Reply All - #168 Happiness Calculator vs. Alex Goldman

Episode Date: October 29, 2020

Alex meets a scientist who has built a tool meant to do the impossible -- measure the world's overall happiness and sadness. Plus, Alex volunteers for a risky and strange experiment. Learn more about your ad choices. Visit podcastchoices.com/adchoices

So, PJ, hello. Hello, Alex. PJ, how has this year been for you in terms of your mood? I mean, you're laughing. It must have been great. This year sucked. This year has been bleak and bad, and winter feels bleaker and badder. Yeah, winter's really freaking me out. I'm actually very scared about what winter's going to be like. How are you doing?

Well, it's funny you should ask. So I read this article last week in the Times about these researchers who are trying to, like, measure and plot everybody's happiness on a day-to-day level. Everyone in the world. Yeah. And they made this thing that they called a hedonometer. Like hedonism. Yes, exactly. Heednism.

Okay. And I was like, okay, I want to talk to these guys. Because first of all, happiness seems like a very slippery concept that I can't even begin to a hedonometer. imagine how one would measure for. And second of all, if they can measure happiness, like, how can this help me a person who is generally very unhappy? So I reached out to these researchers. I ended up talking to one of them. His name's Peter Dodds. Hi, guys. Thank you so much for taking the time to do this. I really appreciate it. It's a pleasure. Yeah. Peter's a data scientist. He teaches at the University of Vermont. Do you think of yourself as a happy person?

No, I think I've friends have described me as melancholy, I suppose. So I think miserable happiness is perhaps a lofty call. But I'm irrelevant. Let me say this. I'm irrelevant. Like, I'm totally irrelevant. I think this is one of the most important things where you study social phenomena at large scale. You have all these little stories from your own life, which really matter, you know, I mean, obviously, to you. But when you're trying to get out and think about how systems work, you have to kind of throw them away. Hold on just a second. My infant daughter just ran up, ran up here. Hi, Pauls. I'm in the middle of an interview. Can you go downstairs? Okay. Please? No. All right. Hold on a second. I have to get it. I love it. This is Peter. Do you

want to say hi, Peter? No. Hi. Why not? I'm sorry. Do you want to take your cookies and take them to Mommy. Bye, Pauls. Sorry about that. So good. So the Hidonameter, this project that Peter came up with the measure happiness, he started working on it in 2007. He and another researcher friend of his, this guy named Chris Danforth,

had this kind of wild idea, which was like, okay, governments are always making decisions based on things like GDP, things that are really easy to quantify. What if we could make our happiness something that literally, we could also quantify. Then we could start making policy decisions based on that. Like the thing that we judge our country on is just like, for the most part, like, is the country making more money this year than last year?

Right. And maybe like, did fewer people die or whatever? And that can mean a lot of different things. Like, it doesn't necessarily mean that people's lives are better or worse. Right. So back when they were sketching out what the hedonometer could look like, the whole time they're thinking to themselves like, what is our data source going to be? And then they hear about this new website that's called Twitter.

And it seems like a really useful place to get a constantly updating stream of people's thoughts. Famously, the website where everyone's fucking happy. We analyze that. The app that I open on my phone when I want to feel better. So we wrote to Twitter in 2008. There are only four people working there and said, hey, this is cool. Could you, like, do you have, you know, could we get some data? And they're like, oh, we made a little research feed, research a feed.

And we've been getting 10% of their tweets. ever since then. And 10% of all tweets, which at that time was practically nothing, is not like 15 million tweets a day. And I'm talking just about the English language tweets. They're building out a bunch of different languages, but so far they've only published data about the English-speaking world. Anyway, what they did is they decided they'd analyze whether the words that people were tweeting were like happy words or sad words. So they took 10,000 words and they asked people on a scale how happy or sad does this word make you feel? Okay.

And they use that to give each word a relative weight. Give me an example of a word. Hold on. So I'm loading up the examples. Hold on. I'm actually going to screen share with you if that's okay. Can you see my screen? Yeah.

