Kriging.

One of the pleasures of my editing work is that it occasionally introduces me to new words, and I’ve just run into one such: kriging. It’s a statistical term, equivalent to Gaussian process regression (whatever that is, and don’t bother trying to explain it to me because even if I understood it at the moment I wouldn’t retain it), and what I want to know, of course, is how it’s pronounced. The Wikipedia article says it’s named after “Danie G. Krige, the pioneering plotter of distance-weighted average gold grades at the Witwatersrand reef complex in South Africa,” and I assume his name would be anglicized as KREE-guh (or, if you prefer, /kri:gə/. But the article goes on to say “The English verb is to krige and the most common noun is kriging; both are often pronounced with a hard ‘g’, following the pronunciation of the name ‘Krige’.” This muddies the waters; the verb form krige would seem to demand a “long i” (/kraɪg/ or /kraɪdʒ/), and “often” implies that it’s (equally? less?) often pronounced with a soft g (/dʒ/). I don’t care either way, and I’m never going to have occasion to say it, but dammit, I want the facts! Any of you know how statisticians actually say this word? (It’s not in the dictionaries — too recent, I presume.)

Comments

  1. Robert Kern says

    I’ve always heard it (and say it) as KREE-ging with a hard ‘g’. I was a geophysics major, and I do a lot of work in the oil and gas industry.

  2. In fact, ‘kridging’ has been around for decades. It’s what you do if you have scattered data points and want to interpolate new points that are consistent, in some explicit sense, with the old ones. It used to be somewhat deprecated because it is computationally expensive– but in the modern world of today, computation is cheap. I’ve always pronounced it (mentally) with a soft ‘g’. But I don’t think I’ve ever actually said it out loud and I don’t think I’ve ever heard anyone else say it out loud.

  3. In the context of the petroleum industry, where kriging is a common method for statistically distributing rock properties in a geocellular model, I’ve always heard it pronounced /kri:gɪŋ/.

  4. I have never heard it spoken. I only remember seeing it as the participle.

    French Wiktionary says it’s pronounced /kʁiʒ/ in French.

  5. I’ve always heard it (and say it) as KREE-ging with a hard ‘g’. I was a geophysics major, and I do a lot of work in the oil and gas industry.

    In the context of the petroleum industry, where kriging is a common method for statistically distributing rock properties in a geocellular model, I’ve always heard it pronounced /kri:gɪŋ/.

    Excellent — that’s what I was hoping!

  6. I took a precision agriculture class in college and there it was pronounced /kri:gɪŋ/ Just like Daniel says.

  7. My favorite kind of kriging (just in terms of name, I can’t speak to its properties) is universal kriging.

  8. Definitely “KREE-ging” (hard g). I’m an oceanographer that does a lot of stats work and that’s been the only pronunciation I’ve ever heard.

  9. Jonathan D says

    Another point for /kri:gɪŋ/. I’ve heard it used by hydrologers.

Speak Your Mind

*