[] Q [Research] Biostatistics concept / design.

Intro. Population vs. Sample
Good sampling -> assume all population appropriately.
(Good representation) -> (Correct generalization)

I. What kind of data? 

1. Qualitative(alphabet) measured:  binary(2), multilevel = Politative?(multiple)
 a. nominal = sex(F M)
 b. ordinal = grade(ABCD)
 - a, b = will be described with frequency vs proportion(or percentage)
 - proportion, percentage: complement rule(success + fail = 1 or 100%)

2. Quantitative(numeric =mathematical) measured: Continuous(uncountable, like height),
                                                                                    Discrete(countable, finite like number of pt.)
 a. interval = temperature(not meaningful zero)
 b. ratio = speed(meaningful zero = absence)
 - a vs b = only meaningful zero or not.

II. how to describe group?

1. Representative of data = 'center'

  - Informative strength = Mean > Median >>>> Mode
  - Mean: all factors matter
  - Median: good for avoiding outlier
 
2. Variability of data = 'how spread'

   - Informative strength = SD > Range(IQR)

III. How dose they apply to?

1. Qualitative = frequency, proportion
2. Quantitative = Mean/SD or Median/Range

IV. Skewed (right skewed = Lt. deviated, Rt. tail = outlier) 

1. Normally distributed = Mean/SD
2. Skewed = Median/Range

[ ] JMP for chapter 5.