There's actually a list of words. So what we're looking at right now are all of the 10,000 words they selected ranked from one, the saddest, to nine, the half. happiest. And the happiest words are like laughter, happiness, love, happy. Can you guess what the unhappiest word might be? Die. So that's not bad, actually. It's right up there. The unhappiest word is a tie between suicide and terrorist. That makes sense. And then the third from the saddest word right now is coronavirus. I'm almost surprised it didn't win. Yeah, me too. But especially since pretty close to the top of the unhappiest words are words like ventilators, ventilator, self-quarantined, sanitizer.

Like, these are among the unhappiest words. It's like a magnetic poetry of everything that's bad this year. Tortured, violence, cruel, cry. Can we go back to the happy side? Yeah. Comedy, jokes, rich. Celebrate weekend. Music, healthy. So they're constantly having to update this list of words, like all those COVID words just got at it. pretty recently, but they also have to remove words from the list, like words that they can't measure anymore because the meaning has changed so much.

One example of that a few years ago was the word thirsty. Thirsty just kind of... I mean, a good word, right? It's a good old word. It means you want to drink something, right? But it moved. And then there was some sort of Twitter, like massive retweetings of people just saying, thirsty as many times as they could in a tweet. And it just overwhelmed the system a little bit.

We're like, okay, we're retiring that word. So they have all these words that are weighted on this happiness scale from one to nine. And to figure out if this thing's actually working and is correctly identifying happiness and sadness, they had the hedonometer analyze a bunch of different texts where the mood is really easy to follow. Like they ran it on the text of crime and punishment. They did it for the count of Monte Cristo. Happy book or sad book? Well, the hedonometer correctly analyzed it for what it is, like a book that starts happy, gets sad, and then gets happy again. This is actually an application that would be useful for me because, like, things that I read and watch really affect my mood a lot.

And so, like, this makes me sound like a child. But a lot of times I'm like, is this going to be sad? Am I going to, I love to. Oh, so you want someone to toss it into the hedonometer to see if it's going to make you feel good or bad at the end? Yeah. And sad in the middle is fine. Just don't leave me there. But yeah, it seems like Peter and his team.

are getting good information from the heat onometer. Okay. So what if, so assuming, assuming that this tool is usefully describing something, what did, what happened when they did it this year? Well, they've run it on Twitter since 2008. So, and so they have this like big map of happiness in the English speaking world for like the last 12 years. So if you go to the heat on a meter website, it's a big graph that kind of looks like

a stock market graph with like spikes of happiness and dips of. sadness over the years. And if you click on specific dates, you can see the most used words on Twitter that day. And one of the things that they learned by graphing all this info is like, globally speaking, we do not have big moments of spontaneous global happiness. The only thing that unites us in happiness is like big holidays. Like for English speakers, Christmas has the biggest effect. It nudges us up a little bit. So Christmas becomes kind of a, a yardstick of sorts. And, you know, this sort of how many Christmases did you get for this, you know, this jump or bump in happiness all the other way down.

So it's, it's useful. So when you and your team see something, you're like, okay, it went up like half a Christmas day or it went down half a Christmas day. It depends on where framing it, but that's not a bad way to talk about things. Yeah. The only other positive event that even slightly moves the needle, the way that holidays do, is when a K-pop star, has a birthday. And Peter knew nothing about K-pop and only through like analyzing

language for several years was like, what is BTS? That's so funny. But the bigger thing that the hedonometer is showing is that we are at the tail end of a long happiness nosedive.

According to Peter, the hedonometer shows that the last five years, we've been getting sadder, and sadder. Between 2016 and now, we have lost a Christmas Day of happiness. As a world? As an English, everybody in the English-speaking world. Before this year, the saddest day on record was the Las Vegas shooting in 2017. We have broken that record multiple times this year. The first time was on March 12th, which was the day that the stock market tanked, the NBA season was suspended. Tons of places started to shut down, and so the hedonometer hit a new record low. It's like,

there's something satisfying about knowing we broke the unhappiness needle. Like, like, because I want to think, the thing I'm telling myself going to this winter is like, well, it's just like, it's like really hardcore training for all the sadness you'll ever experience in life again.

