Research Notes

Fundamental Understanding in Research

Understanding Instruments, Variables, Validity, Reliability, and Accuracy

Written on May 27, 2025

Our daily lives are inseparable from efforts to understand things. How well we understand something or how we comprehend it is generally expressed through measurements or descriptions. From students' understanding levels of their lessons expressed in numerical measurements, how often someone gets distracted by their smartphone, how long they access social media daily, to how much their stomach hurts during class described in levels (from pain to severe pain), and so on. These are all measurements expressed either numerically or descriptively. The effectiveness of feedback given to improve something depends on how well we understand the measurements of that thing.
We often take measurements to make decisions. A doctor measures a patient's temperature, a teacher grades student exams, or a company measures customer satisfaction. To ensure these measurements are meaningful, we need to understand five key concepts.

1. Measurement Instruments

Definition: Tools or methods used to collect data, whether physical devices (like scales) or non-physical tools (like questionnaires).

Examples:

  • Thermometer for measuring body temperature
  • Customer satisfaction questionnaire
  • Digital scales at community health centers

Importance of Appropriate Instruments: Using a measuring cup to weigh rice (instead of a scale) will produce inaccurate data. Instrument selection must match what you want to measure.

2. Research Variables

Definition: Characteristics of an object that can be measured or observed, and can be expressed numerically (quantitative) or categorically (qualitative).

Quantitative Examples:

  • Height (170 cm, 165 cm)
  • Exam scores (scale 0-100)

Qualitative Examples:

  • Gender (male/female)
  • Satisfaction level (satisfied/neutral/dissatisfied)

3. Validity

Definition: The degree to which an instrument truly measures what it is supposed to measure.

Valid Example: Using a thermometer to measure body temperature

Invalid Example: Measuring intelligence with height, Determining someone's music taste through their music play intensity on the JOOX app. This may seem relevant or valid, but there are many conditions that might make someone play certain music on a specific app, perhaps to entertain family at home, etc.

4. Reliability

Definition: Consistency of measurement results when repeated under the same conditions.

Reliable Example: A scale showing 60 kg, 60.1 kg, 59.9 kg in three weighings

Unreliable Example: A ruler that gives different results with each measurement

5. Accuracy

Definition: The closeness of measurement results to the true value (reference/standard value).

Difference from Reliability:

  • Reliability: Consistency of results
  • Accuracy: Correctness of results against true value

An instrument can be reliable (consistent) but not accurate (always off by the same pattern).

Relationship Between Concepts

An ideal research instrument should meet three criteria:

  1. Valid: Measures what it's supposed to measure
  2. Reliable: Provides consistent results
  3. Accurate: Close to the true value

Application Example: Before launching a survey questionnaire, a company must:

  1. Ensure questions are valid (truly measure satisfaction)
  2. Test reliability (consistent answers if repeated)
  3. Verify accuracy (by comparing sample results with direct observation)

By understanding these concepts, we can produce truly meaningful data for both research and everyday decision-making.