Unit 3: Designing a Study

Stratified Random Sampling

We split the population into groups, then randomly select some members from each group to be in the sample.

• Example: high school principal wants to conduct a survey to collect the inions of students. He first splits the students into four groups based on their grades then select a sample of 50 students from each grade to be included in the survey

Cluster Sampling

We split a population into clusters, then randomly select some of the clusters and include all members from those clusters in the sample.

• Example: a company that gives whale-watching hours wants to survey its customers. Out of ten hours they give one da, they randomly select four tour and ask every customer about their experience (the 4 groups are the sample)

Random Assignment

Chance to assign a experimental units to treatments

Replication

Use enough experiment units in each group sot hat any differences in the effects of the treatment can be distinguished from chance differences between the groups

Randomized Block Design

Used primarily in the field of experimental design to control for variability among experimental units.

• Subjects with similar characteristics are designed into blocks and randomly assigned treatments to each block

  • Groups are called blocks

Match-Pair Design

Experimental design that is used when an experiment ONLY has TWO treatment conditions.

• The subjects in the experiment are grouped together in pairs based on some variables they “match” on, such as age/gender. Then within each pair, subjects are randomly assigned to different treatments

Completely Randomized Design

Most basic experimental design.

• Randomization, replication, and reduction of variance

• Randomization: assigning the subjects to the different groups in a random way

• Replication: ensuring there are multiple subjects in each group

• Reduction of variance: removing/accounting for systematic differences among subjects

Systematic random sample

Statistical technique where every nth number of a population is chosen for a sample after an initial random starting point