On “Geek” Versus “Nerd”

To many people, “geek” and “nerd” are synonyms, but in fact they are a little different. Consider the phrase “sports geek” — an occasional substitute for “jock” and perhaps the arch-rival of a “nerd” in high-school folklore. If “geek” and “nerd” are synonyms, then “sports geek” might be an oxymoron. (Furthermore, “sports nerd” either doesn’t compute or means something else.)

In my mind, “geek” and “nerd” are related, but capture different dimensions of an intense dedication to a subject:

  • geek – An enthusiast of a particular topic or field. Geeks are “collection” oriented, gathering facts and mementos related to their subject of interest. They are obsessed with the newest, coolest, trendiest things that their subject has to offer.
  • nerd – A studious intellectual, although again of a particular topic or field. Nerds are “achievement” oriented, and focus their efforts on acquiring knowledge and skill over trivia and memorabilia.

Or, to put it pictorially à la The Simpsons:

Both are dedicated to their subjects, and sometimes socially awkward. The distinction is that geeks are fans of their subjects, and nerds are practitioners of them. A computer geek might read Wired and tap the Silicon Valley rumor-mill for leads on the next hot-new-thing, while a computer nerd might read CLRS and keep an eye out for clever new ways of applying Dijkstra’s algorithm. Note that, while not synonyms, they are not necessarily distinct either: many geeks are also nerds (and vice versa).

An Experiment

Do I have any evidence for this contrast? (By the way, this viewpoint dates back to a grad-school conversation with fellow geek/nerd Bryan Barnes, now a physicist at NIST.) The Wiktionary entries for “geek” and “nerd” lend some credence to my position, but I’d like something a bit more empirical…

“You shall know a word by the company it keeps” ~ J.R. Firth (1957)

To characterize the similarities and differences between “geek” and “nerd,” maybe we can find the other words that tend to keep them company, and see if these linguistic companions support my point of view?

Data and Method

(Note: If you’re neither a geek nor a nerd, don’t be scared by the math. It’s not too bad… or you can probably just skip to the “Results” subsection below…)

I analyzed two sources of Twitter data, since it’s readily available and pretty geeky/nerdy to boot. This includes a background corpus of 2.6 million tweets via the streaming API from between December 6, 2012, and January 3, 2013. I also sampled tweets via the search API matching the query terms “geek” and “nerd” during the same time period (38.8k and 30.6k total, respectively). Yes, yes, yes… I collected all the data six months ago but just now got around to crunching the numbers. It’s been a busy year!

A great little statistic for measuring how much company two words tend to keep is pointwise mutual information (PMI). It’s commonly used in the information retrieval literature to measure the cooccurrence of words and phrases in text, and it also turns out to be a good predictor of how humans evaluate semantic word similarity (Recchia & Jones, 2009) and topic model quality (Newman & al., 2010).

For two words w and v, the PMI is given by:

{\rm pmi}(w;v) = \log\frac{p(w,v)}{p(w)p(v)} = \log p(w|v) - \log p(w) ,

where in this case p(\cdot) is the probability of the word(s) in question appearing in a random tweet, as estimated from the data. For instance, if we let v = “geek,” we compute the log-probability of a word w in the “geek” search corpus, and subtract the log-probability of w in the background corpus.


The PMI statistic measures a kind of correlation: a positive PMI score for two words means they “keep great company,” a negative score means they tend to keep their distance, and a score close to zero means they bump into each other more or less at random.

With that in mind, here is a scatterplot of various words according to their PMI scores for both “geek” and “nerd” on different axes (ignoring words with negative PMI, and treating #hashtags as distinct):

Many people have asked for a high-res PDF of this plot, so here you go.

