses (30%) and

queries (17%). A Twitter content analysis framework for Twitter had been done in the past by

Murat & Ummuhan (2010), but the only information available about their research was an

abstract. The abstract identified 11 categories for European DMO’s usage of Twitter:

announcements, information about local travels, information/news about local business,

information about destination/travel, contests, replies to followers, links to website, social tweets,

encouragement to write and share comments/photos, re-tweets and other. While the categories

fit with the tourism industry, no information about the coding schedule was available to replicate

the coding framework used.

Based on a combination of the coding frames identified by Dann (2010) and the coding

uses from Murat & Ummuhan (2010), a coding schedule with a primary category and a sub-

category was developed. A sample of 800 random Tweets was used to finalize the coding

schedule. Each tweet was coded with a primary category similar to the coding frames used by

Dann (2010); conversational, promotions, status and uncodeable (see Table 2). After the primary

category was coded, the tweets were coded with a sub-category to provide further insight into

how Twitter is used. Once the coding schedule was tested and considered reliable, two

undergraduate students were trained to code the tweets. The two students worked independently

coding the same 1,166 Tweets. The final percentage of agreement between the two coders was

87%, suggesting that the coding classification was reliable.

RESULTS

The first part of the analysis addresses the style of use for each DMO Twitter account and

the second part of the analysis focuses on the analysis of a collection of Tweets to understand

how DMO’s use Twitter.

Stage One. The stage one collection was designed to understand the DMO’s style and

use of Twitter conventions. Using the metric for style of use created by Java (2010), Canadian

DMO’s are using Twitter for information sharing purposes (high followers, low following)

versus information seeking or friendship-relationship purposes. The different level of Follower

(%) when you compare City (90%) to Provincial (75%) DMO’s is interesting (Table 1). It

appears that the city DMO’s are using Twitter to push out content rather than seek information

from their followers.

As a benchmarking exercise, the different DMO’s were ranked according to their

Followerwonk Social Authority. Social Authority is a proprietary measure created by

Followerwonk.com to measure influence on Twitter. The score is based on a collection of

metrics beyond just number of followers or the most active Twitter accounts. The score is based

on retweet rate, time decay and a series of user characteristics (i.e. follower count) optimized

using a regression model. The higher the Social Authority, the more engaging the DMO user is

on Twitter. By ranking DMO’s, similar DMO’s can see which Twitter accounts are the most

engaging. Upon reviewing the findings in Table 1, it is not always the DMO account with the

most followers or most frequent user of Twitter that is ranked higher. You also notice that the

size or popularity of a destination does not necessarily suggest more influential Twitter use. As

an example Tourism Toronto would be considered one of the biggest Large city DMO’s,

however their Social Authority ranking has them in the low category when compared with

others. Depending upon the circumstances, this could be deemed a weakness in their marketing

communications area.

Regular use of Twitter on a daily basis is very different. You have one DMO who

Tweets less than once per day and another that Tweets 17.8 times per day. The overall average

DMO Twitter account has been open for more than three ears and nine months and tweets about

four times per day.

To help DMO’s understand the most active hours for their Twitter followers, the

www.followerwonk.com application was used to collect data. For each DMO Twitter account,

the most active hours of use by followers was collected. The results of the 21 DMO’s were

amalgamated in order to identify and overall most active time of day for Twitter follower activity

(See Figure 1). The findings suggest that followers of Canadian DMO’s are most active between

9 am and 6 pm with more than 50% of the activity occurring during that time. The findings also

suggest that Twitter followers are also still active in the evening when typical DMO offices are

closed, suggesting that solutions to ensure evening Twitter activity should be considered.

