Searches and Engine Results

A research article confuses the number of results and searches.

Search engines are an important means of using the internet. By typing a query into a search engine, the user gets results. From there, people can find related pages to their search terms.

Between volumes and results

How many times do people search for a term (or group of terms)? This is an important question. Google provides a tool, which produces indices of search terms and topics. This is different to the number of results a search term gives.

A published article in The Journal of General Psychology makes a major error. The methods section states:

To locate the amount of search results for a Google query one can set the date or date range in the advanced settings of one’s browser or by using the following query suffix (“query” after:YEAR-MM-DD before: YEAR-MM-DD).

The article confuses the number of results for the number of searches.

Search query “laid off” resulted in 58.5 million Google searches post-pandemic which was an increase of 69% from 2019.

This is not what the researcher intended. (Image: Journal of General Psychology)

There is also the issue of exact matches. Search results for ‘laid off’ includes pages where those two words appear — but not in that order. The number of search results is not the intended measure. The author recognises the severe error, and has contacted the journal.

Redoing the analysis

Using Google Trends, we can find a normalised number of searches in Google. There are limitations:

  1. It is a sample of the billions of daily searches.
  2. The Google Trends index runs from 0 to 100, in whole numbers.
  3. Google Trends excludes searches by very few people and duplicate searches. It also filters out those with special characters.

For this index, 100 represents the maximal search volume for the time (such as day or month) and the location.

For the three search terms: all follow a similar pattern. There is a spike in search activity in March to April 2020. That activity has an irregular subsidence, returning to volumes seen before the pandemic.

Search words could be in a different order. Exact matches show similar spikes. ‘Laid off’ is — by far — the most common query of these three employment-related searches.

Of the three terms, ‘laid off’ is the most common. (Image: R Pubs)

Google Trends offers some insight into what interests people. There may be spikes in search interest for unclear or unspecified reasons. Google Trends is not polling data: it is one data source for analysis among others.

The R code for the graph is available on R Pubs and GitHub.

This blog looks at the use of statistics in Britain and beyond. It is written by RSS Statistical Ambassador and Chartered Statistician @anthonybmasters.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store