This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Mental Health, is properly cited. The complete bibliographic information, a link to the original publication on http://mental.jmir.org/, as well as this copyright and license information must be included.
It has been suggested that some dominant aspects of traditional masculinity are contributing to the high suicide rates among Australian men. We developed a three-episode documentary called
This study aimed to examine the extent to which the
We used Twitter insights data to assess the reach of and engagement with the campaign (using metrics on followers, likes, retweets, and impressions) and to determine the highest and lowest performing tweets in the campaign (using an aggregated performance measure of
The campaign generated a strong following that was engaged with the content of the campaign; over its whole duration, the campaign earned approximately 5000 likes and 2500 retweets and gained around 1,022,000 impressions. The highest performing tweets posted by the host included video footage and occurred during the most active period of the campaign (around the screening of the documentary). The volume of conversations in relation to commonly used hashtags (
The
In Australia, suicide is the leading cause of death in males aged 15 to 45 years [
A number of factors have been found to contribute to higher suicide rates in men. Men are known to choose more lethal methods [
It has been suggested that masculinity—or the rules prescribed by society about how men should live their lives [
Seeking to change the picture on male suicide may benefit from challenges to some of these widely accepted male stereotypes. Discussing dominant masculinity and creating opportunities for redefining help-seeking strategies for men and opening up options for negotiating difficult life events can potentially have significant impacts. However, changing social norms is no easy feat, and requires holistic population-based interventions that are able to reach and engage with the wider community of men from all walks of life, backgrounds, and geographic locations. Developing and testing such interventions have the potential to take the field of suicide prevention forward; at present, only a relatively small number of population-based interventions have been shown to be effective (eg, restricting access to means and school-based awareness campaigns) [
We developed one such intervention, with funding from the Movember Foundation. We collaborated with Heiress Films to create a three-episode television documentary called
From the beginning,
The social media campaign was a crucial component of the overall intervention. There is increasing recognition that social media may have potential in suicide prevention, and may be particularly useful for otherwise hard-to-reach groups such as men [
This study focuses on the Twitter activity that was generated by the
The study addresses the following evaluation questions:
What was the overall reach of the campaign and how did the audience engage with it?
What were the highest and lowest performing tweets and what assets were associated with them?
Did the Man Up campaign increase the volume of relevant Twitter conversations and, if so, was the increase sustained after the show?
What were the main content themes with regard to audience engagement?
We used Twitter data to answer our evaluation questions. These data are easy to access and represent a real-time response, making them ideal for monitoring responses to events, patterns of communication, and general attitudes [
We collected Twitter data from two sources, one via the social media tool
During the campaign, we downloaded weekly data reports from
We harvested original content tweets from a broader period to capture activity in the 14 weeks before the campaign (May 9 to August 14, 2016), the 14 weeks during the campaign (August 15 to November 20, 2016), and the 14 weeks after the campaign (November 21, 2016 to February 26, 2017). Tweets were harvested using the free-of-charge Twitter application programming interface and were included in the dataset if they used the hashtag #MANUP, which was the main hashtag used in promoting the campaign. These data included the full text of each tweet and additional information on when (eg, time and date) and by whom (eg, host, organization, private person, and public person or forum) it was tweeted. Data were stored in an external holding database by a US company called Rackspace. We had access to the data and could download customized datasets throughout the entire observation period (pre, during, and post campaign). A final dataset was imported into Excel, and a subset of that dataset was then imported into the qualitative data analysis software NVIVO Pro V11 developed by QSR International.
All quantitative analyses were undertaken in Excel, and all qualitative analyses were performed in NVIVO Pro V11.
For evaluation question 1, we assessed the reach of and engagement with the campaign by using metrics from the Twitter insights data on followers, new followers, likes, retweets, and impressions (the number of people who saw campaign tweets on their timeline). We calculated frequencies, averages, and percentages for each as relevant, doing so for each of the three phases of the campaign.
For evaluation question 2, we determined the highest and lowest performing tweets in the campaign by ordering all tweets posted by
For evaluation question 3, we used the original content tweets to determine whether the campaign increased the volume of relevant Twitter conversations, aggregating the number of tweets for selected campaign hashtags (
Data sources by time frames and evaluation questions covered.
We calculated the average number of tweets per period and used
Reach of and engagement with the
Indicator | Prescreening of documentary |
During screening of documentary |
Post screening of documentary |
|
Frequent followers, n | 519 | 1355 | 1453 | |
Frequent new followers, n | 519 | 836 | 98 | |
Frequent likes, n | 1500 | 2500 | 851 | |
Average likes,n | 26 | 118 | 41 | |
Frequent retweets, n | 656 | 1300 | 417 | |
Average retweets, n | 12 | 64 | 20 | |
Frequent impressions, n | 423,000 | 436,000 | 163,000 | |
Male | 379 (73.0) | 813 (60.0) | 857 (59.0) | |
Female | 140 (27.0) | 542 (40.0) | 596 (41.0) |
The original content tweets included 46,130 tweets that used the hashtag
Because
Engagement with tweets by asset type for the 20 highest and lowest performing tweets. Data source: Twitter insights.
