subjects
statistics
the arithmetic is rarely the problem. knowing which test, and what the number means once you have it, is.
where it usually goes wrong
the six that cost
the most marks
which test, and why
t or z, paired or independent, chi-square or anova. it follows from your data type, your sample size and what you are actually asking.
what a p-value says
it is the probability of data this extreme if the null were true. it is not the probability the null is true, and that difference costs marks constantly.
confidence intervals
a 95% interval does not mean a 95% chance the parameter is inside this one. the 95% belongs to the procedure across repeats.
conditional probability
p(a|b) against p(b|a). bayes is the bridge, and most “surprising” results are just a small base rate.
sample against population
n minus one is not a typo. it is the correction that keeps a sample variance honest.
regression
correlation is not causation, and r squared is not accuracy. residual plots tell you whether the line was allowed in the first place.
what to post
a photo of the problem, or a sentence about where you got stuck. set the deadline and attach anything that matters — the sheet, your working, the slides.
verified tutors send offers with a price, a rating and a completion rate, and you pick who helps you. the tutor works it with you, so the next one is yours. more questions.
related
post it in the app
statistics, any level.
or scan to get it on your phone