PSYC 3200 Lecture Notes - Lecture 5: Computerized Adaptive Testing, Rasch Model, Standard Deviation
Document Summary
Reliability is usually measured in terms of degree as opposed to all-or-none: very few psychological characteristics are completely consistent. Best viewed as a continuum ranging from minimal consistency of measurement to near-perfect repeatability of results - most psychological tests fall somewhere in between. 95% ci that the true score is within a certain range. First factor is desirable, while the second represents unavoidable error. X = t + e , where x is the obtained score, t is the true score, and e is error. True score is never actually known; we can only get a probability that the true score is within a certain interval - It is impossible to eliminate all measurement error, but test developers try to minimize it as much as they can cannot know with certainty. In a well-designed test, this will be minimal. Test construction: problems can arise from circumstances of administration: room temperature, dim lighting, excessive noise.