Evaluation of biological studies Flashcards
TEACUP
T - testable (Can you research/test it)
E - evidence (Is there any research evidence)
A - applications (Whether it can be applied to the real work? Can it be useful in any way? Any real life uses?
C - construct validity (Can be define and measure the aspects of the theory?)
U - unbiased (Any bias? Is it somehow prejudice or is it universal?)
P - Predictive validity (Does it allow us to predict behaviour? if yes then its strong)
Maguire
- Quasi experiment - naturally occuring variable could not control
- Brain scans were coded so the analysis could be done blindly - avoid researcher bias
- Ethics: MRI does not pose any health risks to the participants and all gave consent
- Quasi-experiment = no cause and effect
- Only male london - difficult to generalise (and it is true that the vase majority of london cabbies are male)
Milner [strength]
S:
* The strength of this study is that it was longitudinal - over 50 years = This means that change could be observed over time.
* Useful in research and understanding of the hippocampus and memory, used to confirm other theories and models. (LoF, MSM)
* Case studies use method triangulation. (Valid and reliable data)
* High ecological validity, no variables were manipulated and HM was observed in his natural environment.
* Milner’s research met high ethical standards of consent, confidentiality and protection from harm.
* Can be used to predict effects of a damaged hippocampus.
L:
* Cannot be easily replicated. However, there are several other case studies of patients like HM - ex. Clive Wearing - which confirm the findings.
* Not generalisable
* Some of the study was retrospective in nature. Means that we do not have a lot of data on HM’s actual cognitive abilities before the accident.
Caspi
- Large sample
- Holistic approach as they examined into both biological and environmental factors in depression
- Natural experiment, with the naturally occurring IV being the length of the alleles.
- Longitudinal research design, research the course of life events and the development of depression over the course of 23 years.
- Used method triangulation (interviews, observations, rating scales and even DNA analysis)
- Gather info from participants, clinicians, their near ones and observers.
- High external validity from investigating real life experiences
- High generalisability of results due to large sample
- study makes the assumption that serotonin causes depression
- information about life-events was self reported. It may be the salience of the negative life events which plays a role in depression (those who are more resilient, may not recall negative life events as easily)
- Gene action is highly complex, and actions of other genes could not be controlled.
- While the stressful life events were standardised as employment, financial, housing, health and relationship, whether or not a participant experienced a certain event as stressful is highly personal.
- lack of ethnic diversity
- Later studies did not show same result - it appears to have low reliability
- There were some participants who did not carry the gene mutation who became depressed; therefore, we cannot say that gene expression along can cause depression
McGuffin
- Interviews are blind to whether the twins were identical or non-identical and also received no information on their mental conditions = No researcher bias
- Triangulation = data from hospital register, family doctors, interviews
- This sample was taken from the Maudsley hospital (London) register
- Lack generalisability
- The research took place nearly ten years after the last patient was added to the hospital register. This means that many of the interviews were retrospective where people were asked to remember back as many as ten or fifteen years (or sometimes even more) and give details of depressive episodes. This has to bring into question the reliability of such memories.
- The sample used was of people who had been depressed enough to have sought medical treatment at a hospital and may not be representative of the general
Yu
- Self report data was supported by haemodynamic changes in the prefrontal cortex using spectroscopy.
- There is no researcher bias as they couldn’t have altered the results as it was mainly dependent on the participants and machines.
- Well controlled
- Good quantitative data
- However, demand characteristics may be present as the novice volunteers don’t have experience and couldve easily believed they felt better because they got an idea what the study was and what is hypothesised to happen when they had to complete a pre and post mood questionaire.
- Small sample size of the sample, it may not be generalised to a large population.
- Lack population and ecological validity
- Infection risk from taking blood sample
Luby
- longitudinal design allow for measuring both IV and DV over a long time period, stronger results
- MRI shows genetic backing
- Data triangulation = high internal validity
- Other studies show strong negative correlation with PTSD and hippocampal size too
- fMRI scanning can be stressful to preschoolers (participants)
- Correlational not causation
- The study sample, because of the larger sample from which it was taken, contained many pre-schoolers with symptoms of depression, limiting generalizability (Cannot apply to a general population)
- The relationships in the mediation model may be bidirectional, meaning that it could be the active response of the child to the caregiver that is showing in the brain changes. This reduces the internal validity of the study.
Sack
S:
* Cause and effect relationship established
* Double blind - no researcher bias
* Placebo - reduce chances of demand characteristics
L:
* potential side effects of melatonin
* Maybe built reliance on it
Schochat
- Had lab data to support
- Lab experiment = highly controlled
- Small sample size = not generalisable
- correlational not causation
- Order effects (fatigue) - participants feel tired because they aren’t doing anything engaging or they are bored.
Wedekind
- Successfully replicated by Jacob et al (2000) and Yamazaki et al show this to be the case for mice as well that male preferred female mice with different MHC
- Double blind experiment - minimize demand characteristics
- Met ethical standards - consent received, all debriefed
- Too reductionist - over simplifies the behaviour of human mate selection by bring it down to the MHC, ignoring cognitive and socio-cultural factors
- Not a representative sample - participants were similar in age and culture
Zhou
- Well controlled lab experiment, in which participants were unaware of the nature of the olfactory stimuli and therefore unable to exhibit any demand characteristics.
- study was counter-balanced to control for order effects such as practice effect
- The reduced effect of the EST on bisexual and homosexual women remains unexplained
- Shift in gender perception of the participants averaged 8%, which is a small change
- Results lack ecological validity because amount of AND and EST were far above than what would be found in the human body naturally. Therefore we cannot know if the smaller amounts that exist naturally would produce the same effect. Not a naturalistic setting.
- Hare et al (2017) fail to replicate. Means result may not be reliable. Researchers found the exposure to putative pheromones had no significant effect of gender perception or attractiveness.
- Not a clear study of sexual attraction but rather if participants received a persons walk as feminine or masculine. Debatable if this is a reliable measure of sexual behaviour.
- Relatively small sample, need larger to determine the reliability of results.
Ronay Von Hippel
S:
* well controlled (attractiveness of female research, time of day, heart rate)
* conducted under highly naturalistic conditions, leading to high ecological validity
L:
* repeated measured design. Order effects (fatigue) could have influenced the findings
* tricks performed cannot be standardized for “difficulty” This questions the validity of the measure of “aborted tries” as a sign of risk-taking
* This could potentially be an example of intrasexual selection, but it is difficult to establish that reproduction is actually the goal of the behaviour
* Ethics: Although the participants agreed to be in the study, they were not aware that they were being observed for “mating behaviour.” This could be seen as embarrassing when they are debriefed and told the real aim of the experiment. They would, of course, have the right to withdraw their data.
Neisser and Harsch
- Case study
- Londitudinal and prospective
- method triangulation - questionnaires and. interviews
- High ecological validity (did not maniupate any variables and the study was not done under highly controlled conditions)
- Natrualistic = good ecological validity
- There are other studies of different events (9/11) - demonstrates transferability of the findings of this study to other situations
- cannot be replicated
- there was participant attrition (participants who dropped out of the study over time)
- Naturalistic = difficult to eliminate the role of confounding variables
- No control over the participants behaviour between the first questionnaire and the secon (no idea how often this memory was discussed or how often the participants were exposed to media about the event
- Demand characteristics may have caused confidence levels to be higher when they were asked to verify their level of confidence - increased ratings to please the researcher or avoid social disapproval for claiming not remember an important day in their country’s history
