RM2- Sampling, Control Measures and Pilot Studies Flashcards
sampling
ways in which researchers select participants for their study
target population
specific group of people from the whole population that you want to study
sample
smaller group that you actually select from the target population to participate in the study
name the 5 different sampling techniques
volunteer
systematic
opportunity
random
stratified
volunteer sampling
participants put themselves forward- self select
systematic sampling
pattern, every nth term
a sampling frame is produced from potential participants
opportunity sampling
researcher directly asks people who are readily available to them to participate, convenience sampling
stratified sampling
random sampling in each strata can be time consuming
the composition of sample reflects proportions of sub groups in the target population
percentage proportion is calculated
random sampling
researcher obtains a list of the target population, participants are chosen by change to take part until enough have been chosen
evaluate random sampling
representative sample
-researcher has no control
-improves population validity
time consuming
-need to make a sampling frame/ list of target population
not guaranteed for a representative sample
-some groups may not be selected
-stratified is better
evaluate opportunity sampling
easy to obtain
-use convenient people
unrepresentative sample
-likely for ppts to have similar characteristics
-reducing population validity
ethical issues
-ppts may feel pressure to take part- abusive power
evaluate volunteer sampling
saves time
-can advertise for a specific/ niche group of people
lacks generalisability
-self selected people
-limiting population validity
not representative
-self selected people
-limits population validity- people from a similiar background may see the same advert
evaluate systematic sampling
avoids researcher bias
-no influence over who is chosen
-more representative
-increases validity
not very representative
-some groups may not be selected
-stratified is better
evaluate stratified sampling
avoids researcher bias
-no influence over who is subdivided into which strata
-accurate reflection of target population
-higher representative sample
time consuming
-needs all details of all ppts in target population
not completely representative
-strata may not reflect all sub groups that exist
-complete presentation is not always possible
control measures
strategies that researchers put in place in order to minimise the impact of extraneous variables
name the 4 types of extraneous variables
participant variables
situational variables
demand characteristics
experimenter effects
why is putting controls in place good for researchers?
-they can establish a greater cause and effect
-more confident that their manipulation of the IV has caused an effect on the DV
-results have higher internal validity
give 4 control measures
counterbalancing
standardisation
randomisation
random allocation
what is counterbalancing?
where the order of the conditions are mixed up
-reduces the order effects (practice and fatigue)
-one group completes A then B, other group completes B then A
what is randomisation?
the organisation of materials random, they are presented to participants in a random order
-avoids systematic errors
-reduces bias as the researcher has no control over the order of items
what is standardisation?
the process in which procedures used in research are kept the same (standard procedures).
-e.g. instructions are delivered by a researcher using a script or reading of a piece of paper or recording to establish standardisation
what is random allocation?
once a complete list of ppts is generated, they are divided by chance into different experimental conditions
-list is worked through using a system until all ppts are place into conditions of equal amounts
control measures for participant variables
-use a large sample to increase representation
-use random allocation
-use repeated measures or matched pairs design
control measures for situational variables
-use a script or written instructions to help standardise procedure
-ensure all ppts are treated in the same way
control measures for experimenter effects
-use a double blind technique
-use random allocation- chosen by chance, no bias or influence
-provide written instructions, no investigator present
control measures for demand characteristics
-use deception (does have ethical issues)
-use an independent groups design
-use distractor questions in self report methods
-use a double blind technique
pilot studies
small, trial versions of proposed studies to test their effectiveness and make improvements.
why are pilot studies carried out?
carried out as a trial run
-assess the feasibility of the research
-helps to improve the validity of the research being conducted
where can problems be identified in pilot studies?
-method/ design chosen
-instructions
-procedures
-materials used
-measurements to be taken
floor effect
task is too hard, participants cannot score at all or they cannot complete the task
-performance is very low
-this assesses the difficulty and therefore researchers can adjust the task
ceiling effect
task is too easy, all ppts virtually achieve full marks or top performances
who are pilot studies carried out on?
target population that are not in the sample/ the actual ppts
what type of credibility to pilot studies add to research?
scientific credibility
why are questionnaire pilot studies important to conduct?
-checks that respondents understand terms in questions
-checks that emotive questions is not used which could invalidate answers
-ensure there are no leading questions- cause bias answers
-ensure it takes a reasonable amount of time, long times can affect response rate and data quality as people lose interested or not have enough time to complete it.
