(Original post by glitterphobia)
I'm also retaking this! completely messed up last time and walked out half-way through. hardly revised AT ALL. been too busy concentrating on biology (which i'm still behind on too).
Here's some basic bits to remember:
Categorical - non-numerical, discrete - listed, continuous - can't list.
Stem&Leaf - discrete/continuous.
Vertical Line - discrete
Histogram - continuous, frequency proportional to area
Mean = sum of x divided by n (in grouped, sum of xf divided by n)
Median = (n + 1)/2th item (Q2 in box and whisker) 1/2n
Lower quartile = Q1, median of lower half, 1/4n (Q3 is same, just upper half/upper quartile, 3.4n)
Interquartile range = Q3 - Q1
Mode = most common (modal class, class with most data when equal width).
Mid-range = half of min + max.
Range = max - min
Sum of squares = sum of (x - x^2) (in grouped, sum of (x - (mean)^2f))
Mean Square Deviation = sum of squares/n
Root Mean Square Deviation = square root of above
Variance = sum of squares/n - 1
Standard Deviation = square root of variance.
Outlier = more than 2s away from mean (or more than 1.5 x interquartile range beyond lower/upper quartile).
P(A) = n(A) / n(E) (n(A) = no. of ways A can occur, n(E) = all poss events).
P(A U B) = P(A) + P(B) - P(A n B) (mutally exclusive, P(A n B) = 0, so just P(A U B) + P(A) + P(B).
P(A) + P(A') = 1
P(B|A) = P(A n B)/P(A)
P event A then event B = P(A) x P(B|A)
If event B independent of event A = P(B|A) = P(B|A') = P(B).
Discrete random variable, X with values r1, r2... and P p1, p2, p3 etc = sum of p1 to pn = sum of k=1 to n P(X = rk) = 1; pi > 0.
Expectation E(X) = sum of rP(x = r).
Variance Var(X) = sum of r^2P(X = r) - (E(X))^2
No. of ways of arranging n objects in line = n!
n! = n x (n - 1) x (n - 2) x... x 3 x 2 x 1.
No. of permutations of r objects to n, nPr = n!/(n - r)!
No. of combinations of r objects from n, nCr = n!/(n - r)! r!
Permutations - order matters, combination - order doesn't matter.
0! = 1.
P(X = r) = nCr x q^n-r x p^r.
B(n, p) expectation of no. of successes = np.
Hypothesis testing, check:
1. Test set up before/after data known?
2. Sample chosen at random/independent?
3. Does it test original claim?
Steps for hypothesis testing:
1. Establish null and alternative
2. Decide significance
3. Collect data (random so items independent)
4. Conduct test
5. Interpret result in terms of original claim.
That's about it.
(my website with video tutorials for MEI)
(all past papers)