Assumptions, Assumptions, Assumptions...

Recently, I was sitting down somewhere in a library and there was a lady speaking on her mobile phone. It seems like she was having difficulty speaking with the person on the other end in English, and so she was looking around for someone to help her to translate for her (into Mandarin Chinese) I guess to help her understand.

After looking around, the lady on the phone saw someone with an East Asian appearance, who was sitting next to me at the time. She then asked the person sitting next to me, in Chinese, what the person over the phone was saying; or words to that effect. However, the person sitting next to me in fact was not Chinese (in fact Korean), and did not understand what this lady on the phone was telling, so she had quite a blank expression. To be honest, I felt that it was quite an amusing situation, but I refrained from laughing.

After a few seconds, the lady on the phone realized the person sitting next to me probably didn't understand, and she saw someone else with an East Asian appearance, who could satisfactorily translate for her. She skipped me for some reason. Maybe because my skin is a bit darker than most Chinese, or maybe because I didn't respond initially when she was talking to the person next to me. I guess my Chinese language skills aren't that good anyway.

Anyway, anecdote aside, this prompted me to write the blog post about assumptions now. I think this type of situation demonstrates both the advantages and the disadvantages of assumptions.

It is true that assumptions are often not 100% accurate. Some say "don't assume", or "if you assume, you make an ass out of me and you". For instance, not all people with an East Asian appearance are Chinese. They may, for example, be Japanese, Korean, Vietnamese, Thai or other races. Furthermore, not all Chinese know how to speak Mandarin (or any Chinese dialect) well, and some of other races can speak Mandarin reasonably fluently (like our former Prime Minister and former Foreign Minister Kevin Rudd). Related to Medicine, it is said that you should not assume that the patient is heterosexual and that their partner would be of the opposite sex. It is clear that if you want to be accurate in your deductions, you can not just assume away.

On the other hand, assumptions may sometimes be a practical way of increasing the probabilities of getting something correct. Going back to the example of the lady on the phone, even though the first person she tried to get to translate for her could not do so, the next person could. Even though the "test" is not perfect, it is more likely that someone who has an East Asian appearance would be able to speak Mandarin than someone who doesn't. So in that way, in the lady's situation, it is more productive to ask people with that appearance rather than just asking any random person in the library whether they could translate for her.

When doing calculations for pharmacokinetics, physics or other sciences, assumptions are often made to try to simplify things down (trying not to simplify too much as to significantly change the final answer) in order to make the calculations more manageable and quick. Finally, it is useful in medicine in terms of screening. Screening tests are best targeted to groups who are at a high risk of having or susceptibility to a certain condition, in order to reduce costs and reduce false positives. In some sense, you are "assuming" that the patient in front of you is more likely to have a certain condition than the average population when ordering the screening test for them, for good reason, but simply based on the probabilities in the population.

So assumptions can be quite an important tool, but care must be taken not to misuse it. Assuming certain things can make things easier for yourself with variable certainty. However, if you need to be more sure, you still have to follow things up further.