Qualitative data analysis is a process of examining non-numeric data in order to reveal themes and patterns of data to meet research objectives. The common examples of qualitative data include interviews, audios, videos and notes from observations of opinions, values, behaviour and feelings of people. Here are the steps to effectively analyse qualitative data.
Transcribe data
The first step of data analysis is transcribing your data because the data you have collected is unconstructed and hence does not make any sense to readers. Transcription makes your data presentable in textual form. You can use various software that have been specifically designed for transcription such as Nvivo, EvaSys and ATLAS.ti.
Organise your data
Once you have transcribed data, the next step is to organise data. Reread your research questions and organise data according to these questions. Use one-dimensional tables to explain only one feature and two-dimensional tables when you are referring to two variables.
Coding
Coding means categorisation of data by assigning a label to sentences, phrases and paragraphs. Specific acts, events, activities, strategies, and any other non-quantifiable element can be coded. You can do either manual coding or use software like Nvivo and ATLAS.ti. Here are the ways to identify themes and codes:
- Look for commonly used words and indigenous terms that are used by respondents. These word repetitions can indicate emotions.
- Look for the frequent use of keywords in sentences
- Codes can be used to explain the actions, conditions, interaction and consequences of phenomena.
- Metaphors used by people to indicate their beliefs and ideology about a particular event.
In short, you can divide your coding into three parts: descriptive coding that summarises the qualitative data in words; in-vivo coding uses the language of respondents; pattern coding is used to find patterns in the collected data.
Following are questions that you should consider while coding the data:
- What are your respondents doing?
- How do your respondents understand about a particular event?
- How do they communicate and what do they exactly do to handle a particular situation or event?
- What did you learn while taking notes?
- What did surprise/intrigue/disturb you?
Data validation
Data validation process ensures the data is valid and correct. The process involves code validation, data range validation, data type validation and structured validation. Data validation is crucial to analyse the validity of outcomes as well as ensuring the consistency between the produced results.
Concluding data analysis
This step requires you to state your findings and establish a link between outcomes and research objectives. You will also highlight pros and cons, study limitations and the areas for further research.
Summing up
Qualitative data analysis can be a backbreaking and frustrating job, but you can carry out it successfully if you use the right tools.