Anecdotal evidence: individual cases that are striking in some way and occur rarely that needs
not to be represented for any large group of cases.
Ex: friends are big fans the show “barney”, this cant be generalized to a larger population. This
is because any group of friends cannot be used as a representative for all college students
Available data: data produced in the past for some other purpose but may help answer a present
Ex: library and internet
-Statistical designs for producing data rely on either sampling or experiments
-The idea of sampling is to study a part (sample) in order to gain information about the
-Data are often produced by sampling a population of people or things.
-Census – attempt to contact every individual in the entire population
-Carefully conducted sample is better than a census
Observational vs experiement
Observational study: individuals are observed and variables of interest are measured but do not
attempt to influence the responses
Experimental study: give out treatments to individuals and observe responses
-The best way to see the affects of a change is to do an intervention where we actually
impose the change
-Intervention: where we impose the change
-Experiments are the only source when understanding cause and affect
-Confounding variable: observational study that examine the affect of a single variable of
an outcome can be misleading when the affects the explanatory variable are confounded
with those are other variables
Ex: observational study: child who attend daycare are rated by their parents as
disobedience, however this is not a cause-effect relationship b/c there can be lurking
variables influenced by their parents
-Experiments require randomization
-Statistical inference: answers specific questions with a known degree of confidence
-Ethics - is the questions asked in order to make an experiment
3.1 Design of Experiments
- Experimental units: individuals on which the experiment is done on.
- When units are humans = subjects
- treatment: specific experimental condition applied to the units
Purpose of experiment is to reveal the response of one variable to changes in other variables.