Evaluating your measurement tool Flashcards
What is content validity
When the instrument used is designed to accurately measure the concepts under study
What is convergent validity
When the results obtained are similar to the results obtained with another previously validated test that measures the same thing
What is correlation coefficient
A test value used to determine how closely one measurement is related to a second measurement
What is divergent validity
When the measurement of the opposite variable of a previously validated measurement yields the opposite result.
What is efficiency (EFF)
Measures the probability of agreement between the screening test and the actual diagnosis
EFF = (A + D / A + B + C + D) x 100
A = true positives
B = # of positive tests
C = # of false negatives
D = true negatives
What is equivalance
How well multiple forms of an instrument or multiple users of an instrument produce the same results
What is homogeneity
The extent to which the properties of any part or variable of an overall data set are the same as any other part.
What is internal consistency reliability
Homogeneity of the measurement instrument. OR a reliability measurement in which items on a test are correlated in order to determine how well they measure the same concept or construct.
What is inter-rater reliability
When you compare the measurements obtained by two different data collectors to make sure they are similar
What is reliability
The consistency or repeatability of the measurement
What is predictive validity
When the instrument used accurately suggests future outcomes or behaviors
Cronbach’s alpha
Represents the internal consistency reliability on a scale of 0 to 1 where 0 is no reliability and 1 is perfect reliability
What is readability
affects both the validity and reliability of an instrument
What is validity
accurate measurement and collection of data
What is stability
the consistent or enduring quality of the measure. It should not change over time and be consistent when tested repeatedly.
Accuracy of a screening test is determined by
the tests ability to identify subjects who have the disease and subjects who do not.
What is sensitivity
the probability that a test subject with a disease will test positive Sensitivity = A / A + C A = true positives C = all who have the disease (including those who test negative)
What is specificity
the probability that a well subject will have test negative on the screen (no disease)
Specificity = D / (ß + D)
D = true negatives
ß = All those who do not have the disease, even if they test positive
What is the positive predictive value
(PPV) tells you what the probability is that a subject actually has the disease given a positive test result
PPV = A / (A + B)
A = true positives
B = number who tested positive
What is prevelance
the amount of illness (# of cases) present in the population divided by the total population
Prevalence = A + C / A + B + C + D
A = true positives
B = # who tested positive
C = # of false negatives
D = True negatives
What is negative predictive value
the probability that a negative screen result represents absence of the disease. NPV = D / (C + D)
D = true negatives
C = False negatives
What is central tendency
The indicator for the center of a dataset as either mode, median, or mean.
What is mode
The most frequently occurring measure(s) in a dataset.
Unimodal - one most frequent value
Bimodal - two equal most frequent values
Multimodal - > 2 most frequent values
What is the median
The middle value when values are listed from least to greatest. Note: if there are an equal number of values the median is the average of the two middle values.
What is the mean
The mean is the average of the values in a dataset (susceptible to outliers)
What is the range
The difference between the least and greatest values
What is the standard deviation
The average distance from the variables mean. (ð)
