Well, not entirely, of course.
But we did notice some unexpected overall similarities between both online publications
with respect to news articles.
We originally set out to see if we could get a handle on the question of whether or not
British publications use
semicolons more than U.S. publications.
But, as the reader can imagine, to do such a topic justice requires a little more effort
than simply finding a few news articles online.
(However, such a piece is in the works.)
Sports and opinion pieces were not considered. Moreover, anything that didn't resemble "news"
was also not considered. Links to stories similar or relevant to the top stories were also ignored.
In total we analyzed 9 top-story articles from the N.Y. Times and 10 top-story articles from BBC News.
The word length distributions of our sample were remarkably similar.
|
Average Word Length |
Median Word Length |
Standard Deviation of Word Length |
| BBC |
4.92 |
4 |
2.67 |
| NY Times |
4.94 |
4 |
2.7 |
Likewise the sentence length distributions of were also quite similar
with the exception that there was substantially more
variation in sentence length in the The NY Times articles.
|
Average Sentence
Length |
Median Sentence
Length |
Standard Deviation of
Sentence Length |
| BBC |
21.25 |
20.5 |
8.56 |
| NY Times |
21.34 |
20 |
12.25 |
The
readability of both samples were also strikingly similar.
| Readability Index |
Grade Level BBC |
Grade Level NY Times |
| Gunning fog |
15.23 |
14.85 |
| Flesch-Kincaid |
11.94 |
11.62 |
| SMOG |
13.94 |
13.61 |
| Coleman-Liau |
12.59 |
12.93 |
| Automated |
13.02 |
13.36 |
Both the median and the average readability score for both publications was 13th grade (college freshman).
Not surprisingly, the average
lexical density of both samples
was fairly high with BBC News and The NY Times respectively scoring 55.99% and 57.94%.
First and foremost, we acknowledge that our "sample" is hardly scientific in the sense that it's non-random.
However, it is repeatable and it would be interesting to hear from readers who have tried
something similar. Do you get similar results when you try the experiment?
We also caution the reader that we shouldn't consider our sample of articles as being representative
of British and U.S. journalism in general. At best our sample is representative of only
BBC News and the New York Times, and even then, it's a quite a stretch to say our sample
is representative of these publications at all.
Our "survey," if the reader prefers to call it that, was very
informal and no real general conclusions can be made upon it.
This is not to say that we wasted our time, however, as the results are quite suggestive.
Our experiment raises some interesting
questions. By taking a truly scientific, random sample of news articles from both these publications,
would we get similar results?
Moreover, do similar patterns hold, both across other news outlets
and other English speaking countries? Namely, Australia and Canada.
(How would we collect such a sample?)
Finally, semicolon usage, contrary to expectation,
was leaner among the BBC articles than The NY Times articles with a rate
of about 13 semicolons per 1000 sentences for The Times and 6 semicolons per 1000 sentences
for BBC News.