Kelly MacKay, Ph.D., Danielle Barbe, Ryerson University Your Festival in 140 Characters or Less*: Exploring Festivals’ Use of Twitter Christine Van Winkle, Ph.D., University of Manitoba Elizabeth Halpenny, Ph.D., University of Alberta This research was supported by the Social Sciences and Humanities Research Council
19
Embed
Your Festival in 140 Characters or Less Exploring Festivals’ Use of Twitter
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Kelly MacKay, Ph.D., Danielle Barbe,
Ryerson University
Your Festival in 140 Characters or Less*: Exploring Festivals’ Use of Twitter
Christine Van Winkle, Ph.D.,
University of Manitoba
Elizabeth Halpenny, Ph.D.,University of Alberta
This research was supported by the Social Sciences and Humanities Research Council
Introduction• In the increasingly crowded field of festivals, social
media platforms offer new channels for attracting, communicating, and engaging festival attendees
• 60% of Twitter’s 200 million active users log in using a mobile device at least once a month
• This study will examine the nature and degree of Twitter use by three popular Canadian festivals, before, during, and after festival production/participation
Festival Backgrounds• Data reported in this study originate from
three popular Canadian festivals that occurred in the summer of 2013: • Pride Toronto• Taste of the Danforth (Toronto)• Edmonton International Fringe Theatre
Festival
Festival Twitter Profiles
Festival Name Twitter Handle
N Follower
s
N Followin
gMember
Since
Total N of
Tweets
Total N of Tweets in
Study Period
Pride Toronto @PrideToronto 14340 262 01/26/20
09 3032 199
Taste of the Danforth
@Taste_Danforth
423 238 06/26/2013 403 292
Edmonton Fringe
@edmontonfringe
7616 1127 06/19/2009 1924 226
Data Collection• Data (i.e., tweets) represent three time points: one
week prior to the festival, during the festival, one week after the festival
• Data on the profiles of the festival Twitter accounts were collected from www.twitter.com and www.followerwonk.com
• Ncapture was used to collect a census of the tweets from each festival’s Twitter handle
• The top Tweets that included the hashtag (#) associated with each festival were collected from www.twitter.com.
• Tweets were transferred to IBM SPSS Statistics for coding and analysis.
• Inter-coder agreement was calculated using Cohen’s kappa (k = .860) based on a reproducibility reliability sample of 125 units
Categories and Definitions for Twitter Content Analysis – Nature of Tweet
Nature of Tweet
Definition Example Sources
Conversational
A tweet that directly addresses another user(s)
by asking/answering a question, involving them
in the Tweet, or using @_________
"@GaryLevyOnline you should totally come by, hang out and tweet with me
for a bit next weekend at #PrideTO!"
Gibbs & Dancs (2013)
Hays et al (2013)
Dann (2010)Java et al
(2007)Promotional A tweet
marketing/promoting an event, activity, contest, website, artist, etc. that
urges the user to partake in an action.
"There are SO many #PrideTO Affiliate Events!
Visit http://t.co/rRGWZ1LMa6 for more info"
Gibbs & Dancs (2013)
Hays et al (2013)
Informational
Any tweet that presents an update or live discussion
of an event, reports news, or provides information, without urging users to partake in an action.
"@PrideToronto Flag Raising today
at noon on City Hall's green roof.
Our special guest is Premier Kathleen Wynne #Topoli"
Gibbs & Dancs (2013)Dann (2010)
Hays et al (2013)
Categories and Definitions for Twitter Content Analysis – Nature of Tweet
Nature of Tweet
Definition Example Sources
Status Answers the Twitter question “What are you
doing now?”
"at the #PrideTO flag raising!"
Gibbs & Dancs (2013)
Dann (2010)Phatic Any tweet containing
statements of greetings to the broader Twitter community, textual
soliloquys/monologues, undirected statements of
opinion, or establish sociability rather than
communicating information or ideas
“We wish everyone a
Happy #PrideTO. We're proud to serve Canada's most diverse
city.”
Dann (2010)Merriam-
Webster (2013)
Unclassifiable
Tweets that do not belong in any of the categories
above
“Dub step diggery do!