Moving up the vertical axis, words become more geeky (“#music” → “#gadget” → “#cosplay”), and moving left to right they become more nerdy (“education” → “grammar” → “neuroscience”). Words along the diagonal are similarly geeky and nerdy, including social (“#awkward”, “weirdo”), mainstream tech (“#computers”, “#microsoft”), and sci-fi/fantasy terms (“doctorwho,” “#thehobbit”). Words in the lower-left (“chores,” “vegetables,” “boobies”) aren’t really associated with either, while those in the upper-right (“#avengers”, “#gamer”, “#glasses”) are strongly tied to both. Orange words are more geeky than nerdy, and blue words are the opposite. Some observations:

  • Collections are geeky. All derivatives of the word “collect” (“collection,” “collectables”, etc.) are orange. As are “boxset” and “#original,” which imply a taste for completeness and authenticity.
  • Academic fields are nerdy“math”, “#history,” “physics,” “biology,” “neuroscience,” “biochemistry,” etc. Other academic words (“thesis”, “#studymode”) and institutions (“harvard”, “oxford”) are also blue.
  • The science & technology words differ. General terms (“#computers,” “#bigdata”) are on the diagonal — similarly geeky and nerdy. As you splay up toward more geeky, though, you see products, startups, brands, and more cultish technologies (“#apple”, “#linux”). As you splay down toward more nerdy you see more methodologies (“calculus”).
  • #Hashtags are geeky. OK, sure, hashtags are all over the place. But they do tend toward the upper-left. And since hashtags are “#trendy,” I take it to mean that geeks are into trends. (I take this one back. The average PMI score for all hashtags is 0.74 with “geek” but 0.73 with “nerd.” The difference isn’t statistically significant using a paired t-test or Wilcoxon test, or practically significant using a common-sense test.)
  • Hobbies: compare the more geeky pastimes (“#toys,” “#manga”) with the more nerdy ones (“chess,” “sudoku”).
  • Brains: the word “intelligence” may be geeky, but “education,” “intellectual,” and “#smartypants” are nerdy.
  • Reading: “#books” are nerdy, but “ebooks” and “ibooks” are geeky.
  • Pop culture vs. high culture: “#shiny” and “#trendy” are super-geeky, but (curiously) “cellist” is the nerdiest

The list goes on. If you want to poke around yourself, download the raw PMI scores (4.2mb) and let me know in the comments what you find. Since many people have asked: I computed PMI for all words appearing in the search tweets with “geek” and “nerd” (millions) and then manually scanned roughly 7,500 words with positive PMI scores for both. The scatterplot contains about 300 words that I hand-picked because they made sense.

(Update: I learned that Olivia Culpo — a self-described “cellist nerd” — was crowned Miss Universe on December 20, 2012. The event was heavily tweeted smack in the middle of my data collection, so that probably explains the correlation between “cellist” and “nerd” here. It also underscores the limitations of time-sensitive data.)


In broad strokes, it seems to me that geeky words are more about stuff (e.g., “#stuff”), while nerdy words are more about ideas (e.g., “hypothesis”). Geeks are fans, and fans collect stuff; nerds are practitioners, and practitioners play with ideas. Of course, geeks can collect ideas and nerds play with stuff, too. Plus, they aren’t two distinct personalities as much as different aspects of personality. Generally, the data seem to affirm my thinking.

I wonder how similar the results would be if you applied this method to the Google Books Ngrams corpus, or something more general instead of a niche media like Twitter. I also wonder what other questions might be answered with this kind of analysis (for example, my wife and I have a perennial disagreement over which word is wetter: “moist” vs. “damp.”).

Finally, when I mentioned to a friend that I was going to write up this post, she said “Well, I guess we know which one you are.” But do we really? I may be a science nerd, but I’m probably a music geek

Update (June 25, 2013): Woah. This has gotten more attention than I ever anticipated. A few impressions. (1) Prior to writing this, I had no idea there was a “geek vs. nerd” holy war in certain corners of the Internet; fueling these flamewars was certainly not my intent. Lighten up! (2) I fear I’ll be better known for this diversion than for any of my “real” research. To be clear: this was a fun way to kill a few hours on a Saturday afternoon, not necessarily my best science. I think the writeup here is sound and self-evident, but I’m the first to acknowledge that there are better corpora, methods, and analysis techniques — which could use a grant, grad student, and/or more than an afternoon — for uncovering this all-important “Truth.” (3) For those interested in the etymologies of “geek” and “nerd,” I found this cool writeup.


