Research data can be categorised as quantitative or qualitative, depending on the methodology and methods used. Quantitative data refers to any information that can be quantified: that is, it can be counted or measured, and given a numerical value. Quantitative data includes numeric values, variables, and attributes. Unlike quantitative data, qualitative data is descriptive: that is, it is expressed in terms of language. Qualitative data includes textual data related to interview transcripts, audio and video recordings, and image materials. So, there are significant differences in analysis and interpretation between quantitative and qualitative research data.
What is data analysis?
Data analysis refers to the process of manipulating raw data to uncover useful insights and draw conclusions. During this process, a researcher will organise, transform, and model the data collected during the research process
One can follow six simple steps to optimise the quantitative data analysis process:
1. Clean Up Your Data
Data cleaning, also referred to as data wrangling, is the process of identifying and correcting or eliminating inaccurate or repeat records from your data. During the data cleaning process, you will transform the raw data into a useful format, hence preparing it for analysis.
Prior to initiating the data analysis process, data must be cleaned so as to ensure that results are based on a reliable source of information.
Useful information on how to process quantitative data files can be found at:
https://www.fsd.tuni.fi/en/services/data-management-guidelines/processing-quantitative-data-files/
2. Identify the Right Questions
Once the cleaning process is completed, a number of research questions will arise which will ultimately uncover the potential of the research data once analysed. Subsequently, it is pertinent to identify the most salient questions one wish to address and answer through one’s analysis.
3. Break Down the Data Into Segments
Breaking down one’s dataset into small defined groups facilitates the data analysis process. Segmenting one’s data will make one’s analysis more manageable, and also keeps it on track.
4. Data Visualisation
Data visualisation is an important facet of data analysis whereby graphical representations of the data collected are created. This makes it easier to identify patterns, trends and outliers.
Creating visuals will also enable findings to be better communicated and conclusions are drawn more effectively. Examples of data visualisation tools include Microsoft Excel and Google Charts.
5. Use the Data to Answer Your Questions
Subsequent to the cleaning, organising, transforming, and visualising your data, the questions outlined at the beginning of the data analysis process need to be revisited. Hence, results are to be interpreted and whether the data has answered one’s original questions needs to be determined.
If the results are inconclusive, try revisiting a previous step in the analysis process. This might be due to the fact that one’s data was too large and should have been segmented further, or perhaps there is a different type of visualisation better suited to one’s data.
6. Supplement with Qualitative Data
While concluding one’s analysis, bear in mind that there might be other solutions to address one’s research questions or support one’s findings. Using a quantitative approach will help one understand what is happening. Subsequently, exploring the possibility of using a mixed method approach by gathering qualitative information will enable a better understanding of why it is happening.
Compared to quantitative data, which captures structured information, qualitative data is unstructured and has more depth. Qualitative data is very often generated through interview and focus groups transcripts, surveys with open-ended questions, observational notes, and audio and video recordings.
There are five steps to follow in order to optimise the qualitative data analysis process:
1. Prepare and organise your data
All data collected, such as transcripts, notes and documents, should be organised in a single place. Also, any sources, demographics and all other information that may help one’s data analysis should be identified. This facilitates accessibility and supports consistent analysis.
Analysing qualitative data is more challenging than analysing quantitative data. Qualitative data can be organised by plotting all data gathered into a spreadsheet; by using specific software such as ATLAS.ti, NVivo and MAXQDA; or by uploading the data in a feedback repository such as Dovetail and EnjoyHQ.
Useful information on processing qualitative data files can be found at: https://www.fsd.tuni.fi/en/services/data-management-guidelines/processing-qualitative-data-files/
2. Review and explore the data
Read data to get a sense of what it contains. One may want to keep notes about one’s thoughts, ideas, or any questions that might arise.
3. Code your qualitative data
Coding is the process of labelling and categorising one’s data into specific themes, and the relationships between these themes. Coding means identifying keywords or phrases and assigning them to a category of meaning. An Excel spreadsheet is still a popular method for coding. However, various other software solutions can help speed up coding. Some examples include NVivo, Dovetail, EnjoyHQ and Ascribe. Advances in machine learning have now made it possible to read, code and structure qualitative data automatically. This type of automated software is offered by Thematic Analysis Software.
4. Analyse your data
This is where one starts to answer one’s research questions. Analysis is the process of uncovering insights through the codes that emerge from the data and identifying meaning correlations. It is also pertinent to note that each insight is distinct, with ample data to support it.
In circumstances where codes are too broad to extract meaningful insights, primary codes should be divided into sub-code. This process, which will improve the depth of one’s analysis, is referred to as meta-coding.
5. Draw conclusions
To conclude one’s analysis process, the research findings need to be reported and communicated. This can be done by preparing a series of charts, tables and visuals which can be generated using data visualisation software such as Power BI, Tableau, and Looker Studio.