In order to answer our research question, we must gather big data, which we will find scraping social media.
Social media platforms
Very popular social media platforms at the moment are Facebook, Instagram, Tumblr and Twitter. We consider the social media channels on which most users are active and share their thoughts and experiences are the most suitable for data scraping. We will be looking for hashtags, which are labels that identify messages on a specific topic.
Currently 41% of all Facebook members (1,5 billion) is an active member (Mander, 2015) and as they are seeing a decline in active members, it may not be as suitable for data scraping. Also from our own experience we see in our environment that most people use Facebook to read and scroll through their newsfeed, rather than updating their status or posting somewhere. However, we will use Facebook for our netnographic research. (Read about our netnography here.)
Tumblr is a popular microblogging platform with 555 million active blogs. (Statista, 2016) Many bloggers post pictures or GIF’s and there are also posts without images, but all posts have in common that they are labelled with hashtags.
Instagram reached 400 million monthly active users in September 2015. (Kharpal, 2015) In our own environment, many people use Instagram to post their daily life pictures, but also special occasions and experiences: It is not for no reason people say a picture is worth a thousand words. Of course, people will also use Instagram to go through their feed, but we reckon most users are active users. A big advantage of Instagram is that the use of hashtags is very popular, making it easy to collect big data.
In short, we are going to collect data from Tumblr and Instagram.
Please note: due to a policy change of Instagram since June 2016, the Instagram Hashtag Explorer is not working anymore. We did not succeed in finding an alternative tool. Therefore, we were not able to scrape data from Instagram, except for #VRgaming. Our apologies. The rest of the data will all be scraped from Tumblr.
Tools of our choice
The tool we will use for Tumblr is the TumblrTool, which can be found at digitalmethods.net. Because digitalmethods is one of Europe’s leading Internet Studies research groups, we believe they have developed the best and reliable tool available for scraping Tumblr.
The same reason applied for why we used the Instagram Hashtag Explorer from digitalmethods.net.
Keywords / Keyphrases & Hashtags
With scraping data and analysing it, we attempt to answer the sub-question “What are the current responses to Virtual Reality?” We will search for the following hashtags, divided in the following subcategories:
- Virtual reality gaming: Our research question looks at how the responses of the Dutch gamers have developed towards VR. We want to focus on the gaming aspect of Virtual Reality for the main research question, hence the VR gaming hashtag.
- Virtual reality: It is important for us to know what the responses in general are towards Virtual Reality. These hashtags are also the most likely to answer our sub-question.
- Virtual Reality Devices: Users are likely to ‘label’ their post with the device the post is about (using / looking forward to /etc.). Since we want to know what they think about Virtual Reality, we hope to find review-like posts when looking for the VR Devices.
Collecting the data
For every hashtag, we chose to pick 20 iterations, which is good for 400 items. We think that any smaller amount will not give a representative outcome and a bigger amount will give too much information to process. We decided to scrape the latest posts, instead of giving a timeframe because we are investigating the current responses. Therefore, the latest posts are indeed the most current.
One big difficulty is the changing of Instagram’s policy. Since June 2016, Instagram has changed its policy, making it impossible to run the Instagram Hashtag Explorer. Therefore, we have only been able to use this tool for #VRgaming. This could be a point of improvement, although it is not digitalmethods’ fault a policy has changed. Nevertheless, there could or should be some way to make Hashtag Exploring on Instagram possible again.
Visualizing the data
We have used the layout Force Atlas 2, because this is a very clear layout which give a nice overview. To enhance the readability and prevent the labels from fully overlapping, we used the ‘prevent overlap’ function. We also applied the function ‘dissuaded hubs’, pushing hubs to the borders.
Furthermore, we used a filter to make our graph clearer. According to the count percentages, most hashtags are at least combined 1 to 3 times. To find the real connections, we therefore decided to filter most graphs on the count of 4, except for #VRgaming which was set to the count of 5. Meaning, we only want to see hashtags in the graph that have been used in combination with the main hashtag at least four times. (in the case of #VRgaming: five times).
Interpreting the data
In the centre we find our main hashtag, #VRgaming. Not surprisingly, the hashtags that are combined with VRgaming are #gaming, #virtualreality, #vr, (thus, virtual reality) #oculusrift, and #htcvive (two devices for VR gaming).
What we found very interesting is how on the right side of the graph most tags can be found related to gaming and there some platforms are mentioned, such as Xbox One and PS3 (Playstation 3), while on the right side there is a cluster that is about design, wearables and Eindhoven with hashtags such as #eindhovendesign, #fashiontech, #manusvr and #vrglove.
