Statistics – Observational Studies versus Designed Experiments (Section 1.2) – Flashcards

Observational Studies Designed Experiments

Changing the amount of an "explanatory variable" affects the value of a "response variable."

I punch you.

How bad will it hurt?

(An observational study measures the value of response variable without attempting to influence any of the response or explanatory variables of the individual). That is, in an observational study, the researcher observes the behavior of the individuals in the study without trying to influence the outcome of the study.

(In Designed Experiments, individuals are assigned to groups where explanatory variables are intentionally changed and the values of the response variables are rewarded). If a researcher assigns the individuals in a study to a certain group, intentionally changes the value of the explanatory variable (remember radiation to rats), and then records the value of the response variable for each group, the researcher is conducting a designed experiment. (Drug testing - Placebo vs. active drub).

(Confounding occurs in a study when you are unable to distinguish among the effects of two or more explanatory variables.) Therefore, any relation that may exist between the explanatory variable and lurking variable may be due to some other variable or variables not accounted for in the study. (Global warming, CO2 levels + Earth's axis. You just can't look at a glacier melting and say "yes, that's global warming."). Confounding is a big problem and is usually attributed to unaccounted or lurking variables.

A lurking variable is an explanatory variable that was not considered in a study, but that affects the value of the response variable in the study. In addition, lurking variables typically relate to explanatory variables considered in the study. (Remember the influenza study. The lurking variables might be age, health status of seniors).

Observational studies do not allow a researcher to claim causation, only association. You just can't say "smoking causes lung cancer."

Are used whenever control of certain variables is possible and desirable. *You wouldn't want to do a design experiment having people smoke cigs. That is why Observational Studies are necessary. They also have lower cost, greater timeliness, and broader range of patients.*

Cross-sectional studies Case-control studies Cohort studies

Observational studies that collect information about individuals at a specific point in time (all at once) or over a very short period of time.

These studies are retrospective, meaning that they require individuals to look back in time or require the researcher to look at existing records. In case-control studies, individuals that have a certain characteristic are matched with those that do now. A disadvantage is that it requires info from the past and responses to be honest. An advantage is that they are inexpensive to conduct and can be done fast.

A cohort study first identifies of a group of individuals to participate in the study (the cohort). The cohort is then observed over a period of time (sometimes a long period of time). Over this time period, characteristics about the individuals are recorded and some individuals in the study will be exposed to certain factors (not intentionally) and others will not. At the end of the study the value of the response variable is recorded for the individuals. *Remember cell phone example (cohort study took over 21 years to complete).

They require many individuals to participate over long periods of times. A disadvantage is that individuals tend to drop out due to the long time frame. They are the most powerful in Cohort studies.

(The collection of data from every member of a population is called a census). US decennial census.