Data, Populations & Samples

Primary Data

Data collected directly first-hand by the researcher for a specific purpose or study.

  • Methods: Direct observations, personal interviews, experiments, surveys, and focus groups.

  • Advantages: Highly specific to the researcher's research question, current, and the methodology is known and controlled.

  • Limitations: Can be time-consuming, labour-intensive, and expensive to collect.

Secondary Data

Data collected by someone else or sourced from existing records rather than gathered first-hand.

  • Methods: Government publications (such as census data), statistical databases, academic journals, books, and public archives.

  • Advantages: Fast and cost-effective to obtain; provides access to large, historic, or population-wide datasets that would be difficult for an individual to gather.

  • Limitations: May not fit the exact research objectives, can be outdated, and the collection methodology or accuracy cannot always be verified.

Categorical Data

Data that describes qualities, characteristics, names, or labels rather than numerical quantities.

  • Characteristics: Classified into non-overlapping groups or categories based on qualitative traits.

  • Examples: Favourite sport, hair colour, eye colour, brand preferences, or nationality.

  • Subtypes:

    • Nominal Data: Categories that have no inherent rank or natural order (e.g., hair colour, favourite sport).

    • Ordinal Data: Categories with an inherent order or ranking, though the exact intervals between ranks are not measurable (e.g., customer satisfaction ratings such as low, medium, or high; academic grade bands).

Numerical Data

Data representing quantities that can be counted or measured numerically.

  • Characteristics: Expressed as numbers on which arithmetic calculations can be performed.

  • Subtypes:

    • Discrete Data: Numerical values that take distinct, separate values, typically obtained by counting.

    • Values cannot be subdivided meaningfully (e.g., a household cannot have 2.52.5 children).

    • Examples: Number of students in a class (2828 students), number of siblings, shoe sizes.

    • Continuous Data: Numerical values that can take any value along a continuous scale or interval, typically obtained by measuring.

    • Can be measured with infinite degrees of precision depending on the tool used.

    • Examples: Distance (15.4 km15.4km), age (14.5 years14.5years), running time (12.38 s12.38s), height (165.2 cm165.2cm).

Populations and Samples

Concepts used to define the scope and participants of a statistical study.

  • Population: The entire group of individuals, objects, or items that are the focus of a study (e.g., all students in a school).

  • Sample: A smaller subset or group selected from the entire population to represent it (e.g., a group of Year 8 students).

  • Census: A complete study or survey that collects data from every single member of the target population.

    • Advantages: Highly accurate and provides complete data with no sampling error.

    • Disadvantages: Can be costly, resource-heavy, and impractical for large populations.

  • Representative Sample: A sample that accurately reflects the makeup, diversity, and characteristics of the broader population.

    • Importance: Critical for avoiding bias and ensuring that sample results can be reliably generalized to the population.

  • Prediction and Inference: An evidence-based conclusion or estimate made about a population by analysing data collected from a sample.