1D - Preventing Error And bias Flashcards
Extraneous Variables
It is a variable other than the IV that has potentially had unwanted effect upon the dependent variable thus possibly affecting results.
Controlled Variables
It is held constant to ensure that the only influence on the dependent variable is the independent variable.
Confounding Variables
It is a variable other than the IV that has definitely influenced the dependent variable, thus the true effects of the IV are unknown, It may be an uncontrolled extraneous variable or a variable that simply cannot be controlled.
Placebo Effect
It is the expectation of the participant that they are receiving the experimental treatment, and they respond according to that belief.
How is Placebo Effect controlled?
Control of this expectation is to give the control group a fake treatment (placebo) so that subjects are unaware of which group they are in.
Experimenter Effect
It is when the expectations of the researcher affect the results of an experiment.
How are Experimental Effects controlled?
Control usually involves the double-blind procedure where the experimenter does not know the group (experimental or control) that the participant is in. It can also use the triple-blind procedure where research assistants are also unaware which group participant is in.
Situational Variables
They are the external factors associated with the experimental setting that may influence responses. E.g. temperature
How are Situational Variables controlled?
Control of these involves making sure that the situation is as constant as possible. E.g Keep temperature same for both groups.
Non-standardised instructions and procedures
They are when the directions and procedures differ across participants or experimental conditions.
How are Non-standardised instructions and procedures controlled?
Control of these involves making sure that there is a script and specific instructions to follow for all experimenters and participants in the research study.
Demand Characteristics
They are cues in the experiment that can lead or bias a participant to giving a particular response.
How are Demand Characteristics controlled?
Control measures include deception, where the experimenter hides the experiment’s true purpose, single-blind procedures, where participants don’t know their group, and double-blind procedures, where both participants and experimenters are unaware of group assignments.
Sampling size and procedures
A larger sample improves population representation, reduces bias from outliers, and ensures fairness through random or stratified sampling.
Counter balancing
Counterbalancing helps control order effects by changing the order of treatments for different groups. Half of the participants do one condition first, while the other half do it second. This is useful in within-subjects designs to reduce bias.
Placebo
A placebo is an inactive treatment (like a sugar pill) used in drug trials to compare results with the actual drug. It helps determine the drug’s effectiveness. It is different from the placebo effect (response), which is when participants experience changes simply because they believe they received treatment.
Single-blind procedures
They are where the participant is unaware of the experimental condition they are allocated.
Double-blind procedures
They are where neither the experimenter nor participant is aware of the experimental condition the participant is allocated.
Standardised instructions, procedures and controlled variables
All participants must get the same instructions and follow the same procedures to reduce situational differences. Experimenters must keep certain variables constant to accurately measure the effect of the independent variable (IV).
