Answer to Question 1
A cohort study involves the study of a specific, well-defined population over time. The same group of people, but not necessarily the same individuals, provide data so the researchers can track changes from one wave to the next.
A cohort sequential design involves studying the same people over time, but at each measurement, a new group is brought into the study that is the same age as the original group was in the first measurement. At each measurement, every group that had been included is retained, but a new group with the same characteristics of the initial group at the beginning of the study starts participation.
Answer to Question 2
When a person is measured on some behavior or attitude, an extreme score could occur because of some type of random measurement error. This type of error is random, so on a second testing, it is not likely to recur. The pattern of an extreme score caused by random error followed by a score nearer the average is called statistical regression, or regression to the mean.
Statistical regression is a problem when a researcher makes two measurements; one of them could involve considerable measurement error, giving a distorted picture of what is being measured. We can overcome this problem by using multiple measurements rather than just two. Any aberrant scores will stand out as unusual compared to the other, more normal measurements.
Sequential decline of whale catches in the Antarctic shows the strong effect of human predators.
The difference between AZT-sensitive versus AZT-resistant reverse transcriptases
Leonardo da Vinci, Study of a Woman's Head or of the Angel of the Vergine delle Rocce.
Difference spectra of mitochondria
These findings are from Colby and Kohlberg's long-term longitudinal study
Life Table for an Annual Plant Cohort