309 Comments Add yours

  1. Eliza Gomez says:

    Another useful way to track the use and frequency of each word is through the Google Ngram Viewer. Here’s a graph tracking Geek vs. Nerd in googles literary corpus from 1980-2008.

    One 1989 book says “A geek is a carnival performer who bites the heads off live chickens and snakes.”

    Geek surpasses Nerd in frequency in late 2000, and we see the emergence of such books as “The Geek Handbook: User Guide and Documentation for the Geek in Your Life”


    Love from an English nerd

    1. Eliza Gomez says:

      Whoops just read the last paragraph and saw the link to Ngram. But actually take a look at the usage in context, it’s pretty wild.

  2. nivangio says:

    Fantastic!! Once more statistics and text mining confirm the real difference between geeks and nerds.

    As I always said, I am definetly a nerd but never a geek

    1. Timelord says:

      Some are simply pioneers 😉

  3. angeli says:

    “I may be a science nerd, but I’m probably a music geek” That was exactly my conclusion upon reading this article. Also, lol at cellists.

  4. Cassandra says:

    Nerd all the way. Except in the case of music, where the stacks of records in my studio and the pile of synthesizers in my basement indicate I cover the board on all sides. Including vegetables. Excepting cellist.

  5. hiya says:

    Damp is “wetter” than moist

  6. ibolyabodor says:

    Thanks for this article. It made me realise that there is nothing wrong with me willing to learn more statistics…

  7. JimK says:

    Regarding the moist vs damp: to my mind damp is a surface condition and moist applies to a volume. Now, is this a geeky or a nerdy interpretation?

    1. Peg Pearson says:

      nerdy! i like this distinction. i think of damp as a description of something I would expect to be dry, a surface condition, whereas moist is a degree of wetness and refers more to the thing itself . I personally think of moist as being wetter than damp.

  8. Mike says:

    Why did you use a paired t-test instead of an unpaired one?

    1. burrsettles says:

      Because there are two measurements for each #hashtag: the PMIs for “geek” and for “nerd.” The paired t-test gives us a more meaningful measure of whether the PMI for one is systematically higher than the other.

  9. Paul Crowley says:

    Would be lovely to have a higher-res version of the geek/nerd plot!

    1. burrsettles says:

      I just added one, check the image above.

  10. Migraine says:

    What’s not on that chart, and should be, is Shakespeare. I am both a Shakespeare geek and nerd (although my geekiness and nerdosity is not limited to that). I’m not alone. And there is a surprising degree of overlap between that Bill and various other Bills.

    1. wifosaurus says:

      Bill is surprisingly absent from Twitter, it would seem. That, or there were too many variations of his name to show up on the chart. Bill, William, Shakespeare, misspellings of Shakespeare (so many), Billy, Will, #BillShakespeare, #WillShakespeare, #WilliamShakespeare…who knows how many others.

  11. wifosaurus says:

    I wonder if you could provide a larger image of the chart for those of your readers that wear #glasses. I’m guessing there are a lot of us!

    1. burrsettles says:

      Check the link, I just added one.

  12. I’ve seen the same argument made, but with the terms reversed. Of course, they didn’t have the dark wizardry that is math on their side, so I’m inclined to believe your assessment.

  13. I love it! My wife and I now agree that I am definitely a nerd!

  14. krakowian says:

    So… nobody’s referenced the song “Pencil-neck geeks”? In that song, there’s no difference between the two–it’s pretty old, though–70s or 80s. Back then, I don’t remember making any distinction between the two. I wonder when this started?

  15. I’d love to see some kind of Geek vs Nerd test based on selection of keywords. It’s probably out there already somewhere..

  16. William Dwyer says:

    Thanks. I think… I’ve always wanted to be able to tell people that I’m a geek or a nerd but never knew which term applied. Now I know I can extoll my nerdness (or is it nerdity?) with confidence!

  17. seancallanan says:

    As someone who is unashamedly a Sports Geek (that is my company), enjoyed the article even if I am an oxymoron.

