by Liz Walter

Has there ever been a time when we’ve been so dependent on data? All over the world, people are anxiously looking at graphs and charts tracking the progress of Covid-19. In this, the first of two posts, I look at the language associated with the word data itself. My next post will cover words and phrases used to describe what the data shows. While this language is particularly relevant at the moment, I hope you will find it generally useful too.

A note about the word data to begin with. This word is of Latin origin, and in Latin it is plural, with the singular form datum. Most English speakers operating in a non-scientific context now treat it as a mass noun and make the verb agree accordingly: The data shows that … However, some people prefer to use it as a plural noun, and the plural use is common in scientific contexts: The data show that …

Regular readers know that I often focus on collocation, and there are many nice collocations connected with data. When we get data, we collect, gather or compile it, and the most typical prepositions are on or about:

The questionnaires enable us to collect data on customer satisfaction.

They have set up a database to compile data about ship movements in the area.

When we try to work out what the data shows us, we analyse (UK)/analyze (US) it, and when we give the results, we often say that the data shows or indicates something:

We use statistical models to analyse the data.

NASA climate data indicates that global warming is continuing.

Of course, the quality of data is extremely important. Good data is often described as accurate, reliable or (a particularly nice collocation) robust:

We need more accurate data on the causes of death.

The data was not sufficiently robust to draw firm conclusions.

Typical adjectives to describe poor data are faulty, incomplete or weak:

Hospital ranking systems were based on faulty data.

The authors of the report admitted that their data was weak.

In my next post, I will present some of the language used to talk about what data shows, particularly when numbers increase, decrease or remain the same.