Statistics
-descriptive measures computed from the data of a sample.
Parameters
-measures computed (estimated) from the data of a population.
Statistical estimation
-process by which estimates of population parameters are generated from sample statistics with minimal bias.
Hypothesis testing
-making a conclusion about a hypothesized difference or relationships using observations from the sample.
Random Error
-error that varies unpredictability from one measurement to another.
Systematic Error
-error that has similar values from one measurement to another.
a
Which systematic error is based on how we choose our sample?
a) sampling method
b) observation/instrument influence
c) confounding
b
Which systematic error is based on human error or calibration?
a) sampling method
b) observation/instrument influence
c) confounding
c
Which systematic error is based on not accounting for the effect of a third variable?
a) sampling method
b) observation/instrument influence
c) confounding
Statistical distribution
-a type of distribution that is defined by some theoretical probability distribution.
-describes the way random variables are expected to behave.
empirical distribution
-when values are taken from the actual data and calculated to determine the distribution.
z-scores
Which statistical distribution type is used to identify outliers?
Central Limit theorem
-states that given a sufficiently large sample size, the sampling distribution of the mean of a variable will approximate a normal distribution regardless of that variable's distribution in a population.
will equal the population mean
In the CLT, the mean of all sample means _________.
is equal to the standard error of the mean
In the CLT, the standard deviation of sampled means ________.
declarative, relationship between 2 or more variables, testable
What are characteristics of a good hypothesis?
2-tailed test
-the statistical value we are measuring can be lower or higher compared to the placebo or control
1-sided test
-We are certain that the statistical value we are measuring will be higher or lower in those who are exposed to the variable than in the control group.
Type 1 error
-rejecting the null hypothesis when the null hypothesis is actually true.
-false positive
Type 2 error
-failing to reject the null hypothesis when the null hypothesis is actually false.
-false negative
Beta
______ error is closely related to power.
Power
-the probability of correctly rejecting the null hypothesis when false
tells us the likelihood of being correct; identifying an effect that actually exists
Why is power analysis important?
Alpha
________ is the threshold for rejecting the null hypothesis.
reject null
p<a
fail to reject
p>=a
p-value
-quantifies how unusual the observed results would be if the null hypothesis were true