  18. mathiasmag says:

    Is e meaning of the words consistent over time? If we took some other data the spanned many years and partitioned it in five year slices, would we then see the meaning change over time? I imagine geek has become much more positive over the last 10 years while nerd may have begun to even more take on the meaning of just being socially awkward without requiring any of the other traits described.

  19. Thomas says:

    I really don’t consider your analysis to be all that accurate. In fact, judging by the way you wrote the article (and correct me if I’m wrong), you already had your conclusion set before performing the experiment.

    Yes, according to your data the geeks tend to focus on collections, but I find the evidence suggests that geeks are more *practitioners* of their subjects than nerds. For example, both cosplay and #cosplay appear deep in the geek spectrum. Cosplay is inarguably a form of “practicing” a fandom.

    Nerds are not necessarily achievement oriented, but specifically *academic achievement* oriented. The presence of entrepreneur and #kickstarter on the geek side demonstrate a strong achievement orientation in the geek spectrum as well as a firm emphasis on the consideration of ideas.

    My own conclusion from the data is that nerds tend to think and study their subjects academically while geeks tend to focus on entertainment and actively involve themselves in their subjects culturally. For example, a Dr Who geek would dress up as a blue phone box and go to conventions. Nerds tend to focus on intellectual subjects and study the trivia of their subject. For example a Dr Who nerd would know every fact and figure and obscure canon reference for the series but may never have a costume or participate in any convention or event. Both would likely cite me for not spelling out Doctor Who.

    Just as a side note, could this study have been done 10 years ago, the results would have been significantly different. “Geek” has become pop culture, and thus many of the traditionally “geeky” key words have migrated closer to the nerd side of the spectrum as more fashion-centric keywords have invaded on the geek side.

    I would like to see a broader study done with different data sets. I don’t use twitter, pinterest, or google so, using me as an example of a class, there is a large subset of geek/nerd culture that would not be included in such studies. I do use facebook (it’s how I found this study), but I don’t believe facebook has any open data sets to work with. If you could combine data sets from Google, Yahoo! and Bing that may provide a more comprehensive working set.

    I conclude by suggesting that when speaking of my entertainment interests and hobbies (movies, music, games etc) I am a geek; when speaking of my intellectual interests and hobbies (languages, cultures, religions, meaningless research like this) I am a nerd. What does that make me? A “gerd”? A “neek”? Perhaps I’m a greek, if you grock.

    1. burrsettles says:

      Actually, I like this interpretation and I don’t think it’s that far off from what I had imagined… perhaps I just didn’t articulate as well. Hence why I consider myself more of a science nerd but a music geek.

      1. I actually thought it was a fine distinction. I’m not entirely sure “practicing” fandom (via cosplay) can be considered equivalent to practicing what the cosplay/fandom actually is.

        For example, I often cosplay as a Klingon [cellist] with my Klingon band, which isn’t the same thing as actually writing/performing Klingon music or analyzing Klingon music theory (activities which I also do). While the latter might be closer to being a form of [nerdy] practice, I rarely feel like the former is.

        All that just boils down to the issue that, well, Klingons are a fictional race anyway so how can my Klingon music activities and practice really be, well, a form of practice? For example, Cosplaying a Klingon warrior isn’t the same thing as practicing (i.e. training) to be a Klingon warrior.

        Then there’s simply the uneasy relationship I have with various fandoms. Sometimes I wonder how different the world might be if all the folks putting so much energy into a culture centered around fictional cultures would actually learn about ‘non-fictional’ cultures–especially those in their own regions. I’m often bemused that folks know more about me-qua-Klingon than me-qua-Thai.

    2. Fantastic alternative thinking. Thank you for rounding whilst sharpening my own perspective on the matter. You and Burr both appear gracious hosts of conversation. Appreciate the educational folly!

  20. Peg Pearson says:

    interesting comparison of common usage of the terms. I haven’t read all the comments but maybe there is a degree of introversion or extroversion in the equation. Nerds are simply into their own thing whereas Geeks are more in your face about their obsessions?

  21. I’d say I’m a music nerd, but a geek in most other interests.

  22. A geek cares too much about things no one else knows about, a nerd knows too much about things no one cares about.

    For instance, a grammar nerd would rewrite the above sentence so that it doesn’t end in a preposition. A grammar geek would go on a tirade about the fact in ends in a preposition. It is possible to be both.

