PSYC 2360 Lecture Notes - Lecture 6: Observational Error, Internal Consistency, Scantron Corporation
Document Summary
Reliability and validity: random (ex: misreading a question on an exam, filling in the wrong scantron bubble, systematic (ex: construction in the room next to us during exam, random error= chance fluctuations in measurement. These cancel each other out over time. Unlikely that you would make the same random error twice. Random error does not affect the average, only the variablility around the average: systematic= measured variable is influenced by other conceptual variables. These do not cancel out over time. This error will cause observed scores to systematically increase or decrease. Systematically increase/ decrease the score creates a systematic bias. Worry about this much more than random error. Reliability: extent to which a measure is free from random error, assess reliability: Extent to which scores on the same measure, administered at. Ex: scores a 3 on the anxiety scales 4 separate times. Limitations: participants can learn your scale, ****some concepts.