Some time ago I pointed out to the second year 2+2 Early Years students that a literature review should be concered with the secondary sources rather than primary ones. Please see short and helpful explanations about their differences: http://library.uwsp.edu/guides/webtutorials/primary.htm and http://www.libraries.iub.edu/index.php?pageId=1483.
However, thinking about application of those definitions to Early Years, it seems to me that it may not always be clear what sources are primary and which are secondary.For example, is the BECTA's report a primary or secondary source? (Probably yes, as such reports normally refer to what has been learned previously and conclusions are drawn for formulating recommendations, rules and policies.) Should a government policy document be analysed as part of a literature review? (Probably not if it does not explain on what knowledge those prescriptions are being founded. Essays and dissertations are more appropriate forms for discussing those documents.)
No doubts that almost anything can be mentioned in a literature review, e.g. to illustrate discrepancies in a policy with what researchers and practitioners were saying. However, the fact of those discrepancies is probably less interesting for a literature review. The problem of those discrepancies should be discussed in other parts of a research paper.
Therefore, the writings which constitute "scholarly literature" (those texts which explore, test, analyse etc. facts, statements and conclusions) should be regarded as the focal point of a literature review.