Measurement that is... Flashcards
Valid, Accurate, and Reliable yields data that are…
…most useful for advancing scientific knowledge and guiding data-based practices.
Accurate and Reliable yields data that are……
…meaningless for the purpose for which measurement was conducted.
Valid and Reliable yields data that are…
…always wrong. Save data by adjusting for consistent measurement.
Valid and Accurate yields data that are…
…sometimes wrong. Can not separate good data from bad.
Validity (of measurement)
The extent to which data obtained from measurement are directly relevant to the target Bx of interest and to the reason(s) for measuring it. #RelevantDimension
Reliability (of measurement)
Refers to the consistency of measurement, specifically, the extent to which repeated measurement of the same event yields the same values. #SameResultsRepeatedly #IOA
Accuracy (of measurement)
The extend to which obsreved values, the data produced by mesuring an event, match the true state, or true values, of the event as it exists in nature. #TrueValues #Calibration
Threats to Validity
- Indirect measurement
- Measuring the wrong dimension
- Measurement Artificats #DiscontinuosMeasures #PoorlyScheduledMeasurementPeriods #Insensitive/limited
Threats to Accuracy
Human Error #PoorlyDesignedMeasurementSystems #InadequateObserverTraining #ObserverBias #Obseverdrift #MeasurementBias #ObserverReactivity
Threats to Reliability
Human Error #PoorlyDesignedMeasurementSystems #InadequateObserverTraining #ObserverBias #Obseverdrift #Measurement Bias #ObserverReactivity
3 Required Validity elements
- Measuring directly a socially significant target Bx
- Measuring a dimension (rate, duration) of the target B relevant to the question or concern about the Bx
- Ensuring that the data are representatie of the Bx’s occurrence under conditions and during time sthat are most relevant to the quesiton or concern about the Bx.
If the measurement is not valid the accuracy is…
moot.
Accuracy trumps
reliability.
Treatment Drift
Occurs when the applicaiton of the IV differs from the way it was applied at the stuy’s outset,
Observer Reactivity
Influences the data reprted by an observer that results from the observer’s awerness that others are evaluating the data
Measurement Bias
Nonrandom measurement error; a form of inaccurate measurement in which the data consistenly overestimates or underestimates the true value of an event.
Indirect Measurement
Occurs when the Bx that is measured is in some way different from the Bx of interest; considered less valid than direct measurement beacuse inferences about the relation between the data obtained and the actual Bx of interest are required.
Prevent Observer Drift by
by retraining observers
Prevent Observer Bias
with Naive observers
Prevent Reactivity
with unobtrusive recording
