What does statistical significance mean?
The extent to which a result deviates from the null hypothesis and is not due to sampling errors (if a difference cannot be explained by chance, it's most likely real)
What is the purpose of hypothesis testing?
To detect significant differences
What does a null hypothesis generally/usually mean?
Assume there's no difference between the parameters or samples being tested, any apparent difference is due to chance
What does the alternative hypothesis generally/usually mean?
There is a real difference between a parameter or samples being tested that is unlikely to have occurred by chance
What is the 'alpha level' and what level do we almost always use?
The point at which it's highly probable that any apparent differences are real, in social sciences (and this class), we use 0.05
What is a type I error?
The null hypothesis is TRUE, but we accidentally reject it
What is a type II error?
The null hypothesis is FALSE, but we fail to reject it
What is the critical statistic? What do you use to find it?
The point (value) at which we reject or support the null hypothesis, found using a t-table
What are the components of a t-table and what important value of our test is found using it?
T-table uses the 1. degrees of freedom (DOF) and 2. level of significance (alpha level) to find the critical statistic of our test
What is a t-distribution?
A series (multiple) of distributions where the exact shape of each is determined by it's degrees of freedom
What are 3 characteristics of a t-distribution?
1. Is a series of distributions that varies in shape based on it's degrees of freedom
2. As the sample, N, increases, the t-distribution looks similar to a Z distribution
3. Areas under the curve change as N changes
What are degrees of freedom (DOF)?
Represents the number of scores that are free to vary when calculating each statistic (is analogous to sample size [n])
At what point does a t-distribution look like a normal distribution?
At 120 degrees of freedom, a t-dist. is equal to a normal dist., it stops changing shape and has the same properties
What is a one-tailed test?
Used to indicate directionality and test hypotheses that are greater or lesser than the null hypothesis
What is a two-tailed test?
Used to test hypotheses that are different or not equal to the null hypothesis, no directionality indicated
What are the pros and cons of using a one or two-tailed test?
One-tailed test are easier to reject the null hypothesis, but that also means an increased chance of a type I error.
Two-tailed tests have a smaller critical statistic and are more difficult to reject the null hypothesis, but that means there's less chance of an error