October 28, 2014 Memoir, non-fiction, Psychology, Science 19 ★★★★

DataclysmTitle: Dataclysm
Author: Christian Rudder
Source: Edelweiss
Links: Bookshop (affiliate link) |Goodreads

Summary:ย This was a very light, accessible look at data analysis which answers some interesting, but often obvious, questions about how we date and how we describe ourselves online.

As one of the creators of the dating site OkCupid, author Christian Rudder has a fascinating dataset to play with. In combination with data acquired from other data-collecting websites (Facebook, Google, etc), he’s able to ask and answer some very interesting questions. For instance, who do people want to date? And, more interestingly, how does this compare to who they say they want to date? Does the way people describe themselves and the way that people respond to them vary by ethnicity? By age? Even questions that people might not answer accurately can begin to be answered here.

There were a lot of things I liked about this book. The author came across as personable, funny, and enthusiastic about the questions one can answer about big data. His humor was occasionally too negative for my tastes, involving putting one group or another down, but the rest of the time, his sense of humor worked for me. I liked that the author made sure to point out the limitations in his data set (most obviously, anyone involved must use the internet and is almost certainly single). The graphs didn’t work in my eARC, but his descriptions were clear enough that I could get by without them, which I think speaks well of the clarity of the author’s descriptions in general. I thought he did a great job explaining the challenges data analysts face in way that could be interesting and accessible to a general audience.

The downside to this book, for me, was that it was very light. Most of the data analysis seemed simple and obvious. It wasn’t any more complicated than making graphs and the answers seemed obvious too. For example, I didn’t find it at all surprising that most men would prefer to date women under 21 while on average women’s preferred partner’s age increases as they age. One exception was his comparison of words used most often by different ethnicities compared to people of other ethnicities, which I thought was sometimes surprising and also very clever. Overall though, in term of both data analysis techniques and questions addressed, this book felt like a very light version ofย The Signal and the Noise. This isn’t necessarily a bad thing. The questions addressed were interesting even if the answers were ofent intuitive and the lighter introduction to data analysis could be a great place to start if it’s a subject you’d like to know more about.

Do you ever read nonfiction related to the work you do? Do you find it hard to find books that strike that right balance between being too academic and too light for you?



19 Responses to “Dataclysm”

  1. Leah @ Books Speak Volumes

    Hmm, I’ve been really interested in this book, but now I’m not so sure? I like the light tone, but I think I might want some more complex analysis.

    Eww, I’m so skeeved out that most men want to date women under 21. That says some really creepy, sad things about our culture.

    • DoingDewey

      If you’re interested in the topic, I’d definitely recommend The Signal and the Noise more highly. Although it deals with more complex data analysis, I thought it was also pretty accessible.

      To be fair to older men, there are probably evolutionary reasons for men to be more interested in dating younger women who can reproduce more readily while women might be interested in older men who can provide for them. Perhaps this will disappear eventually? It was also true that most older men listed themselves as being interested in older women so it seems like they’re at least mostly aware that it’s not ok for them to be interested in such young women. So, I guess it could be worse ๐Ÿ™‚

    • DoingDewey

      It might not be. It was a fun, light read, but there weren’t any fun facts that surprised me so much I wanted to run and share them with someone. It was only ok.

  2. tanya (52 books or bust)

    Interesting that you compared it to The Signal and The Noise because those two books sit together in my mind as well. For me the lightness of Dataclysm might be a good thing, as too much data makes my head spin.

    • DoingDewey

      They’re definitely similar! I think a big part of the differences was that The Signal and The Noise actually devotes a decent amount of time to explaining the data analysis, while Dataclysm was mostly just graphing things to share some fun facts. I definitely think this could be more enjoyable for some people, but I’d recommend it more highly if the fun facts had been less predictable.

  3. Laurie C

    I’ve read a couple of librarian memoirs and liked some better than others. It’s important that the author has a sense of humor, but probably that probably makes it harder to deal with weightier topics. I think reading people’s reviews of this book is probably enough for me to get the idea of it!

    • DoingDewey

      I agree! I think it’s very unusual for an author to be able to successfully pull off humor about weightier topics and I’m sure it’s a challenging thing to do. I think reading reviews is all I’d recommend. This was fun, but it definitely wasn’t a book I want to run out and convince everyone to read.

  4. Catherine

    I’m the worst at non-fiction and so do have to keep it pretty limited. Too academic and I’m out. I tend to do better with narrative non-fiction and books about topics I like- like fashion and food. Which I guess, about sums me up! Oh and books, of course. ๐Ÿ™‚

    • DoingDewey

      I definitely have some go-to nonfiction topics myself! I’m trying to read more broadly, but most often find myself sitting down with books on science (especially biology) or women in history. One of my favorite things about nonfiction is how there’s something for everyone, no matter what their passions are ๐Ÿ™‚

  5. Audra (Unabridged Chick)

    Great review — been considering getting this one to read for work — popular non-fiction is usually a good springboard for me, but that fluffy/light walks a fine line between helpful and not helpful.

    • DoingDewey

      That’s a good point! I think this would be helpful for learning some basic stats concepts and it’s not bad for learning fun facts, but I wouldn’t pick it up to learn anything complex about data analysis.

  6. Zezee

    That’s a great question you posed. I don’t work in technology so I appreciated how this book was written but I did wonder if those who work in tech would find it annoying or too simply to bother with.

    • DoingDewey

      I definitely didn’t find the simplicity annoying. I think I’m only likely to find simplicity annoying if it leads to inaccuracies and the author did a good job avoiding that. I did find it pretty simplistic, which is part of why I’m not raving about it, but I did enjoy it and I’m not sorry I spent the time on it, especially since it was a nice, easy read ๐Ÿ™‚

  7. Allison @ The Book Wheel

    I listened to this on audiobook while at work and loved it – it was the perfect pace. Like you, I found some of the analysis a bit on the light side but it does make it approachable to a wider audience who may not know the difference between stated and revealed preferences ๐Ÿ™‚

    • DoingDewey

      It’s true! While I might have preferred something more complicated, I’m expect there is an audience for a book that looks at the topic from a more introductory perspective. And it does cover some interesting topics too ๐Ÿ™‚

Leave a Reply

CommentLuv badge

This site uses Akismet to reduce spam. Learn how your comment data is processed.