Yong Ho Kim, April 29th 2003
1. What are some criticisms of trait theory in general?
Possible criticisms include that subjects can falsify answers because the result of a personality test can be personally important to them, that very often the questions are culture-specific, and that subjects across different cultures don’t necessarily present the same main personality traits (Kalat points out that in the Chinese there is only an “loyalty to Chinese traditions” trait instead of “agreeableness” and “openness to experience”. This could be interpreted as meaning that in China a high agreeableness (being loyal to other people who follow the Chinese tradition) and being closed to experience (not trying other non-Chinese things) have always correlated together and could be lumped into one category.)
It turns out that given personality tests chunk groups of people together, it will forego other less prominent differences in personality, considered by the creators of the big five as overlapping or unimportant. However, ignoring small individual differences can directly lead to stereotyping, which isn’t always a desirable result. Of course, there is the payoff that the more traits we add and intend to measure through tests, the less parsimonious the test becomes. It is possible to do the reverse and decrease the number of traits even further arguing that they still overlap (Eysenck), even though it might generalize the descriptions even more.
2. Evaluate your lab section’s choice of traits.
Overall it was a good sample, but “experimental” and “curious” seemed to overlap in meaning. To experiment, one needs to be curious. For “motivated” and “diligent”, it seems like one corresponds to a mental disposition while the other is a behavior. Isn’t it the case that being motivated will lead the person to work harder, thus being perceived by others as diligent?
Also, “social” and “shy” were two degrees of the single trait. Hence, we would be using more titles than necessary, and making the accuracy of the test lower. Overall, it was about a good number of traits.
3. Evaluate your lab section’s choice of questions to evaluate those traits.
Questions such as “during the past week, I cried more than twice” were aimed at too narrow audiences to make any sense out of the results. No male subjects responded yes to that question, but it is possible that if the frequency had been reduced to once per month the number of respondents had been over zero (still including those who cry twice), and make a conclusion out of it, since a result of zero doesn’t tell us what the lower limit is.
4. What are some of the difficulties associated with developing personality questionnaires?
The problem lies in the fact that the questionnaires have to be created with human subjects in mind. If the test is too long, it will discourage subjects from finishing or volunteering to work on the test. But the more questions the questionnaire carries, the higher accuracy can be expected from it.
5. What are some of the difficulties associated with administering personality questionnaires?
Especially if subjects are acquaintances of the experimenter, they might hide those qualities deemed undesirable and emphasize those desirable. Also, often the pattern of answers subjects give is highly dependent on the social and emotional context in which the subject was situated at during the specific time and place of the day at which the subject took the questionnaire. Also, it is a written test, so people must sit down or at least stand still, which excludes an important population of Macalester College who are often running from class to class. (This is not a implicit reference to the “running boy”.)
6. How could personality evaluation be improved?
It is necessary to hide the “socially unacceptable behavior” tag from the questions as much as possible, in order to prevent social hindrance at the moment of responding questionnaires. Detaching the questions of positive remarks is also important, the idea being that questions should sound rather neutral. This has been done for the current test, but still “People I don’t know make me nervous” carries a strong negative implication that should be corrected. (I suggest, “I am mostly friendly towards people I know”). The conclusion seems to be that this is a embedded problem for self-administered tests.
Also, the format of the question could be changed so that it could, for example, be read off a tv screen or heard from a cassette recorder. Or done orally individually. If this were to be done, it would increase the repertoire of subjects to be included in the pool.
On a final note, I thought the effectiveness of personality tests could be improved if subjects didn’t realize that it was a personality test that they were taking at the moment of taking them, (so that they can be less conscientious and less censoring while answering questions) but this seems impossible given the access college students have to Psychology courses.