Like, if you can get through this, it's like, you've beat the extreme weight course. And like, just having scientists confirm what I knew, which is that this was the most miserable year, just feels nice.

So that was just the first time that the Hedonameter record was broken this year. It was broken again in May. So the 25th of May was the day simultaneously the Amy Cooper event in Central Park happened. You remember the woman who... Oh, that was awful. She, like, called the cops on a black man who was a birdwatcher and then falsely claimed he was threatening to assault her. And he was just asking her to leash her dog.

That was a bad day. That was also the day that George Floyd was killed. Right. And so five days later, Sunday, May 31st, according to the hedonometer, was the unhappiest day in the entire English-speaking world since 2008. Oh, so it wasn't the day of his death. It was the day that it had sunk in for everybody or everybody had heard about it. Yeah. I mean, it's interesting just seeing it, like this huge spike of sadness that we all felt at the same time.

Wow. Now that you say it, it's like, I also remember distinctly it was like, for once there was one topic on the internet for a while. for like quite a while. It was like there wasn't nobody was like oh by the way I want to pick a fight about this thing or whatever. It was really like one topic

and it was an extremely sad topic. Yeah and that feeling that we all had that's actually exactly what the hedonometer showed us because usually when there's a tragic event it takes a day or two days to get back to sort of the normal average

happiness on the hedonometer but in the case of George Floyd the needle driven dropped, and it took an entire month. But returning to normal isn't really saying a lot because the baseline of the year is just so sad. It's just so stark for me to watch, to look at this graph, like, to actually extend it out

and see the past, I guess, four years. And it's just, the negative spikes seem so much deeper and longer lasting. I mean, we keep setting these records, right? So the bombing in the Boston Marathon, you know, that was a record at the time, right, for us. The Pulse Nightclub in Orlando. And then the mass shooting in Las Vegas, you know,

that was so extreme. But then, yeah, then we get to the COVID-19 and George Floyd, like finding new, new depths, essentially for this kind of collective well-being, You know, this is a traumatized population. So obviously, a lot of very hard stuff has hit all of us in a pretty short amount of time. And it's, like, very depressing to think about. But in another way, it's, like, kind of comforting to be able to quantify at least a little bit how bad 2020 has been.

And I was looking over all this data, and I was like, I wonder if there's any way to, like, see if I line up with this. Like, is my happiness affected, too? Like, am I also becoming sadder? You know, this is the rare reporting question that I feel like I could probably answer for you, as can anyone who works with you. Well, Peter said that as a policy, they don't analyze individual people's data. Like, they don't even gather individual data because they don't want the hedonometer to be a surveillance tool. But he said that if I wanted to give him my data to analyze, he would be willing to make an exception. But you delete all your tweets.

Yeah, there's no proof. My tweets auto-delete every two weeks. Which is, that is like the one kindness I perform for the world. But I was like, I was like I wonder what I could measure. Your relationship through your own privacy is a beautiful thing. Yeah, I'm wondering what kind of data set would work for me. I mean, maybe the emails I send, but probably not because I mostly just send work emails. They're not going to be fun.

What about my text messages? I bet that would be good. Yeah, that's the stuff. That's the stuff. That's high quality. So I scraped the last year's worth of outgoing text messages from my phone. Absolutely psychotic. 13,660 text messages.

I dropped them in a spreadsheet. The word cloud. Okay. And I gave them to Peter. And in addition to being a professor at the University of Vermont, he also has this company called Quaka with these other researchers, Andy Reagan and Chris Danforth. And they worked on, they parsed, they parsed the data I gave them. So just some basics, right? 13,000, close to 14,000 messages, about 40 a day.

