Mr.Tea Wrote:I love how funny it is that I have an illogical conclusion.  When the information you bring to the table is people not understanding how to read a question and making no reference to a personality testing, but general polling.  To address that article:  Most surveys do the multiple questions.

It's not about people not understanding how to read a question, it's about the fact that one sentence can be interpreted differently (legitimately or not), and that even very similar combinations of words can have quite different connotations/associated meanings. For instance, in the example that I quoted, the second question explicitly uses the phrase "unborn child" which can cause the reader to view an embryo as a fully human entity, and this can influence the reader to then feel morally obligated to protect the "fully human" "unborn child" and prevent it from being killed (it's debatable about whether or not it is a fully human child). Whereas the previous question does not by itself predispose the reader to feel that moral obligation. That's just one example of a result of difference in wording between two superficially similar questions. It's not really a feasible option to be able to completely get rid of this "wording of the question" effect, and having a number of these sorts of questions in a questionnaire can actually as you said, help to determine the consistency of a person's answers.

Oh, and it shouldn't have to be said the wording effect is applicable to naive's personality test.

Mr.Tea Wrote:

Quote:One way that researchers assess whether respondent fatigue created problems in a long questionnaire is by including consistency checks: Questions from an early section of the questionnaire are repeated in a later part (either in identical form or alternatively phrased) so that the reliability of responses can be checked.

I agree with this statement, but I don't think it's just a measure of respondent fatigue.

Mr.Tea Wrote:If people don't respond correctly to the first time as desired then there is something wrong with the survey.

There's not necessarily a correct answer to questions in personality questionnaires. If you mean they don't respond consistently, there could be a variety of reasons why. Perhaps two questions which seem the same, are actually asking different things. Or asking the same sort of thing in a different way/ with a different perspective that can result in a different answer (maybe the new view changed their mind?). And Linguistics is not as simple as we think. The same word might have a different meaning between two people (albeit probably slight).

Mr.Tea Wrote:If you are going to make a survey ridiculous long probably want to break it up and conduct separate target sensing session.  Common sense and experience point out people's attention span. Marketers already know this length.  I actually have some marketing research experience (ten years ago), but what do I know.

Just because you had that experience doesn't automatically make your opinion right. Not that I mean you don't know anything.

Some things are less effective and accurate to determine with fewer parameters/measures. In the case of this personality test, an adequate level of reliability and validity wouldn't be possible without a substantial amount of questions.

Mr.Tea Wrote:In reference to you personality test-http://cobweb2.louisville.edu/faculty/regbruce/bruce//rflct600/mount1.htm
"One should be careful when interpreting the results of a personality test. Just because someone tests one way does not necessarily mean that they are that way. People are made up by a combination of several personality traits. There also is the chance of incorrectly labeling a person. Employers should be aware that there is no hard evidence that personality tests, like the Myers-Briggs Type Indicator test, are accurate measures of one's personality."

Yeah, like I said: "it might not be 100% accurate... (but the) test can give a rough indication of personality".

Mr.Tea Wrote:"Inaccurate test results could prevent someone from getting a job. Using personality tests to match a person with an occupation also is a flawed practice. The validity and consequences of personality tests should be taken into account when they are given to someone. "

Don't disagree with this, however again, the test can give a "rough indication" of some jobs which may be suitable for a person.

The rest of the article seemed to be about either the writer reflecting on how the test results did not match his own opinion of his personality (which just fits the idea that the test is not 100% accurate), or about him simply rewording his earlier points.

Mr.Tea Wrote:As I read the linked personality test there were statement I neither agreed nor disagreed with, but it was a yes or no question.

Fair criticism, that's one of the problems with yes and no questionnaires. Questionnaires using the Likert scale (agree, strongly agree, neither agree or disagree, etc.) are more effective.

Mr.Tea Wrote:Most leaders understand that the if you want to know someone's personality you observe them or challenge them and see they respond.  Oh ya, overtime people's personalities may change by being exposed to different stressors and breaking points.

I agree with your last sentence, although I think it should, in a way, also be applied to your first sentence. People will react to the challenges differently, according to the different stressors (internal or external), the differences in their current mood, etc. etc. etc. There are heaps of factors which can influence how they respond to a challenge. Their response isn't necessarily an indication of their personality. Even less so then this test, which is considerably more controlling of those variables.