9.3.2 Practical Issues In The Design And Implementation Of Research Flashcards
What is the IV?
The one thing that you change.
What is the DV?
The change that happens as a result of the IV.
What is operationalising variables?
How you will define and measure a specific variable.
What are participant variables?
Related to individual characteristics of each ppt that may impact how they respond. E.g. mood, intelligence
What are situational variables?
Factors in environment that can unintentionally affect the results of a study. E.g. temperature, noise.
What is experimenter bias?
Where the experimenter influences the results in order to portray a particular outcome.
How can demand characteristics create bias?
Clues given to the ppt about tehe point of the experiment.
How can interpretation of data create bias?
The results may be interpreted wrongly; for example if subjective data, such as qualitative, is gathered.
How can inter-rater reliability create bias?
Can’t decide what the results are, so can’t make a conclusion
How can triangulation create bias?
Information and details can be lost as it is passed from one person to another.
How can social desirability create bias?
Ppts lie, which ruins the results so they look better to others.
How can interviewer bias create bias?
The results are bended to fit what the interviewer expected to see.
How can experimenter bias create bias?
Can’t reach a conclusion.
Sherif generalisability
Low because small sample; 24 boys, all aged 11.
Sherif validity
Low ecological validity, ppts spent time away from home with a group of boys they have never met before, which is unnatural and may have created unnatural results.
Sherif subjectivity/ objectivity
A lot of the results was observation, which is subective.
Raine Generalisability
High generalisablility, a variety of ppts were used; mix of men and women, mix of mental health disorders and non mental health disorders.
Raine validity
A CPT was used which could have been considered artificial; the ppts were completing an unnatural task. low ecological validity.
Raine reliability
PET scans were used which are reliable because they can be replicated to test for consistency.
Raine Subjectivity/ objectivity
Brain scans are an objective form of quan data.
Raine ethics
The use of brain scans is unethical because it can cause physical harm, such as radiation.
Watson and Raynor generalisability
Low generalisability because only one ppt was used.
Watson and Raynor validity
High ecological validity because he was too young to be affected by the use of an artificial environment.
Watson and Raynor subjectivity/ objectivity
Albert was observed, which produced qual data. This is subjective because they had to interpret the behaviour of Albert.
Baddeley generalisability
High generalisability because men and women were used. However, they all came from Cambridge uni which means the results can’t be generalised to people who are of a different intelligence level.
Baddeley validity
Low ecological validity. Recalling word lists is an unnatural task, that could result in unnatural responses.
Baddeley reliability
A standardised procedure was used, which means it is reliability because it can be repeated to check for reliability and credibiltiy.
Baddeley subjectivity/ objectivity
quan data was gathered, which is objective, making the results more reliable.
Rosenhan generalisability
A variety of hospitals across a variety of different states in America, meaning results can be generalised to a wide population.
Rosenhan Validity
High ecological validity because to the ppts, a person coming into the hospital complaining of hearing voices is a normal situation, which means normal results/ responses.
Rosenhan Reliability
There may have been observer bias because the pseudos may have not recorded what they considered was important, and information may have been missed out.
Rosenhan subjectivity/ objectivity
Qual data was gathered which makes it subjective because the pseudos could have exaggerated/ assumed something as more extreme than it actually was.
Rosenhan ethics
The ppts didn’t know that they were involved in a study, which is deception. They were also given no right to withdraw. This creates practical issues for the research.
Loftus and Palmer generalisability
In experiment 2, 150 students were used. This is high in generalisability because a large sample was used, however they were all the same age range, which decreases the generalisability.
Loftus and Palmer validity
Lacked mundane realism and ecological validity because ppts watched a video rather than an actual car crash.
Loftus and Palmer subjectivity/ objectivity
The data collected was quan which means it is objective.
Why does objectivity/ subjectivity matter?
Objectivity ensures that bias by experimenter or interpretation doesn’t affect results-scientific. Too narrow as objectivity explains what is seen, but doesn’t go into depth to explain why.
Why the type of data used matter?
Having an objective type of data makes the results more credible than a subjective type of data.
Why does data collection matter?
Know how credible or reliable the data is
Why does internal validity matter?
It provides levels of accuracy which increased credibility. Increased accuracy means more effective research for society-increases usefulness of applications.
Why does external validity matter?
Should be high so that results can be representations of real life situations. Leads to useful contributions to society. Eg Raine use of info in court cases
Why do standardisation and replication matter?
Easily replicated to test for consistency, check reliability of results. Low mundane realism too standardised, unrealistic which leads to unnatural response
Why does reliability over time matter?
To prove consistency over time and relevancy for society. It is still applicable. Studies are done in the past could not be easily replicated now (w and r)
Why does scientific status matter?
Can lead to practical issues because scientific measures can not always be used. High control over extraneous variables as environment is controlled.
Why does generalisability matter?
High generalisability means the results can be generalised too a wider population. Low generalisability means the results can’t be representative of a wider population which means a lower application.
