For each of the following scenarios, indicate (a) whether randomization is present, and (b) whether sufficient replication has been made.
An experiment is conducted to determine if a certain type of genetically engineered corn seed will outproduce traditional corn seed. Fifty rows of Farmer Brown's field are sown with the genetically engineered corn seed and 50 rows of Farmer Grey's field are sown with the traditional seed.
An experiment is conducted to see whether a new diabetes treatment will outperform a traditional treatment. Two mice each are randomly assigned to each of the treatment and control groups.
Solution
The 50-row sample is probably enough replication. However, randomization is lacking. Perhaps Farmer Brown is a better farmer than Farmer Grey, in which case, this lurking variable or confounding variable will confound the results.
Here, the randomization is fine, but two mice each is probably insufficient replication to uncover any strong statistical results.