Adding items to a knowledge test can increase reliability without requiring the added items to be redundant with existing items¶

Claim · All claims
Evidence · 1 study · theoretical r3 · q1
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q1 i? Because the error variance term in alpha depends on the number of items, reliability can be improved by lengthening the test, requiring only that added items are not much less reliable than existing ones. → Steffen Zitzmann 2025

Evidence¶

Steffen Zitzmann 2025¶

Steffen Zitzmann, Gabe A. Orona. (2025). Why We Might Still be Concerned About Low Cronbach's Alphas in Domain-specific Knowledge Tests. Educational Psychology Review. https://doi.org/10.1007/s10648-025-10015-5

q1 · i? · theoretical · r3

Theoretical argument from the alpha formula: since error variance depends on the number of items K, reliability "can be easily improved by expanding the test's length." The authors argue this is a feature over mere interrelatedness, and that added items can also improve generalizability and content coverage.

"Increasing reliability via the addition of items does not require that an additional item is well predictable by existing items. It only requires that this item is not much less reliable than the other items."

Discussion¶

Related Claims¶

• Theoretically defined test reliability computable from a single administration via parallel-half vector lengths and angle exceeds split-half reliability but falls below Cronbach alpha — related
• Cronbach's alpha equals the reliability of the scale score under classical test theory, domain sampling, and generalizability theory, and should be interpreted as a reliability coefficient — related
• A high Cronbach's alpha across items measuring distinct knowledge aspects indicates the test is not working as intended — related
• Alpha of knowledge tests is substantially related to the number of items in the test — a narrower finding that bears on this claim
• Cronbach's alpha does not generally indicate the reliability of knowledge tests because knowledge is a heterogeneous construct — related