from paper to article

Now that you’ve submitted your paper to the ASAs, how can you turn it into a publication? Two ideas. First, if it is nifty and about social movements, please consider submitting it a special issue of Mobilization that I’m putting together. Deadline is Friday, January 11th.

Second, you should stop calling it a “paper” and start calling it an “article.” Seriously. You might also want to ditch wishy-washy words like: seeks, attempts, looks, and presents. Instead, have hypotheses, analysis and results with consistent, positive effect sizes. And certainly delete the word preliminary. At least that is what the data suggests we find.

While procrastinating from doing my ASA submissions, I downloaded all the ASA abstracts from the last ten years. It turns out that the search feature allows some fairly common words and you can trick it into returning your results 5,000 at a time. I extracted the abstract text of presentations since 2007. This totals 18,710 abstracts. I think this is just about everything that was submitted and accepted to a section session, regular session or roundtable. I then pulled the abstracts for articles published in the last five years in seven general interest sociology journals (Sociological Inquiry, Sociological Perspectives, Social Science Research, Sociological Spectrum, Social Forces, American Sociological Review, American Journal of Sociology and Social Problems.) from Web of Science. This totals 1,616 published articles.

The analysis part is pretty simple.* For all words that were used in at least 50 different abstracts, I counted what proportion of ASA presentation abstracts had that word, and what proportion of published articles had that word. If a word appears in much higher proportion of ASA presentation abstracts than in abstracts from published articles, you might want to think about revising the way you present your contribution or look for different publication venues.

So here’s the list of words that you might want to avoid in revising your paper if you are interested in publishing in a general interest sociology journal:

Word Presentation/Publication ratio
preliminary 27.8
paper 6.6
transnational 6.5
my 5
seeks 4.3
medical 3.6
towards 3.4
attempt 3.4
care 3.3
space 3.3
goal 3.3
business 3.2
look 3.1
see 2.8
purpose 2.8
will 2.7
constructed 2.7
discussion 2.6
concepts 2.5
discourse 2.5
transformation 2.4
i 2.4
come 2.3
order 2.3
practice 2.3
taking 2.3
attempts 2.3
emergence 2.3
systems 2.3
his 2.3
year 2.2
city 2.2
presents 2.2
like 2.2
debates 2.2
production 2.2
project 2.2
observation 2.1
means 2.1
since 2.1
need 2.1
understand 2.1
narratives 2.1
explores 2.1
ethnographic 2
focused 2
successful 2
itself 2
developing 2
issue 2
participant 2
explored 2
seen 2

Here’s some words you might want to add:

Word Publication/Presentation ratio
article 6.2
panel 2.9
net 2.4
increases 2.3
size 2.3
blacks 2.2
positively 2.2
neighborhoods 2.1
composition 2
consistent 2
crime 1.9
lower 1.9
rights 1.9
married 1.9
strongly 1.9
hypotheses 1.9
adolescents 1.9
longitudinal 1.9
rates 1.8
inequality 1.8
stratification 1.8
analyses 1.8
surveys 1.8
substantial 1.8
test 1.8
find 1.7
whites 1.7
results 1.7
adolescent 1.7
black 1.7
attainment 1.7
effects 1.7
suggesting 1.7
variation 1.7
associated 1.7
neighborhood 1.7
associations 1.7

Feel free to make sense of the difference in the comments. Me, I’ll be busy revising our paper article. It turns out it screams presentation.

* If this analysis looks familiar, I did a similar one last year where I compared abstracts in high and low visibility journals.

11 Comments

  1. Posted January 10, 2013 at 1:04 pm | Permalink

    Can you run it treating n.e.c. as a word? That is, is it good or bad to use words that were used in less than 50 abstracts?

    Like

    • Posted January 10, 2013 at 1:21 pm | Permalink

      I haven’t looked at that yet, but my guess would be that it would show up as a bad thing, but it might not be. There is a lot of papers at the ASAs that are nontraditional sociology and the authors don’t necessarily aspire to be in a general interest sociology journals, and you might just be picking this up.

      Doing this sort of analysis for high and low visibility journals might be more informative. Or accepted/rejected from the same journal. My instinct would be that high visibility journals use rare words more frequently.

      Like

  2. Posted January 10, 2013 at 1:15 pm | Permalink

    word to avoid: transnational

    did you mean transitional? though i’m a big fan of avoiding transnational too.

    Like

    • Posted January 10, 2013 at 1:33 pm | Permalink

      transnational it the one. It could be those projects are aimed at demography journals.

      Like

      • Posted January 10, 2013 at 1:38 pm | Permalink

        transnationalism is the hot trendy topic in migration studies, and as there are a couple of big asa sections that deal with it directly and secondarily, it’s not surprising. it’s also a big trend in migration and demography journals… so i don’t know if it’s really a word to avoid for those reasons. in fact, other than the first two, i don’t see why any of those other words are really to be avoided. they seem perfectly acceptable.

        Like

    • Posted January 10, 2013 at 1:51 pm | Permalink

      One way to read these lists are, “hey–what appears in general interest sociology journals is not reflective of the methods and topics that many sociologists are interested in.” Alternatively you could say, “Most good sociology is published in places besides this set of journals so don’t just mimic what mainstream soc journals do.” Another way to read the list is more instrumental: “I want my work to appear in a general interest sociology journal. How can I best frame my research?” I wrote the post adopting the third POV, but I’m very sympathetic to the other arguments.

      Like

  3. Posted January 10, 2013 at 2:20 pm | Permalink

    Seems like there are some words that possibly co-occur frequently together in the journal list. Possibly rerun the analysis looking for the most commonly occurring 2- and 3-grams?

    Like

    • Posted January 10, 2013 at 9:05 pm | Permalink

      In my quick version adding in some of the top bigrams, “results suggest”, “results show” and “results indicate” were the only ones that popped up, and all were twice as likely to be found in a published articled.

      Like

  4. Posted January 10, 2013 at 5:29 pm | Permalink

    This is brilliant! I have my suspicions about the multiple meanings of “panel,” but other than that, I’m ready to change my language whole hog.

    Like

    • Posted January 10, 2013 at 9:17 pm | Permalink

      Thanks.

      Take your pick: panel survey; panel study; panel census; panel data; panel corrected; panel models; or wood paneled station wagon. I know which one is for me.

      Like

  5. Posted January 11, 2013 at 9:22 am | Permalink

    Sharp post. A less compelling analysis: relationship between submission time and acceptance rate.

    Like

Follow

Get every new post delivered to your Inbox.

Join 875 other followers

%d bloggers like this: