Research Writing

Content analysis method Sources Features How to do exxamples

Content analysis method is a set of techniques that make it possible to explore themes and verify predefined hypotheses. In addition, it seeks to understand what is behind the production and reception of the message.

Beginning researchers face difficulties in putting the content analysis method into practice. After completing an exhaustive data collection stage, the question arises: How to interpret so much information?

Even if the work has books and renowned scientific articles as a source, the researcher needs to decode the content and make adaptations to the studied reality. 

What is content analysis method?

Content analysis is a method used to analyze qualitative research data . Among the main authors who talk about the subject, it is worth highlighting Laurence Bardin, who studied the subject in depth and published an exclusive book talking about the methodology. 

Between 1940 and 1950, E. Berelson, together with P. Lazarsfeld, elaborated a definition for content analysis that is still accepted today:

Content analysis is an investigation technique whose purpose is the objective, systematic and qualitative description of the manifest content of the communication.

Bardin, in his work on Content Analysis, improves the definition of the methodology, stating that it is a set of communication analysis techniques, which employs systematic and objective procedures for describing the content of messages. Even after exposing the definition, Bardin considered it “insufficient” to demonstrate the specificity of the technique.   

Content analysis has semantics as its main pillar, that is, research to understand the true meaning of a text. It is an exhaustive study that interprets between the lines, figures of speech and reticence. 

Source of content analysis

Although it is a modern technique that was introduced in the 20th century, content analysis has an ancient origin. She was born with hermeneutics (art of interpreting) sacred texts.

According to Bardin, the first name that really illustrates the history of content analysis is that of H. Lasswell, who analyzed the press and propaganda in mid-1915. The objective of the investigation was to measure the sensationalist impact of the articles. 

World War II created interest in political science in the United States. As a result, content analysis began to be used to verify subversive advertisements in newspapers and magazines.


  • Seeks the true meaning of the message;
  • Compare messages considering different receivers or different situations involving the same receivers;
  • Guided by standardized measures;
  • Seeks an objective, systematic and qualitative description;

How to do content analysis?

Documents, field diary, interviews and videos are just a few materials that can be submitted to content analysis. Here’s a step-by-step guide to conducting the investigation:

Pre-exploration of the material 

After conducting interviews and observations, it is necessary to evaluate all material created and organize it. The researcher must consider all the documents that respond to his research problem and disregard what does not fulfill this role. 

The readings of the collected material do not aim at systematization, but at understanding the main ideas and meanings. It is on this basis that the organization takes shape.

According to Bardin, the principles that guide the organization phase are:

  • Perform floating readings (first contact with data collection documents);
  • Build the corpus (complete collection of information on a given topic);
  • Formulate content hypotheses (suggest explanations and arguments, which will be validated or discarded);

The choice of material should follow the principles of completeness (do not leave documents out), representativeness (possibility of generalization), homogeneity (no particularities) and adequacy (it is important to comply with the research objective).


A previous reading of all collected material is carried out. Thus, the researcher is able to discard interrupted interviews or with problems in capturing the audio. 

Exploration of the material

Selection of analysis units

Coding starts with the registration unit, that is, what will be analyzed by the researcher. In qualitative studies, the research questions act as a guide in the investigation. 

These units of analysis can be words, sentences, phrases, paragraphs or themes. Events, objects and characters are also considered units of record.


Through an interview, a study asks people about the use of instant messaging applications. Every time the interviewees say the word “Whatsapp” or “Zap”, the researcher clips the speech and records it. Thus, he creates a unit of analysis.

In coding, in addition to considering the unit of record, it is also important to check the unit of context.


The word WhatsApp alone means nothing. To understand the meaning, it is necessary to analyze the sentence and the paragraph from which the term was extracted, that is, the unit of meaning. 

Categorization and sub-categorization

At this stage of the analysis, there is a concern to classify the constitutive elements of a set of differentiation. The researcher will look at your record units and turn them into categories. 

A category is a set of data with similarities that appear in different contexts or situations. It must always be defined objectively and consistently. 

The grouping can be semantic (meaning of the codes), syntactic (according to the organization of the sentence structure), lexical (word position) or expressive (what the interviewee wanted to say). It is important to think about this criterion so that there is a standard in content analysis and, therefore, a scientific process in the investigation. 

There is no magic formula for defining categorization. However, the researcher must consider the theoretical foundation , intuition and experience. 

in their theoretical knowledge, guided by their competence, sensitivity, intuition and experience.


One of the challenges of the study is to verify the purpose of using WhatsApp. For this, the researcher decides to create three categories: work, communication with the family and source of information.

Treatment and interpretation

The researcher needs to make an effort of interpretation to understand in depth the material produced by data collection. For this, he must make use of inference

The act of inferring does not mean creating subliminal assumptions, but seeking grounding in theory and in concrete situations that involve the production and reception of the message.

To make the interpretation, the investigator needs to consider an inference variable, the analyzed material, the date and the description/interpretation. 


Inference variables Analyzed material Dates Description or hypotheses/interpretation
Dissemination of fake news about COVID-19 Oswaldo Cruz Foundation Research April 2020 Significant increase in health-related fake news reports
Age group that uses WhatsApp the most Deloitte Research 2016 The age group that most uses WhatsApp is the population aged 45 to 55, surpassing the young audience (18 to 24 years old)

Inference identifies specific characteristics of the message. The procedure takes place as follows:

  • Who? (Source, issuer)
  • Because? (Coding process)
  • What? (Message)
  • To what effect? (Decoding process)
  • For whom? (Receiver)

The main qualitative analysis and interpretation strategies are:

  • Pairing : Comparison of collected data.
  • Historical analysis: how the studied phenomenon behaved over time (changes and permanence).
  • Construction of an explanation: the explanation starts from the units of meaning.

The infographic below was based on the content analysis process created by Bardin in 1977:

How to present a data analysis?

It is important to create a chapter in the work to present the data analysis. In the introduction, list the categories and technique used in the methodology.

Each category can be a chapter sub-item. When presenting the category, make a short summary of a descriptive nature and based on theory.

Always start the paragraph with the collected data and not with the theoretical study. Then comment on the information and use a theoretical interlocution.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *


Back to top button