Hacker News Reader

Comment by ZeljkoS | Hacker News Reader

Yes, most tests are bad, but there is a quick advice that improves them immensely: never ask questions that are not at least Level 3 in Bloom's taxonomy of knowledge: https://blog.testdome.com/blooms-taxonomy/

I cofounded a technical screening startup, and despite our efforts to educate our customers that enter their questions on our platform, 95% of their custom questions are bad questions. They tend to ask trivia questions, that can quickly be googled, instead of asking work-sample questions (which we at TestDome.com prefer). I think they learned that from years of paper tests in school, it is very hard to unlearn.

Just to give you an idea, for testing web programmers we suggest questions where you need to find bugs in HTML code (https://www.testdome.com/questions/html-css/inspector/17629), while our customers would ask questions like "What does CSS stand for?"

View on HN

Thanks for the article.

The advise against level 5 questions is interesting. I found them to be very useful for answering if a candidate has really applied in the field before, especially comparing two big fields against each other. For example asking if they prefer sql or nosql. Anyone who doesn't answer some variant of "it depends" has probably not had enough hand on with either of them. But regardless if the candidate has a strong opinion or not, and regardless if it's the same opinion as yours, the answer is going to give you endless amount of openers for dialogue and follow-up questions on how they reached that conclusion and in which projects they used such techniques and how it worked out.

As the articles notes, the questions can easily be misperceived as checking if the candidate has the same opinions as yourself, or you becoming outsmarted by eloquence. Since you are yourself asking and evaluating the questions you have to ensure this doesn't become the case, by putting opinions aside and only asking about fields which you yourself deeply understand. Not sure if it might be off-putting for a candidate to get this type of question though, if they don't understand the deeper purpose of the question? Should you give a disclaimer that you are not actually not interested in their final opinion?

I guess that "putting opinions aside and only asking about fields which you yourself deeply understand" is actually very very hard to consistently apply in practice ?

"Why is Pluto no longer classified as a planet?" You've passed the "categorization" level, but might have actually failed the real "understanding" level, for which "We have discovered many large((r)) objects beyond it." might actually be a more correct answer!

After the discovery of dozens of new trans-neptunian "planets", categorizing Pluto as a "real planet", while it probably had more in common with them became problematic.

The best definition that we had to come up with to cleanly separate "real" planets from the others was therefore quite unlikely to include Pluto!

can anybody comment on the difference between "understanding" and "applying" in that taxonomy. The examples provided don't seem very good to me.

is it the difference between understanding a language and not speaking it, perhaps? following a conversation, but not being able to participate in it?

I have learned Bloom's taxonomy with comprehension instead of understanding. So if in an algorithm's class, you were taught that algorithm X was exponential time, then the exam question at the comprehension level can be, "describe why does it take exponential time to run algorithm X".

An applying level question could be, "Find the running time of algorithm Y (which was never discussed in class)."

Understanding is being able to read python code with no reference book/manual.

Applying is writing python code with no reference book/manual.

In both cases doing it properly means you are reasonably fast while being confident about the correctness (either that you correctly predicted the program's behavior or that you wrote bug-free software).

Reply on news.ycombinator.com