Quick Review Chapter 9 & 19 Flashcards
Why do you suppose it is a called a goodness fit test?
we are testing how “good” of a “fit: we have between the data we collected and the results we would have by chance
What is a contingency table?
table that shows the frequency of occurance of outcomes of one level of the independent variable across the levels of the other independent variable
What is a double blind study?
it is a study in which neither the ppl in the study or the researchers know which condition the partcipant was assingned
what is a marginal total?
the total of the rows and the collums and the grand total
how do you use a chi square test to perform test on proportion
you convert porportions to frequencies
Risk ratio: another name for it
relative risk
what does it mean to say that a risk for cancer among long term smoers and non smokers is 24.2
long term smokers have 24.2 more likely to get cancer then a non smoker
In a scatterplot, which variable is plotted on the absiscassa (x axis)
if there is an obvious indepedent variable, that goes there- if not its pretty much arbitrary
the regression line should be interpretted as
the line that gives the best prediction of Y for a given value of X
A corelation is -.81 is significantly lower then the correlation of +8.1 (true or false)
false- the sign only tells us the direction of the relationship
The covariance, by itself is not a satisfactory stattistic for preseting the degree of relationships between variables because
it varies with the variance of the underlying variables as well as the degree of the relationship
spearmens correlation coeffecient:
a pearson correelation coefficient
list thing things that affect the relationships btw two variables
range restricions, heterogeous subsamples, and non linear relationships
We would use a two tailed test of R if
we wanted to reject the null if the same was to positive or too negative
a point biserial correlation applies to data where
one variable is a dichotomy and the other is continous
