4 postsRe: Topic 2 DQ 1
Qualitative research produces a variety of data, from a variety of sources. Data sources may be personal interviews (written or recorded), surveys, questionnaires, official documents or observation notes. To complicate matters, more often than not, there are numerous respondents or participants and multiple researchers. To extricate and code data from multiple data sources can be difficult, but made much easier if the data is organized appropriately. (Katherine B.2017)
The vast majority of qualitative data is “Unstructured Data,” which includes documents, photographs, audio, and video.
The simplest things we can do to improve the usability of unstructured data for analysis are:
- Convert it to a structured schema that can be evaluated with quantitative methods.
- Make it simple to find.
On the first point, we can feed documents to full-text search engines such as Lucene, which make data retrieval simple. We can also design full text search engines to execute faceted searches, allowing us to attach Metadata facets (e.g., Author, Media Type, Creation Date, etc.) to enhance our quantitative research. The same search engine was used. (Bensal P and others…. 2010)
On the second point, there are a variety of methods for converting qualitative Unstructured Data into Structured Data (which may be quantitatively examined). But it all relies on what you want to do with the Structured Data and how you get it. You can, for example, create n-grams (continuous sequences of words) and then analyze those n-grams to identify what the most common terms are within a subset of texts.
You might wish to have someone manually transcribe all consumer references of a product when evaluating footage. There are already Machine Learning algorithms that can transcribe and recognize speech.
Machine Learning and Deep Learning programs that can extract usable and reliable quantitative data from qualitative data will be extremely important in the future of analytics. However, manual methods such as employing Amazon Mechanical Turk, or a combination of both, are equally viable options for extracting Quantitative Structured Data from Qualitative Unstructured Data.
Using 200-300 APA FORMAT with references to support this discussion,
Qualitative data has been described as voluminous and sometimes overwhelming to the researcher. Discuss two strategies that would help a researcher manage and organize the data.