Table 1

DMO Twitter Use Characteristics

Relative Tweets

Followers Social Social Total Days on per

Provincial/National DMO's Followers % 2 Authority 3 Ranking Tweets Twitter Day

Alberta 81,934 99.2 70 High 7,172 1,807 4.0

British Columbia Tourism 43,042 98.4 67 High 8,275 1,775 4.7

Canadian Tourism Commission1 16,258 58.3 55 Medium 11,386 1,629 7.0

Nova Scotia Tourism Agency 11,167 82.6 55 Medium 7,837 1,524 5.1

Tourism Ontario 7,377 73.3 53 Medium 4,135 869 4.8

New Brunswick Tourism 6,192 67.3 53 Medium 6,270 1,173 5.3

Manitoba Tourism 7,534 78.5 50 Medium 2,832 1,221 2.3

Newfoundland Labrador Tourism 12,431 62.6 49 Medium 4,390 1,509 2.9

Saskatchewan Tourism 12,552 97.2 47 Medium 3,137 1,546 2.0

Tourism PEI 5,991 53.0 39 Low 3,139 1,551 2.0

Quebec1

1,654 73.0 38 Low 4,482 785 5.7

Average Provincial/National DMO's 18,739 76.7 52 5,732 1,399 4.2

Large City DMO's

Tourism Calgary 13,615 88.3 64 High 7,321 1,621 4.5

Tourism Hamilton 7,067 98.5 62 High 15,802 886 17.8

Tourisme Montreal 1 32,173 98.8 59 Medium 3,641 1,392 2.6

Tourism Vancouver 39,945 98.7 57 Medium 6,142 1,718 3.6

Tourism Winnipeg 8,261 89.3 54 Medium 5,374 1,545 3.5

Ottawa Tourism 10,396 99.6 51 Medium 4,580 1,368 3.3

Tourism Toronto 5,105 94.3 39 Low 5,350 1,499 3.6

Edmonton Tourism 3,699 85.9 34 Low 1,969 1,590 1.2

Quebec City 1 1,307 60.9 21 Low 600 1,105 0.5

Average Large City DMO's 13,508 90.5 49 5,642 1,414 4.5

Average All DMO's 16,385 82.9 51 5,692 1,406 4.3 1 Rankings only looked at English Twitter Accounts

2 Followers % is a clacultion of Total Followers / Total Followers + Total Following

3 Social Authority is a measure of Tiwtter influence calcuated by www.followerwonk.com

Figure 1: Followerwonk Canadian DMO Most Active Hours

Stage 2. While stage one was designed to identify general style and use of Twitter by

Canadian DMO’s, stage two was designed to reveal the most common types of communication

used. The coding of 1,166 Tweets identified that the Twitter could be considered a promotional

platform for DMO’s because almost 60% of Tweets are about the promotion of the destination

(See Table 2). Beyond promotional messages, almost 24% were conversational and 15% were

status messages. These results are in contrast to previous studies that suggest Twitter is used for

conversations or to communicate status (Dann, 2010). This difference in use could be related to

the purpose driven use of social media by business entities versus the individual driven use of

social media by non-business users.

Upon further investigation through the use of sub-categories, more understanding of how DMO’s

use Twitter can be identified. Within the conversation category, almost 70% of Twitter activity is

related to a response to user communication. Based on this finding, it could be argued that

Twitter is a platform to interact and have conversations with followers. By far the most used sub-

category within promotion was destination (43%) and future event (28%). While the use of

Twitter to promote the destination could be expected, the use of it for promoting future events

identifies a unique use of Twitter by DMO’s. When considering Twitter use, event producers

need to use their DMO partners to help get the message out in the community through Twitter.

his event use of Twitter was also present in Status tweets. More than 60% of status Tweets

related to event use.

One dominant theme in the use of Twitter by Canadian DMO’s is the presence of links to

other content on the Internet. From the 1,166 Tweets coded, almost two out of every three tweets

provided followers with a URL link to an external source. The tweets were frequently linking

with websites (21%), event website (5%) or photo websites (17%).

Upon review of the different links from Twitter, more than 10% of links were not

working (see Table 4). Further investigation revealed that the majority of links not working were

a result of retweeting content. Links were frequently placed at the end of the tweet and the last

remaining characters in the link were cut off because of the 140 character limit on Twitter. When

Table 3

DMO Tweet Coding Results; Sub-categories (n=1,166)

Category Sub-Category # Tweets % of Tweets Description Sample

Conversational Action 11 1.8 Activities involving other Twitter users or implying the

user to take an action.

@2012blogger you can contact @Tjerven with your

info

Query 8 1.3 Questions, question marks or polls. @ChingRae What are your plans this weekend in

Calgary?!

Referral 22 3.5 An @response which contains URLs or

recommendation of other Twitter users. (Excludes RT

@user).

