12.5: Randomized Experiments
The randomization process involves assigning study participants randomly to experimental or control groups based on their probability of being equally assigned. Randomization is meant to eliminate selection bias and balance known and unknown confounding factors so that the control group is similar to the treatment group as much as possible. A computer program and a random number generator can be used to assign participants to groups in a way that minimizes bias.
Simple randomization refers to random assignments based on a single sequence. The selection of a person to a group remains completely random. Flipping a coin is the most common and primary method of simple randomization.
It consists of randomly assigning participants to different groups to produce an equal sample of participants. This method ensures that sample sizes are balanced across groups over time. Each block is small and balanced, with predetermined group assignments, so the number of participants in each group is always similar.
In stratified randomization, covariates are controlled and balanced. This method allows participants' baseline characteristics (covariates) to be balanced between groups. To estimate the impact of a covariate on a dependent variable, the researcher needs to identify specific covariates. Participants are assigned to the appropriate block of covariates based on stratified randomization, which creates separate blocks for each combination of covariates. After all, participants have been identified and allocated into blocks, and simple randomization is used to assign them to one of the groups within each block.