IB Docs (2) Team
Common exam errors
This page explains eight common misunderstandings and mistakes that are often seen in student exam responses. The following list is not exhaustive, but making sure that you don't make these mistakes will improve your overall performance with regard to critical thinking.
For each "mistake," be sure that you understand why it is not a good response - and what you could write instead!
This is a common approach to evaluation - always write about generalizability! The problem is, this is often not a good strategy.
In many studies, the research is qualitative. For example, in Brown and Harris's study of women with depression, the goal was never to generalize to men. Therefore, the statement above is totally irrelevant.
Biological studies are particularly problematic. Maguire's study, for example, only used taxi drivers from London. To argue that this cannot be generalized to taxi drivers in other countries (or even cities) does not make sense as the human brain would not work differently in the city of London.
This may be correct, but it depends on the study. When discussing biological research it is often not the case that simply because the sample is made up of Americans, there may be a cultural bias.
If you want to think about cultural bias, think about how the culture may actually affect the results of a study. For example, a collectivistic culture may be less willing to share emotions, Americans tend to choose extremes on Likert scales, or individualistic cultures may be less likely to conform to the behavior of the group when carrying out a focus group study.
Never simply state that because the participants belong to a certain culture, there must be a cultural bias.
First, it is important to note that a "laboratory experiment" is not really a type of experiment, but rather where an experiment is done. An experiment may also be carried out in the field.
When discussing ecological validity, always look at the nature of the study. Does the use of controls in the lab result in a situation that does not account for variables that would most likely affect behaviour outside of the lab? Is the situation artificial?
A good example is MRI studies. The MRI takes a scan of the brain to determine its structure. Although these studies are done in a lab (e.g. Maguire), the study has ecological validity as the environment in which the study is done does not produce results that are not true outside of the lab.
This is another common error. Although it is true that the lab environment helps to control for extraneous variables, increasing the internal validity of the study, simply the fact that something is carried out in a laboratory does not mean that causality can be determined.
What helps us determine causality?
- The independent variable is manipulated by the researcher.
- Participants are randomly allocated to conditions.
- There is a high level of control over external variables.
All three of these conditions must be met in order for us to consider a causal relationship. Lab experiments such as Sharot, Maguire, and Berry's study of Temne and Inuits do not fulfill these requirements and do not show causal relationships.
Although this could be correct, it demonstrates a clear bias in one's understanding of research. Reductionism can be both a strength and a limitation of a study or theory.
Reductionism can help a researcher to find a singular cause of a behaviour. A holistic approach cannot do this. On the other hand, a reductionist approach may oversimplify a behavior, resulting in a finding that may not recognize how the variable interacts with other factors to produce behaviour.
This is not only on exams, but this appears also in evaluations on IAs.
It is not a requirement that groups be balanced. In fact, groups are rarely balanced in research. The reason for this is that participants are randomly allocated to conditions. In addition, participants have the right to withdraw, so when a study is published, attrition rates may affect the sample size.
The "balance" only matters when a variable may clearly affect the findings. In that case, there should be a matched pairs design. If mathematics ability, gender, or length of time as a taxi driver may affect the findings, then it is important that these groups are equally represented in each group. But if you are going to make this argument, you need to explain why this is important. Simply stating that it is problematic because they aren't "balanced" is never a good response.
This is a common error which is actually a problem with terminology. Ecological validity has to do with the environment in which the study was done. The nature of the sample is the population validity of the study.
When students write this, they are confusing the terms ecological validity and external validity. External validity is when one considers both the ecological and population validity of the study to determine the extent to which the findings of a study will hold true over variation in samples, settings, and over time.
Although technically not an error...
This is a common strategy when applying critical thinking in essays; the problem is that is rarely relevant. If you are writing an essay to discuss the effect of one neurotransmitter on behaviour, then discussing the ethics of animal research is not highly relevant.
My advice is - only write about ethical considerations when asked to do so in the question.
You could, however, discuss ethics in the following ways to demonstrate critical thinking:
- Discuss how ethical considerations mean that we are reliant on animal research, which is limited in generalizability.
- Discuss how the right to withdraw has an effect on longitudinal studies, meaning that certain types of participants may drop out over time, potentially skewing the findings of a study (e.g Marmot's Whitehall study)
- Discuss how obtaining consent may have influenced the behaviour of the participants, leading to demand characteristics - for example, in Fagot's home observations of parents playing with their children.
Twitter
Facebook
LinkedIn