  23. I was with you as a probability nerd until you revealed yourself as a stats *geek* with that frequentist t-test. Do you have a MODEL that says PMI should be Gaussian-distributed? Well? Do ya??

    1. burrsettles says:

      Touché, but the 0.01 diff isn’t significant under a Wilcoxon test, either. 🙂

  24. Clara says:

    Thank you so much for accidentally helping me with my MA thesis!
    A friend pointed out this article to me for fun reading. Then I saw your link to your “real research” and I thought, well ok, maybe a nerd has something cool to contribute to my extremely nerdy MA thesis in linguistics. I’m writing about pragmatic speech barriers in the communication of humans and machines with a special focus on AI. And voilá, you’ve written several really cool and helpful articles about machine learning! So thank you, thank you, thank you:-)

    Love from a German SciFi nerd and linguist:-)

  25. owenmp says:

    Nice and interesting chart, however you should delete autism and autistic. Use of the words geek or nerd in reference to the medical terms autism and autistic may increase the undeserved stigma patients of autism endure.

    1. Thomas says:

      The presence of these words indicates an association, whether good or bad, between autism and geek/nerd-ism. There’s actually evidence to support a medical association between them (i.e. geeks/nerds tend to fall into the autism spectrum more than the general public).

      But whether or not that is the case, to delete the terms would be tampering with his research results, which is counter-productive to science. If you don’t want autistic persons to have a stigma, then don’t stigmatize them.

  26. SFG says:

    Can you have negative PMI; ie, words that are negatively correlated with geek or nerd-dom? What do they look like? Sports? Fashion? Religion?

    1. burrsettles says:

      Yes, negative PMI is possible: that implies that the occurrence of “geek” (or “nerd”) in a tweet decreases the chance to seeing the other word. I link to a spreadsheet in the article that has a bunch of the raw PMI scores, including negative ones.

  27. hswerdfe says:

    perhaps you should mention “data noise”, or limitations or something caused by the short 1 month timeframe. My understanding is that Olivia Culpo (Miss Universe Dec 2012), is a self described “Cellist Nerd”.

  28. Brian Paone says:

    Stick with real science, Whoever Wrote This Abomination. It’s abundantly clear that you eggheads will never understand. 😉

  29. Dan says:

    Nice. I have always taken nerd to mean someone who had short comings in the areas of social interaction and Geek to be some one who is intensely interested in some subject. So frequently some one is such a geek that he/she has little time for socializing and thus are a nerd too. But then you get some one who is very socially apt but also “in to” something. And of course you have people who are just not very social.

  30. Jonathan says:

    As a cellist and a nerd, I shall not complain about that occurrence.

  31. Nick says:

    This seems to me a lot like the difference between weight and mass or precision and accuracy!

  32. Waldo Smith says:

    As an earlier comment notes, a geek was originally a carnival sideshow performer. Modern geeks may raise urban chickens and could morph etymologically into zombies. There is a famous old movie.

    “:How can a man sink so low? ” – Nightmare Alley (1947)

  33. Martin says:

    your color code should represent the continuum too. It only discriminates between geek and nerd but it doesn’t represent the intensity. According to colors, for instance “vegetables” is as nerdy as “cellist”. You should implement gradients

    1. burrsettles says:

      I tried that, but (a) it was hard to find a legible color scheme and (b) it was becoming hard to implement for an afternoon diversion. So I went with a background gradient instead.

  34. Ben says:

    I am suspicious about this data.
    Apparently the nerdiest thing in the list is: “miniskirt”

    1. burrsettles says:

      I found that funny, too. I haven’t had the chance to look for the tweets responsible to see the context, but my guess is that there was a spike of tweets in December about “nerdy schoolgirl” type fashion, which tipped it well above the baseline probability of “miniskirt” in the 2.6M corpus (similar to “cellist” as explained above). For instance, nerd mini skirts on Etsy.

