Quasi Experimental, Correlational and Descriptive Designs Flashcards
What are the general issues of non-experimental designs?
Third variable issue (hard to anticipate what the effect of the third variable may be, also hard to establish is variable A is caused by variable B)
Varying reliability of measures (have to be sure the measures are reliable, confuse low reliability and a lack of relationship between 2 variables.
What are non-manipulation designs?
No randomisation
No manipulation, more realistic
When can non-manipulated designs be used?
Manipulation isn’t possible
Naturalistic setting
Establish an association
Temporal change
Developing/refining measures
Explanatory models
Natural variation
What are quasi-experimental designs?
No randomisation
cause and effect relationship
Uses DV and Quasi-independent variables
Quasi-independent variable: variable that naturally varies and isn’t directly manipulated by the researcher
What are the advantages/disadvantages of a quasi-experimental design?
+ determine cause and effect relationships and generalisation of findings to the real world
- Lack of precise control over natural events, influence of extraneous variables
What is a correlational design?
Indicates strength of relationship amongst variables
Relationships aren’t always causal
No manipulation or randomisation
What are the strengths and weaknesses of a correlational design?
+ estimates strength and direction of relationship amongst variables
- doesn’t permit the determination of cause and effect relationships, no causation
What are lab experiments?
Manipulation of IV
Measures DV
What are the strengths and weaknesses of a lab experiment?
+ determine cause and effect relationship
- data obtained in an artificial lab
What are cross sectional designs?
Collect information in the presence of one or more variables, as they exist in a defined population in one point of time
Study people who differ in age very often
Data analysed to determine the distribution of one or more variables
What are the strengths and weaknesses of a cross sectional design?
+ age differences, not expensive, little time to conduct
- extraneous differences between cohorts rather than true changes
What is a cohort?
group of people of the same age who are exposed to similar cultural environments and historical events as they are growing up
