- Shuffle
Toggle OnToggle Off
- Alphabetize
Toggle OnToggle Off
- Front First
Toggle OnToggle Off
- Both Sides
Toggle OnToggle Off
Front
How to study your flashcards.
Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key
Up/Down arrow keys: Flip the card between the front and back.down keyup key
H key: Show hint (3rd side).h key
PLAY BUTTON
PLAY BUTTON
15 Cards in this Set
- Front
- Back
|
Gaussian distribution
|
arranged symmetrically around the mean
mean=median=mode |
|
Gaussian distribution:
1sd 2sd 3sd |
68% within +/- 1 SD
95% within +/- 2 SD 99% within +/- 3 SD |
|
Standard Deviation
|
average distance an individual value from the mean
|
|
Standard Deviation Formula
|
1. xi-mean & square that
2. add all the means together |
|
Skewed distribution
|
mode>median>mean
mean is skewed by outliers |
|
Skewed distributions - how do you measure data since it's not Gaussian
|
Rank data & calculate percentile rank
ex: 5% (below 2.5 & >97.5) |
|
Determining Differences between data sets (difference between the performance of two methods)
|
1. difference between means
2.difference in dispersion of data 3.correlation of determination 4. bias plots |
|
Differences in mean - definition (what are you looking for?)
|
whether two methods produce the same average value
Least useful parameter gaussian - t-test Non-gaussian - Wilcox-Rank-Sum |
|
Differences in dispersion of data -definition
|
a new method is more or less reproducible than the current method
F-test |
|
Degree of correlation between two methods
|
least squares methods of linear analysis
|
|
least squares method
|
calculate an equation for a line in which the sum of the squares of the distances of each point from the line is at a minimum
new data = y axis reference = x axis y intercept = bias slope = the result from method y is x% of the result from method x |
|
correlation coefficient (r)
|
calculate who well the points fit to the line
1 to -1 |
|
Bias plots (Bland-Altman plot)
|
y axis - bias between new method & reference method
x axis - result from reference method ideal method- percent bias similar at all concentrations |
|
Reference ranges
|
Convenience samples - 100-120 people
Validate a previously est reference ragne - 20 people, no more than 2 fall outside the reference limits |
|
Partitioning
|
Refers to separating the sample into subgroups - based on features (gender )
|
