Graphic depicting water data.

Hydrometric Grammar – Is Data Singular? Are Datum Plural?

Good stream hydrography requires careful attention to technical detail.

We care deeply about the precision of our data. So why would we be tolerant in the precision with which we define data?

At issue is matching a verb with the noun. According to Wikipedia, “the word data is the traditional plural form of the now-archaic datum”. From Latin roots, data means “something given,” which implies a fundamental unit granularity.

In modern use the word ‘data’ is used as a mass noun, like water, using the singular tense. I am definitely not modern. I don’t have tattoos, body piercings or dependency on personal electronic devices. I am stuck with the historic, scientific, use of the word data as the plural of datum.

This is a bit of a problem because I never use the ‘archaic’ word datum to identify a single element of data. I will use the terms ‘value’, ‘unit value’, or ‘data point’ for a single unit of data.

I reserve the term ‘datum’ to identify a horizontal plane of reference for a gauge. The offset value for a gauge datum can vary through time (e.g. due to land subsidence) so the term is often associated with a set of values of which only one is valid at any given time.

Nonetheless, whatever the name given to a single unit of data – it is a named thing. This is quite different from anything that qualifies for use as a mass noun. There is no name given to identify a single element of water. It is always called water whether it has the mass of a single molecule or the mass of an entire ocean.

It may be a small thing but I like the integrity that the historic use of the word data brings to any discussion.

I know I am in a decreasing minority as instant messaging and thumb typing change the way we communicate. One thing I am sure of is:

…data that are accurate are better than data that is wrong.

  • Stu Hamilton
    Posted at 1:35 pm, June 2, 2014

    My concern about the trend in usage of the word data is that it reflects a societal change in the understanding of data. The use of data as a mass noun implies an abstract concept. The use of data as a collective noun implies a discretely granular concept.

    I am not arguing that an abstract conceptualization of data is not, occasionally, useful. I am commenting on the rapidity with which the collective noun is being abandoned by society as a whole.

    Every data analyst is acutely familiar with the influence even a single value in a dataset can have on the information content of the whole. The concept of data as a population with inherent variability, where the spread is typically more informative than the central tendency requires a collective, not a mass, noun.

    The use of the mass noun is most prevalent amongst people who don’t work with, or adequately understand, data. Increasingly, it seems, data is merely an abstract blob to most people.

    Data, in an abstract sense, is used to justify an opinion. Data, in a collective sense, are used to systematically discover the truth.


  • Tobi Gardner
    Posted at 11:55 am, June 24, 2014

    Well stated Stu. I agree with your underlying message of sticking to the fundamentals. There is no reason to change from the correct use of the term data, as you’ve outlined above.

Post a Comment