Asset type | Frequency, n (%) | Sum of reactionsa | Sum of retweets | Sum of replies | Sum of likes | Sum of media views | |
Video | 12 (60) | 14,069 | 240 | 28 | 395 | 12,856 | |
Graphic | 5 (25) | 1599 | 148 | 26 | 293 | 542 | |
Link | 2 (10) | 2205 | 43 | 3 | 62 | 2014 | |
GIFb | 1 (5) | 897 | 2 | 0 | 6 | 872 | |
Total | 20 (100) | 18,770 | 433 | 57 | 756 | 16,284 | |
Link | 18 (90) | 62 | 2 | 0 | 42 | 0 | |
Graphic | 1 (5) | 3 | 0 | 0 | 2 | 1 | |
Video | 1 (5) | 3 | 0 | 0 | 1 | 0 | |
Total | 20 (100) | 68 | 2 | 0 | 45 | 1 |
aAggregate of all engagement including video views.
bGIF: Graphics Interchange Format .
Hashtag frequencies on Twitter before, during, and after the campaign based on original content tweets.
We identified a subset of 2093 tweets that included the hashtag
RT @ManUpTVSeries: Great to see the conversation getting started. #ManUp #SmashTheStigma #itsokaytotalk
@newlz in @HuffPostAU Talking. Listening. Sharing. These are the tenets that now drive me. #ManUp #weneedtotalk
@username Except if you’re a male victim of #domesticviolence - then you get told to #ManUp and discriminated against #Reality #ABCManUp
Several content themes were identified in the tweets: expressing emotions, mental health issues with the subthemes of mental health and suicide, men’s issues with the subthemes of being a man and fathering and raising boys, help-seeking with the subthemes of providing options for help and other help-seeking, personal stories, and supporting others.
RT @ManUpTVSeries: The need for men to be emotionally honest is greater than ever. Interesting #blog post from @3DMathW. #ManUp
RT @username: Not ashamed to admit a few tears have been shed watching #ManUp over the last few weeks #ABCManUp
RT @ManUpTVSeries: The strong silent type might be sexy in films, but it’s unhealthy in real life. #ManUp #itsokaytotalk
Content themes of the 1876 campaign-related tweets. Data source: Original content tweets.
Theme | Description | Tweet count |
Retweet |
Original tweet |
|
Expressing emotions | Tweets that include topics such as speaking up, opening up, talking about uncomfortable issues, breaking down stigma, and crying | 710 (100.0) | 550 (77.5) | 160 (22.5) | |
Mental health | Tweets that indicate mental health, depression, or posttraumatic stress disorder, or include materials and links that discuss these topics, or use the hashtag mental health | 410 (100.0) | 300 (73.2) | 110 (26.8) | |
Suicide | Tweets that indicate suicide or suicide prevention, or include materials and links to sites that discuss these topics, or use the hashtag suicide | 182 (100.0) | 140 (76.9) | 42 (23.1) | |
Being a man | Tweets that discuss the concept of masculinity or challenge the concept of masculinity | 165 (100.0) | 125 (75.8) | 40 (24.2) | |
Fathering and raising boys | Tweets that encourage discussion about what it means to be a father and raising boys | 82 (100) | 63 (77) | 19 (23) | |
Providing options for help | Tweets that provide information on help services and encourage their use | 165 (100.0) | 147 (89.1) | 18 (10.9) | |
Help-seeking other | Tweets mentioning other content on help-seeking (ie, not about providing options for help) | 96 (100) | 79 (82) | 17 (18) | |
Personal stories | Tweets relating personal stories, written in the individual’s own voice, and revealing detail about the person (not commentaries or statements) | 101 (100) | 61 (60) | 40 (40) | |
Supporting others | Tweets about providing support to others, support options and general advice | 78 (100) | 63 (81) | 15 (19) |
The theme of
Massive shout out to @GusWorland. As sufferer of PTSD for 8 years as Ex-Cop #ManUp really hit home hard. Congrats mate. #ManUp #SpeakUp
RT @username: Depression is an illness people can help you recover from #ManUp on ABC at the moment is great. #mentalhealthweek
Suicide has touched so many lives, I’m tearing up already #ABCManUp #manup #blackdoginstitute #lifeline #beyondblue
RT @ManUpTVSeries: How suicide can become “contagious” to other at-risk young men. Important piece in @DailyMailAU. #ManUp #SpeakUp
Inspired by the #ManUp TV series, we have a chat about what it means to be a MAN https://t.co/0a6wqtimAP #Movember #mentalhealth #goodcause
RT @ManUpTVSeries: “Our ideals of #masculinity have shifted.” @MichaelGLFlood is one of our #RealAussieblokes. #ManUp…
RT @OliShawyer: This ad made me cry. I’m covered in Tatts. I ride a Harley. And I’m crying. Try tell me that’s weak. #manup #speakup thank you @gusworland
Tweets on
RT @ManUpTVSeries: We need to have a hard look at how we raise our boys. #ManUp #raisingboys #ChildHealthDay @harkin_tom
RT @username: #ABCManUp #ManUp Let’s start helping boys from a young age. Dads need to give them cuddles, talk about feelings. #natural
RT @Top_Blokes: Providing boys with positive older male mentors is important to keep them safe and alive #ABCManUp #ManUp
The theme of
