Influence of Data on Decision-Making

Without data, you’re just another person with an opinion. — William Edwards Deming

Image from iStock.

To what extent does data impact decision-making, for multimillion-dollar corporations and the average person? What is the difference between data-driven and data-inspired decisions? In today’s post, we find out what it means to be data-driven or data-inspired, get a quick overview of the two main types of data that analysts utilize in their workflow, and get a few takeaways for the week.

The Coca-Cola Company conducted a taste test of a prospective new product on 200,000 subjects in 1985. The participants preferred the new product flavor to Pepsi. This test resulted in the launch of a new Coca-Cola product. This scenario depicts a data-driven decision, as the company made its conclusion based on data acquired directly.

Again, passenger X has many Ride-Hailing applications (Uber, bolt, in-driver) on their phone. For a journey, they compare pricing across the various apps before deciding the exact app to use for that trip. They make a data-inspired decision by comparing pricing and finally selecting one.

NB: A data-driven decision employs specific data to make judgments, but a data-inspired decision considers several data sources to explore an alternative.

What kinds of data are being considered? Can various types of data accomplish the same outcomes? The most common forms of data used by analysts include

Quantitative data

Designed in Canva

Quantitative data is the numerical data obtained through surveys, polls, and systematic interviews. They are the specific and objective measurements of numeric facts. It is usually the ‘what’, ‘how many’, or ‘how often’ of a given dataset. Examples include a company’s monthly revenue and the ages of students taking a test.

Qualitative data

Designed in Canva

On the other hand, qualitative data is the subjective and explanatory assessment of the quality and features of an object. It usually answers the ‘why’ questions of an analytics quest. Consider the Twitter community’s reactions to the Ghana-Uruguay match and graphs and charts demonstrating the depreciation of the Ghanaian cedi versus the dollar as examples of qualitative data.

None of these two is more important than the other because they complement each other during analysis. Qualitative data gives an in-depth explanation of the mystery presented by the quantitative data.

From the table below, considering the conversation we have had so far, can you identify some quantitative columns? How about the qualitative ones? The quantitative columns include GPA, Age, and other numerical columns. The gender and ID columns are examples of qualitative data. After categorizing the various columns, we can tell the number of pupils in the higher and lower echelons of the class using the quantitative columns. The qualitative gender column, on the other hand, will aid in appreciating the distribution of males to females in both of the class sections, and perhaps the justifications for such a distribution. Through this example, we can see how both qualitative and quantitative data contribute to the data analytics process.

figure from here.

Notice also that I considered the ID column as qualitative instead of quantitative. The simplest explanation for this choice is that the ID is unique to the student and represents an identity or attribute of the data point.

Differences between Qualitative and Quantitative Data

Designed in Canva by me

Weekly Take-Aways

1. Data in itself has no value until someone gives it a narrative. Considering the table above, I lifted from an article online. I only rationalized it based on my understanding of the attributes (column names); it could be a false narrative.
2. Data analysis takes data, turns it into information (by comparing it to other data sources; contextualizing it), and knowledge consumes the information and applies it.
3. Often, we don’t have access to all the data we need.
4. Your skills and knowledge will be the most critical aspects of the analytics process. Make use of logic and creativity along the way.

Resources:

1. https://www.coca-colacompany.com/company/history/the-story-of-one-of-the-most-memorable-marketing-blunders-ever
2. https://www.g2.com/articles/qualitative-vs-quantitative-data
3. * https://www.statisticssolutions.com/archival-data-finding-an-appropriate-dataset/
4. Ask Questions to Make Data-Driven Decisions — Week 2
5. Canva