  35. nerdintraining says:

    A post on Gizmodo brought my to this post which sent me to your google scholar page which lead me to your home page which lead me to Duolingo which I never would have found without your post. What a wonderful thing the internet. A truly wonderful thing.

  36. D.Thomas says:

    100% agree! And the SImpsons reference is perfect! I’ve never gotten into a debate over this, but the media always confuses the two and it quietly irks me. Thank you!

  37. AJ says:

    Your statistics show correlation, not causality as no variable was manipulated a priori… (i.e. phrases such as “Collections ARE geeky” and “Academic fields ARE nerdy” cannot be stated, only that they are correlated…) Thoroughly interesting read though! 🙂 #nerdforlife #geekforever

    1. burrsettles says:

      Good point. So, if you have any ideas about how to do a controlled study that manipulates hundreds of thousands of word variables with real human subjects, I’m sure the entire field of distributional semantics would love to talk to you about it. 🙂

  38. mattmoyle says:

    As someone who could justly be accused of being a nerd, a geek, AND a sports fan… I’m thinking of the frequent reference to the practitioners of advanced analytics (especially in baseball) as “sports/stats/numbers geeks” or “[…] nerds,” pretty much indiscriminately as far as I can tell. (Though I haven’t done anything close to this kind of study, I just read too many sports blogs when I should be doing my more conventionally nerdy work.)

    Loved this piece, in any case.

  39. Sarah H. says:

    Would it make sense to try it without distinguishing the hashtags, i.e. to include both spiderman and #spiderman as the same thing? I don’t understand this technique to understand well enough if it would matter…and maybe these things are sufficiently different that your way is better justified.

    BTW, this is Sarah Hamersma, who you may or may not remember from Madison. My husband Rob somehow came upon this. You are internet famous!!!

    1. burrsettles says:

      Hi Sarah… of course I remember you guys! Collapsing hashtags is a fine thing to do… I think I mainly wanted to see if there were subtle distinctions between a word and its hashtag, or to see if hashtags in general were more associated with one over the other. In the end, I think separating them probably just introduced noise into the analysis. 😦

  40. StarCo says:

    When I saw the Simpsons characters I wondered what I am, since I’m like the guy at the left, dressed like the one at the right. When I saw the graphic reality smacked me in the face. I cover almost all of it! LOL.

  41. JD Kael says:

    “Moist” is wetter than “Damp”

  42. Henry Marx says:

    Have you considered partnering with a linguist? You might be able to co-author a paper. Semantics isn’t really my thing, otherwise I could probably suggest someone.

  43. James says:

    Well, if you think you got a lot of attention before, I came here via a news article reporting on your study via NBC. You are getting national media attention now.

    I wonder where my wife, a former software engineer but now public librarian, would fit into this? Or me, a disabled vet electronics technician but now an editor?

  44. vinnatron says:

    Fascinating stuff! A very interesting read – I always find the geek/nerd divide a very interesting area, and I love the idea that the nerdy side contains more abstract terms than the geeky side which is full of ‘stuff’ as you put it. Well done on this.

  45. Hatsune Mikk says:

    People used to call me a geek than a nerd, but than i took a nerd test and a geek test (2 separate ones and I was apparently more of a nerd than a geek.

    1. KJM says:

      Nice. Where are these geek and nerd tests you mentioned? Wonder if the results (and the words used in those tests) would correlate with the PMI results in this post?

  46. Kharn9 says:


    Also the first word I saw on the graph, cracked me up.

  47. Becca Ne' says:

    I always thought of myself as a geek (sort of a generalist though…), but your detailed evaluation paints me as walking the line between geek and nerd. I always wondered why even my geeky friends think I’m so weird. Thanks for the clarification!

  48. hopoffthis says:

    Reblogged this on Hopoffthis.com and commented:
    Geek Versus Nerd is an awesome read.

  49. suzannadanna says:

    I keep saying that I want a neon sign ‘Geek’ flag for my office once I become a professor, as I fly that flag proudly. However the fact that I am an academic (and got excited over this data) puts me squarely in the ‘Nerd’ camp. As usual, I find myself hard to codify!

  50. Professor Frink says:

    What a nerd. Poster now on wall being analyzed by tens of people.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s