When people test their hypotheses, they seek exploratory data from the
population they are making predictions on. There are several different ways to go about
collecting this data- by a census, sample survey, experiment, or by an observational
study. All four of these methods are used for different situations.
Census- a census attempts to contact every individual in the population.
Sample Survey—a sample survey is contacting a portion of the respective population to
Experiment- an experiment deliberately imposes some treatment on individuals of the
population in order to observe their responses.
Observational study- an observational study observes the individuals and measures
variables of interest, but unlike an experiment it doesn’t attempt to influence their
My hypothesis is concerning college students of private schools and the
relationship between the number of hours they study and their grades.
a) How would I conduct a census? A sample survey?
b) How would an experiment and observational study differ in this situation?
a) A census would be contacting every student who goes to a private college in
America. However, a sample survey would be randomly picking several
hundred students from different colleges throughout America.
b) An experiment would be selecting a certain individuals through a sample
survey and imposing a certain amount of study hours and recording their
grades. An observational study would be selecting individuals through a
sample survey, and categorizing their study hours and then record the
respective grades. (causation can only be inferred from an experiment, not an