Lesson # 23

Front 1

Placebo

Back 1

fake treatment or no treatment

Front 2

Observational Study

Back 2

A study where researchers observe individuals and record info of interest. (sample studies are observational)

Front 3

Experiment

Back 3

a study where researchers impose treatments and measure responses of individuals.

Front 4

Definitions

Back 4

Population : entire group of individuals about which the researcher wants info
Variable: Characteristic measured in the sample.
Sample: part of population that is measured.
(missin some definitions here)

Front 5

Non-Probability Sample

Back 5

A sample chosen using personal judgement to determine which individuals are in sample or not.Probability of having an individual in sample is not available.(This is bad sampling)
Examples:
Convenience Sample
Voluntary Response Sample
Mall-Intercept Sample
Quota Sample

Front 6

Def of non-probability samples

Back 6

Convenience : as said
Mall-Interecept: relevant questions to subjects and for which answers can be generalized.
Voluntary Response: subjects respond themselves
Quota Sample: interviewers select individuals themselves.

Front 7

Probability Sample

Back 7

sample chosen to use some random device that gives each member a chance of being selected. Probability can be computed.
Examples:
SRS, Multi-stage, stratified.

Front 8

SRS Explanation

Back 8

SRS: every possible SAMPLE has equal chance of selection.
Use computer generated software or random digit table or some valid device.

Front 9

Random Digit Table

Back 9

1. Each digit in table has same probability.
2. Knowledge of one part of table gives no info of other part.
STEP #1 Subjects assigned numbers
STEP # 2 Select a line from the Table for each member.
STEP # 3 All no. larger than Population size are eliminated.
STEP # 4 Repeated numbers are deleted
STEP # 5 Start from step # 1 until you have desired sample size to investigate

Front 10

Stratified Sample

Back 10

1. Classify individuals into groups (strata)
2. Choose Probability Sample within each stratum
3. Combine samples from strata to form a complete sample.

Front 11

Multi-Stage Sample

Back 11

For three stage sampling:
#1 probability sample of groups is selected
#2 probability of sample of sub-groups is taken from each selected group
# 3 units are sampled from within each selected sub-group.

Front 12

Further study on multi-stage

Back 12

1. When you increase stages in multi-stages you are increasing the number of times you will sample in each group.
2. Diff b/w multi-stage and stratified sample is that in stratified you would sample from every group. But in multi-stage you select certain groups and then sample from them. In every stage in multi-stage we "sample diff things in each stage."