A factor that affects all measured variables and explains the correlations between those variables.

A factor that affects all measured variables and explains the correlations between those variables.

• A. Unique variance
• B. Common factor
• C. Component matrix
• D. Compound symmetry

Answer: (B)

In factor analysis, a common factor is a latent variable that influences all the observed measures and accounts for the correlations among them. Each observed variable also has unique variance—part that is specific to that variable plus measurement error—that is not shared with others. The idea is that the shared variance across variables is driven by these underlying factors, which is why the variables tend to covary together. A common factor explains why multiple measures move together, while the unique variance explains why they don’t. The other terms refer to parts of a variable’s own variance or to methods/assumptions rather than a single latent source that ties all measures together. So the description fits a common factor. For example, a general intelligence factor can influence many cognitive tests, producing the correlations observed among them.

In factor analysis, a common factor is a latent variable that influences all the observed measures and accounts for the correlations among them. Each observed variable also has unique variance—part that is specific to that variable plus measurement error—that is not shared with others. The idea is that the shared variance across variables is driven by these underlying factors, which is why the variables tend to covary together. A common factor explains why multiple measures move together, while the unique variance explains why they don’t. The other terms refer to parts of a variable’s own variance or to methods/assumptions rather than a single latent source that ties all measures together. So the description fits a common factor. For example, a general intelligence factor can influence many cognitive tests, producing the correlations observed among them.