We do the same things we did for the hedonimo. We lowercase them and break them up. It's a good, it's a good amount. He basically wrote me like an academic paper about my texts. God. It's called a narcissistic dream. It's called a consideration of the text messages sent by Mr. Alex Goldman from 29, from October 16th, 2019 to October 15th, 2020.

It's like a real step forward in the field of Goldmanology. We're finally doing some field work. He found a lot of heart emojis. That's for your wife. A lot of LOLs. That's for me. You have a good supply of eggplant emojis? I'm just going to say that.

That is a surprise to me. That is a surprise. I'm just going to leave it. So that's just true. I have to say, I wasn't super nervous about this until we actually started diving into the data and my heart rate.

I just definitely rising. So it's hard to break up some of these, like emojis, can I just tell you? Imoges are the most ridiculous things in the world. So some things that didn't break up. There's one, which has got some words in it. And it's like,

Tori, poop, poop emoji, ass, dollar sign, like a flying dollar sign twice, maybe two goats, and then a couple of smiley faces. That's like a blob. And I'm just going to leave it as a blob. That's something that I wrote? Yes.

And then he found a bunch of what he calls stretchable words, which is where you add like a, where like I spelled bad with nine A's, spooky with five os. That's usually when you're unhappy. They're like things are bad. And then he said, I found a poop with 13Os and a poop with 59os.

That's usually when you're either happy or manic. That's totally right. That's totally right. I don't know why he didn't consult me. So Peter actually broke out for me my texts between 3 a.m. and 6 a.m. He was like, he was like, you don't text. Country I don't want to visit. Well, what he said, what he wrote was, you don't text so much during this time because of the whole sleep thing.

But when you do, it's not a happy time. So number one with a bullet is terrible. followed by never shit asshole shot which is me misspelling shit never shit asshole shot not sorry murder you sound like a serial killer with GI problems it's so weird listen to the so better shows up in there and then it's mean wars suffering crime hated dying hates it just it's like what am I talking about about in the middle of the night.

I kind of know what you're talking about. I'm just that DJ's not in there. Okay, so here's a question for you. So what do you think your love to hate ratio is? I would say that for every... That's the ratio. For every one love, there's five hates, I would say. So it's actually four to one love to hate.

See? Wow. I did take out one that was Lovecraft because I... Presume you'd be watching HBO. Yes, that I had. But yeah, yeah, yeah. But the shape of my happiness. The shape of my happiness by Alex Goldman.

One man travels through himself. It's my self-help. My forthcoming self-help book. Okay, so Peter started with October of last year. He zoomed in on that month and told me that, like, compared to Twitter, I was actually doing pretty good. I can talk you through the words a little bit. Yes, please. So October last year, you know, it's above, right?

So you've got more love and like and life and soup. You've got these kind of words. You've got negative words, right? It's a mixture. But you were still managing to be above. Meaning October of 2019, according to the Hedonameter, I was happier than most people. Wow.

I don't remember those months being happier or less miserable than any other time. Like, for me, everything kind of smears together. It's really hard to remember when the genuine unhappiness of 2020 started. But according to the hedonometer, I was my happiest in October and November of last year. And I was like trying to think back to what was going on. And I feel like I was talking to Fia about it. And Fia has a much better memory for what I was doing last year than I do for some reason. And it might be because I was working on, thank you for me for

noticing the story about John and Santa Fe, which was there was a lot of fun. I had a lot of fun going on a trip. You got scared of the weed store. Very scared of the weed store. But so far this, it's like you're most, just you are personally most unhappy when you don't have an idea for a story and your happiness when you're in the reporting for a story. The reporting for that was in October. And also in October, I remember I went to this wedding and the cat skills and I spent the

