Left or Right of me.

[rah]

It is difficult to construct validation questions that are obscure enough to avoid recognition. People are not stupid.

 

[Manfred Belheim]

If you're trying to gauge an overview of someone's political opinions, then it would seem hard to do that without asking questions about their political opinions. And if their aim is to give the impression that their political opinions are different to what they actually are, then "giving the answers I would want to be seen giving" seems too obvious to even be considered as any sort of clever manipulation.

But also why even bother doing any of that when getting the answer you want is as easy as...

Economic Left/Right: -479.0
Social Libertarian/Authoritarian: Æ©Ð

 

[rah]

Sometimes it's not conscious.

 

[Hrothbern]

It is difficult to construct validation questions that are obscure enough to avoid recognition. People are not stupid.

And if you throw in another kinds of questions as a base load distraction ?

Your for example political questions mixed with a set of cognitive questions and a set of memory association questions... still good distinctable for the test taker, but because of alternation, the switching mindset and a time pressure, encouraging the test taker to fall back to use moreover the same approach for all the questions.
And nothing forbids to weave in specific detectors.
Nothing forbids to interpretate hesitation delay times
etc

 

[rah]

That's one method. But to really eliminate it you need multiple tactics. And to be fair, I think some people are too smart to eliminate it with 100% certainty.

 

[Akka]

It absolutely does. If someone is able to make the test spit out a result that matches their preconceived notion of their political beliefs then it isn't a real test at all, more like a game.

That's nonsensical. A test is supposed to be taken on good faith. How exactly do you plan to make a test that gives the right results when someone gives the wrong answers ?

 

[Hrothbern]

That's one method. But to really eliminate it you need multiple tactics. And to be fair, I think some people are too smart to eliminate it with 100% certainty.

Are you involved in designing such tests ?

if so.. I think you have a great job !

 

[rah]

I'm more on the integrity of the final data side but I have worked on questionnaire design and implementation. But lately mostly just telling people what to do.

 

[Hrothbern]

I'm more on the integrity of the final data side but I have worked on questionnaire design and implementation. But lately mostly just telling people what to do.

Isn't that the lazy end up of any career
"helping" other people.

 

[Hrothbern]

I cannot stop now

Do you, combined with high speed exposure, also use discarding tactics ? allow people to skip for example 20% of the questions, and analyse the skipped ones as well ?
and ofc measure time per question, skipped question as well.

 

[Lexicus]

That's nonsensical. A test is supposed to be taken on good faith. How exactly do you plan to make a test that gives the right results when someone gives the wrong answers ?

You are missing the point. It is possible to make a test where the connection between the answers and the results is more obscure and thus more difficult to game in the fashion I described. It is of course impossible to make a test foolproof in this regard but that doesn't mean some tests aren't better than others.

 

[Hygro]

Maybe our fellow's test re-taking was him, issue by issue, deciding to figure out what he could reconsider to be what he believed in while continuing to be... I guess... what he also believed in.

 

[rah]

I cannot stop now

Do you, combined with high speed exposure, also use discarding tactics ? allow people to skip for example 20% of the questions, and analyse the skipped ones as well ?
and ofc measure time per question, skipped question as well.

High speed exposure is not a common tactic due to the nature of current testing. Unfortunately most surveys are being done on-line due to cost and speed of results.
Which actually sucks due to all the weaknesses. The attention span on-line is critical. Anything that will discourage the completion is eliminated.
The biases of only on-line has been a big topic of discussion and how to make up for it, but all the other methods are just way too expensive in comparison.

 

[Manfred Belheim]

There doesn't seem much point in the thread if you're just going to assume people are lying or trying to game any tests they take. I don't know why that would be assumed in this context though.

Sometimes it's not conscious.

Isn't that a good thing though? Or you mean the manipulation isn't conscious?

 

[rah]

Yes, the manipulation isn't conscious.

The thread was to test perceptions, not test results. We've seen them before. I'm still shocked I ended up being the third from the right end. Certainly didn't expect that. but that was part of the purpose of the thread.

 

[inthesomeday]

All the right wing folks didn’t come and register themselves, is the problem. @Manfred Belheim where would you put yourself in this list?

 

[rah]

I know leaning right isn't popular here so I can understand the reluctance, but I figured more would own it.

 

[El_Machinae]

Maybe our fellow's test re-taking was him, issue by issue, deciding to figure out what he could reconsider to be what he believed in while continuing to be... I guess... what he also believed in.

Yeah, we're missing the output. I wonder if he changed his stances on some topics in order to more closely sit with the conclusion?

For example, if I were a Christian, I could interpret the 'eye of the needle' metaphor according to my internal wishes. But if I then learned that there was a much better idea of what Jesus meant, it would potentially change my moral framework. After the dissonance.

 

[Manfred Belheim]

All the right wing folks didn’t come and register themselves, is the problem. @Manfred Belheim where would you put yourself in this list?

No idea. I'm not familiar with the political leanings of most people on here to be honest, the handful of radical communists aside. I posted my political compass score though, so go with that. It's not going to be any less accurate than me just guessing.

 

[Hygro]

No idea. I'm not familiar with the political leanings of most people on here to be honest, the handful of radical communists aside. I posted my political compass score though, so go with that. It's not going to be any less accurate than me just guessing.

But you guessing or asserting or anything is waaaaay more fun.