Oct 16 Bio Reading Statistics

From Noop
Jump to: navigation, search
  1. Categorical vs categorical: chi squared
  2. Categorical vs continuous: t-test
  3. Continuous vs continuous: linear regression

Standard Error Bars

Standard error of the mean indicates uncertainty in a calculated mean. It measures precision.

Confidence interval is the range a value is expected to fall within.

SEM = SD/sqrt(sample size)

Chi-Square

Used to compare observed with expected (theoretical) data to determine whether certain things are due to chance or are significant.

T-tests

Used to determine if there is a significant difference between means.

Analysis of Variance (ANOVA)

Used to compare means of 2+ sets of data and determine which variables affect which.

Regression and Correlation

Done to determine the slope of the relationship (if it exists) between two variables.

p-values and significance

  1. Specify null hypothesis
  2. Calculate test statistic
  3. Determine probability of getting by chance a test statistic at least as large as the one calculated.

P-value indicates probability of rejecting null hypothesis. A P value of 0.01 means a 1% chance that the null hypothesis has wrongly been rejected. 0.05 is generally considered to be enough.

Both-And Rule

The problem of two random events (before they happen) is the probability of each multiplied together.

Either-Or Rule

When there are multiple ways for an event to occur, add the probabilities as long as they're independent.