index.md (7656B)
1 --- 2 title: "Post types and sentiment from my Instagram feed" 3 description: "Does Instagram spark joy? A look at what I'm seeing and how I feel while using this platform." 4 date: 2022-07-13T12:28:22-04:00 5 draft: False 6 knit: (function(input, ...) { 7 rmarkdown::render( 8 input, 9 output_dir = file.path(Sys.getenv("HUGO_ROOT"), "content/posts") 10 ) 11 }) 12 output: 13 md_document: 14 variant: markdown 15 preserve_yaml: true 16 categories: 17 - Data Science 18 - Science 19 tags: 20 - Psychology 21 --- 22 23 I've been critical of social media platforms (in spite of my continued 24 use of them), but I've traditionally defended Instagram. I knew that it 25 had major problems (most galling, [internal research showed that use of 26 the platform was hurting teenage girls' body image, and Meta suppressed 27 the 28 findings](https://www.wsj.com/articles/facebook-knows-instagram-is-toxic-for-teen-girls-company-documents-show-11631620739?mod=article_inline)), 29 but browsing IG has nevertheless been a pleasant experience for me. 30 Photos of my friends' pets, children, and vacations are *nice things*, 31 and being exposed to them *made me happy*. 32 33 Recently, Instagram instituted changes to the app's main feed, which 34 were [announced by Adam Mosseri, the head of Instagram, on 35 Twitter](https://twitter.com/mosseri/status/1521589403671355392). I feel 36 like Instagram stopped being as fun for me shortly after these changes 37 were rolled out, so I collected a bit of data. 38 39 ## Method 40 41 Over the course of four days (July 9--12, 2022), I only opened the app 42 once per day, on my phone. I looked at the first 50 posts on my feed, 43 and logged a couple features, described below, about each one. I also 44 gave each post a subjective "sentiment score" using a five-point scale 45 from -2 (for posts I actively disliked) to 2 (for posts I quite liked), 46 with a score of 0 being neutral. 47 48 ## Results 49 50 What does Instagram show me? Let's first look at "poster category" 51 counts, which shows who the posts come from. 52 53  54 55 Most of the 410 accounts I follow are my friends'. I also follow a few 56 *artists* (a combination of musicians, visual artists, and authors), 57 *organizations* (primarily political and community orgs), *businesses* 58 (all of which are run by friends---a couple bakers, a couple shop 59 owners), and a small handful of *meme accounts* (e.g., "dnddads"). 60 Subjectively, meme accounts are over-represented on my feed; I follow 61 hundreds of actual people and only around five meme accounts, but the 62 latter account for over half as many posts as the former. 63 64 The biggest issue here is that the majority of posts come from *accounts 65 I don't follow*. I understand that Instagram is an ad-supported 66 business, but I see more ads than posts from friends. Worse still, I see 67 almost twice as many "suggested posts" as I do posts from friends. 68 69 What types of content are represented by these posts? 70 71  72 73 Mosseri announced that Meta wanted to make video a bigger part of the 74 Instagram experience, and they seem to have delivered. Of the 200 posts 75 I logged, the majority were video. Photographs, which the app was 76 originally designed for, account for a little over a quarter of the 77 posts, with other types of image (drawings, illustrations, etc.) 78 accounting for the rest. 79 80 What's the relationship between *poster category* and *post medium*? 81 82  83 84 There's a tight correspondence between what medium is employed for a 85 post and who's posting it. Videos are being pushed by accounts I don't 86 follow, while my friends continue to prioritize photographs. 87 88 Since post medium and poster category are so tightly linked, I'll focus 89 on looking at the sentiment by poster category---it would be difficult 90 to tease apart how I feel about photos vs. videos and how I feel about 91 posts from accounts I do vs. do not follow. 92 93  94 95 These box-and-whisker plots show the distribution of sentiment scores 96 for each post category. The heavy vertical line indicates the median 97 score, and the boxes extend from the first through third quartiles (the 98 "interquartile range", or IQR). The "whiskers" cover the full range of 99 scores, unless any observation is identified as a potential "outlier". 100 Outliers are defined as observations with values more than 150% of the 101 value of the IQR above or below the third or first quartile, 102 respectively. For example, I was bothered by a photo of a dead bird 103 which was posted by a friend, and gave it a score of -1. The first and 104 third quartiles of sentiment scores for posts from friends are 1 and 2, 105 providing an IQR of 1. Since -1 is less than 1 (the first quartile) - 106 1.5 (150% of 1, the IQR), that score is an outlier. 107 108 I'm not surprised that I like posts from the accounts I follow more than 109 those from accounts I don't. It's disappointing that suggested posts 110 aren't much nicer to see than ads; these are supposed to make the 111 platform more "immersive" and "engaging" and fail to achieve that. 112 113 In an attempt to quantify the "experience" of browsing Instagram, I 114 looked at the cumulative sentiment scores across each of the four days. 115 There's no doubt that the actual dynamics of my affect, to the extent 116 that it can even be quantified, is more complicated than +2 + -2 = 0. 117 However, this is a good first approximation of what it feels like to 118 scroll through a feed. 119 120  121 122 Each of the four days has a negative cumulative sentiment score, which 123 indicates that I see more posts that I dislike than posts that I like. 124 There's also a suggestion of a pattern, in which my sentiment stays 125 relatively flat and positive over the first 20--30 posts, and then 126 drops. 127 128 I wonder if this is the result of an intentional design decision by 129 Instagram. If they have an internal model of what posts I might like, 130 they could certainly time them in such a way to encourage longer 131 stretches of engagement, and they have the data to figure out what the 132 optimal timing would look like. On the other hand, this is a purely 133 subjective score that I didn't attempt to carefully calibrate, so it's 134 worth considering that I might just get tired of my feed after a couple 135 dozen posts. 136 137 ## Conclusions 138 139 I don't know if it's worth an attempt to use my experience to formulate 140 suggestions to Instagram (or their would-be competitors). I'm 141 approaching middle age, and I don't spend much money on the internet; 142 Instagram can get stronger engagement and much better ad conversion 143 rates from people who aren't like me. Perhaps I'm an unprofitable user 144 and this is their attempt to nudge me off the platform? 145 146 However, this has been a useful exercise for me, because it suggests a 147 few things I could change right now to improve my experience: 148 149 - Stop following meme accounts and organizations (these posts provide 150 a neutral-to-good experience, but they potentially crowd out posts 151 from friends and artists that I would enjoy more) 152 - Stop scrolling my feed after the first 20 or so posts (cumulative 153 sentiment appears to begin dropping after this point) 154 155 Most importantly, it calls into question whether I should even continue 156 using the platform at all. The maximum cumulative sentiment that I 157 experienced was only six (which corresponds to seeing just three things 158 that I really like), and that was only achieved on two of the four days 159 that I measured this. As much as I enjoy seeing photos from my friends 160 (and the data show that I really do enjoy this), it's hard to conclude 161 that the experience of using Instagram is joyful.