1.2
Come vengono classificati i dati: i dati categorici
Una variabile, solitamente indicata con lettere maiuscole come X e Y, rappresenta una caratteristica o un valore che può essere determinato per ciascu…
Data, a scientific term used for the collection of observations and measurements, forms the basis for all the statistical analyses and inferences.
Data can be classified based on whether it can be measured or not. For example, consider different hair colors. One cannot measure hair color in liters or kilometers but instead can group them into categories such as black, brunette, or red.
Such data sets are called categorical data or qualitative data; they cannot be measured or counted but can be labeled or put into different categories.
Another example is human blood, which is grouped into four different types: A, B, O, or AB.
In certain cases, categorical data can be ordered in a particular fashion; such data are called ordinal categories. For example, the size of coffee cups—small, medium, large—or the height of trees in a forest—short, medium, tall—can be arranged in the order of increasing size.
View the full transcript and gain access to JoVE Core videos
Q1: What is categorical data and how does it differ from other data types?
Categorical data, also called qualitative data, cannot be measured in units like liters or kilometers. Instead, it is grouped into categories or labels. For example, hair colors such as black, brunette, or red are categorical. Unlike how data are classified numerical data, categorical data represents characteristics that can only be labeled or organized into distinct groups rather than counted or measured numerically.
Q2: What are examples of categorical data in scientific research?
Common examples of categorical data include blood type (A, B, O, or AB), party affiliation (Republican, Democrat, Independent), hair color, age group, and sex. These variables represent characteristics that cannot be measured numerically but instead are divided into distinct categories. Each observation falls into one category, allowing researchers to organize and analyze populations based on these qualitative attributes.
Q3: What are ordinal categories and how do they differ from regular categorical data?
Ordinal categories are categorical data that can be arranged in a meaningful order or sequence. Examples include coffee cup sizes (small, medium, large) or tree heights (short, medium, tall). Unlike nominal categorical data, ordinal categories have a natural ranking, though the differences between categories cannot be measured numerically. This ordering reflects a progression from one level to another.
Q4: How are variables and data related in statistical analysis?
A variable is a characteristic or measurement determined for each member of a population, typically notated by capital letters such as X or Y. Data are the actual values of those variables, which may be numbers or words. A single value is called a datum. Variables provide the framework for collecting and organizing data during statistical analysis and research.
Q5: Can you measure the difference between ordinal categorical responses?
No, the differences between ordinal categorical responses cannot be measured numerically. For example, in a cruise survey with responses ranked as excellent, good, satisfactory, and unsatisfactory, the responses are ordered from most to least desired. However, you cannot quantify the exact difference between excellent and good or between any two consecutive responses.
Q6: Why is it important to classify data as categorical in research?
Classifying data as categorical is essential because it determines how data are analyzed and interpreted. Categorical data requires different statistical methods than numerical data. Recognizing whether observations represent categories or measurable quantities guides researchers in selecting appropriate analysis techniques and drawing valid conclusions from their data.
Q7: What is the difference between nominal and ordinal categorical data?
Nominal categorical data has no inherent order, such as blood types or hair colors. Ordinal categorical data can be ranked in a meaningful sequence, like survey ratings or size classifications. While both are categorical, ordinal data conveys additional information through its ordering, whereas nominal data simply assigns observations to unordered categories.