Content analysis

Content analysis is the process of reviewing any type of media to evaluate what it says and how it says it. While it is often thought of in the context of literary content analysis—examining written works to understand the author's meaning—the term has broader implications. It can be applied to the analysis of such diverse sources as books, newspapers, television programs, movies, survey results, social media posts, and advertising.

Background

The practice of content analysis has existed for centuries. Ancient people began the practice of hermeneutics, the study of wisdom books and sacred writings such as the Bible and Quran to discern their meaning. Works of literature, including plays, have also been studied for meaning and symbolism. However, the practice of content analysis as a systematic activity by researchers is a relatively new discipline.

This change was largely due to the improvement in ways to conduct content analysis. Prior to the middle of the twentieth century, content analysis required laborious study and effort. The document or source would have to be reviewed, and notes taken. These notes would then need to be reviewed and compared to find areas of similarity, repeated words and concepts, etc. New notes would then be made of these similarities. The work would be re-examined to see if there were other instances that fit into the categories developed by the earlier reviews and any new material would be noted and compiled into the categories. After this was repeated several more times, the categories would be compared, analysis would be done, conclusions would be drawn, and the information would be prepared in a way that could be shared, such as an essay, study, or report.

Even after computers were invented, the process was still somewhat tedious. Information still needed to be input manually by people who would read the source material and select the necessary data. Early computers could tabulate the information, such as the number of times a certain word or combination of words appeared, but this was returned on punch cards, which required another person or persons to interpret the data.

The effectiveness of content analysis took a huge leap forward in the late 1950s and early 1960s. By this time, computers were able to do more than serve as expensive calculators adding up occurrences of certain bits of information. At the same time, researchers were beginning to look for more types of data, such as repeated concepts and how words were used in relation to other words. As computer technology increased in ability and affordability, it became easier to analyze content, and the applications for content analysis grew. In the twenty-first century, nearly every kind of media is analyzed, from books, music, and movies to newspaper headlines and political speeches to posts on social media such as Facebook, Twitter, and Instagram.

Overview

While there can be some variations depending on the content being analyzed, there are two basic types of content analysis: conceptual and relational. Conceptual analysis refers to identifying and quantifying (determining the number and frequency) of various concepts or ideas expressed in the media. Relational analysis compares these concepts and looks for ways they are the same or different, and how they are interconnected.

A researcher examining a text for concepts might start with a specific idea of what to look for, such as the number of times an author uses concepts of good versus evil. The researcher might also begin the process with no set idea of a concept and search instead for patterns; for example, a researcher looking for trends in social media might start a list of each unique topic that is found and count any repeated topics. Sometimes this analysis is self-contained, such as when a specific book or body of work by an author is the subject of the analysis. Other times, it can be broader, such as identifying how many times a political candidate addresses a particular issue in speeches over the course of several months. In this form of analysis, the researcher is only interested in quantifying the information and determining how often each word, concept, etc., appears in the media under examination.

Once the conceptual data has been compiled, it is possible to begin relational analysis. The study of the media now moves to looking for relationships between the concepts that have been identified. For example, the researcher may seek to determine if a book contains more times when evil triumphs over good or vice versa, or if social media posts include more mentions of product X or its competitor product Y. Discovering the way the identified concepts interrelate can help researchers draw conclusions based on the type of media studied, from literary themes to the success of political and advertising campaigns.

Content analysis can be a very useful tool for understanding published and broadcast media, surveys, and other information sources. It provides insight into facets of culture that can be relevant both in the present and in historical context. This form of research provides a nonintrusive way to study how people interact (for example, reviewing public social media posts versus watching people in person). It also provides concrete data, such as a specific count of the number of times a concept is referenced, rather than a subjective opinion of the content.

There are some drawbacks, however. Despite the assistance of computers, it remains a time-consuming process to analyze the facts and draw conclusions. There is also the possibility of bias being introduced when relationships are drawn between the concepts. For instance, counting how many times a politician mentions a particular issue does not automatically take into account how many of those references were positive or negative.

The effectiveness and reliability of concept analysis is improved when researchers are careful to select concepts to study that are a reliable gauge of the work in question. The rules of the analysis—what is being studied and how the search parameters are defined, for instance—should be clear and repeatable. In other words, if another researcher uses the same rules and concepts, the data should be the same. One who is reviewing data derived from content analysis—such as those who read published studies—should be aware of the possibility that any data can be skewed by the approach taken in analyzing that content.

Bibliography

Bengtsson, Mariette. "How to Plan and Perform a Qualitative Study Using Content Analysis." NursingPlus Open, vol. 2, 2016, pp. 8–14, www.nursingplusopen.com/article/S2352-9008(16)00002-9/pdf. Accessed 13 Jan. 2017.

"Content Analysis." Business Dictionary, www.businessdictionary.com/definition/content-analysis.html. Accessed 13 Jan. 2017.

"Content Analysis." Pew Research Center, www.pewresearch.org/methodology/about-content-analysis/. Accessed 13 Jan. 2017.

"Content Analysis." School of Information, The University of Texas at Austin, www.ischool.utexas.edu/~palmquis/courses/content.html. Accessed 13 Jan. 2017.

"Content Analysis." The Writing Studio, Colorado State University, writing.colostate.edu/guides/guide.cfm?guideid=61. Accessed 18 Sept. 2024.

"Content Analysis: Introduction." University of California, Davis,psc.dss.ucdavis.edu/sommerb/sommerdemo/content/intro.htm. Accessed 13 Jan. 2017.

"The Ten Steps to Content Analysis." University of Surrey, libweb.surrey.ac.uk/library/skills/Introduction%20to%20Research%20and%20Managing%20Information%20Leicester/page‗74.htm. Accessed 13 Jan. 2017.

"What Is Content Analysis?" Terry College of Business, University of Georgia, www.terry.uga.edu/management/contentanalysis/research/. Accessed 13 Jan. 2017.