This paper is a comparing study of the retrieval effectiveness of the search engines Google, Google Scholar and Scirus. The aim is to find out how good they are at retrieving relevant academic material in the research-field of Library and Information science. The thirty search questions where based on actual information needs collected from exams within the field of Library and Information Science. This method was used to prevent that none of the search engines were given an advantage because of construction of the information needs. The first twenty retrieved documents on the retrieval lists are examined for academic content and relevance. The methods of measuring the effectiveness of the search engines are Precision, Relative Recall and Jaccards Index. Academic content both relevant and non-relevant material for the information need is judged and the result is presented in percent. Binary scale is used for judging the relevance of the retrieved documents. Guiding principles are being followed when it comes to judging whether a document contains academic content or not. Citation only, book links, inactive, duplicate and mirror links are all considered as being irrelevant. Scirus gets the highest scores, thereafter Google and Google Scholar gets the lowest score. All three search engines don’t retrieve the same relevant material frequently so it could be advisable to use all three for greater coverage.