What is Conventional Content Analysis in Qualitative Research? Step-by-Step Guide
Conventional content analysis is an inductive method of qualitative research in which you develop codes, categories, and themes from textual data instead of from preexisting theories.
This method involves first immersing yourself in the textual content in order to identify codes, patterns, and trends by counting the frequency of specific words or phrases. Then, you infer new insights from the frequency counts as they relate to the phenomenon you are studying.
When should I use conventional content analysis?
Inductive research methods like conventional content analysis—a bottom-up approach where you develop codes as you analyze data—are used when the existing theories or literature on the phenomenon are limited or fractured. Or when you plan to study a new phenomenon—just as you would with other inductive methods such as latent content analysis.
In contrast, directed content analysis is an example of deductive research methodology—similar to manifest content analysis, and opposite to latent content analysis—where preexisting theories or studies help focus and guide your research questions and coding scheme.
To summarize the use cases of conventional content analysis, you can employ this methodology in the following research scenarios:
There is little or no information about the phenomenon under study.
You want to describe a new phenomenon you are investigating.
You want to understand the unique lived experiences of a social setting without imposing preconceived theories or categories (with an example to follow).
Unique differences from other methods of qualitative analysis
There are two key differences between conventional content analysis and other forms of inductive analysis—such as grounded theory and thematic analysis.
First Difference
First, code frequency is central to the analytical process of conventional content analysis. Uniquely, this method considers frequency as a core marker of relevance.
The frequency—also referred to as intensity—of words, themes, and categories are often recorded in a table format and clearly explained in the final write-up. In thematic analysis and grounded theory, frequency counts can be used as a tool but it is generally frowned upon to equate frequency with any direct importance or relevance to a study.
Second Difference
Second, grounded theory and thematic analysis—as well as phenomenological research—share a similar inductive approach to conventional content analysis. All start with immersion in your data in order to identify key concepts and begin creating your coding scheme.
However, these three methods go beyond conventional content analysis in order to offer a theory or a nuanced understanding of the lived experience of people through interviews and direct observation (Hseih and Shannon, 2005). That is to say that the results are largely driven by direct interaction with people rather than at arm's length through textual content.
Typical sources of content in conventional content analysis
Conventional content analysis is generally applied to textual data and there is no widely accepted list of specific content mediums you must use in your research.
Though not an exhaustive list, conventional content analysis usually analyzes:
News articles
Research papers
Transcribed documentaries
Transcribed news stories
Social media posts or blog posts
Poems (as in the example below)
Some content mediums are used more often than others. Transcribed interviews, for instance, can be used in conventional analysis, but they are used more often in grounded theory, thematic analysis, or phenomenological research where the goal is—as mentioned—to analyze lived experience of people through interviews and direct observation.
When you don’t have access to study participants nor the time or resources to conduct interviews or direct observation, content analysis is an advantageous method. Access to textual data can offset this lack of resources and offers another scenario where content analysis is used.
Frequency is key: an example of conventional qualitative content analysis in practice
Oftentimes, one of the best ways to understand a specific research method is to consider a study conducted by applying that approach. The following study exemplifies a typical research scenario where conventional content analysis is best-suited to the researcher’s goals.
Overview
In this conventional qualitative content analysis example—Primary school students’ poetic malaria messages from Jimma zone, Oromia, Ethiopia: a qualitative content analysis—Kebede, Hayder, Girma, Alemayehu, Abebe, Sudhakar, and Birhanu (2021) explored twenty poems produced by students aged 12 to 17 years to understand the malaria messages conveyed in them.
The research was conducted in rural areas of Africa where malaria is endemic as part of a larger intervention and education approach to quell the spread of the disease.
By analyzing students' thoughts and experiences with malaria in the form of poems, the content depicted their communities’ experiences, perceptions, beliefs, and emotions in regard to malaria. The study also analyzed their perception of ways to prevent, transmit, and treat the illness.
Though just primary school students, the authors of the poems were engaged in developing and disseminating messages to schools and broader communities. The researchers state this is an effective community approach as schools are a “traditional societal setting crucial for people to interact and communicate with and share information about different individuals.”[1]
Using frequency counts in conventional content analysis
Codes, categories, and five central themes were derived inductively through the poems.
First, the existing concepts within the poems were identified and quantified through immersion in the selected poems. Then, the contents were interpreted to infer meaning from the “dominating words” that appeared at the highest intensity throughout the poems.
The researchers selected 22 words and phrases (categories within the larger themes mentioned below) by their dominance—i.e. frequency, intensity, ocurrence—in the poems. There were 602 total message counts of those terms tallied between the 22 categories. [Table 2]
For instance, the “most dominant” message presented across the poems was the “perceived severity” of malaria under the thematic message “perceived effectiveness of measures”—accounting for 101 of the 602 (16.8%) of the specific words or phrases collected.
As another example of dominant messages, insecticide-treated nets (ITNs) were mentioned 66 times throughout the 20 poems. Specifically, 17 “caring for nets” messages and 49 “ITNs” messages. This coded message was a sub-category of two of the five larger thematic messages: “knowledge about malaria” and “perceived effectiveness of measures”.
The intensity of this term—along with the researcher's analysis of it within the context of the poems—showed that ITNs are a proven first-line defense. Further, many community members often misuse or simply do not use the government-provided nets altogether. Misuse included using the nets as animal traps or not caring for them properly (stitching rips, cleaning the nets).
Further highlighting the core importance of frequency counts in content analysis, the researcher's counts were tabulated in the final study results in order to display the frequencies of their occurrence across all of the poems—shown in Table 2.