weekend up there and it was really nice. And in November, I went to, for things. I went to Florida with my kids and we got to hang out on the beach and I think that accounts for like a lot of happiness points because the kids really love it and them being happy makes me happy. People should have on their like this would I would not I would consider consenting to this happening automatically on my phone where it would just be like hey you've you seem unhappy what's going on and it would just kind of like you're like you were a lot happier last November and also you were like walking out. outside more. Just like, I don't, that voice in my head is really underdeveloped, like noticing when I'm sinking or noticing why I wasn't sinking before. Yeah. Totally. Okay, so December, I'm guessing you go down. He gave me a chart. Yes, December, it goes down a bit and continues to go down. Well, then you're past the holidays and then it's just deep winter, January, February. And then you're like, thank God soon it'll be March.

Right. I'm able to go outside a lot. Mixed it up with my friends in various restaurants. Catch a movie in a movie theater. So it might shock you, but March was a bad month for me. Then you have mad and coronavirus and rejected and hated and hate and painful and cried and stressed and death. Those things are in there. It was going down. But what surprised me is that last month, September, was the worst month of the year for me so far. And what I think the Hadameter was picking up on is the fact that Harvey is having a hard time with school, which just started.

I was very worried about it. I was constantly texting Sarah about it. And October, which we've only got half of the month, it's still worse. You started above Twitter. You were like half a Christmas day above, but now you're down. You're with all of Twitter, the one big mass of Twitter. You're at the same level. Yeah.

And I guess, you know, if we linearly interpolate, you're going to go down below Twitter. So it's going down, man. Peter says that I am not alone, that Twitter is in this crazy downward trajectory on the heat autometer this October. I'm looking at the word cloud for some of these days, and it's prisons, thugs, protesters, prisoners. Do you have any idea to what that is attributed?

So that one is Nigeria, actually. Yeah, that's a... So this is because of... global, that one got pushed in there. So that's like, points to, like, just how there's everything is going on. Because the first stuff is Trump, coronavirus, there's the debate with Biden, you know, led into that. Just, just a maelstrom of stories. And of course, it's precipitated by Ginsburg's death.

Like, it just looks like it's like, it just looks like we're falling off a cliff. Which doesn't feel great. I think, like, I don't know. I think on the one hand, a lot of the factors that are making things hard are just completely out of our control. And it is nice to know everyone's in the same misery. And at some point, everyone won't be in that same misery. I keep thinking of that poem that's like, laugh in the whole world, laughs with you, cry and you cry alone. And it's exactly untrue.

If you laugh, you're laughing, according to this machine, if you laugh, you are laughing by yourself. And if you cry, the whole world is crying with you. So, wait, is this like an advertisement for depression now? Yeah, what are you saying? Ask you to talk to her about depression. Yep. I mean, honestly, I sort of feel like the big thing I get out of this is, you know, you're a person who complains a lot about the world and the things in it that are bothering you.

I'm feeling, like, I don't really like this description of myself. but like the thing that this proves is like you are having a harder time with this than most people like the thing that is affecting everybody it is affecting you more yeah and there's a part of me that's like what right do i even have like i've got like like i'm employed my family i know you like you're as like a person with a conscience you're like legally obligated to note these things But, like, as a person who has to experience the world, like, you can't tell your unhappiness that other people have it worse, really. Like, I've never found that to be a particularly useful strategy. It's, if anything, you're just like, and screw you for feeling this way.

Like, I don't know. For me, it's like, maybe there's nothing you can do with it. For me, it just makes me want to keep an eye out for you. Oh, thank you for keeping an eye out for me. No, thank you for telling me that I complain all the time. Is this, this, this is not, surely, this is not new information. No, it's not. After the break, nice warm bath.

