Collection of Data

1
Collection of Data

• Jim Bohan
• Bohan.educ_at_gmail.com

2
Three Types of Studies

• 1. Survey
• A study in which the researcher gathers data by
  asking for responses from subjects.
• 2. Observational Study
• A study in which the researcher observes
  behaviors of the subjects.
• 3. Controlled Experiment
• A study in which the researcher imposes
  treatments on the subjects.

3
Methods of Data Collection

• Census
• Studying ALL subjects of the population of
  interest.
• Sample
• Studying a proper subset of the subjects from
  the population of interest.

4
Issues with Sampling

• The purpose of sampling is to generate a proper
  subset of the population that is representative
  of the population.
• The major concern with sampling is
• Bias
• Bias is a systematic effect that skews all of
  the data values in a sample.

5
Types of Sampling INVALID

• Convenience Sampling
• Choosing the subjects in the sample by
  convenience.
• Voluntary Response Sampling
• Subjects are included in the sample on the basis
  of their volunteering to be included.

6
Valid Types of Sampling SIMPLE RANDOM SAMPLE
7
Valid Types of Sampling STRATIFIED RANDOM SAMPLE
8
Valid Types of Sampling SYSTEMATIC SAMPLE
9
Valid Types of Sampling CLUSTER SAMPLE
10
Valid Types of Sampling TWO STAGE SAMPLE
11
The Key to VALID SAMPLING

• Subjects are chosen by the application of a
  probability rule that is, based on RANDOM
  SELECTION

12
Controlled Experiments Vocabulary

• Experiment
• A study in which the researcher imposes
  treatment(s) on the subjects.
• Controlled Experiment
• A study in which groups receive different
  treatments whose effects are compared.
• Units
• The subjects who participate in the study.

13
Controlled Experiments More Vocabulary

• Subjects
• The term applied to human units.
• Control Group
• The group who receives either no treatment or a
  placebo, a treatment that causes no effect.
• Treatment Group(s)
• The group(s) who receives the treatment(s).

14
Controlled Experiments More Vocabulary

• Explanatory Variable
• The variable to which the researcher assigns
  values in the study the independent variable.
• Response Variable
• The variable that measures the effect of the
  value of the explanatory variable the dependent
  variable.

15
Three Requirements of Controlled Experiments

• Comparison
• Randomization
• Replication

16
Confounding The Problem

• Two variables are CONFOUNDED when the effects of
  the explanatory variable cannot be separated
  among the treatment groups.
• A LURKING VARIABLE is a variable that is not
  included in the study but may be effecting the
  results of the experiment.

17
Confounding The Solution

• The effects of confounding can be minimized by
  RANDOMIZATION.
• The effects of a lurking variable should be
  spread uniformly among randomized groups.

18
Basic Experimental Designs

• Completely Randomized Design
• Randomized Block Design