3

I've noticed that sorting by date, and relevance has different numbers of results. Why is this?

Edit: Here is a reproducible example:

Here is "foo" sorted by date (15,000 references)

https://scholar.google.com/scholar?scisbd=2&q=foo&hl=en&as_sdt=0,5

enter image description here enter image description here

Here is "foo" sorted by relevance (300,000 references)

https://scholar.google.com/scholar?hl=en&as_sdt=0,5&q=foo

enter image description hereenter image description here

3
  • Which aspect of Google Scholar are you seeing this behavior in? I can't figure out which piece of the UI you are talking about.
    – jakebeal
    Commented Jan 5, 2016 at 3:38
  • I edited to provide example Commented Jan 5, 2016 at 4:18
  • Thanks for the revision, it's a more complete question now.
    – i am me
    Commented Jan 5, 2016 at 4:22

1 Answer 1

6

(I'll repeat an earlier answer of mine, which is relevant here, although the question is not a duplicate. Nevertheless, the question itself is relevant. Why are you interested in how many Google Scholar results a particular search yields in the first place?)

The "approximate" ("about") number of hits Google reports is completely worthless - in all Google searches. To see why, look at this number on both the first and the tenth page of Google hits:

First results page

Tenth results page

When I just did this, I got "approximately 15,200,000 hits" reported on the first results page... but only "approximately 96 hits" on the tenth page.

Given the opacity of Google's search algorithms, which likely have a lot of sampling built in, it's not overly surprising to me that you get surprising results even if you only sort your results in a different way.

In addition, your results will probably vary depending on your search engine bubble (another reason why this number is useless).

2
  • Note: if you click "repeat the search with the omitted results included" on page 10, and page 1 and page 10 both say "about 15,500,000 results." (I don't know how many pages it says this for--at least to page 22.)
    – Kimball
    Commented Jan 5, 2016 at 15:18
  • I'm looking at about 20 to a couple as there's only a couple dozen cites, to begin from based upon the original paper on the topic.
    – i am me
    Commented Jan 5, 2016 at 22:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .