Stats: Lecture 2
Basic Concepts
Degrees of freedom (df): number of data points free to vary
o How many observations you need to know before having all information about a
sample
o i.e.
We have four data points. How many data points do we need to calcula
Messy Data
For most analyses, its wise to assume we violate assumptions to some degree
We dont care whether we violated them we care about whether departure from the
assumptions is sufficient to have an impact on our inferences
Avoiding
o To avoid messy d
Contrasts
Matrices
o We are dealing with multiple variables so calculations get complicated
Calculations are going to get more complicated so we now need to rely on
matrix algebra
o Matrix algebra is used
o *Just need to know what a matrix is
o Matrices:
Lecture 6: NHST for Discrete Data
Binomial Distribution
o Binomial = when we have two outcomes
i.e. success vs. failures
Heads vs. tails, red card vs. black card
o Recall the pmf (probability mass function) and cmf (cumulative mass function)
o Recall th
Lecture 7: Effect Sizes
Just because something is statistically significant does not mean it is practically or
clinically significant
o You should use effect sizes in conjunction with NHST; effect sizes give you an
indication of the magnitude of the effec
Lecture 8: Power
4 basic possible scenarios at the end of a study:
ActualStateofAf
TrueNull
Decision
Failto
correct
False
typeII
RejectNull
RejectNull
typeIerror
o If the null hypothesis is true and we reject the null, this is a type 1 error
.05 = probab
Lecture 5: t-tests
Recall z-tests (review)
o When we compared a single person to a population mean, we used:
Z = score sample mean/sample SD
o When we compared a sample mean to a population mean, we used:
Z = (Xbar pop mean) / pop SD (or SE)
o We used t
Lecture 3: Confidence Intervals
Confidence Interval: gives a range of values that have a set probability of containing the
population parameter; the P of the CI containing that parameter is the confidence level
o i.e. a 95% CI has P=.95 of containing the
Stats: Lecture 1
Basic Concepts
Variable: a condition or characteristic that can have different values
o Examples of variables: depression, conscientiousness
o Independent variable (IV): variable that is manipulated and has an effect on a
dependent variab
Lecture 4: Null Hypothesis Significance Testing
NHST Procedure
o (1) State null and alternative hypotheses
o (2) Set criterion for rejecting null hypothesis
o (3) Collect sample and compute statistics
o (4) Interpret results
Null and Alternative Hypothese