KML News Blast

Google Scholar goofiness

Google Scholar is not a research database but people use it because it appears to provide the ease of "one-box" searching while producing scholarly results. Unfortunately, it lacks the most basic quality control found in research databases, so do be aware of its limitations when you use it.

To see it's goofiness revealed, go to the advanced search screen in Google Scholar, enter search term(s), and then set date limits of 2011-2025. Of course, there shouldn't be any items for those dates yet, but you will get results. Browsing your results will reveal several problems:

1. Google searchbots look for numbers that look like years, but the searchbots will select numbers that have nothing to do with the date the item was published. These date mistakes are obvious when using the ridiculous publication dates as search limits, but the ridiculous results demonstrate that date range limiting in Google Scholar is hazardous.

2. Author names are frequently incorrect, because the searchbots can't truly recognize names and Google doesn't provide quality control to find and correct the errors. For example, do a search for bullying and limit to 2011-2025. Scan down the results and see Google's attempt to identify the authors names -- K. Theme, GI Procedures, D Learning, P Steer, etc. Among the first 10 results, only one correctly identifies an author.

3. Only one of the results in the search on bullying is scholarly.

Most of the time, you won't be aware of these problematic results, because Google Scholar will display at the top of its results list the items from known scholarly publishers such as Science Direct, etc., where the quality and consistency of publisher data makes it possible for Google to mine the data more accurately. Given that most people scan the early screens and never reach item 15,000 or item 30,000, they will see more accurate authorship and mostly scholarly items.

Library research databases structure the information about articles. That makes it possible to refine searching accurately by date, author, title, source, etc. in order to produce more focused search results than are possible in Google or Google Scholar.