Flashcards in Statistics III Deck (29):
What is the basic idea behind survey sampling?
simple random sampling
If respondents from small geographically local clusters are chosen, it is called ...
... cluster sampling.
How is the procedure, of which the following is an example of, called?
We could randomly choose a number of women from a population, and separately randomly choose a number of men from the population, where the numbers are chosen so that the proportions of males and females are the same as in the population.
What is the correlation coefficient?
the normalized version of the covariance
What does the correlation coefficient show by its magnitude?
By its magnitude it shows the strength of the linear relation.
What does "normalization" (in its simplest case) mean?
Adjusting values measured on different scales to a notionally common scale (often prior to averaging).
What does μ stand for?
the mean of the population
What does σ stand for?
the standard deviation
How is the letter σ called?
How is the letter μ called?
What's the greek symbol for the "mean"?
What's the greek symbol for the standard deviation?
What does it mean to standardize values?
To standardize values around their mean and a standard deviation of 1.
What does x-Dach stand for?
the arithmetic mean
How to calculate the standard score of a raw score x?
z = (x - μ) / σ
What does z represent in the calculation of the standard score of a value x?
The absolute value of z represents the distance between the raw score and the population mean in units of the standard deviation. (negative or positive)
About covariance: What does s = 0 mean?
no association, statistical independence
About covariance: What does s > 0 mean?
About covariance: What does s < 0 mean?
What do we know about the mean and the standard deviation of a z-score?
μ(z) = 0
σ(z) = 1
About correlation: What does r = 0 mean?
no correlation, statistical independence
About correlation: What does r > 0 mean?
About correlation: What does r < 0 mean?
What tells us the coefficient of determination?
The amount of the variance of y that can be explained with y's linear dependency with x.
r = 0.5
r² = 0.25 -> 25 % can be explained
Is Cor(x, y) = Cor(y, x) true?
Is Regression(x,y) = Regression(y,x) true?
What does regression(x, y) tell us?
The influence of x on y.
In a regression(x, y) how are x and y called?
x ... predictor variable
y ... response variable