When it comes to collecting data, you'll encounter two types of data, quantitative and qualitative. Depending on the type, you may have to collect and organize the data differently. Knowing the difference can help you form the right questions to ask to get the answers you want. In these lessons, we'll learn how the collection of one data point may provide you with a completely different type of insight compared to another data point. Qualitative data describes or characterizes a subject and allows you to provide more context about it. This type of data is often non-numerical and enables you to categorize a subject based on identifiers, such as specific properties, attributes, and labels. Qualitative data enables the user to understand and explore a particular subject in depth and determine trends of traits or qualities. It answers questions, such as who an organization's customers are, what issues they have, what they care about, and how to better serve them. To describe an organization's customers, the organization may collect data on customer characteristics, such as gender, native language, home address, or occupation. This type of information provides more context about the customer, as well as the ability to group customers together based on similarities across several traits. When describing an organization's products, they may collect data around appearance, color, shape, manufacturer, and source of origin, which enables them to better understand their products, find patterns, and draw insights. For example, if a store selling table starts noticing customers purchasing wooden tables tend to be men and marble tables tend to attract women, they might consider tailoring advertisements according to this observation. We can also categorize qualitative data into binary data, nominal data, and ordinal data. Binary data is data that indicates a quality in terms of yes or no, true or false, a positive or negative result, or right or wrong. Nominal data can be used to label something. It can also be called named or labeled data. And ordinal data describes qualitative data that categorizes something in a particular order or on a ranging scale. For example, ordering customers by grouping their age or credit scores. There are certain data collection methods that are more effective at collecting qualitative data, such as interviews, focus groups, and surveys. Directly interacting with an individual or a group of individuals to gather information is one of the most common types of qualitative data collection methods. Questions are often open-ended, and the data collector can capture different perspectives from interviewees about a certain subject. Observations. Qualitative data can be collected by observing subjects and taking down data points. Data collectors can observe subjects in real-time, through video recording, audio recordings, or photo images. Case studies and longitudinal studies. For more complex data collection about a subject, case studies and longitudinal studies may be used. Case studies are deep, intensive analysis, which would include capturing multifaceted influences on the subject. Longitudinal studies observe a single subject or a group of subjects over a period to collect qualitative data points. Qualitative data is critical in providing context about an organization's business and how to better serve their customers. Multiple points of qualitative data are required if they want to investigate certain subjects through multiple lenses for patterns, trends, or insights. Deriving value from qualitative data may only require a small sample of data, which makes it great for exploration into various elements that may influence a particular business question. For example, before an organization can attempt to improve its customer engagement experience, it must first figure out who its customers are. Capturing a few data points to characterize their customers can quickly provide a tangible perspective of who they're serving. Moreover, because qualitative data is often collected with more open-ended responses, the collection process may remove self-imposed research bias that may influence data collection. One of the biggest drawbacks of relying purely on qualitative data is that qualitative responses could be subjective and open to interpretation, which can lead to inconsistent insights and conclusions. Consider an organization trying to describe their customers based on their occupation, and the customer thinks of themselves as someone who works with numbers every day. When information about their occupation is collected, would a customer log their occupation as an analyst, data scientist, business analyst, data analyst, or data entry specialist? Customers who may essentially have the same job functions and responsibilities may provide very different data. How would a data collector interpret the differences? The challenge of inconsistent qualitative data also comes up when categorizing products. Let's look at laptop computers being sold by a large warehouse. Should they be categorized as a laptop, computer, media device, gaming device, electronic device, mobile device, or a workstation? The qualitative data describing a laptop computer might vary depending on the specific objective and context the data collector may have. In addition, qualitative data alone may not be sufficient to drive meaningful action. Let's say a political campaign successfully collects data on the native languages used in Los Angeles, which include Chinese, Spanish, and English. But this isn't enough information to tell you which language you should focus most of your messaging in. You may assume it's English based on personal experience, but what more information do you need to be sure? Collecting data from different subjective perspectives opens a lot of data variety. In addition, most collection methods of qualitative data tend to be time-consuming and resource-intensive, making it difficult for most people to collect it at scale.
