CRD (Completely Randomized Design

All experimental units assigned random among all treatments

BLOCKING

Goal: create homogenous groups Benefit: reduce variation

RBD (Randomized Block Design

Experimental units put

into homogeneous blocks. The random assignment of the units

to the treatments is carried out separately within each block.

Matched Pairs

A form of blocking in which each subject

receives both treatments in a random order or the subjects are

matched in pairs as closely as possible and one subject in each

pair receives each treatment, determined at random

Experiment

Experiments have treatments (5 pts) and

name the treatment

Observational Study

Observational studies do not have treatments

Purpose of Blinding/Double Blinding

to reduce bias

Blind

When the subject does not know which

treatment is being administered

Double Blind

When the subject nor the

evaluator does not know which treatment is being

administered

Volunteer Response:

when only those that choose

to participate do usually because they have an

emotional connection (feel strongly

Response Bias

give a response that is not truthful

because of the wording of the question or because

they are uncomfortable to respond truthfully

Undercoverage Bias:

Where a certain group of the

population is left out

Selection Bias

Where a certain population is

overrepresented in the sample

Advantage of using a

Stratified Random Sample Over an SRS

Guarantees that each STRATA will be

represented (10 pts)

2. Reduces variability in a SRS of same size.

(10 pts)

Does correlation imply causation?

Observational studies do NOT show causation (4

pts) Only well-designed (4 pts) controlled (4 pts)

experiment (4 pts) shows cause and effect

(causation). (4 pts

Why use a control group?

Used to evaluate the effectiveness of the

treatment(10 pts) by

2. REDUCING the effect of confounding

variables (10 pts)

Confounding Variable

A variable that is not controlled in the experiment

(outside factor) that is effecting the response

variable rather than the explanatory variable is

effecting it. Describe how it effects both groups in

context.

𝐑𝐚𝐧𝐝𝐨𝐦𝐢𝐳𝐚𝐭𝐢𝐨𝐧

Purpose: To reduce bias

Conclusions you can make:

For Random Sample (RS): Generalize the

conclusions to the population

For Random Assignment (RA): Cause and Effect

RS is collecting the subjects randomly. RA is assigning

the treatments to the subjects randomly.

SRS

Number the entire population, draw numbers from a

hat (every set of n individuals has equal chance of selection)

Stratified

Split the population into homogeneous groups,

select an SRS from each group.

cluster

Split the population into heterogeneous groups

called clusters, and randomly select whole clusters for the

sample. Ex. Choosing a carton of eggs actually chooses a

cluster (group) of 12 eggs.

census

An attempt to reach the entire population

convenience

Selects individuals easiest to reach-improper

way to sample introduces bias