Data Analytics: Developing Rigorous Analysis

This blog is part 4 of a 5-part series providing insight into Edgeworth Analytics’ approach to data analytics for businesses. In our last post we discussed “Collecting and Preparing Data.” Check back soon for “Data Analytics: Communicating Data and Results”

You have a business question you want answered and you have collected and cleaned the necessary data, now it’s time to get analyzing. But where should you start? Analytics can range from simple descriptive analysis to more advanced statistical and predictive analysis. The type of analysis you chose to deploy will depend on the question you are trying to answer. However, using descriptive statistics to understand your data and what it tells you is usually the best place to start.

Descriptive and Graphical Analysis
Descriptive and graphical analysis are a good first step in any data analytics project as it helps you better understand your data, identify outliers, and explore trends. Descriptive analysis generally includes measures of distribution, central tendency, and dispersion, while graphical analysis allows you to visualize data over time, by category, or other displays of your choosing. Insights gleaned from these initial analyses can be explained simply to audiences and can help inform more advanced statistical analysis.

Assessing Relationships
Much of the work we do as data analysts has to do with assessing relationships. What factors lead to higher customer retention? Does more training lead to fewer manufacturing plant injuries? Why are profits in the Eastern Region higher than other Regions? One common tool to assess relationships is regression analysis. A regression is a statistical tool to describe the relationship between a “dependent” variable (e.g. injuries) and one or more “explanatory” variables (e.g. training). The world is a complicated place and most outcomes are the result of more than on factor. Regression analysis allows us to isolate the effect of the factors we are trying to measure.

An Iterative Process
Now that you have come up with an answer to your question using descriptive analysis and statistical models, you’re done right? Not quite. As we discussed in a previous post on “Crafting the Fundamental Business Questions,” your initial analysis may lead to additional questions or areas for further study. Analytics is an iterative process and your initial insights may lead you to dig deeper into the data and its relationship to your business. It is also important to test alternative models and theories. Ask yourself, do my results hold up to reasonable changes to my assumptions or methodologies? Are there other explanations that need to be explored? Evaluate your results to make sure they are complete and capture the whole story.

Pitfalls to Avoid
Finally, data and analysis can sometimes be misleading. Its important to think critically about your own analysis and analysis that is presented to you. Below are a few common pitfalls to look out for when working with data and analytics.

Contact

Karuna Batcha

Services

Related Capabilities

Share on facebook
Share on linkedin
Share on twitter

Contact

Karuna Batcha

Subscribe to our Newsletter

This website uses cookies to improve functionality and performance. By continuing to use this website, you agree to the use of cookies in accordance with our Privacy Policy.