Click on one of the following to learn
How to interpret the results of

• A simple search

  (Click on "A simple search" to close) Whenever you perform a search in the corpus you will get a KWIC concordance (i.e. Key Word In Context). The KWIC concordance reflects tendencies in the English language. It is not a source for what is right and what is wrong, but it is a fairly reliable source as to what the tendencies are. For instance, you will get a concordance containing the string "he don't", which is an ungrammatical form. You will also find that this form is mostly found in direct speech and/or slang. If you search for "he doesn't" you will not only get many more hits, but you will also see that it is distributed across most text genres and registers. Other tendencies that may be discovered include word forms, spelling, national varieties, to mention a few. Here are some examples: If you type in the search strings "dreamed" and "dreamt", you will get the following concordances displayed (the Key Word is shown in bold): These two concordances tell you that the verb dream has two past tense and past participle forms in English. Both dreamed and dreamt have been found in such contexts. The concordances also tell you which of the two is the more common form in the corpus; in this case dreamed (with 121 hits vs. 19). The results from the corpus indicate how words and expressions behave in similar contexts outside the corpus. At the end of every line there is a code written in red. The code tells you what kind of text the concordance line has been taken from. If you click on the code you will get to the Corpus text header page, which gives you more information about the text. By clicking on the code link you will for instance find out that the form "snuck" is typically used in American or Australian English. This is not to say that it is never used in e.g. UK English, but it indicates that it is more typically used in the two other varieties of English. If we click on the penultimate code link, we get the following information from the OIE corpus header. Read more about the Header here (Click on "A simple search" above to close)

• An advanced search

  (Click on "An advanced search" to close) As was the case with simple searches, you will also learn about the behaviour of words in context by observing the concordance of an advanced search. If you search for "do* bad*" to find out whether the most acceptable expression is "I'm doing bad" or "I'm doing badly", you will find out that the adverb badly is required in a similar context: The first two hits are not relevant to us, while hits three and four show a different context than the one we are after (does and doing are followed by the adjective bad, but instead of telling us HOW someone is doing, it modifies the noun things. The remaining hits are all similar to "our" context, and it can be concluded that "I'm doing badly" is likely to be the most grammatically acceptable expression. If a word or an expression is NOT found in the corpus, this does not mean that the word or expression is non-existent in English. It may, however, indicate that it is not a commonly used construction. If you are to use such a construction in your own writing you should take care to verify its use in other reliable sources. The sorting function enables you to stray away from the default KWIC result by sorting concordances by left word or right word. This will in many cases make it easier to spot exactly what you are looking for. If you want to find out which word most often follows rancid, sort concordances by RIGHT word and you will immediately see that it is butter. Similarly, if you want to find out which preposition is most likely to precede "terms of", sort concordances by LEFT word, and you will see that it is in: Whether you perform a simple or an advanced search you can find out more about the text from which a result line has been taken by clicking on the red code at the end of the line in the concordance. Read more about the corpus header here. (Click on "An advanced search" above to close)