Threats to Validity
Confounds
extraneous variables that systematically vary or influence both the independent and dependent variables
- may have a different confound in your experimental and control groups.
- a major threat to internal validity
- where do confounds come from? the experimenter, participants (individual differences, demand characteristics and participant reactance), and the effect of time.
third variable problem
- the best way to account for systematic confounds is usually to manipulate the independent variable.
- pseudo/false experiments are most susceptible
- importance of control groups
- importance of isolating the variable of interest.
control groups
- you want to manipulate only the variable of interest between groups/conditions
- the challenge is to keep everything else constant
- almost impossible to isolate and remove all other variables
- need to try to keep all other variables outside of your control constant and matched in experimental and control group
Internal Validity Threats
from the Experimenter
- experimenter bias is a confound that undermines the strength of a causal claim.
- the bias of the experimenter may influence the way a dependent variable is scored.
- the experimenter may behave in a way that influences the participants and confounds the results of the experiment
- not always intentional
- previous knowledge and ideas can create tunnel vision in the experimenter
- double-blind procedures neither the experimenter nor the participant are aware of the conditions
- prevents both unintentional and intentional bias when interpreting data.
Internal Validity Threats
from the Participant
- the way a participant can influence the validity of the results.
- the individual differences that are systematic can interfere with the causal relationship you are investigating.
participant expectancy
- participants identify the purpose of the study > behave in a certain way as a result
individual differences
- differences in characteristics between all individual participants, including animals.
- when there is a systematic issue with individual differences there is a threat to internal validity.
- need to try to control for these differences, but they are impossible to completely remove.
click to edit
participant reactance bias
- when a person modifies their behaviour because they are participating in a research study.
- a feature of the study suggests what the purpose of the study is.
- this changes your ability to measure the behaviour of interest.
participant reactivity
- good subject role: tries to produce the response the researcher wants.
- negative subject role: tries to produce the response contrary to what the researcher wants.
- apprehensive subject role: overly concerned about their performance.
- faithful subject role: follows instructions and ignores suspicions
demand characteristics
- a feature of the study suggests what the purpose of the study is > the participant changes their behaviour in response.
- surveys and scales are prone to this > difficult to hide the purpose of the study in questionnaire research
solutions
- unobtrusive observation: hidden video camera
- indirect measures: you can't just ask people things, you have to go about it indirectly > waiting for them to notice.
- deception/confederates: dummy participants
Time-Related
Confounds
maturation: the effect of time on participants
- short term: mood, tiredness, hunger, boredom
- deal with this by: counterbalancing the order of tests, controlling for the time of day, designing experiments of a reasonable length, including breaks.
- long-term: age, education, wealth
- difficult to control, though only important for long longitudinal studies.
- random assignment and sampling helps reduce.
artifacts: something that is ever-present in all groups being tested and stays constant
- reduces external validity
- prevents you from generalising your results.
mere measurement effect: being aware that something is observing or measuring your behaviour may change the way you behave > hawthorne factory
-important for external validity as it undermines the ability to generalise our results to the wider population.
- similar to participant reactance, except that it affects all subjects in the experiment
history effects: the effect of a period of time may make an entire sample biased
- the data is influenced by the moment in time
- can't generalise these findings to a wider population or different contexts
selection bias: participants who have a biased interest in the topic of research or the outcome of the study volunteer
- important to use a random sample
- doesn't eliminate all problems but reduces the likelihood of systematic biases.
- compulsory poll- census
- reduces sampling bias
- produces results and data that are more widely acceptable
- less susceptible to biased groups
- still susceptible to demand characteristics
non-response bias: problem for experiments that involve voluntary sign-ups or surveys
- people do not response when they're not interested > lose a large sample of the population
- undermines external validity > limited population means that the results cannot be generalised to a wider population.