Proper data analysis is the creation of information from raw data. Data analysis requires the skill to collect, measure, transform, and create meaningful information. Data in and of itself will not provide any meaning unless it can be delivered in a proper way. This article will establish some questions that any data analyst should ponder.
Is the data meaningful? Data analysis starts with collecting the right data to analyze. The data should pertain to the goals and objectives of the analysis. If the data does not provide meaning to the analyst than it can not be converted to information to an audience. Make sure that the data in use will provide the needed results.
Is the data measurable? It can be said that the first step to success is defining an objective. Data analysis requires objective measurable facts. Without concrete measurable data the analyst will not be able to see whether success is achievable. Make sure the data can be defined and quantified. Even subjective observations can be measurable to a certain degree. This step might require some creativity but it is important to data analysis.
Is the data transformable? The data analyst needs to be fluent in the important tools of the information age. The proper tools will allow the analyst to sift through data quickly and achieve the desirable results. Proper data analysis tools include database administration, data mining, operations research, artificial intelligence, machine learning, neural networks, and much more. The data analyst need not be an expert in each area but have a good understanding. Proper data transformation can lead to meaningful information for the analyst's audience.
Is the data beneficial? This is probably the most important question to ask in data analysis. As one of my former managers used to say "Does it pass the smell test?" In other words, is the data analysis presenting itself in a meaningful way to its intended audience. Remember that data is only data until it becomes information. Examine the data analysis repeatedly to make sure its meeting desired objectives.
Use peer review to help Always double and triple check results Always be learning new methodologies
Don't think you have all the answers