While the key steps of content analysis may vary among scholars, Hansen and Cottle have proposed a six-step process consisting of definition of research problem, selection of media and sample, defining analytical categories, construction coding schedule, piloting the coding schedule and checking reliability, as well as data preparation and analysis. This paper looks at the first three.
Definition of the research problem
According to Hansen and Cottle (100), this is the premise on which the final deductions such as media roles or social phenomena and textual characteristics aspects of a body of text under scrutiny are based on.
Defining a research problem may involve first developing a succinct conceptualization of the problem and later defining its aspects and categories in regard to the nature of the content to be analyzed (100). A well defined research problem should help the analyst stay focused on the subject of research, all the while concentrating on the significant aspects that are of interest (100).
Selection of media and sample
This involves first selecting a particular media coverage (e.g. newspaper, television or radio, magazine, cinema etc) followed by drawing of a “representative sample” from the body of the selected media. According to Hansen and Cottle, the choice of a particular media may be influenced by the geographical reach, size of audience and type, political stance, accessibility and availability of the research material required (101).
Overall, the choice of the media would depend on the nature of the research problem (101). During sampling specific issues, dates or periods may be used. For more specific analyses, Hansen and Cottle recommend the coverage that occurred before and after the events be considered. Otherwise, for broad and general coverage, the sampling is more open but care should be observed to avoid bias of any sort.
The final stage of sampling involves filtering the gathered material for “relevant content” (104). The relevant content should match the research problem and the literature review undertaken.
However, determining the relevancy rests upon the analyst who should ensure that the “representative aspect” is not distorted during this process (104). At the end of sampling, the content analyst, sorts the exact content (e.g. article, report etc) that relate to the subject of interest (105).
Defining analytical categories
In this stage unique text characteristics or “dimensions” that relate to the research question are identified. The analyst could use generic categories common to most analyses such as ‘medium’, ‘date’, ‘position within the medium’,’size’,’type/genre’, among others.
Hansen and Cottle consider sound knowledge on the content to be analyzed critical to developing appropriate and relevant categories (105). Hansen and Cottle have discussed four categories common to sociological-oriented analyses. These are:
Actors/source/primary definers and their attributes
Hansen and Cottle assert that this cohort hold the key to understanding media roles in social representation and power relationship in society (108). However, they argue that each of this actors’ influence will vary depending on the context/scenario he/she is presenting the information and that such variations should be taken into account.
This entails breaking down a general subject under study into specific sub categories
Vocabulary or lexical choices
This aspect pertains to the vocabulary and lexical aspect of the text under scrutiny. A content analyst may be interested in their occurrence or symbolic meanings.
This has been defined by Hansen and Cottle as “an attempt at classifying coverage in terms of value or judgment …”
Hansen, Anders, and Cottle Simon. Mass Communication Research Methods. [ed],[city],[Publisher],1998.Print