Equations
Desired outcome/ All possible outcomes
Probability
Data from a subset of the population
Sampling
You want to investigate whether there is a significant difference in the mean test scores between two schools, School A and School B. You collect test scores from a random sample of students from each school.
Independent t-test
Multivariate tests have how many dependent variables?
2+
What p-value are we looking for when we look for significance
Less then .05
_
(X - X)
Deviation
Our capacity to detect a significant effect
Power
You want to investigate whether there is a significant difference in the mean test scores among three different teaching methods: Method A, Method B, and Method C. You collect test scores from students who were randomly assigned to one of the three teaching methods.
One way ANOVA
If a correlation coefficient is -.7 what does that mean?
Strong negative correlation
What does correlation NOT Equal
Causation
_
X-M/ SD
Effect size
A numerical value (or values) that defines the center of a distribution of data
Central tendency
You are conducting a study to examine the effectiveness of a new memory training program. You recruit a group of participants and measure their memory scores at three different time points: before the training program (pre-test), immediately after completing the training program (post-test 1), and one month after completing the training program (post-test 2).
Repeated measures ANOVA
Which is a “stronger” correlation or regression?
Regression
What does the null hypothesis tell us?
There is no statistically significant change or difference
SS time/ Df time
Means square time (repeated measures ANOVA)
How “spread out” the data are within the distribution
Variability
You conduct a study investigating the effectiveness of a new medication in reducing blood pressure. A group of participants is measured for their blood pressure levels before taking the medication (baseline) and after two weeks of taking the medication.
Paired samples t-test
How is beta (β) different then b?
Beta does not vary as a function of the scale (standardized)
Which two measurement types get combined and get called “continuous”
Ratio and Interval
(I wrote it on the whiteboard, I'm not paying $20 just to use the squared function)
Variance
The actual size of the mean or difference between means
Effect size
The average score on the SAT practice test nationwide is 1000. You want to determine if a group of 50 students from a particular school district performs significantly differently from the national average.
One sample t-test
List 1 limitation of using a correlation
Outliers, Range restriction, Heterogeneous subsamples
What does a normal distribution presume? (all 3)
The more samples drawn from the population the more “normal” the distribution
Values will be normally distributed in a population
We can use the normal distribution to determine probability