Experiments and Studies

Observational Study

a study based on data in which no manipulation of factors has been employed

What can't we do in an observational study?

Conclude causation because of outside variables

Experiment

A research method in which an investigator manipulates one or more factors to observe the effect on some behavior or mental process -find cause/effect relationship

Confounding Variables

factors that cause differences between the experimental group and the control group other than the independent variable

What are the three kay factors of an experiment

Control Group, Randomization, and Replication

Control Group

the group that does not receive the experimental treatment.

Randomization

a process of randomly assigning subjects to different treatment groups to promote equality and similar groups

Why is randomization important

It allows for homogenous groups and generalizations to be made to the population

Replication

repeating the essence of a research study, usually with different participants in different situations, to see whether the basic finding extends to other participants and circumstances

Experimental Units

the individuals on which the experiment is done

Explanatory Variable

What is the experiment (Independent Variable)

Response Variable

What is the response (Dependent Variable)

Treatments

What is being changed (Similar to Explanatory)

Why is it important to have a control group?

To have something to compare the results too.

Completely Randomized Design

the treatments are assigned to all the experimental units completely by chance

Example of Completely Randomized Design

Random Allocation -> 2 Groups -> 1 Treatment for each group -> Compare the results (be specific with everything)

Block Design

the random assignment of units to treatments is carried out separately within each block (There are pre-known factors/grouping)

Example of Block Design

Random Assignment

Group Men -> Random Assignment -> 2 Groups -> 1 Treatment per group - > Compare results with other Group

Group Women -> Random Assignment -> 2 Groups -> 1 Treatment per group - > Compare results with other Group

Matched Pair Design 1 (2 Treatments) Example

Random Allocation -> 2 Groups -> Both treatments in each group (4 total) -> Compare Results

Matched Pair Design (Paired) Example

Participants are paired before random assignment into groups (same format as MPD 1)

Does replication reduce variation?

Yes, a larger sample produces a more accurate estimate.

Does the control group reduce the effect of confounding variables?

Yes, changes due to variables can be observed through comparison

Can randomization reduce the effect of confounding variables and reduce variation?

Yes, homogenous groups.

Does Blocking reduce the effect of confounding variables and variation?

Yes, it improves the randomization process and produces more homogenous groups

Does a Matched Pair design reduce variation?

Yes, using the differences measured between treatments

What is the end goal of an experiment?

To best determine the effect of the treatment on the experimental units.