The process of the qualitative analysis

Finishing the course of qualitative methodology, I wrote some notes about the qualitative analysis, and will not be more fit for a blog entry.

If one wanted, as it is sometimes useful to work on methodology, to reduce and to simplify to the maximum the things, well you can say that the qualitative analysis is composed of 3 moments, which we will briefly touch on below:

  1. The time of extraction of information.
  2. The time of construction of relations.
  3. The time of interpretation of the sense.

The time of Extraction of Information.

This is a time that is easy to overlook, especially when it requires little time to do the analysis (which is very common in practice). The fact that we are predisposed to find meaning in the speech of the people, a skill that is integral to operating in the daily life, makes the job of removing systematic qualitative information is often not taken into account: Promptly, and not a few times during the execution of the soil, generate interpretations, and conclusions. For the same reason, a moment of extraction systematic is even more necessary: in The end, the problem is not so much that the interpretation of ‘immediate’ is wrong as it tends to be extremely incomplete.

It is always necessary a close reading and careful of the text (or image or map or whatever we are discussing). Paragraph by paragraph, sentence by sentence, and in general unity of sense by the unity of sense. A thorough reading should be reflected, where possible, a detailed coding. Although the codes later on usually are grouped is relevant that in the first reading to be very attentive to the various meanings that appear in a speech given.

In a single sentence there are many meanings and oppositions, and to see how each one unfolds and what role they occupy in relation to the text is relevant. A few times it is possible to analyse in depth and detail each element of the speech of the interviews, but in the measure in which one is about at that level is that it understands what the subject wants to say. Do not forget that a good working hypothesis in relation to the discourse analyzed is that everything in the speech is significant.

The time of construction of relations.

A set of codes does not represent an analysis. The codes are related to each other: Since they say together until you have logical relations between them (of opposition, implication, part / whole etc). Once obtained the codes then the next time it requires them to play among themselves.

It is important in this sense to emphasize the following: Although we are presenting the moments sequentially (first to be extracted, then relationships), in the analysis there is real feedback: it may Well be that one reason why I can relate a code with others is simply that this was not completely adequate (some nuance of meaning is lost when you encode it that way, met under a code phrases that perhaps it would be better to separate, and vice versa). Now, do not forget that the difficulty can be in itself part of the discourse, and therefore the code may well remain as such -but check how to set the codes by itself does not cease to be relevant in the task of understanding the speech that we analyze.

How to establish relationships between codes differs between the multiple forms of analysis. Now, the following observations may have some utility on these relationships

  • The relationship can be direct (in a phrase are the two codes) or indirect (there is a relationship of meaning between codes that do not appear in the same sentence); however the relationship should always have evidence in the text (which may be one or several appointments). If the analyst attempts to exemplify the relationship and has difficulty doing so that implies that it is useful to review that relationship
  • The relations should cover the various areas. If I have several patterns of relationships between codes in which each one covers a certain area it is necessary to do to play these schemes to each other. What is not possible is simply to leave separated the various parts of speech. In principle, one of the ideas that serves as the base-case scenario is that we are before a discourse-integrated.

The time of interpretation of meaning.

To be able, finally, to understand a group it is necessary to interpret the whole of what is said. The central question is: from what perspective, from what position it is reasonable to say all the things you say? If the idea of the moment relational is to build a schema that allows to see how they integrate the various elements; the idea of the interpretation is to find the actor, the speaker who makes such integration possible. Implies think about who would be the one that would build relationships that way. In other words, interpreting involves, in some way, to pass from the text that an actor says.

This involves two things which is important to avoid when doing the interpretation:

  • The interpretation is not the summary of what was said. You could say that I’ve ‘summed up’ the information provided by the group at the time of extraction (in which I reduce everything that is said to a smaller set of codes). In the summary there is an actor who ‘produce’ the speech is the speech occurred; and the interpretation is, in some sense, search for the actor behind the speech, find the producer and not the product. It is the job of the actor who makes a speech is a speech (including its coherence and its conflicts)[1].
  • The interpretation should remember who is the speaking subject. And it is important to differentiate between what the subject expresses and what he says on the subject of themselves through their expressions. In particular, the subject expresses observations of various situations and actors. It is not the task of the analyst to get to the level of those observations, because those observations have no evidence. Your evidence is in relation to the speech of the subject, not what the subject is talking about. If group X says ‘what happens is that in And happens ta thing’ we know nothing of, And we know what that X says Y. By the way, what you say And may be perfectly adequate (the groups that we investigated are as observers and the researcher, and have their own ways of obtaining knowledge of the social life); but this is not necessary. In fact, the misunderstanding between groups (which is what X says, And is not necessarily suitable for understanding And) is something that regularly occurs in the field of the social life (and of which researchers are not exempt, since they are also are a few X that say some things about, And their investigated). But beyond that the speech of X says y, the speech of X always says something about X, and that is what we are interested in.

An admonition is final.

What we have done is a simplification, a gross simplification, of the process of analysis will Never more remember that in the real analysis are not followed, these moments sequentially, and which are confused with each other. But, recalling the Wittgenstein of the Tractatus: it is sometimes necessary to climb up a ladder that, once you arrived at the point that interests us, is useful, but to reach the ladder was required. This differentiation is, by the way, does not pass beyond to be an attempt to ladder.

It is central, I think, to remember in this regard that the work of research, if done well, requires a continual going to and fro between moments. The meaning and sense is not read directly, or at one time, but that requires watching what they do and analyzed continuously and only returning on the own steps you can, finally, obtain an interpretation that gives proper account of what the subjects say.


[1] incidentally, this, like all things that have been raised in these notes, it implies a theory. Not be amiss to mention that there are theories that have the relation the other way round: That the speech is produced by the actor. Enter the discussion and critique of them would take us in another direction. But for the sake of mentioning the different perspectives you can not fail to mention such a position.

Begin typing your search term above and press enter to search. Press ESC to cancel.

Back To Top