Hello folks,
In every company you go as a data analyst, you would be given variety of data to process. Task of yours would be to find the right business insights out of it. Will divide this into two part as it would take much time and effort of your to keep on concentrating for whole. Make sure you go through part 2 too, so that you gain the whole knowledge. As they say, half knowledge is more dangerous than no knowledge.
Let’s dive into the ocean,
Regardless of what type of data analysis you’re conducting, the process is generally the same and doesn’t vary much. It would be easy for you to understand, if I explain you through my personnel experience as an intern and some real industry examples. The example that I’ll walk through is that of our employee engagement survey, but you could imagine that this process applies to just about any data analysis that you’re going to conduct as an analyst. The first thing you want to do is ask. You want to ask all of the right questions at the beginning of the engagement so that you better understand what your leaders and stakeholders need from this analysis. The types of questions that I generally ask are around, what is the problem that we’re trying to solve? What is the purpose of this analysis? What are we hoping to learn from it? After you’ve asked all the right questions and you’ve wrapped your arms around the scope of the analysis you need to conduct, the next step is to prepare. We need to be thinking about what type of data we need to answer those key questions. This could be anything from quantitative data or qualitative data. It could be cross-sectional or points in time versus longitudinal over a long period of time. We need to be thinking about the type of data we need in order to answer the questions that we’ve set out to answer based on what we learned when we asked the right questions. We also need to be thinking about how we’re going to collect that data or if we need to collect that data. It may be the case that we need to collect this data brand-new. So we need to think about what type of data we’re going to be collecting and how. For our employee engagement survey, we do that via survey of both quantitative and qualitative questions. But it may actually be the case that for many analyses, the data that you’re looking for already exist. Then it’s a question of working with those data owners to make sure that you are able to leverage that data and use it responsibly. After you’ve done all the hard work to collect your data, now you need to process that data. It begins with cleaning. This to me is the most fun part of the data analytics process. We can think of it as the initial introduction or the handshake, hello, to your data. This is where you get a chance to understand its structure, its quirks, its nuances, and you really get a chance to understand deeply what type of data you’re going to be working with and understanding what potential that data has to answer all of your questions. This is such an important part, too, where we’re running through all of our quality assurance checks. For example, do we have all of the data that we anticipated we would have? Are we missing data at random or is it missing in a systematic way such that maybe something went wrong with our data collection effort? If needed, did we code all of our data the right way? Are there any outliers that we need to treat differently? This is the part where we spend a lot of time really digging deeply into the structure and nuance of the data to make sure that you’re able to analyze it appropriately and responsibly. After cleaning our data and running all of our quality assurance checks, now is the point where we analyze our data, making sure to do so in as objective and unbiased a way as possible. To do this, the first thing we do is run through a series of analyses that we’ve already planned ahead of time based on the questions that we know we want to answer from the very, very beginning of the process. One thing that’s probably the hardest about this particular process, the hardest thing about analyzing data, is that we as analysts are trained to look for patterns. Over time as we become better and better at our jobs, what we’ll often find is that we can start to intuit what we might see in the data. We might have a sneaking suspicion as to what the data are going to tell us. This is the point where we have to take a step back and let the data speak for itself.
Further will continue in the next part. stay tune!
Watch out for “Example of Data Process Part 2”
Link: Example of Data Process Part 2
Thankyou.