Statistics: Random Sampling

Observational Study

Observes individuals and measures variables of interest but does not attempt to influence the response. Implying no treatment

Experiment

Deliberately imposes some treatment on individuals to measure their response. Placebo is a treatment; placebo is not an explanatory variable.

Subjects

When experimental units are human beings

Experimental Units

The smallest collections of individuals to which treatments are applied

Confounding variables

Occurs when 2 variables are associated in such a way that their effects on the response cannot be distinguished from each other.

4 components for a GREAT experiment

• Replication: Use enough experimental units in each group so that any differences in the effects of the treatments can be distinguished from chance differences between the groups

• Random Assignment: Use chance to assign experimental units to treatments. Doing so helps create roughly equivalent groups of

  experimental units by balancing the effects of other variables among

  the treatment groups

• Compare 2 or more variables

• Control Group: Keep other variables that might affect the response the same

  for all groups.

Random assignment

The experimental units are assigned to treatements completely by chance

Blind

If either the subjects or the researchers (not both) know who is receiving the treatment

Double Blind

If neither the subjects nor the researchers know who is receiving which treatment

Statistically significant

When an observed difference between groups is too large to be explained by chance

Hat Method

• write down (no. + thing) on equally sized pieces of paper

• Place in a hat and SHAKE IT

• Draw out (how many ever is needed) pieces of paper with no replacement

• These ___ will get ___

Random Digit Table

• Number all (things) 000 - (whatever no.)

• Select (no. that is needed) unique 2/3/4-digit numbers, ignoring 000 to (no. after) — 999

• Those ___ will get ___

Random Number Generator

• Number all (things) from (sample size; i.e. 1 - 600)

• Select (how many is needed) unique numbers between 1-600 inclusive (TI-84 (randint: 1, 600,1) )

• Those ___ will get ___