All finds of text searches are saved in a special ‘text searches’ part of the Node Explorer. Text searches provide references simply to the text units found; that is, they do not provide the context for the unit. Context can be provided by ‘spreading’ the results of a find, which provides a specified number of additional text units around each of the finds. This is referred to as system closure in NUD*IST. The found text, including any spread to text, is coded at the new node, placed in the ‘text searches’ (or ‘index searches’ in the case of an index search) area of Node Explorer. This means that the text found as a result of a search can be used to do further analyses such as further searches. The node can be renamed and moved to the node tree, or it can be merged with an existing node. The node can also be used to restrict the scope of further searches. If, after due consideration, you are sure that it doesn’t help with the analysis, then delete it.
Index searches These are a particularly strong feature of NUD*IST and are of particular value when you want to move beyond simply establishing a set of categories, and what there is in each of them, to explore interrelationships between categories. As a simple example, you may have set up two or more separate nodes which you may suspect actually refer to the same category or concept. Doing a ‘union’ search gathers all text units coded at the different nodes together and lets you easily explore whether or not they are effectively the same thing. If you decide that this is the case, the results of this search can be used to merge them together. As a different example, you might be interested in contextual relationships between codes, perhaps whether text units with a particular coding are closely followed (say within three text units) by a particular second coding.