Observational Design
a research technique where you observe participants and phenomena in their most natural settings
Experimental Design
the process of carrying out research in an objective and controlled fashion so that precision is maximized and specific conclusions can be drawn regarding a hypothesis statement.
Why do you only change one variable?
To make an inference about one factor influencing the other, you want to change only the hypothesized factor between the groups you are comparing and hold all other factors constant
Independent/treatment variables
one or more likely casual variables that you can manipulate-> varied in systematic way across experimental units in experimental group
experimental group
a group that has been subjected to a hypothesized casual factor
control group
a group that has the baseline condition, providing a comparison to experimental group to asses the effect of treatment
Potentially confounding variables
other variables that are held constant between groups; factors that could interfere with the experiments outcome
Dependent, or outcome, variables
something that depends on other factors.
why do we want conclusions to be generalizable
experiments help us uncover potential cause and effect relationships underlying a general pattern
Scope of Inference
can we say something about the world in general and not just a specific circumstance
Replicates (replicate control/experimental treatments)
- achieved by applying treatment to multiple experimental units: ex. individual student, Petri dish
- what inference do you want to draw → individual, population, or community
- allows you to reduce uncertainty + improve generality of results
increase scope of inference → repeat the experiment (why?)
- natural variation = always among experimental unit
- increase confidence in results → assign experimental units randomly to groups if possible → won’t differ by something that could confound results
- using replicates increases the chance that differences you observe among groups/treatments = due to experimental manipulation and not some other confounding factor
Hypothesis
proposes an explanation for a pattern → ,ust make clear the evidence that would support/disprove the explanation
falsable
should be able to show the hypothesis = wrong through testing
