Crash Course Psychology #2 is a dive into Psychological Research & Experimentation.
For context, Crash Course inspired me to learn the basics of psychology, so I’ve made it my mission to watch the entire Crash Course Psychology playlist and paraphrase each episode in my own words. This journey wouldn’t have been possible without the Crash Course team, so many thanks to them! To showcase what I learnt, here is my personal paraphrase of episode 2:
Crash Course Psychology #2’s Introduction: Intuition & Hindsight Bias
- Your intuition is only right around half the time.
- If you predict something, like a potential danger, and it turned out to be right, this reinforces trust in your intuition. But if you’re wrong, you probably won’t think about it again. We call this the Hindsight Bias or the I-Knew-It-All-Along phenomenon.
- It means our intuitive sense more easily describes what just happened, than what will happen in the future.
We also tend to perceive order in random events, which can lead to false assumptions. For example, if you flip a coin five times, you have equal chances of getting all tails or alternating heads and tails. But we see the five in a row as unusual, a streak, thus giving that result some kind of meaning even though it doesn’t have any. This is why we have the methods and safe-guards of psychological research and experimentation, and the process of scientific inquiry. To prevent these mistakes when studying our minds!
The Scientific Method
In most ways, psychological research is no different from any other scientific discipline. For example, step one is always figuring out how to ask general questions about your subject and turn them into measurable, testable theories. We call this operationalising your questions. This is how the scientific method works;
- Question and a theory
- In science, a theory is what explains and organises lots of different observations and predicts outcomes. This is not a hunch.
- Testable prediction A.K.A. Hypothesis
- When you come up with a testable prediction, that’s your Hypothesis.
- Test with a replicable experiment
- Once you have your theory and hypothesis, you need a clear and common language to report them with. This allows other researchers to replicate the experiment.
Replication is key. You can watch someone display a certain behaviour once and it won’t prove much. However, if you keep getting consistent results, especially when you change factors (like the subject, the situation, etc), then you’re probably on to something.
Case Studies
- Replication is a problem when it comes to one popular type of psychological research: case studies, which take an in-depth look at one individual. Case studies can sometimes be misleading, because by their nature, they can’t be replicated, so they risk over-generalising.
- Still, they’re good at showing us what CAN happen, and end up framing questions for more extensive and generalisable studies.
- They’re also often memorable and great story telling devices for psychologists to use to observe and describe behaviour, BUT you would still have to look at lots of other cases to determine that conclusively.
Naturalistic Observation
Where researchers simply watch behaviour in a natural environment. The idea is to let the subjects just do their thing without you trying to manipulate or control anything. Basically spying.
Both case studies & naturalistic observation are great at describing behaviour, but they’re very limited in explaining it.
Surveys and Interviews
Another method to collect behavioural data, which is by asking people to report their opinions and behaviours. Surveys are a great way to access consciously held attitudes and beliefs, but they have limitations;
- How you ask the questions can be tricky; subtle word choices can influence results. Different words elicit different reactions.
- Who you ask is also important due to Sampling Bias. To fairly represent a population, aim for a Random Sample where all members of the target group have an equal chance of being selected to answer the question.
Drawing Conclusions
Once you’ve described behaviour with any of the three above methods, you can (A) start making sense out of it and (B) attempt to predict future behaviour. One way to do this is you can look at how one trait or behaviour is related to another, this is called a correlation. For example, one trait may have happened before another and it makes sense that the former possibly caused the latter; that’s a correlation. However, it’s also possible that the two traits were completely unrelated and were just coincidentally near each other. Hence, correlation is not causation. Additionally, there could be other factors unbeknownst to us, as we are limited to only what is seeable before us. Be mindful, we don’t know what we don’t know.
It’s tempting to draw conclusions from correlations, but it’s super important to remember that correlations predict the possibility of cause-and-effect relationships, but they cannot prove them.
Experimentation
To really get to the bottom of cause-and-effect behaviours, you have to experiment. Experiments allow investigators to isolate different effects by manipulating an independent variable, and keeping all other variables constant. This means investigators need at least two groups: (A) The experimental group that gets their variable manipulated, and (B) The control group that does not get manipulated.
- Just as surveys use Random Samples, experimenters also need to apply the same concept by randomly assigning participants to each group to minimise results from skewing.
- Sometimes one or both groups are not informed about what’s actually being tested. This is so that researchers can compare results between groups that are affected by substance, with groups affected by placebos, for example.
- Additionally, often even researchers themselves are not aware which group is experimental and which group is control. This is so that the researchers themselves don’t unintentionally influence the results through their own behaviour. We call this a double blind procedure.
Watch the end of Crash Course Psychology #2 to see all the above put into practice!
A long thought for Crash Course Psychology #2
I personally think the Random Sample technique is insufficient in overcoming the Sample Bias obstacle, if we wish to be full proof that is. That is why we best describe it as a tool to minimise skewed results, not to prevent them.
- [Problem] The flaw of the random sample technique is that there is still the possibility of picking up a group anomaly within the random sample. Hence, this would defeat the purpose of the random sample to begin with.
- [Counter Argument] While I argue that there is always that possibility, I acknowledge that there is also a difference between a possibility and a probability. Hence, I do admit that the random sample technique is currently our best approach that we have so far.
- [Proposed Solution] If I had to propose a solution, I would suggest intentionally mixing the sample size with humans that contrast each other. Making the conscious effort to maximise group diversity.
- [Counter Argument] However, I also acknowledge that the researchers are still humans, not robots. My proposed solution does not take into account that the researchers themselves may tamper with sample groups with their own selection bias.
Danniel’s thoughts on Crash Course Psychology #2
- Surveys and Interviews are indeed a great way to access consciously held attitudes and beliefs, and their limitations were mentioned. They were (A) how you ask the questions, and (B) who you ask them to. However, I think there is another limitation with surveys: Response bias. Additionally, participants could intentionally lie.
- Hank mentions in the video at 7:19 that sometimes one or both groups are not informed of what is actually being tested. He used the example of to test the affects of substance versus placebo. I think this is a superb example. I can imagine participants behaving differently if they know they’re consuming substance, whether intentionally or unintentionally.
- Hank mentions in the video at 7:30 that even researchers themselves don’t know which group is experimental and which group is control so that they themselves don’t unintentionally influence the results through their own behaviour. I think it’s also worth mentioning that even without tampering the results through behaviour, researchers can still perceive un-manipulated results with bias. Thus, even with untampered results, conclusions and analysis can still be vulnerable to bias interpretation.
Enjoyed this learning of Psychology? Test your knowledge against these quick custom Kahoot! quizzes I’ve made based on the episode above: This is the easy mode and this is the hard mode for Crash Course Psychology #2.
Also, do check out what else Psychology related I’ve learnt from my Psychology blog!
Credits for Crash Course Psychology #2
Original Content & Media by Crash Course
Content Consumed and Paraphrased by Danniel Iskandar
Paraphrase Proofread by
Paraphrase Reviewed by