Recap
The researchers generated codes, categories, and themes by immersing themselves in 20 poems. Then, they organized messages into codes, categories, and themes by how frequently they appear within the text. Next, they defined and analyzed all three. Finally, the final write-up tabulated frequencies of occurrence across the poems and offered a conclusive narrative of the study.
What is the step-by-step process of conventional content analysis?
Now, you better understand how conventional content analysis is applied in the field. In order to apply this method in your own research, we outline the steps followed by researchers—in the malaria study and others—to utilize this approach.
1. Collect your data
The first step in any research method is to collect your data. In conventional content analysis, data is primarily collected through textual content like news articles, research papers, poems, transcribed news stories, and sometimes transcribed interviews.
2. Immerse yourself in the data
The second step is to read your collected data together multiple times to compile and make sense of the categories within it. As with all qualitative methods of research, content analysis is an iterative process. You should read the data repeatedly to achieve immersion and obtain a sense of the whole picture (Tesch, 1990).
3. Identify codes that capture key concepts in your data
Read your transcripts several times in order to inductively identify the words or phrases that embody the main concepts—and that connect them within the content. These high-frequency words and phrases help structure your coding scheme and pick out larger thematic concepts. Finally, just as the researchers did with the malaria poems, assign codes to each message you decide to track in your frequency counts.
4. Code the remaining transcript
Next, make notes and memos about your first impressions, thoughts, and initial analysis through bracketing—for yourself and other researchers involved in your study. Delve makes this easy with the ability to record notes directly to your codes.
From here, you assign codes to the new words or content you identify that don’t fit into existing codes. Repeat this process until your content is completely coded. You want clear, reproducible results for your study so it is important to reflect on the consistency of your code and your process to this point.
The malaria study provides guidance at this step of the process. Especially if you will conduct your study with other researchers:
“As much as possible, the investigators reported the actual meanings of the contents in the poems with minimal interpretation bias. To ensure this, they maintained subjective neutrality and bracketed themselves to provide consistent interpretations transpiring during the coding process instead of their own interpretations. Peer debriefing and daily interactions among each other maintained the credibility of the findings.”
5. Group the codes
After coding your transcript, the next step is to organize your codes into categories and subcategories based on how they relate or link to one another. As you continue, categories emerge that reflect more than one key idea. These emergent categories are used to organize and group codes into meaningful clusters (Coffey & Atkinson, 1996). The clusters then become themes—like the five thematic messages from the malaria poems—that help organize your code for your final narrative write-up.
6. Define each code, category, and theme
Now, you clearly define each code, category, and theme that emerged from the previous steps in preparation to report your findings. Depending on the purpose of your study, you might also decide to identify the relationship between categories—such as ITNs and caring for nets—further based on how they relate to one another in the content. This is an option if you find their concurrence provides supporting insights for your study.
7. Prepare your narrative
Finally, report your findings, including the definition of each code and category. To reinforce the trustworthiness and validity of your study, you can also explain the perceived relationship between the categories you identified as they relate to the larger theme of your study. The narrative should include a summary of how your findings contribute to the topic of interest and suggestions for future research on the subject.
A tangible example from the malaria study helps clarify this final step and reiterates the main difference from other qualitative methods like thematic analysis. The researchers state that "to assess the intensity of the underlying messages and the extent to which they were conveyed, the identified message contents were quantified in a frequency table considering a total count of 602 codes as a denominator."
Essentially, they state that frequency counts are core to their findings and make it clear by tabulating their results in the final write-up. A thematic analysis paper—or any other method of qualitative research—would not assign relevance to this information.
The best software for conventional qualitative content analysis
With Delve’s qualitative content analysis tool, code counts are instantaneous and organized through the simple drag-and-drop feature. You can easily track how prevalent codes and categories are in your data and even create bracketing memos and notes to refer back to later—both important steps that are highlighted in the guide above.
Advanced code frequency & co-occurrence matrices—with Delve
Delve also provides advanced code frequency through code co-occurrence matrices. These matrices show how frequently codes overlap, and how codes correlate to descriptors or attributes you identify. With this information, you can pinpoint the overarching themes of your study.
Cloud-based, collaborative, and cost-effective
Delve is cloud-based, collaborative, cost-effective, and easy to learn. It includes free tutorial videos, responsive customer support, and flexible payment options. Start your free trial today.
Not convinced? As researchers ourselves, we understand the pitfalls of most CAQDAS software. See why researchers like Tamra are switching to our easy-to-use, drag-and-drop coding software.
References
Abbas J, Aman J, Nurunnabi M, Bano S. The impact of social media Lon earning behavior for sustainable education: evidence of students from selected universities in Pakistan. Sustainability. 2019;11(6):1683.
Kleinheksel, A. J., Rockich-Winston, N., Tawfik, H., & Wyatt, T. R. (2020). Demystifying Content Analysis. American Journal of Pharmaceutical Education, 84(1).
Tesch, R. (1990) Qualitative research: Analysis types and software tools. Falmer, New York.
Coffey, A., Beverley, H., & Paul, A. (1996). Qualitative Data Analysis: Technologies and Representations. Sociological Research Online, 1(1), 80–91.
Kebede, Y., Hayder, A., Girma, K. et al. Primary school students’ poetic malaria messages from Jimma zone, Oromia, Ethiopia: a qualitative content analysis. BMC Public Health 21, 1688 (2021).
Cite this blog post:
Delve, Ho, L., & Limpaecher, A. (2022c, December 21). Step-by-Step Guide: What is Qualitative Conventional Content Analysis? https://delvetool.com/blog/conventional-content-analysis