LESSON 38 - METHODS OF DATA COLLECTION: TYPES OF DATA || SOURCES OF DATA

Data collection is the make-or-break step in research: without it, researchers can’t answer research questions, test hypotheses, or produce conclusions, recommendations, and suggestions for further study. Choosing the right data collection approach starts with three decisions—what kind of data is needed (numerical or narrative), which collection methods fit that data (questionnaires, observation, interviews, or document analysis), and how the resulting data will be analyzed (since the data type directly shapes the analysis plan).

“Data” is defined as quantitative or qualitative information about a variable gathered from research subjects. Variables are measurable characteristics that vary across people, objects, or events, and the information collected about those characteristics can appear in many forms: images, numbers, words, figures, facts, and ideas. The lesson also distinguishes between a tool and a method. A tool (or instrument) is the device used to collect data—such as an interview guide—while a method is the technique for collecting data—such as conducting interviews. Instrumentation goes beyond designing the instrument; it includes how the instrument will be administered under real conditions.

The lesson then breaks data into two major types. Quantitative data are numerical and mathematically computable, split into discrete quantitative data (whole numbers only, like counts of people or animals) and continuous quantitative data (values along a continuum, like height, weight, or distance that can include decimals). Qualitative data are non-numerical and categorical, typically textual or descriptive. Examples include interview transcripts, reports or minutes, emails, field notes, and video or audio recordings, as well as photographs. Qualitative data are discussed alongside nominal and ordinal measurement scales: nominal for naming categories and ordinal for ordering categories where rank differences are not necessarily equal.

Measurement is treated as a broader concept than data collection or tools. It includes assigning scores, numbers, meanings, and descriptions to represent the characteristic of interest. At a minimum, measurement requires defining terms operationally and identifying the scale of measurement. Four scales of measurement are highlighted: nominal (names), ordinal (ordered ranks), interval (no absolute zero), and ratio (has an absolute zero).

Finally, the lesson outlines three sources of data—primary, secondary, and tertiary. Primary sources are original materials created at the time of an event or by someone who directly experienced it (e.g., handwritten manuscripts, government documents, public records, photographs). Secondary sources are later accounts based on primary sources, used to interpret or evaluate them (e.g., textbooks, biographies, analytical articles). Tertiary sources compile or digest other sources (e.g., dictionaries, encyclopedias, directories, guidebooks, manuals, Wikipedia, and similar reference digests). The primary-versus-secondary boundary isn’t always clear; classification depends on how and why a source is used, not only on what it is.

When using secondary data, the lesson emphasizes that it must be available, relevant, accurate, and sufficient for the research question. Across both primary and secondary work, data quality depends on validity and reliability, piloting instruments before full administration, and collecting data only with authorization—typically involving ethics approval and permission from those controlling access to participants. The lesson closes by reiterating four core data collection methods—questionnaires, observation, interviews, and document analysis—setting up the next session on questionnaires.