@annora Some great restaurants: @The_Woods_,

Rook & Raven, and Il Secondo! @VisitSaskatoon

will have more ideas. Enjoy your visit!

Response 90 14.4 Replying to a Tweet. A reply can have mutiple @

symbols.

@AndrewFRESHFM @953FreshFM Thanks for the

chat! :)

Promotion Attraction 49 7.8 Any tweet that promotes a touristic attraction

(museums, parks, zoos etc.)

Did you know? @CalgaryZoo penguins eat 1,200

kg's of fish per month. #unknownyyc #yyc

http://t.co/wq9gbZXv9P

Contest 34 5.4 Any tweet that offers details about a contest or

announces the winners of a contest

Your chance 2 win a @ResortsOntario getaway: Stop

by a Travel Centre to enter & find out more:

http://t.co/Lwp1MQGFc9 #DiscoverOntario

Destination 177 28.3 Any tweet that focuses on the overall promotion of the

tourist destination or multiple attractions/itineraries.

Just launched: our U.S. marketing campaign. Toronto

the Untamed Metropolis. http://t.co/1Pa0k4yyUT

Future Event 113 18.1 Any tweet presenting information about a future event

(festivals, concerts, exhibitions etc.)

#Luminato2013 in Toronto celebrates music, dance,

theatre, visual arts + more. Make a date for it June 14-

23 http://t.co/k2sb6yKHRJ ^ML

Local Business 36 5.8 Any tweet that presents Information/news about a

local business that would not be considered a touristic

attraction

Sweet. Aunt Sarah's Chocolate Shop in Trinity is now

open for the season. http://t.co/oiAueAcaEN ^EK

Status Event 51 8.2 Any tweet which presents updates of an event

occuring in the moment that was tweeted.

#winnipeg getting ready for @pridewinnipeg #pride

#lgbt #pridefestival. http://t.co/GRxLcHaEui

Referral 33 5.3 Any tweet that is not adressed to another Twitter

user, but makes reference to something non-

promotional. Includes retweets which make reference

to another Twitter user or provide links.

Hungry? @offthebeatenpal has been eating her way

around #Alberta. Take a look at what she found -

http://t.co/TSHHIPwGbR

Weather 1 0.2 Report of weather conditions. #HamOnt #Weather today - A mix of sun and cloud,

chance of showers and a risk of a thunderstorm with a

high of 25, feels like 32.

Table 4

DMO Tweets and Links

Twitter Links Tweets %

No Link 432 37.0

Link to Website 253 21.7

Link to Event Website 63 5.4

Link to Photo 200 17.2

Link to Blog 91 7.8

Link to Video 39 3.3

Links Cut/Not Working 88 7.5

DMO’s retweet, it is recommended they shorten it through bit.ly or include the link in the middle

of a message.

A popular category for linking Tweets was the sharing of photographs with followers

through Twitter. More than one in four Tweets with a link directed the Twitter follower to a

photography. The most popular photo sharing service used was the system built into Twitter.

Linking photos to Instagram was more popular than linking them to Facebook, which shows that

DMOs understand that their followers have time constraints and quickly browse their feeds. One

interesting pattern of use identified was the retweeting of pictures by followers while they are at

the destination. By taking this strategy, DMO’s are promoting the destination itself, seen through

the eyes of their travellers.

CONCLUSION

Twitter is an emerging communications platform for Canadian DMO’s. Canadian

DMO’s are using Twitter for information sharing purposes (high followers, low following)

versus information seeking or friendship-relationship purposes. The frequency, followers and

influence of Twitter by Canadian DMO’s varies from organization to organization. This initial

benchmarking of Twitter will help Canadian DMO’s understand their use in comparison to

similar organizations and can help drive results by providing users with actionable insights

related to their use. A comparison with other Twitter users can help Canadian DMO’s

understand time of day and frequency of use patterns for Twitter.

The analysis of 1,166 Tweets identified that the primary use of Twitter is for promotional

purposes and that almost two in three Tweets are designed to link followers to other content on

the Internet. Two common uses of Twitter are the sharing of photographs and linking

information related to future and current events happening in the destination.