RT @ManUpTVSeries: #RealTalk for a sec: if you or a mate are doing it tough please call @LifelineAust on 13 11 14 #ItsOkayToTalk #ManUp
The tweets in the theme
RT @ManUpTVSeries: #ManUp survey: over 56% of men would rather manage themselves than seek professional help. #weneedtotalk #questmh
RT @ManUpTVSeries: Shame could be a big reason why…some men [don't] ask for help. Beautiful #blog from @drmwroberts #ManUp #NoShame
Another key theme—
Absolutely opened my eyes to the daily struggles of both genders. #ManUp I gave my fiancé a big hug after watching that tonight #ABCManUp
I’ve lost 3 mates to suicide. Wish I noticed what they were going through. Don’t #Manup, seek help cuz there’re many out there #ABCManUp.
When I was younger, everything I did was bulletproof #RealAussieBlokes #ManUp #exercise.
The final theme was
Powerful stuff @ManUpTVSeries #ManUp #ABCManUp We have a way to go to support our young men on their journeys. It’s a tough world we live in.
Sometimes the most important thing is just to listen. @BeardedGenius in @JOE_co_uk. #ManUp #SpeakUp #ListenUp
Across all themes and subthemes, the majority of tweets were retweets rather than original tweets, indicating high levels of engagement. Proportionally, the highest percentage of retweets was for
We evaluated the extent to which the
Certain elements of the campaign were particularly successful. These included tweets relating to the CSA Gus created on screen that encouraged men and boys to reject the constraints of traditional masculinity and speak up if they were facing tough times, as well as tweets featuring the trailer and episode teaser videos. The conversations generated by the campaign aligned with its major themes of expressing emotions, mental health issues and suicide, being a man and fathering and raising boys, help-seeking, personal stories, and supporting others. Again, related to the CSA release, the most discussed theme was expressing emotions.
The large number of positive comments indicated great acceptance and endorsement of the documentary. Many tweets welcomed open discussion of masculinity and male suicide and embraced the call for men to open up and express their emotions. There was a sense that for some men, questions on male identity and masculine norms had been bubbling beneath the surface, and the campaign gave men permission to articulate these thoughts and emotions. For others, ideas around changing the way we look at masculinity and its link to suicide appeared to be new, thought-provoking, and even challenging. These differing perspectives added to the richness of the discussion.
The Twitter campaign occurred as part of a strategic multimedia campaign. It was rolled out around the documentary via three phases, each of which aligned with a specific goal, and it was one component of the broader campaign. A significant proportion of the content released by
As noted earlier, there is still much that is unknown about what works and what does not work in suicide prevention. There are relatively few interventions for which there is indisputable evidence of effectiveness [
Both datasets that we used here had certain limitations, and these should be considered in interpreting our findings. In the case of the Twitter insights data, the key limitation relates to our measurement of success. We used the standard metrics of numbers of followers, likes, and retweets, and we created an aggregate measure, which we termed “reactions” (retweets, replies, likes, profile clicks, URL clicks, hashtag clicks, expanded click, follows, and views) to rank tweets in terms of their performance over the duration of the campaign. The way we aggregated “reactions” is open to challenge, although, as noted above, we felt that it was a democratic approach. In addition, the fact that we monitored tweets’ performance over the duration of the campaign disadvantaged tweets from earlier in the promotion cycle as these had less exposure because of lower numbers of followers and generally lower engagement with the campaign. This comparison could be improved by monitoring the performance of each tweet over the same duration (eg, for the first 2 weeks after it was posted) and creating some sort of performance per follower weighting, but “leveling the playing field” in this way was beyond the scope of our current endeavors.
In the case of the original content tweets, the main limitation relates to the way in which we were able to capture tweets relating to the campaign. We monitored the use of the hashtag
There are also limitations associated with using Twitter data in general. These data present something of a skewed picture because they can only represent those who are active on Twitter. In Australia, only about 19% of Internet users use Twitter, and a majority of these are relatively young [
The
Australian Broadcasting Corporation
community service announcement
The authors would like to thank the Movember Foundation for supporting the development and evaluation of
None declared.