Descriptive stats + Log Models

Test retest (Validity)

use same process with same sample multiple times

Parallel forms

same sample second set of questions similar to the first set

Convinience sampling

most accesible

simple random

randomly allocated

• equal chance

systematic

random starting point but equal intervals

stratified

divides populations in groups with similar characteristics (strata)

quota

sample size proportionate to actual population

median formula

middle value

• (n+1)/2

Outliers

x< Q1 - 1.5(IQR)

X> Q3 + 1.5(IQR)

Linearisation methods

1. y vs. logx

2. logx vs. y

3. logx vs. log y

Linearisation - y vs. logx

linear

y=mx+b

Linearisation - logy vs. x

exponential

logy= b m^x

• y=10^b (10^m)^x

Linearisation - logx vs. log y

Power

logy = b*logx^m

• y=10^b (10^x)^m