Experimental Methods [Definitions] Flashcards
(30 cards)
Describe a Lab experiment
- Controlled and artificial conditions
- Researcher manipulates IV and measures DV
- Contains both experimental and control conditions
- Random allocation
Describe a field experiment
- Natural environment
- IV is manipulated, DV is measures
- Participants are usually less aware they’re being studied
Describe a Quasi experiment
- IV is a naturally occurring difference between people
- Used when researcher is interested in a difference that already exists
- DV measured in a lab
- Not a ‘true’ experiment because there’s no random allocation of deliberate manipulation of IV
Describe a natural experiment
- No deliberate IV manipulation, take advantage of a ‘naturally occuring’ IV
- Used when impractical to manipulate the tested variable
- No random allocation
- DV is tested in a lab, field or online
Describe the location of a Lab
- Most scientific location
- Artificial environment ‘set up’ to allow maximum control of variables
- Access to specialist equipment
- Common place for tests
Describe the location of a field
- Natural setting
- May or May not be new to participants
- Research methods mostly carried out in the field
Describe the location of online
- Internet as a tool to conduct research
- E.g surveys and experiments
- Social media used to find samples/ conduct research
- Mass/global reach
What is opportunity sampling?
- A sample of people who are avaliable at the time of the study being carried out
- E.g using students in the library at 11am.
What is self-selected sampling?
- People who put themselves forward for the study
- e.g responded to an advert on a newspaper, social media, etc.
What is random sampling?
- Every member of the target population placed into a hat/random generator
- Everyone has an equal chance of being chosen
- Identifying everyone in target population
What is systematic sampling?
- Target population/sampling frame needed
- Every nth person is selected from the list: participants are selected at fixed intervals
What is stratified sampling?
- Population is classified into categories (strata)
- Sample should represent the demographics of the population.
- Random selection st the end.
What is quota sampling?
- Classification into strata, in a representative format
- However, researcher makes final choice of who is selected at their convenience
What is snowball sampling?
- Used when targer population is not easy to access.
- Researcher uses personal contacts, asks them to find contacts of their own.
- Extends the chain of participants
Difference between each hypothesis
DIRECTIONAL: Increase, decrease, more, less, higher lower. ‘Significant increase in happiness in college students with 3 or more pets than college students with less than 3 pets.’
NON-DIRECTIONAL: Difference, amount. ‘Significant difference in happiness between college students with 3 pets or more, compared to students eith less than 3 pets.’
NULL: ‘There will be no Significant difference between the amount of happiness a college student has and the amount of pets they own. Any differences will be due to chance factors.’
Issues of Validity
- Unrepresentative sample [E]
- Artificial [I]
- Researchers and other factors influence participants response [I]
- Ppts trying to work out the aim of the study and acting accordingly [I]
- Ppts worried about the impression they give to others so change their behaviour [I]
Dealing with problems of validity
Internal Validity [If test measures what it intended]
- Standardisation to remove extraneous variables
- Using a single or double blind design
External Validity [Whether the findings can be generalised]
- Wide variation of participants in the sample
- Study represents everyday life [both task AND environment]
What is Face Validity? [Ways of assessing internal validity]
- If the method used seems to measure what was intended
- Ask a lay person/non- expert if the study appears to measure what it intends to
What is Content validity?
- Does the method actually measure what it intended to?
- Ask a panel of experts if the study actually measures what you intended
What is construct validity?
- Is the method actually measuring all parts of what we are aiming to test?
- Make sure all parts of the test are defined (e.g what is aggression)
What is concurrent validity?
- How well does the measure agree with already existing measures?
- Test the ppts with an already established test of known validity
- If the test and the example have close agreement, it has concurrent validity
What is predictive validity?
- Is oir measure linked with a future behaviour?
- Following up our participants to see if future performance is similar to the performance on our measure.
Issues of reliability
- Only one researcher
- Research has not been replicated
- Instructions mot given in the same way
- ppts do not have the same experience as no Standardisation
- Items in the questionnaire are not answered in the same way
Dealing with the issues of reliability
- Use more than one researcher
- Repeat the study on another occasion using the same participants
- Standardise the procedures so all participants have the same experience
- Standardise questions and possible answers
