Primary Data: Information collected firsthand by the researcher specifically for the current study. This ensures the data is tailored to the research question and the collection method is fully understood, though it is often time-consuming and expensive to obtain.
Secondary Data: Data that has already been collected by another party (e.g., government databases, internet archives, or research journals). It is cost-effective and quick to access, but the researcher must verify its reliability, accuracy, and relevance to their specific needs.
Bivariate Data: Involves pairs of values collected for each subject to investigate the relationship between two variables. For example, recording both the age of a vehicle and its annual maintenance cost allows for correlation analysis.
Multivariate Data: Involves sets of three or more variables collected for each subject. This is used in complex studies, such as tracking a patient's weight, blood pressure, and cholesterol levels simultaneously to understand health outcomes.
| Feature | Discrete Data | Continuous Data |
|---|---|---|
| Nature | Countable | Measurable |
| Values | Specific points (gaps exist) | Any value in a range (no gaps) |
| Examples | Number of pets, Shoe sizes | Temperature, Mass, Time |
| Precision | Fixed by the unit | Limited only by the tool |
The 'Number' Trap: Do not assume all numbers are quantitative. For instance, a Zip Code or a Student ID is a number, but it functions as a label (Qualitative/Nominal) because calculating an 'average zip code' is meaningless.
Discrete vs. Continuous Identification: Ask yourself if a 'half-unit' is possible and meaningful. If you can always find a value between two other values (like m and m), the data is likely continuous.
Secondary Data Verification: When using secondary data in an exam scenario, always mention the need to check for bias, age of the data, and the original collection methodology.
