Medical Biostatistics 1
by
Mode is not sensitive to outliers –> measures only the most frequent observed value
Central Tendency
- Center of Normal distribution
- Mean: avg
- Mode: most frequent observed value
- Median
- Odd = Middle number is Median
- Even = Avg of Middle two numbers
- Chart: Mode is the heights point in the chart
- Chart: Positive Skew / Negative Skew (Mode [top] > Median > Mean)
- Chart: Mode least likely to be affected by outliers, only affected if common number is changed
Dispersion Measures
- Standard Deviation
- Variance
- Standard Error of the Mean
- Z-score
- Confidence interval
Sd
How dispersed is the data set? Two datasets can have the same mean, median, mode but can be different in how much they are spread out (far away from the mean).
- first get difference between each data point and mean
- Squared of the sum (remove neg signs)
- Sum of those difference
- +-1Sd == 68% of population
- +-2Sd == 95% of population
- +-3Sd == 98.7% of populationb
Variance
It is Sd squared.
Standard Error of the mean
How precisely you know the true population mean. How close we are to the true mean. The higher sample number, the less SEM (closer to true mean)
Z-score
Equal to the number of Sdeviations you are away fromt he mean
- 0 == mean
- +1 == 1sd > mean
- -1 == 1sd < mean
Confidence intervals
CIs are for estimating population mean from a sample dataset (which is a subset of the population and not the entire population)
- Formula:
CI = Mean +/- (1.96 x StdErrMean)
Confidece Intervals vs Standard deviation
Sd is for a given dataset and reflects the mean range within that sample
where as
CI does not describe the sample, it is an inferred value of where the true mean of the population might lie.
tags: biostatistics - formulae