Reactions to Putin’s birthday on Telegram
I’ve been wondering for the last while whether the way users react to messages on social media can in any sense be a good guide to sentiment. There are many reasons to think not. For the case I’ve been looking at, the arguably interesting data comes from a channel, TASS, with a comparatively small following; it’s almost certain that anti-war people hate-reading the feed account for a decent number of negative reactions; and just a couple of days ago, in my last post, I wondered whether there wasn’t some as-yet-not-really-explained funny business going on in whether and how this channel allows reacts.
Nevertheless, and more in the spirit of adding data rather than making an argument, here I just want to note some reactions on TASS today, because it’s a special day, Putin’s 70th birthday. Taking a look at them is not only inherently mildly interesting, in my view, by looking at what seem to be patterns in the reactions might enable one to sort of work out a hate-read quotient which one could then ‘subtract’ to get a sense of the non-hate-read reaction. I’ll also suggest that even if negative reactions are subject to trollish behaviour, the absence of positive reactions isn’t, and so work for a future time is to consider how positive reactions have risen or fallen, and I’ll end with a couple of instances that are somewhat suggestive. With that said, some screenshots. The only thing to note is the reactions: the text in each case is similar as described below.
This was posted around four hours ago. It doesn’t say much of interest — it’s advertising a long-form article of facts and figures about Putin, and points out that he’s had four terms as president, visited a bunch of countries in that role. The thumbs up — thumbs down ratio is not exactly amazing!
Others are slightly better. Consider next these three from earlier in the day:
This post advertises the same special report, and we see here the ratio of positive to negative is very roughly 4:1, something that seems relatively constant:
It’s noteworthy, I think, that we see this constant ratio. Is that natural? Presumably the plausible explanation of it is that it’s the same group of people reacting each time (it would be a weird coincidence if there were three different non-overlapping groups that happened to react in the ratio 4:1), but does one tend to react to the same story multiple times? And again:
This is a couple of hours ago: and we see again roughly the same ratio, albeit not the same raw numbers, presumably because people haven’t gotten a chance to react to it yet.
Is there any information to be gleaned from this? I’m not sure. I remain puzzled why TASS would even bother allowing negative reactions, and so I think, even if it’s not interesting as a window to the general population’s sentiment, its mere existence is noteworthy (As I noted previously, some, like Ramzan Kadryov, disallow negative while allowing positive reacts; others disallow any emojis; others allow emojis but disallow comments (which Kadryov allows).
But finally: even if it were the case that negative react data were too easily messed with to be worth caring about, nevertheless the lack of positive react data, especially if it were to manifest in the bigger channels like Solovyov or Kadyrov, would be arguably interesting. That’s work for another day which seems profitable, but let’s just note, continuing an approach I used the other day, that there seems to be an on-going flipping of sentiment driven by a decrease in positive reactions on TASS’s daily link to its rolling news feed. Here’s some recent ones, where in each case the text is identical and just providing a link to the rolling news:
My previous post looked at this a bit more, but I’m inclined to think that if studying reactions is profitable, then the thing to do is look for trends in positive reactions that aren’t susceptible to trolling; and to do so for bigger channels than TASS. But that’s for another day.