#entertainment #PrideTO”
Gibbs & Dancs (2013)
Dann (2010)
Categories and Definitions for Twitter Content Analysis – Purpose of Tweet
Purpose of Tweet
Definition Example Sources
Information Sharing
Any tweet that provides information to followers about a particular event,
subject, idea, etc.
“Join us for @PrideToronto flag raising on Jun 24 at
12pm at City Hall. Lunch provided. All welcome.”
Java et al
(2007)
Information Seeking
Any tweet that asks/requests information
from follower(s)
“What are you most looking forward to this
weekend at #PrideTO?!”
Java et al
(2007)Engagement/Relationship Building
Any tweet used to engage, build a
relationship with, or express appreciation to a
follower(s)
Happy First Day of #Summer and
#PrideTO! Are y'all ready for this?
Java et al
(2007)
Other Tweets that do not belong in any of the categories
above
Today's DOMA ruling is a historic step forward for
#MarriageEquality
Findings - General• A total of 870 tweets were captured• 82.9% of tweets originated from festivals and 17.1%
from patrons using the “#festivalname”• 21.4% were retweets• The majority (60.3%) of tweets were retweeted 3
times or fewer• The three festivals produced 65.8% of the tweets• 55% of all tweets contained links
• 53.4% of links were to photos• 27.4% of links were to websites• Less frequent links were to videos (8.6%),
Facebook (7.4%), and contests (2.3%)
Findings - General• Nature of the tweet content
• 28.5% was promotional• 26.2% was informational• 24.9% was conversational• 13.7% was phatic• 6% was status• Less than 1% was not classifiable
• Purpose of the tweet• 69.8% was for information sharing• 22.2% was for friendship/relationship building• 4% was for information seeking • 4% was for “other”
Frequency of Festival Related @ Mentions
Mention Type Frequency %Festival Partner/Promoter 217 43.1Attendee/General Public 154 30.6Festival + Partner 57 11.3Festival + Attendee 23 4.6Partner + Attendee 23 4.6Festival 17 3.4Other 13 2.6Total 504 100
Findings• There was no significant relationship found between the nature of the
tweet and the stage of the festival• Differences across the festival stages were evident in other variables:
• Retweets• User Mentions (@________)• Links• Purpose
Pre
During
Post
28.7% of tweets
68.3% of tweets
3% of tweets
Patterns Pre-Festival• Retweets were more likely to occur pre-festival than in
any other stages (52.2%)• Festival + attendee mentioned together was twice as
likely to occur before (68.2%) vs. during (31.8%) the festival• “RT @darrylwolk Looking forward to @Taste_Danforth!
• The inclusion of links in the tweet was lower than expected before the festivals (23.6%)
• Links to websites were higher than expected pre-festival (38.5%)
• Information seeking purposes were more likely to occur before the festivals (45.7%)
• Relationship building purposes more likely to occur before the festival (55.5%)Pre During Post
Patterns During Festival
• The majority (68.3%) of tweets took place during the festival
• Retweets were less than expected during the festival (45.2%)
• Links to photos were most likely to occur during the festival (84.1%)
• Information-sharing purpose was more likely to occur during the festival (69.4%)
Pre During Post
Patterns Post Festival• 3% of tweets were post-festival• The inclusion of links were more likely to occur
after the festival (4.6%)• Friendship/Relationship building purposes were
less likely after the festival (1%)• “Other” purposes, tweets that did not belong in
any of the categories, were more likely post festival (5.7%)
Pre During Post
Conclusion• Consistent with the “in the here and now” essence
of Twitter• There was no significant relationship found between the
nature of the tweet and the stage of the festival, however other aspects did vary
• Pre-festival tweets reflect some anticipatory content:• “@kerryrwerry: YAY!! The Pride Parade is next
Saturday!! I'm so pumped! It's gonna be awesome! @PrideToronto”
• Differential use of links across festival periods• Website links pre-festival: Twitter used to
promote and direct people to the festival and related events
• Higher photo links during festival: In the moment happenings to promote specific events or actions at the festivals
Conclusion
• Festivals are not taking advantage of the unique opportunity to communicate with attendees via social media post festival
• Post visit communication has been noted as a valuable opportunity to maintain contact and encourage visitors to continue processing their on-site experience
• Festivals could consider information seeking and relationship building tweets after the festival to evaluate and encourage repeat attendance
• Further research should explore why this is not occurring and examine these and other marketing implications