Ch. 5: Experiments and Observational Studies

Disasters in Experiments

-Confounding variables -Interacting variables -Placebo, Hawthorne, and Experimenter effects -Ecological validity

Disasters in Observational Studies

-Confounding variables and the implications of causation -Extending the results inappropriately -Using the past as a source of data

Reasons for Observational Study

-Unethical or impossible to assign people to a particular treatment ex. drug use, disease, education -Some traits are inherent and cannot be randomized ex. sex, age, race

Advantages of Case Control Observational Studies

-efficiency -reducing potential confounding variables

Case Control Observational Study

A group that does not have the particular characteristic or attribute as the group of interest being studied

Control Group

A group that does not receive the treatment

Which of the following defines what is meant by a control group in an experiment? -a group that is used by researchers to monitor how the experiment is going -a group that is handled identically to the treatment group(s) in all respects except that they don't receive the active treatment -a group that is handled identically to the treatment group(s) in all respects except that they are controlled to a greater extent than the other groups, providing baseline data -none of the above

A group that is handled identically to the treatment group(s) in all respects except that they don't receive the active treatment

Which of the following influences on the experimental units can bias the results of an experiment? -placebo effect -Hawthorne effect -not being blinded -all of the above

All of the above

Which of the following is a potential complication of observational studies? -confounding variables -improper extension of the results -the improper conclusion of causation -all of the above

All of the above

Randomized Experiment

Create differences in the explanatory variable and examine the results in the response variable

What does randomization mean in terms of experiments? -The researchers randomly select which treatments they will include in the experiment. -Each of the experimental units is randomly selected to participate. -Each of the experimental units is randomly assigned to a treatment. -none of the above

Each of the experimental units is randomly assigned to a treatment

Blinding

Either the researcher and/or participants do not know which treatment they are receiving, intended to eliminate bias

Which of the following is not a type of observational study? -introspective study -case-control study -prospective study -retrospective study

Introspective study

Placebo

Looks like the real medication but has no active ingredients

Repeated Measures

Multiple measures on the same participants ex. before and after

Double Blind

Neither the researcher nor the participants know which treatment the participant is receiving

Matched Pairs

Not the exact same participants but each participant in one group is paired with one participant in the other group ex. twins, couples, medical (sex, weight, height)

Confounding variables are a bigger problem in an (blank) study

Observational

(Blank) can only measure and analyze if confounding variables are related to response variable

Observational studies

Observational Studies

Observe differences in the explanatory variable and notice whether these are related to differences in the response variable

Confounding Variables

Occurs when it is difficult to tell if the difference in the outcomes is due to a treatment (or specific variable) or due to some other unknown variable (confounder) -related to the explanatory variable -affects the response variable

Interaction

Occurs when the relationship of one of them to the response variable depends on the other one

Retrospective Observational Study

Participants are asked to recall past events, reviewing old records

Prospective Observational Study

Participants are followed into the future and events recorded

Hawthorne Effect

People behave differently just because they are part of an experiment

Placebo Effect

People get better because they think they are on medication

Experiments use (blank) to reduce the effects of confounding variables and other sources of bias that are naturally present in observational studies.

Randomization

(Blank) can control for some confounding variables

Randomized experiment

An observational study that asks participants to recall their first childhood memory is a (blank) study

Retrospective

Effect Modifier

Subgroup variable that modifies the effect of the explanatory variable on the outcome

Ecological Validity

The methods, materials, and the setting of the study must approximate the real world that is being examined

Single Blind

The researcher know which treatment the participant is receiving but the participant does not know

Experimenter Effect

The researcher's bias impacts the study

Block Design

Treatment randomly assigned within each block (group)

Interacting Variables

Two explanatory variables (X) the result in different relationships with the response variable

Random assignment deals with what aspect of an experiment? -who the results can be applied to -whether or not a cause and effect relationship exists -what the actual statistical results are -none of the above

Whether or not a cause and effect relationship exists

Random sampling deals with what aspect of a study? -whether or not a cause and effect relationships exists -who the results can be applied to -what research question is being asked -what the actual statistical results are

Who the results can be applied to