Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Problem
As an example, suppose we conduct a national
survey. We might divide the population into An auto analyst is conducting a satisfaction
groups or strata, based on geography - north, survey, sampling from a list of 10,000 new car
east, south, and west. Then, within each stratum, buyers. The list includes 2,500 Ford buyers, 2,500
we might randomly select survey respondents. GM buyers, 2,500 Honda buyers, and 2,500
Toyota buyers. The analyst selects a sample of
Cluster sampling. With cluster sampling, every
400 car buyers, by randomly sampling 100
member of the population is assigned to one,
buyers of each brand.
and only one, group. Each group is called a
cluster. A sample of clusters is chosen, using a Is this an example of a simple random sample?
probability method (often simple random
(A) Yes, because each buyer in the sample was
sampling). Only individuals within sampled
randomly sampled.
clusters are surveyed.
(B) Yes, because each buyer in the sample had
an equal chance of being sampled.
Note the difference between cluster sampling
(C) Yes, because car buyers of every brand were
and stratified sampling. With stratified sampling,
equally represented in the sample.
the sample includes elements from each
(D) No, because every possible 400-buyer
stratum. With cluster sampling, in contrast, the
sample did not have an equal chance of being
chosen.
(E) No, because the population consisted of
purchasers of four different brands of car.
Solution
Sample survey. A sample survey is a study that Observational studies do not feature random
obtains data from a subset of a population, in selection; so generalizing from the results of an
order to estimate population attributes. observational study to a larger population can be
a problem.
Experiment. An experiment is a controlled study
in which the researcher attempts to understand Causal inference. Cause-and-effect relationships
cause-and-effect relationships. The study is can be teased out when subjects are randomly
"controlled" in the sense that the researcher assigned to groups. Therefore, experiments,
controls (1) how subjects are assigned to groups which allow the researcher to control
and (2) which treatments each group receives. assignment of subjects to treatment groups, are
the best method for investigating causal
In the analysis phase, the researcher compares relationships.
group scores on some dependent variable. Test Your Understanding
Based on the analysis, the researcher draws a
conclusion about whether the treatment Problem
(independent variable) had a causal effect on the
Which of the following statements are true?
dependent variable.
I. A sample survey is a type of experiment.
Observational study. Like experiments,
II. An observational study requires fewer
observational studies attempt to understand
resources than an experiment.
cause-and-effect relationships. However, unlike
III. The best method for investigating causal
experiments, the researcher is not able to
relationships is an observational study.
control (1) how subjects are assigned to groups
and/or (2) which treatments each group (A) I only
receives. (B) II only
(C) III only
Data Collection Methods: Pros and Cons
(D) All of the above.
Each method of data collection has advantages (E) None of the above.
and disadvantages.