Google vs. Google Scholar
Google is a well known search engine that add almost anything it can find on the web in its search results. The disadvantage in using Google in a literature search should be obvious - it gets lots of information of very dubitable quality.
Google Scholar is also well known and heavily used. It should not be ignored but neither should it be your first resource. Google Scholar looks specifically for papers, books, conference presentations, etc. BUT it still does not distinguish between scholarly and not.
Ranking of results - Google uses a very complex algorithm to determine the order of the results it will list. The problem is that only some Google employees know the algorithm. In some informal research, people have found that articles with lots of citations to them do show up at the top of the list but so do articles that very few people have ever cited and that are, therefore, likely to be less important.
Literature Databases vs. Google Scholar
Literature databases such as SocIndex use published criteria to determine the publications it is going to index. The database will often distinguish between types of literature so you can look specifically for peer-reviewed articles or theses and dissertations if that is what you need. The search engines for databases are designed by experts - usually librarians. These database search engines have been designed to help you create sophisticated searches that will help you get to the literature you want and have to look at relatively little irrelevant literature.
Key to a successful literature search
Use multiple databases. Use all the databases that you have access to that might contain information on your topic. You can certainly include Google Scholar in that list, but you must spend more time and effort evaluating your results from Google Scholar than from other databases where there are experts evaluating what goes into the database in the first place.
Peer-Reviewed journals give the highest quality information while magazines and newspapers offer lesser quality articles. Below are descriptions of each type of publication and the information contained in them.
Journals are scholarly, reliable information. You still need to evaluate the information, even the best experts make mistakes.
Reay, Mike. 2010. "Knowledge Distribution, Embodiment, and Insulation." Sociological Theory 28(1):91-107. doi: 10.2307/25746215. (all the examples below come from this article)
Peer reviewed journals are a special class - they are the creme de la creme of information sources.
What does peer review mean?
Articles submitted to a journal.
The journal send the articles out to 1 to 3 experts in the field without identifying the author.
The experts review the article and send it back to the journal with recommendations for improvements or approval.
The article is sent back to the author with the recommendations and a tentative approval or rejection of the article. The reviewers remain anonymous.
If improvements are made and there was a tentative approval, the article is resubmitted and published.
In other words, the information is vetted by experts before it is even published.
Most databases allow you to limit your results by peer-reviewed journals.
Example
Keller, A. G. 1916. "Sociology and Science." Nation 102(2653):475-78. (all examples below come from this article)
