Mapping #climatechange

I expect to find most people tweeting about #climatechange in the Northern hemisphere, 0° to 90°latitude, since that is where English is more prevalent, and because I feel as though people in this area are more aware of climate change. I also expect the Western hemisphere, -90° to -180°longitude to have more tweets about #climatechange, however this does leave out most of Europe and Australia, two places where I expect to find a lot of tweets. I think my hometown, Boston, 44°, -70°. I expect some people to be tweeting from this area, but not many just because this is such a world issue, literally.

I have 69,923 tweets to date, with only 60 of them being mappable, which is 0.09%.

Yau writes about how data shapes these data visualizations, and that the visualizations are meant to communicate data to meaningful conclusions. My data might underrepresent non-English speakers, since I am only searching for #climatechange, which is obviously an English word/phrase. People in, say, Brazil may be talking about climate change on Twitter, but it will not be included in my data unless they use specifically #climatechange. Most of my geolocation tweets seem to be in the Northern hemisphere, as I expected, and a lot of tweets in the Western hemisphere. So I expect most of my tweets to be in the Northwestern quadrant, AKA mostly North America, which is mostly English-speaking. Upon closer look, most of the tweets I found are in English, which is no surprise. I expected a bit more tweets to be in more popular languages, such as French or Italian, and a surprising number of tweets in Thai…? Now that I look closer at the Thai tweets, they are all retweets of the same tweet. The fact that most of the tweets are in English again over-represents English speakers and underrepresents non-English speakers.

Based on how boyd & Crawford define Big Data, I do not feel as though I have Big Data. My data, while rich, should not be considered Big Data since my numbers do not speak for themselves, which is a concept boyd & Crawford present. Even the simple fact that I am scraping data from Twitter is not representative of “all people,” as “…it is an error to assume ‘people’ and ‘Twitter users’ are synonymous: they are a very particular sub-set,” (boyd & Crawford, 669). Additionally, of my collected over the past two weeks, only 0.09% of that data is mappable, so it is certainly not Big Data.

#Energy had more tweets than #climatechange because #energy is a common term and can explain a lot more than just environmental issues, such as working out or supplements. #KeystoneXL, not surprisingly, had far fewer tweets, since it is a specific topic, far more specific than #climatechange, which is a global issue.

https://togan1.carto.com/builder/a5d3a8a4-f614-11e6-8987-0e05a8b3e3d7

CartoDB must have changed a lot from last year, so I don’t see any of these markers or paintbrushes or simple maps…

SF 2010 SF 2000 SF 1990 SF 1980

 

I mapped out the 18-24 age category in the 1980, 1990, 2000, and 2010 census in San Fransisco, a city where a lot of tweets about #climatechange were coming from. I chose this age group because as of now, when it comes to climate change, the Millennials (also known as Generation K) are the generation who will have to deal with climate change, and the one who can combat it, which is why I expect them to be the ones talking about climate change most on Twitter. What I found for San Fransisco, though, is that over the years, the distribution of 18-24 year olds has remained relatively uniform from 1980-2010, however, there seems to be a lot of young people in this city.

I read two articles on climate change that had unusual linkages. One was on how climate change and birth weights may be correlated. The article looked to examine “…the effect of extreme temperatures on birth weight, by race,” (Deschênes 212). The article did not examine anything on age, per se, but I’m guessing if a baby’s birth was affected by climate change (and realized by more people), then more people would be talking about climate change. Another article, “Climate Change and Human Evolution” focuses on just that – how climate change affects human evolution. Similar to the first article, if climate change has an effect on humans and their development, then surely more people will be talking about it, on platforms such as Twitter, using #climatechange.

Graham says software-sorted geographies means people’s decisions are more based on computers than we think. Going back to the two sources I found and how I am trying to track who is talking about climate change and where, I realize the more people realize how severe climate change is as a world epidemic, the more people will talk about it and share their views on the Internet. This, in turn, shows how people’s decisions and opinions can be easily shaped based off of where they gather their information, especially on the Internet, without even realizing it.

 

Behrensmeyer, Anna K. “Climate Change and Human Evolution.” Science 311, no. 5760 (2006): 476-78. http://www.jstor.org.ezproxy.trincoll.edu/stable/3843397.

 

Deschênes, Olivier, Michael Greenstone, and Jonathan Guryan. “Climate Change and Birth Weight.” The American Economic Review 99, no. 2 (2009): 211-17. http://www.jstor.org.ezproxy.trincoll.edu/stable/25592401.

 

3 thoughts on “Mapping #climatechange

  1. I can’t believe you had so few mapable tweets. You had so many tweets about climate change and it is such a global issue. You had a couple thousand more tweets than me overall, but I had a thousand more capable tweets. That could be that with #energy people are using it in a context that is not to do with climate change and searching for green energy. It could be that some of my mapped tweets were about how a banana gives you #energy (I’ve actually had a few of those). You’re mapping of San Fransisco was very interesting and how you looked at age population. I too thought age was the most important category to examine, because energy and climate change are an millennial’s issues and that age range is also the most popular one on twitter.

  2. I find it really cool that you framed your population mapping based on the 18-24 year olds given that the audience of #climatechange directly effects this population and required them to be active about the environment in which they will grow up in and undoubtedly change. I wonder how much of a proportion of that population actually is involved in environmental activism and keep up with topics related to climate change. For instance, measuring the amount of 18-24 year olds that follow other twitter pages that are environmentally centered or talk about climate issues. In looking at the effects of climate change and how it is shaping things like evolution, I wonder how it relates to the development of the pipeline project and our natural energy resources used to date.

  3. I too was quite disappointed by the lack of correlation between my 76000 tweets and my 53 mappable tweets so I can certainly empathize with your frustration. I do find your remark about the potential language barrier disparity in hastags referring to climate change very interesting. I suppose because my hashtag is a place and not a term, it did not occur to me that different languages would have misrepresentative effects on the geographical representation a hashtag’s data. I also find your commentary on Graham’s article very apt. I wonder too about how people measure their own actions in the climate change debate based on internet activism. This is to say, who is denouncing climate change digitally but not recycling their plastic water bottles in real life?

Leave a Reply

Your email address will not be published. Required fields are marked *