Understanding Scores in Classical Test Theory

Classical Test Theory (CTT) is a framework used in psychometrics to understand and improve the reliability and validity of psychological tests. A core concept in CTT is how we represent a person's score on a test. In this theory, the score a person actually gets on a test, often called the obtained score, is considered to be made up of two parts: their true ability or trait related to what the test measures, and some amount of random error.

The Fundamental Equation of Classical Test Theory

The relationship between these components is expressed by a simple equation:

$$X = T + E$$

Where:

• \(X\) represents the obtained score, which is the score the person actually gets on the test.
• \(T\) represents the true score, which is the hypothetical score the person would get if there were no errors in measurement. It reflects the person's actual level of the trait being measured.
• \(E\) represents the error score, which is the portion of the observed score that is due to random error in the measurement process. This error can be positive or negative.

In the context of Classical Test Theory, the term used for the obtained score \(X\) is the observed score.

Analyzing the Options

Let's look at why the other options are not the correct answer:

• True score: As explained, the true score (\(T\)) is a component of the obtained score (\(X\)), not the obtained score itself. It's the error-free part.
• Standard score: A standard score (like a Z-score or T-score) is a transformed score. It is derived from an observed score but is not the observed score itself. Standard scores express a person's performance relative to a group mean and standard deviation.
• Error score: The error score (\(E\)) is also a component of the obtained score (\(X\)). It represents the discrepancy between the observed score and the true score.

Therefore, based on the fundamental definition in Classical Test Theory, the obtained score, which is a combination of the true score and the error score, is referred to as the observed score.

Revision Table: Classical Test Theory Terms

Additional Information on Classical Test Theory

Classical Test Theory is foundational in psychometrics, especially for understanding measurement reliability. Reliability in CTT is defined in terms of the proportion of the variance in observed scores that is attributable to variance in true scores. A test is considered more reliable if the error component is smaller relative to the true score component. While simple, CTT has some limitations, notably that test parameters (like difficulty) are sample-dependent, and person parameters (like true scores) are test-dependent. More modern theories like Item Response Theory (IRT) address some of these limitations.

Understanding the distinction between observed score, true score, and error score is crucial for interpreting test results and evaluating the quality of psychological and educational assessments. The observed score is what you see; the true score is what you'd hope to see ideally; and the error score is the noise in the system.