4.3.1 Frequency distribution
Firstly, we’ll look at the frequency distribution, which will tell us about how many times each unique value in our observations has occurred. Frequency distribution is relevant especially for when our data contains a small number of unique values, which may be either quantitative or qualitative. This table is similar to what you might see in a histogram, which we’ll talk about in a later section of this toolbox. Frequency distribution is likely the most commonly used type of analysis in data on humanitarian and development contexts.
A frequency distribution table is really valuable in communicating how often different observations have occurred in our dataset. For example, imagine that we’ve surveyed 50 people to ask about whether they would prefer cheese, pepperoni, or veggie pizza. We’d calculate the frequency that each of these options has occurred to see which was the most popular.
In our case study, we have done frequency distribution on a number of different variables to assist in understanding the humanitarian context of the target population. One example can be seen below in the responses to the question: “What is the principal source of drinking water for members of your household?”.
|Principal water source||Frequency||% of total|
|Piped connection to house||14||15%|
|Bottled water, water sachets||11||12%|
|Unprotected hand-dug well||4||4%|
|Rain water collection||6||7%|