Stats test 1 Flashcards
Population
consists of all individuals or objects of a particular type
Descriptive statistics
consists of procedures used to organize and summarize the important characteristics of a set of measurements
Inferential Statistics
allows us to draw conclusions based on information contained in a sample
Probability
forms a bridge between descriptive and inferential stats and gives us an understanding of the properties of a population based on its distrubtion (deductive reasoning)
Steps of Hypothesis test
1.) determine parameter 2.) formulate null hypothesis 3.) determine test statistic 4.) From Ha 5.) set alpha 6.) compute 7.) determine Reject or DNR
Type 1 Error
rejecting the null hypothesis when Ho is true
Type 2 Error
Accept null hypothesis when Ho is false, denoted by beta
P value
the smallest level of significance at which Ho would be rejected, the smaller the value the more contradictory it is to Ho
P values for z test
Upper (1- phi(z))
Lower phi(z)
Two tail 2[1-phi(|z|))
Pooled t
two populations that are normal with equal spreads, determining if centered on same place
Testing Proportions
p hat (sample proportion) n*Po >10 and n(1-Po)>10 for test to be valid
Confidence Interval
range where parameter is expected to fall, it is the measure of the degree of reliability of the interval
Interpretation of CI
95% of all confidence intervals constructed will cover the true mean
Precision of CI
increase n, to make interval smaller, making alpha bigger will result in a bigger CI
Margin of Error
based only on sample size
T curve and Z curve
Z curve is a t curve with infintie degrees of freedom
Factor
a variable that is being changed and results observed
Level
a value that is assigned to change the factor (I)
Treatment condition
the set of conditions for a test in an experiment (factors and their levels
Replicate
a repeat of a treatment condition (J)
Randomization
treatment conditions are run in a chance order to prevent any build up of results
Orthogonal array
a simple way of putting together the treatment conditions so that the design is balanced and factors can be analzed
Interaction
two or more factors that together produced a result different than their separate effects
Assumptions
I treatments have distrubtion N(mean, variance) with similiar variances their sum is 0, Each Xij is normal and IID,
Evil error
random error, dont know where it came from (eij)
Alpha i
deviation- spread from mean
Alternative description of ANOVA
Xij= mu+alpha(i)+ eij
Fixed Effects
alpha