The current study examined the information contained in the DMO tweets, a future study

could investigate the extent to which content categories can predict Twitter use by DMO

followers. Such a study could examine whether followers are more interested in information

about future events or conversations with other users and the extent categories of use predict

frequency and time spent using Twitter. Results from a user focused study could provide DMO’s

Table 6

Most Common Photo Sharing Sites Used

Twitter Links Tweets %

Twitter 117 58.5

Instagram 36 18.0

Facebook 21 10.5

ow.ly 17 8.5

Flickr 5 2.5

Tumblr 1 0.5

Other 3 1.5

with insightful information about information Twitter users seek. This knowledge would help

DMO’s adjust their use of Twitter to better meet the needs of followers.

REFERENCES

Aharony, N. (2010). Twitter use in libraries: an exploratory analysis. Journal of Web Librarianship, 4(4), 333-350.

Boyd, D., Golder, S., & Lotan, G. (2010). Tweet, tweet, retweet: Conversational aspects of retweeting on twitter. Paper presented at the System Sciences (HICSS), 2010 43rd Hawaii International Conference on.

Canadian Tourism Commission. (2013). 2012 Annual Report. Vancouver, British Columbia. Dann, S. (2010). Twitter content classification. First Monday, 15(12), 1-12. Dwivedi, M., Yadav, A., & Venkatesh, U. (2012). Use of Social Media by National Tourism

Organizations: A Preliminary Analysis. Information Technology & Tourism, 13(2), 93-103.

Golbeck, J., Grimes, J. M., & Rogers, A. (2010). Twitter use by the US Congress. Journal of the American society for information science and technology, 61(8), 1612-1621.

Hambrick, M. E., Simmons, J. M., Greenhalgh, G. P., & Greenwell, T. C. (2010). Understanding professional athletes' use of Twitter: a content analysis of athlete tweets. International Journal of Sport Communication, 3(4), 454-471.

Hansen, A. (1998). Content analysis Mass Communication Research Methods (pp. 91-129). New York: New York University Press.

Heverin, T., & Zach, L. (2010). Twitter for city police department information sharing. Proceedings of the American Society for Information Science and Technology, 47(1), 1-7.

Honey, C., & Herring, S. C. (2009). Beyond microblogging: Conversation and collaboration via Twitter. Paper presented at the System Sciences, 2009. HICSS'09. 42nd Hawaii International Conference on.

Jansen, B. J., Zhang, M., Sobel, K., & Chowdury, A. (2009). Twitter power: Tweets as electronic word of mouth. Journal of the American society for information science and technology, 60(11), 2169-2188.

Java, A., Song, X., Finin, T., & Tseng, B. (2007, August 12,). Why we twitter: understanding microblogging usage and communities. Paper presented at the 9th WebKDD and 1st SNA-KDD 2007 workshop on Web mining and social network analysis., San Jose, California.

Kwak, H., Lee, C., Park, H., & Moon, S. (2010, April 26-30, 2010). What is Twitter, a social network or a news media? Paper presented at the 19th international conference on World Wide Web (pp. 591-600). ACM., Raleigh, North Carolina, USA.

Leung, D., Law, R., van Hoof, H., & Buhalis, D. (2013). Social media in tourism and hospitality: A literature review. Journal of Travel and Tourism Marketing, 30(3-22), 3-22.

Mitchell, A., Rosenstiel, T., & Christian, L. (2012). What Facebook and Twitter Mean for News. from http://stateofthemedia.org/2012/mobile-devices-and-news-consumption-some-good-signs-for-journalism/what-facebook-and-twitter-mean-for-news/

Naaman, M., Boase, J., & Lai, C. H. (2009). Is it Really About Me? Message Content in Social Awareness Streams.

Schultz, B., & Sheffer, M. L. (2010). An exploratory study of how twitter is affecting sports journalism. International Journal of Sport Communication, 3, 226-239.

Smith, A., & Brenner, J. (2012). Social networking: Twitter update 2012. Washington, DC: Pew Internet & American Life Project.

Sotiriadis, M. D., & van Zyl, C. (2013). Electronic word-of-mouth and online reviews in tourism services: the use of twitter by tourists. Electronic Commerce Research, 1-22.

Twitter. (2013, September 30). Who's on Twitter. Retreived from https://business.twitter.com/whos-twitter

Xiang, Z., & Gretzel, U. (2010). Role of social media in online travel information search. Tourism management, 31(2), 179-188.