When searching on Instagram for #eindhovendesign again, it does not require too much effort to find the first post with #eindhovendesign combined with #vrgaming and #manusvr. There is a technical University in Eindhoven, where they organize a (technical) designerweek every year. Doing a quick desk research on #manusvr also leads to a website from a start-up company in Eindhoven, that has developed the first consumer virtual reality glove. The results of this designer week and of the Manus VR are also posted on Instagram, which explains the cluster.
On the down side of the graph, there are some hashtags that display what users think about #vrgaming. We can see here #awesome, #nerd and #instagood. Both #awesome and #instagood refer to a positive response, while #nerd can either be sarcastic, or meant serious. We cannot assume this is a negative response, but we would not categorize it as positive. Out of this complete graph, we have found two positive and one uncategorized responses.
#VR is often connected #VirtualReality. We expected this to happen, as VR is short for Virtual Reality. What we noticed is that even though in this hashtag there is no ‘gaming’, it is connected with gaming-hashtags, such as #gaming, #games and #gameplay. We can learn from this that VR is also often associated with the gaming industry.
Other hashtags that stand out are again the VR devices, #OculusRift, #HTCVive, #Samsung, and even #GoogleCardboard. The remaining hashtags do not provide any information on how people perceive VR, which perhaps is a limitation of this chosen keyword. Our main learning point from this graph remains that #VR is associated with gaming, which is a confirmation for us that the gaming side of Virtual Reality is indeed where the public interest is.
Before visualizing this graph, we expected the outcome of #VirtualReality to be more or less similar to our previous hashtag, #VR, as it is just an abbreviation of Virtual Reality. However, we actually found some connections that we did not find with #VR. Again, there are several hashtags of the devices (brands) connected. (#OculusRift, #HTCVive, #Samsung)
Just as in our previous graph with #VR, people have associated #VirtualReality with gaming hashtags, such as #gaming, #gamer, #videogames, but also #nintendo. Nintendo is the world’s largest video game company and we think the assiocation between Nintendo and #VirtualReality might be another indication that Virtual Reality indeed has a certain connection with gaming.
What we see here in the graph of #VirtualReality are connections with more professional related things, as #architecture, #business strategies, #business, #marketing, but also #Elon Musk. The relation between architecture, business strategies, business or marketing with virtual reality can be explained by thinking of the other possibilities that virtual reality provides. The relation between Elon Musk, the big business magnate, CEO of SpaceX, Tesla, co-founder of SolarCity, Paypal, and more (Bio., 2016) and Virtual Reality is a little harder to make. When consulting the Internet, we can see several news items pop up about Elon Musk, saying he believes we are all living in a simulation. The news item dates back from 3 June, 2016 and this data scraping is done on 5 June, 2016. Since Tumblr is a microblog, it is most likely people have posted some blogs about this statement by Elon Musk. This explains the relation between #Elon Musk and #VirtualReality.
This graph is about #OculusRift, the device “where it all started” for the majority of the society, and hashtags it is associated with.
Previously, we have described how so far all hashtags have been connected to the VR devices. We also wanted to look at the VR devices, in the hopes of retrieving information on how Tumblr users respond to Virtual Reality.
Knowing that, it is no surprise we see four dark blue hashtags in the core of the graph that are interconnected: #OculusRift, #Oculus, #VR and #VirtualReality.
Oculus Rift was developed with gaming intentions, which explains why there are so many gaming hashtags on the right side of the graph. This time not only #gamer, #videogames, #playstationvr, but also some games pop up: #fallout4, #minecraft, #starwars.
On the top side of the graph, there are some tags related to Google Cardboard, the cheapest “VR device” available, which is more or less a counter movement against the very high priced Oculus Rift. The other available VR devices, such as HTC Vive and Samsung Gear VR are also mentioned here.
When we look at some other hashtags, we find #kids #react, #scary, #funny, #fun, #horror, almost like there has been a sentence made out of hashtags. We think this is the case and this can be explained. In doing our research, we have come across all kinds of information and have seen some “people react to”-videos. We suspect that the hashtags are coming from a blog with possibly a “people / kids react to Oculus Rift”. Nevertheless, we can still use these hashtags in building towards an answer to our sub-question.
The hashtags #scary and #horror indicate that most likely a horror game was played and the people found it scary. At the same time, #fun and #funny are positive responses to the Oculus Rift. In the top of the graph, there is also #vrhavingaproblem, which we code as a negative response.. When we look at the colour of the node, the darker blue the more often it is used/associated with the core hashtag #OculusRift. This means that people have associated #fun and #funny more often with #OculusRift than #vrhavingaproblem. More hashtags should be related to how people experienced the Oculus Rift /VR to get a better overview of people’s experience, but from what is available now, it seems people mostly had a positive response to the Oculus Rift / VR.
