Precision

Research Requirements and Precision

Measurements can be made with various degrees of precision, and how precisely a measurement should be made depends on the research requirement. Precise measurements are always better than imprecise ones, but more precise measurements are not always superior to less precise ones. The description of a residential location as “123 5th Street” is more precise than “inner-city neighborhood.” The description of a historic house built “in March 1876” is much more precise than “in the late 19th century.”

More precise measurements such as “123 5th Street” are not always necessary or desirable. If the description of residential locations such as inner city, inner suburbs, or suburbs satisfies the research requirement, the more precise measurements (i.e., complete residential addresses) necessitate that the researchers identify whether these addresses are located in the inner city, inner suburbs, or suburbs.

An understanding of the degree of precision is also required to direct an efficient data collection. For example, the work experience of people is measured in years or months, but not in hours. The preparation time of students for final exams is measured in hours, but not in minutes or seconds. Any attempt that measures the work experience of people in hours or the student preparation time for final exams in minutes is wasted.

Less precise measurements are not always inferior to more precise ones. In the case of people's work experience, the measurement in years or months is better than hours. Similarly, the measurement of student preparation time for final exams in hours is better than minutes.

Precision and Accuracy

Precision is an important criterion of measurement quality and is often associated with accuracy. In experimental sciences. including social and behavioral sciences, there is low precision, low accuracy; low precision, high accuracy; high precision, low accuracy; and high precision, high accuracy. The measurement with high precision and high accuracy is certainly a perfect measurement that can hardly be made. The best measurement that can be made is to come as close as possible within the limitations of the measuring instruments.

It is easier to be accurate when the measurement does not aim at producing a precise result. In the meantime, if the measurement aims at obtaining a precise result, then it is likely to produce an inaccurate result. The most commonly used illustration to exemplify the difference between precision and accuracy is dart throwing. Dart throwing is neither precise nor accurate when the darts are not clustered together and are not near the center of the target. It is precise but inaccurate when the darts are clustered together but are not near the center of the target. It is accurate but not precise when the darts are not clustered together but their average positions are the center of the target. It is both precise and accurate when the darts are clustered together in the center of the target.

...