click below
click below
Normal Size Small Size show me how
Stats - Chapter 3
statistics
| Term | Definition |
|---|---|
| statistic | sample |
| parameter | population |
| mean | don't use if you have outliers |
| median | use if outliers |
| mode | use if you have nominal/ordinal data |
| sample | statistic (x bar) |
| population | greek mu |
| affects mean | unusually large or small data values (outliers) |
| mode types | none, one, two, many |
| midrange | (lowest dv+ highest dv)/2 |
| raw/grouped | use 1 var |
| modal class | the class with the highest frequency |
| weighted mean | x= L1 f=L2 |
| range | highest d.v. - lowest d.v. |
| standard deviation | the square root of the variance |
| (s)tandard deviation | 1 var stats |
| (s^2)ample variance | s(all decimals) then ^2 |
| what is consistent | most data in one class |
| CVAR | (s/x)x100 = ?% |
| higher CVAR % | higher variation |
| range rule of thumb formula | s=range/4 |
| range rule of thumb rules | the distribution is unimodal and roughly symmetric when you don't have raw data |
| use to find lowest/highest value in data set | x (+/-) 2s |
| Chebyshev's theorem | 1- (1/k^2) = ? x 100 = ?% |
| Chebyshev's rules | "at least" any shape graph |
| follow to find percentage of values that fall between two values | step one: find "k" step two: find % |
| find "k" | k= d/s or 1-(1/(k^2))=% |
| empirical rule rules | only normal(bell-shaped) distribution |
| empirical rule table | 1=68% 2=95% 3=99.7% |
| percentile | (((# of values below X)+0.5)/total # of values) x 100= ?% |
| c(position) | =(n x p)/100 n-sample size p=rank |
| c rules | If 'c' is not whole #, round up to the next whole number If 'c' is a whole #, use the value halfway between the "c" and "c+1" values |
| interquartile range | IQR = Q3 - Q1 |
| Q2 | median of data set |
| Q1 | P25 |
| Q3 | P75 |
| measure position outlier(s) | <Q1 - 1.5(IQR) >Q3 + 1.5(IQR) |
| boxplots | min, Q1, MD, Q3, max |