Objectivity, Sensitivity, Validity and Reliability PDF

Title Objectivity, Sensitivity, Validity and Reliability
Author Madison Barton
Course Motor Behavior
Institution University of Utah
Pages 1
File Size 34.9 KB
File Type PDF
Total Downloads 63
Total Views 147

Summary

Motor Behavior ...


Description

Objectivity: no matter who does measurement, the measurement will be correct. More objective = more you can capture in value or number what is going on, and have definite true response; more subjective = more inconsistencies between people who measure! ! Sensitivity: depending on type of measure we use, the better able we will be to describe the amount of error we have in terms of measurement; highly sensitive = accurate in error reporting,; less sensitive = less able to describe the quality of the error and all we look at is existence of error ; did you make it or miss it! ! Validity: ! • Extent to which a tests (instrument, questionnaire, etc) measures what the researched intends to measure ! • Does your test strategy actually measure what it was supposed to measure ! • Construct validity - what represents what is necessary to be skilled; ex: typing skills - how fast you can type tests ! • Face validity - idea that it is so obvious what you're measuring is a representation of what you wanted to measure. Measuring exactly what ability your trying to test ! • How it is grouped around what target was! • Take into account all trials, and average them, where does value fall - direction can cancel out ! ! Reliability: ! • The extent to which a measurement is repeatable under the same conditions ! • Consistency ! ◦ intra-individual variations ! ‣ Momentary changes in internal state ! • Bored, tired, trouble paying attention! ‣ Systematic changes ! • Alter game strategy, extra practice ! ! Describing Outcomes: Error! • A reality ! ◦ When you miss (don't achieve the target), what degree did you miss by?! • Types of error ! ◦ Constant error (CE)! ‣ ! ◦ Variable error (VE)! ◦ Total variability (E) - also known as total error ! ‣ Degree how much scores vary from the target; standard deviation units! ‣ How do we figure it out?! • Xi]==! • Xi - whatever trial we are looking at ! ◦ Absolute error (AE) - individual ! ◦ Absolute constant error (ACE)- group ! !...


Similar Free PDFs