Topic 2 DV Flashcards
What two scientific goals make up Correlation research?
Description and Prediction
What two scientific goals make up Experimental research?
Explanation and Control
Define causation
One factor DIRECTLY affects another factor
To show causation what must be demonstrated?
- Changing the 1st thing produces change in the 2nd
2. There is no other cause for the change
Define a population
Members of a specific group
What is a study population defined by?
Defined by the purposes of the experiment
Define a sample
Small subset of the population that represents the population
How is a representative sample achieved?
Through random sampling where each member has equal chance of selection
What is descriptive statistics?
Summarises the data collected from the sample
What is inferential statisitcs?
Generalise from the sample to the population
What two things must be considered when you can’t directly measure topic of interest?
- Property of interest - what you are trying to measure e.g. intelligence
- Dependent variable - a measurable value that must indirectly reflect the property of interest e.g. score on IQ test
(the indirect measure is the operational definition of your property of interest)
What is validity?
A dependent variable is valid if it measures what it’s supposed to
What is a dependent variable?
A measurement taken
What is recorded
Depends on what the participant does
What is a threat of validity?
Confounding
Poor Operational definition -> invalid dependent variable
What is reliability?
A dependent variable is reliable if under the same conditions the same results are found and contains the minimum of measurement error
True or false
If a measure lacks reliability it will also lack validity?
True
What is bias?
A biased dependent variable is consistently inaccurate in one direction e.g. always high, always low
What is the ceiling effect?
When a task is too easy that all the results are very high
What is the floor effect?
When a task is too hard that all the results are very low
Describe a nominal scale
- Categorises data WITHOUT ORDERING
- Numbers substitute for names e.g. gender, 1=F 2=M
Describe an ordinal scale
- Categorises data and ORDERS
- Bigger means more
- distance between the points is NOT CONSIDERED EQUAL
e. g. placing 1st 2nd 3rd
Describe an interval scale
- Categorises, orders, and establishes an equal unit of measurement in the scale
- We KNOW VALUE BETWEEN POINTS on the scale
- Distance between points on the scale considered equal
e. g. degrees celsius
Describe a ratio scale
- Categorises, orders, and establishes equal unit in the scale
- CONTAINS A TRUE ZERO POINT
- Allows ratio statements e.g. ‘twice as big’