#HTCVive is another VR Device that is already becoming very popular, thanks to exposure in videos of for example IGN. The hashtags that are most associated with HTC Vive are, #VR, #VirtualReality, #HTC, #Oculus and #Gaming. Logically, the device is associated with Virtual Reality itself. Looking at the green coloured hashtags, we again find a lot of gaming hashtags (mostly on the downside of the graph). #gamer, #videogames, #game, #gaming, #ps4. Interesting enough, there are some hashtags that have to do with google, namely #googlecardboard, but also #playstore and #googleplay. This might be because the HTC Vive can be paired with your HTC phone and has developed apps for that.
Another thing that stands out for us is #art and #tiltbrush. Google has launched an app for the HTC Vive called Tilt Brush, which allows the user to paint in 3D space with virtual reality. (For more information on how this looks like, check this introduction video for the Tilt Brush by Google)
#SteamVR is a hashtag that is associated with HTC Vive, because Steam is an entertainment platform and gaming community that works together with HTC Vive.
Again we see #react, which is probably associated with HTC Vive because of the “People react”-videos, as earlier explained. Unfortunately, this time, we find no specific emotions or responses towards HTC Vive near the react-hashtag.
Unfortunately, we also see #vrhavingaproblem again: a negative response. Nevertheless, we see #thefuture and #thefutureisnow in the right side bottom, which are more often used than #vrhavingaproblem. These hashtags are used to indicate excitement about how the future is so nearby already. We consider these hashtags as positive responses.
The first thing that we noticed is how this graph looks different with more distant clusters than the other graphs, while we have used the exact same layout (with ForceAtlas 2) and have had a filter from the count of 4 (for 400 items).
What we can see here is #SamsungGearVR being surrounded by #Samsung (the brand), #Oculus (another VR Device), #VR, #GearVR and #Virtual Reality. On top of that and closely around that, we see a lot of hashtags that are related to Samsung itself. #SamsungGalaxy, #SamsungGalaxyNote4, #Galaxys7edge, etcetera. At the end of February, around the time of the Mobile World Congress (#mobileworldcongress2016, #mwc2016), Samsung announced all pre-ordered Galaxy S7 Edge come with a free Gear VR. (Miles, 2016)
Besides that, the Gear VR is another device that makes use of a telephone as a screen, unlike the Oculus with a build-in screen. That is most likely the reason why there are many Samsung telephone models as hashtags.
On the right side, we can find a cluster about blogging: #bloggerstyle, #blogger, #blog, etc. As mentioned before, Tumblr is a microblogging platform, which could clear up why there are hashtags about it. However, another explanation could be that a blogger has reviewed the Samsung Gear VR for example. There are no signs of this scenario, but we cannot eliminate the option.
On the left side, there is a cluster about different types of fans: #applewatchfans, #samsunggearvrfans, #virtuarealityfans, etc. Although it is unclear to us what an explanation for this cluster could be, these hashtags show support for Virtual Reality and the Samsung Gear VR. Again, positive responses to VR.
In the middle we see #PlaystationVR, with close around it #Playstation, #PS4, #gaming, #sony. These hashtags are all connected to #PlaystationVR, because the PlaystationVR is a device developed by Sony Playstation, intended for gaming in Virtual Reality on the Playstation 4. This also explains all the gaming hashtags. In previous graphs, we also found many gaming hashtags, however, the PlaystationVR will only be intended for gaming. That makes the connection even stronger between Virtual Reality, PlaystationVR and gaming.
The PlaystationVR will be released October 2016, which is the reason of #October #2016 #nerdnews. (Playstation, 2016)
On the outside of the graph, the games Until Dawn, Uncharted and Star Wars Battlefront are mentioned (#untildawn, #starwarsbattlefront, #starwars, #battlefront, #uncharted4, #uncharted) Close to the core there is #4theplayers, the slogan of Playstation 4.
Two other game related hashtags can be found on the right down-side of the graph, #tgs and #tgs2015. Doing a little desk research, shows that tgs stands for Tokyo Game Show: “an annual convention where publishers and developers around the world show off the hottest upcoming video games.” (IGN, 2016)
What was very noticeable for us, was that other VR Devices are not so prominently present in the core of this graph, while with other graphs most of the devices were all in the centre. We can still see #Oculus and #HTCVive, but it is pushed more outwards.
Another thing that stood out is the amount of positive associations with and excited responses to PlaystationVR. In the graph we can see #awesome, #thefutureisnow, #beyondreality, #gameon, etc. These hashtags clearly are positive and show excitement towards the PlaystationVR.
All graphs showed some positive hashtags. However, these few positive hashtags are not enough evidence of a positive response towards Virtual Reality. A big limitation about data scraping is the sorts of data you gather. Although hashtags are very useful in labeling images and text, it does not allow us to thoroughly analyse the information.
We need to perform netnographic research, to confirm our current findings and to gain more insights in the discussions and responses of the people.