fundamental skills lecture 4 statistics 2 Flashcards
how is the sum of squares calculated?
the difference of each observation from the mean, then square each difference to eliminate any negative signs and then add them all together
how do you find the variance?
the sum of squares divided by the number of observations (n)
formula of (x1-x)^2+(x2-x)^2 / n
how do the find the standard deviation?
by square rooting the variance
what does a general linear model describe?
a relationship between your response and explanatory variable.
what else does a general Linear model allow?
helps partition the amount in the total variation into the one explained by your variable and the amount that remains unexplained.
what is the TSS?
The total variation in response variable composed of the unexplained residual SS and the explain SS (explanatory variable)
in the example…Can I explain variation in
egg size using the different bird colonies. which are the response variable and what is the explanatory variable
the response is the egg size
the explanatory is the different bird colony
the data we want to understand the variation is the response and the data we use to account for the variation explanatory variable
what is the response variable?
the y-variable and the dependant variable. in a general linear model its always a covariate
what is the explanatory variable?
the x-variable and the independent variable, its a covariate or a factor
what is a null hypothesis?
the explanatory variable cannot account for variation in response variable
what is a alternate hypothesis?
the explanatory variable can account for variation in response variable
what is the p-value?
the probability of the significance statistic being that extreme or more if the null hypothesis is true
what does the r-squared value mean?
R-sq is the proportion of variation explained by the model
it shows how much the explanatory variable explains the variation in the response variable
when do you reject the null hypothesis?
when the p-value is less than 0.05.
what is type 1 error?
when you reject a null hypothesis that is actually true.
