EMB 3 ; Hypotheses testing, p values and CI

Question 1

SAMPLING VARIATION

Answer 1

• means of different random sample will be different
• if sample is large enough it will be distributed NORMALLY
• thus sample & population mean will equal (almost)

Question 2

how to calculate sampling variability

Answer 2

standard error

Question 3

accuracy vs precision

Answer 3

accuracy; measures how representative the study is of the whole population.

precision ; measures the VARIATION between sample statistics

Question 4

standard error

• what is it?
• formula
• whats it used for

Answer 4

SE; measures the PRECISION of the sample mean as an estimate of the population mean

formula ; SE= ;
SD over;
square root of n

used?
to estimate population mean from single sample

Question 5

standard deviation
- what is it?
- formula?
whats it used for?

Answer 5

what?

• measures the variation in the POPULATION
• tells us how far away an individual observation or measurement is away from the population mean

?

Question 6

confidence interval

• what is it
• formula

Answer 6

what?
- gives us a range in which we are confident that the population difference lies within the range (95%)

formula ;
mean symbol ; line then a triangle below

mean ±1.96 SE

Question 7

REFERENCE range & CI ;

are they the same?

Answer 7

NOOOOOOOO;

ref range (&SD) ; measures how much variation there is between individual observations and population mean

formula ; mean ± 1.96 sD

CI(&SE); measures how precise an estimate the sample mean is of the population mean

Formula; mean± 1.96 sE

Question 8

PROPORTION
formulas change
- SE
- CI

Answer 8

SE;
SQAURE ROOT ; p(1-p)
sqaure root n

CI changes follows above

Question 9

always easier to do what with hypotheses

Answer 9

to find evidence against them

Question 10

p values

• what are they?
• what do they mean
• how to calculate

Answer 10

what?

• they give strong/weak evidence against null hypotheses
• smaller ; greater the evidence against null hypotheses

aka ; the lower the chance of getting a difference as big as the one observed of the H0 is true

• greater; weaker the evidence

calculate?
- calculate test statistic or z score ;

difference in means over ;
SE of difference

then look up log tables to convert z score to p value