Expirament Design Flashcards
(20 cards)
experiment design
a technique to efficiently assess the effect of multiple inputs, or factors, on measures of performance, or responses
Probability Sampling
Every individual has a known non-zero chance of being selected
Convenience Sampling
Picked based on ease of access
Randomization
Assigning participants to experimental or control group purely by chance
Matching
Pairing or grouping participants based on shared characteristics before assigning them to groups
Internal Invalidity
A lack of confidence that the observed effect is truly cause by the independent variable
Did the treatment really cause the effect?
External Invalidity
A lack of confidence that the results can be generalized beyond the study sample or setting
Can findings be applied to other people/settings/times?
Experimental Studies
a study in which a treatment, procedure, or program is intentionally introduced and a result or outcome is observed.
One shot case study
There is no pretest
identifies the extent and nature of cause-and-effect relationships between two or more variables
Classical Experiment
requires a control group
uses random assignment, an experimental and control group, as well as pre- and posttesting
Static group testing
- no randomization/treatment
- a software testing method that examines a program – along with any associated documents – but does not require the program to be executed
Natural Experiments
an observational study where researchers observe naturally occurring events or changes, rather than actively manipulating them, to study the impact of a treatment or intervention
One group pretest-postest
a research design where a single group of participants is assessed before and after an intervention, with no control group for comparison
Strength to Experiment
- causal clarity
- can be repeated
- reduces bias
- controlled environment
Weakness to Experiment
artificiality
- not reflective of the real world
The classical experiment design helps guard against
the sources of internal invalidity
An experiment examines the effect of the ____ variable on the _____ variable.
independent
dependent
one group pretest-postest
- survey before the treatment
- admins the treatment
-survey after the treatment
In experiment research is to find causal relationships
- the representatives of a sample is less important