by Sarah Mason
There are four types of analytics: Descriptive, Diagnostic, Prescriptive, and Predictive. Selecting the right type requires knowledge of analysis and the needs of finding information to problem solve. Most analysis is gathering information and generating knowledge in response to a need usually based on a problem. With four types, each is distinct. However, Descriptive and Diagnostic are often confused.
A descriptive statistic is a summary statistic that quantitatively describes or summarizes features from a collection of information, while descriptive statistics is the process of using and analyzing those statistics.
Diagnostic analytics is branch of advanced analytics that seeks to answer the question, “Why did it happen?” by using complex algorithms. It is includes techniques such as drill-down, correlations, root-cause and failure analysis
When starting with data, producing summary statistics is a starting point if not the end goal. The initial is taking a survey and determining if there needs to be data cleansing or preprocessing which describes the data. This is also a snapshot of the current state as presented in the dataset. Taking it farther and seeking out a reason for a specific value or correlation with an event is diagnostic analysis, or diagnosing a problem.
A requirement of analytics is to be able to clearly state the question or problem to investigate through data methods. There are the four types of analytics with descriptive, or summary, being the most common and basic. Diagnostic Analysis is the focus on why after a summary that uses more effort and focus. The result is a listing of details that often become actionable items for change and current state metrics.
Sarah Mason is a Healthcare Data Analyst and Founder Sarah Mason Consulting LLC.