The collocates page shows the most frequency words that occur within four words (left or right) of the "node word" (e.g. bread in this example), and the collocates are grouped by part of speech.

1. Click on any word in the collocates list (e.g. butter, loaf, slice, banana) to see collocates for that new word.
2. The frequency of the node/collocate pair. For example, bread and butter occur within four words of each other 18101 times in iWeb.
3. The Mutual Information score. By default, the collocates are sorted by frequency [2], but you can change this to Mutual Information score [6].
4. Click to see the node/collocate words together in the KWIC (Keyword in Context) display
5. Indicates whether the node word occurs primarily before or after the collocate. The darker the color, the more frequently the node word occurs on that side of the collocate. For example, the cell before butter is highlighted, because bread / butter (e.g. bread and butter) is much more common than butter / bread. Likewise, the cell after loaf is highlighted, because loaf / bread (e.g. loaf of bread) is much more common than bread / loaf.

Note: click on Advanced Options to see [6] and [7].

6. Choose whether to sort the collocates by frequency (the default) or Mutual Information score.
7. If sorting by frequency, the minimum Mutual Information score for the collocates. The higher the number, the "tighter" the association between the node word and the collocate, and vice versa. If sorting by Mutual Information score, then this is the minimum frequency of the collocates.