Problems with Experiments Flashcards
two problems that threaten the internal validity of an experiment
2
demand characteristics
investigator effects
demand characteristics
5
a participant guesses the nature of the experiment and what behaviour the experimenter is looking for or what they expect to find
they change their behaviour, not behaving as they usually would and ruin the validity of the results
some participants may want to be helpful and please the experimenter which results in them being overly cooperative and thus behaving artificially
others react in the opposite way and display the ‘screw you’ effect where they deliberately behave in a way that spoils an experiment — e.g. if they are given two memory tests at different times of the day, they may guess that the study concerns how time of day affects performance, this may lead them to try and perform the same on each test to show that time of day does not affect performance
many are unsure about what to do so actively look for clues as to how they should behave in that situations, these clues and demand characteristics which lead the participant to guess the nature of the experiment
example of demand characteristics
7
Orne et al (1964)
participants had to sit in a room on their own for 4 hours
one group of participants were asked at the beginning of the study to sign a form releasing the experimenter from responsibility if anything happened to them during the experiment
this group was also given a panic button to push if they felt overly stressed
the other group were given no information to arouse their expectations
the first group showed extreme signs of distress during isolation which can be explained in terms of expectations created by the situation, signing the form and having a panic button meant that they expected to feel distressed
this demonstrates how expectations and behavioural cues can guide participants behaviours — he said that “cues convey the experimental hypothesis to the participant and become determinants of their behaviour”
investigator effects / experimenter bias
4
any impact the researcher may have on the behaviour of the participants and therefore the results
e.g. experimenter bias (if they’re trying to prove a theory they believe in), appearance, personality, body language, verbal communication
direct effects = a consequence of the investigator interacting with the participant
indirect effects = a consequence of the investigator designing the study
investigator effects examples
5
may include cues from an investigator that encourage certain behaviours from participants, they might do this so that their expectations or hypothesis is fulfilled
investigators may unconsciously encourage participants — they may spend more time with one group of participants or behave more positively towards them
research has shown that male investigators tend to be more pleasant and friendly with female participants than with male participants
even the way an investigator asks a question may lead a participant to give the answer that the investigator wants — similar idea to leading questions investigated by Loftus
the investigator experimental design effect, the way an investigator designs a study can impact the results — investigator may operationalise the measurement of certain variables in a way that makes the desired outcome more likely or they may limit the duration of the study for the same reason
how can these problems be dealt with?
3
single blind design
double blind design
experimental realism
single blind design
2
the participants are unaware of the aims of the study and of which condition of the experiment they are receiving
prevents them from seeking cues about the aims and reacting to them, therefore this eliminates demand characteristics
double blind design
4
neither the experimenter or the participants know what condition they’re in, the research aims of the study or the hypothesis
experimenter is less likely to produce cues about what they expect because they are unaware of the aims of the study
eliminates potential bias, demand characteristics and investigator effects
e.g. used with placebos in drug trials
experimental realism
2
if an experimenter makes a task real and engaging enough, the participant will pay more attention to the task that the fact that they’re being observed
makes behaviour more natural and reflective of real life
participant variables
5
characteristics of individual participants that affect the results
act as extraneous variables if independent groups design is used, can be controlled in repeated measures and matched pairs
may include age, intelligence, motivation, experience, gender and so on
should not be controlled unless there is reason to believe that these characteristics are significant and relevant
EXAMPLE = Eagly (1978) suggested that women may be more conformist than men, therefore having more women than men in one condition of an experiment might affect the results and mask the effects of the IV
situational variables
2
features of a research situation that may influence participant behaviour and affect the results
EXAMPLE = order effects (participant performance may improve due to the order of the tasks, can involve the practice effect and the boredom effect), means that the results may be due to order of tasks rather than the IV
