Content analysis is a method of quantitative analysis of text documents widely used in social sciences and humanities. Its essence is to reliably determine the meaning and direction of a specific text message by counting semantic units.
Instructions
Step 1
Any documents containing text messages can be used as an object of research. In particular, newspaper articles, public speeches of public and political figures, books, answers to questionnaires, diaries, letters, official statements, etc. Content analysis can be carried out both manually and automatically. The second option is used to study large arrays of textual data and requires computer technology and special statistical programs.
Step 2
To conduct an independent content analysis, without the use of automatic processing tools, first of all, it is necessary to determine the array of data with which the work will be carried out. For example, if it is planned to analyze the coverage of a regional election campaign in the press, then the required sample will be all newspaper publications on this topic for the selected period.
Step 3
The second step in the content analysis procedure is the selection of semantic units that are directly related to the problem under study. Individual words, names, phrases that carry a key semantic load can act as a semantic unit. For example, in the context of the election campaign, such units can be the names of candidates, the phrases "modernization of the economy", "development of small business", "struggle for power", etc. Moreover, the selected semantic units should be characteristic for all studied texts.
Step 4
The next stage is one of the most important in the entire process of content analysis. It is a codification of units of text. Its essence lies in the development of rules for correlating semantic units with a list of categories of analysis. The result of the codification stage is the development of a codifier, which includes not only a list of observed indicators, but also data about the document in which they are present. If we are talking about newspaper articles, then the name of the publication, city, release date, format, number of pages, page of placement, and the like will be taken into account.
Step 5
After the formation of a sample of messages, the selection of semantic units and the creation of a codifier, they proceed directly to the very analysis of the texts. In practice, this is expressed in the compilation of a dictionary in which each observation (semantic unit) belongs to a certain type or class in accordance with the rules of the codifier. After that, a quantitative calculation of the use of all semantic units is made. An important point is also attributing specific assessments (positive, negative or neutral) to key references. In other words, a fairly fine ranking is required. Scaling by pairwise comparison or the so-called Q-sort method are traditionally used as ranking methods. You can learn more about both of these techniques in textbooks on applied sociology or political science.
Step 6
The content analysis procedure ends with a quantitative calculation of the data obtained and the calculation of the arithmetic mean of the scale for each case. Then the resulting average scores are ranked in a certain way.