Okay, so Alex. Mm-hmm. One of the things sort of the hedonimeter has me thinking about, I guess, is like, the idea that you could kind of be telling yourself a story about your, relentless unhappiness, but you could miss a moment of happiness that happens within it. You know what I mean? Yeah, unless you're like super conscious and deliberate about trying to be happy or like being like, today is the day. You know, let me tell you something about my wife. She will, she like goes the extra mile just to make her life like two like iodas better every day. Like every day I come downstairs

and there's like half a lemon and I'm like, what's going to with a lemon? She just likes to put lemon in her water. She like wants to make her life better. Two things. One, I think most functional people are like your wife. While I agree, I find it weird. I think anybody else listening is just like, wow, the world looks really different from inside depression all the time. We're like, who are these weird freaks that aren't just like trying to make their lives miserable all the time? So this actually leads into what I wanted to talk to you about, which is like, I'd be like, What, like, as I said, very terrified of this oncoming winter. And I've been trying to notice, like, I've been trying to notice the things that I try to do to make myself happy that are actually just depressed.

That, like, they get depressing real fast. Like, playing video games for an hour, cool. Playing video games for nine hours, like, less cool. And I've also been trying to notice, like, the small kind of, like, lemon water type things, like, little things that you do for yourself that actually do make you just incrementally, incrementally happy. Like, they're not raising you, like, a Christmas standard deviation, but they're just giving you, like, point one more on whatever day you're on. You know, I have to say, revealing this somehow feels, like, more vulnerable to me than exporting my text messages.

Yes. To a spreadsheet and sending them to a researcher. Well, because either it's like, it's so, it's like literally this weird little way you're trying to show yourself a tiny bit of love, or it's such a basic thing that it's, like, embarrassing to admit that it's kind of a big deal. I've become a bath guy. I've become a candle guy. Like, scented candles?

Yes. I have to say, if we can buy in our respective joys, we could become more powerful than anyone's ever imagined. Sented candles and a bath? When did bath guy start? I don't know. It started last month. I was like, what if I take a, like, what if instead of, like, just getting in the shower to clean myself, I will, I, like, sit in the tub for a while and just do nothing?

Listen to podcasts, play some Matchington Mansion, just chill. Wait, you're playing, you're taking a bath and you're playing your shitty iPhone game? You know it, brother. I like how you managed to put ketchup on everything. No, I mean, that's, it's nice. It's like, it's, it is the closest to the, like, complete, like, brain shutoff time that I had when I used to have a commute. When I had to sit on a train forever and just not think about anything and, like, just be, just exist. Read a book.

That is totally fair. I think in the spirit of this endeavor, I'm not going to make fun of you playing Magic Kid Mansion. Well, that ship's kind of sale. But I'm asking you because I actually wanted to do, I wanted to ask our listeners. I feel like our sort of forays into not being miserable all the time are, they are, they're tentative first steps that most. They're Bush League. I think we're the JV team and you want to see what varsity self-care stuff looks like.

Yeah. I'm like kind of curious for like what people do a little bit selfishly, like heading into a hard season and a little bit like as a gift for everyone else heading into a hard season. Like the stuff you figured out the like making your bed in the morning or taking a walk like really like fundamentals of human happiness 101 To see like what's working for people That's actually a pretty good idea. Okay, so What's our email and what should the subject line be? Our email is reply all like gamutmedia.com

Our subject the subject line should be I don't know Varsity self-care Varsity self-care And I think what we want is for people to record a voice memo describing the thing. I mean, bonus points if you want to record it while doing it. But yeah, just send us what's working for you. Repaal is hosted by PJ Vote, Emmanuel Jochi, and me, Alex Goldman. Our show is produced

this week by Shruti Pinnamenei, Fia Benin, Damiano Marquetti, Anna Foley, Jessica Young, and Lisa Wang. Our executive producer is Tim Howard. We were mixed by Rick Kwan. Fact-checking by Michelle Harris. Our intern is Mohini McGauker. Our theme song is by the Mysterious Breakmaster Cylinder, additional music production by Mari Romano. Special thanks this week to Chris Danforth and Andy Reagan. Matt Lieber is at least a Christmas Day of Happiness. You can listen to our show on Spotify or wherever you get your podcasts.

Thanks for listening. We'll see you soon.