QC Equations Flashcards
Sensitivity
(TP / (TP + FN)) * 100
Specificity
(TN / (TN + FP)) * 100
Positive Predictive Value
(TP / (TP + FP)) * 100
Negative Predictive Value
(TN / (TN + FN)) * 100
Efficacy
((TP + TN) / Total n) * 100
Prevalence
((TP + FN) / Total n) * 100
Bias
Average of the differences between new and old methods (new - old)
SD Diff
SD of the differences between new and old methods (new - old)
-note: NOT difference of SDs
How is R calculated from R^2?
Square root R^2
What value of R is acceptable?
R >= 0.95
r = 1 is perfect correlation r = 0 is perfect randomness
Proportional Error (PE)
(Slope - 1) x 100
F Test
Variance (larger) / Variance (smaller)
Variance = SD^2
F Critical
Look for degrees of freedom on F table
-rows and columns for degrees of freedom correspond to n - 1 for each data set
Paired T Test
(Abs (bias) * SQRT (n)) / SD diff
T Critical
Look for degrees of freedom on T table
-degrees of freedom = n - 1
Passing F test value. What happens when F test fails?
F Test < F Critical
Failing F test means we do not need to do the T test
Passing T test value
T Test < T Critical
The Y-intercept of a line is also known as…
Constant Error
Coefficient of Variation (CV)
(SD / mean) * 100
Standard deviation (SD)
Square root ((E (x - mean)^2) / (n - 1))
% Error
Abs ((measured - assigned) / assigned) * 100
Median
Middle number of organized data set (smallest to largest)
Mode
Most frequently appearing number
How to determine the probability of X between two Z-scores
(Difference between values of the 2 Z-scores) * 100
How to determine the number of values that fall between two values using the Z score
(Difference between values of the 2 Z-scores) * total n
-note - don’t need to convert probability into % since we are multiplying by it
Variance
SD^2
Z score
(x - mean) / SD
1 SD probability
68.2%
2 SD probability
95.5%
3 SD probability
99.7%