A couple (of) things – now with added stats!


Two subjects in one post?! Sweet mercy…

Something occurred to me a few weeks ago when I was watching one of my favourite YouTubers, Mamrie Hart (if you have no idea who that is, you need to get some of her in your life by clicking here!). She was making a drink, as she does, and mentioned that I’d need to use “a couple things.” This isn’t the first time I’ve heard this phrase, although I would tend to use “a couple of things.” What struck me as the key factor between these two forms was the variety of English that was being used: American English and British English respectively.

Mostly to satisfy my nerdy curiosity, I decided to look into this a little further, specifically via the ‘poll’ feature on Twitter. I asked this question with these options:

BrE and AmE speakers: which one sounds ‘right?’ Select your variant & then whichever sounds ‘right.’

  • BrE – A couple of things
  • AmE – A couple of things
  • BrE – A couple things
  • AmE – A couple things

After 24 hours, the responses were as follows:

Screen Shot 2016-03-30 at 10.15.46

From this screenshot, it seems like you can identify a few things pretty clearly. However, statistics can be used to manipulate views in an almost effortless manner, mostly by inference, and often by a complete misunderstanding of interpreting the raw results. To exemplify this, here are a few statements that may be said about the data that are technically true, but may lead to misleading inferences (an example for each is shown in parentheses):

  1. Nearly 10 times as many BrE users participated than AmE users (so AmE users aren’t as interested in participating)
  2. The highest selected response was “a couple of things” (meaning it is ‘more correct’ than any other available option)
  3. Over 95% of BrE users prefer the option with ‘of’ (so the majority of BrE users in the population agree with this)
  4. The majority of AmE users prefer to omit ‘of’ than to keep it (and this carries the same inferential force as the previous bullet point)

It’s possible to increase transparency by breaking these down a little more (this time with more appropriate inferences shown in parentheses):

  1. Due to the sampling method and the geographical links held in the social network, nearly 10 times as many BrE users participated than AmE users (so the sample is not wholly representational of both populations)
  2. The highest selected response was “a couple of things” for BrE users, although all other responses were selected at least once (meaning that there appears to be a strong preference for “a couple of things” from the BrE perspective, but sampling issues may affect this; also, more selections is not equivalent to being ‘more correct’)
  3. 26 BrE respondents preferred the option with ‘of’ (so in replacing the percentage with the actual number, it can be seen that the sample number is minuscule in comparison with the potential population number!)
  4. When looking at all responses, out of those who identified as AmE users, 7% chose to omit ‘of,’ but 6% chose to keep ‘of’ (therefore reducing the magnitude of difference between these options)

Of course, there’s no such thing as a perfect methodology. In this case, there are many flaws in addition to those mentioned above: I cannot verify that the respondents reacted truthfully to the question; 24 hours is far too small a timescale to perform an analysis such as this in-depth; it is not open to other varieties of English (e.g. Australian English); and there is no indication whether or not L2 users of English can participate, and in turn, how this variable may influence the results. Also, the term ‘right’ is extremely subjective (but maybe I was aiming for that?)…

Nonetheless, it provides a good initial delve into this particular linguistic phenomenon, and perhaps a bit more backing to look into it in more detail, with a more robust methodology, and to look for other patterns. For example, is this limited to ‘couple (of)’ or are there other collocations where this difference between varieties occurs? Also, a fellow linguist from across the pond (who has recently become Dr Morris!) asked me if I could use the phrase in a sentence, leading to the influences of context and co-text. I was thinking of the phrase

To complete this task, you’ll need a couple (of) things

although it’s possible that it could be used (for instance) as a response:

‘What are you thinking about?’

‘A couple (of) things.’

And, of course, the other message from this post: never blindly trust a reported statistic!


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