Qualitative vs. Quantitative: Which Method Should You Use?

Any analytics project, whether analyzing for compliance in cannabis regulations or measuring market dynamics over time, begins through understanding the data available. Data can be placed broadly into two categories: quantitative and qualitative. These categories determine whether statistical and/or mathematical techniques are appropriate for the research.

Quantitative data encompasses any information represented in a numeric form, such as integers for a count of plants, or continuous intervals or ratios, like a measure of weight or a percentage. Reliable quantitative data has more precision, leading to more defensible results (e.g., there are two plants per square foot). Quantitative data can be used in a variety of ways, including mathematical and statistical analysis, and is crucial in the admissible areas of public policy and law.

Qualitative data, on the other hand, provides context about the quality of the object in question. It may come in the form of a description of an object or relate to a particular characteristic. For example, a study may want to consider a cannabis dispensary’s proximity to transportation hubs or examine the effects of growing cannabis indoors versus outdoors.

Statistical analysis with qualitative data requires methods like transforming the data and using approximate indicators related to the quality of the information being analyzed. For instance, to see the influence of a business being close to a major roadway, we can take a number of approaches. One would be to identify the distance from the retail location to a highway, while another would be to make a dummy Boolean variable indicating whether the location is (or is not) within a specified distance from a highway. Dummy variables are a simple and widely accepted method for including qualitative data directly into statistical analysis. When trying to construct a statistical model, analysts can often find creative ways to include qualitative factors in their work.

While a simple understanding of the different types of data helps, subject matter expertise amplifies decision making by already knowing how you would like to treat your data. For example, when analyzing conversion efficiency from dry flower to concentrate, I know to keep in mind what the possible ranges are. Equally as important, using the earlier example, I need to make a reasonable induction that access to transportation arteries would affect retail activities.

Combining knowledge of the available data and expertise on the subject matter enables development of suitable measures and accurate models for any task.

Stay tuned for more blog posts from members of the NCS research team in a new series called Nerding Out.