The article leads with a poll that asks the reader about agreement with statements whose meaning depends on ill-defined politicized words. I’ll cite the first question:
Select the statement (a or b) that you agree with the most. a) Many of the unhappy things in people’s lives are partly due to bad luck. b) People’s misfortunes result from the mistakes they make.
A poll like this can be seen as either a poll on respondents’ definitions of the words, or as being intended to use the meaning assumed by poll’s author (we are not told which one it should be; I think the respondents are not supposed to notice the distinction). The second option is too misleading for this particular poll, as there is a wide distribution of people’s preferences about these words and we don’t know very much about the author. The first option is particularly annoying when you notice how some words don’t have settled meanings, forming statements that don’t have well-defined truth value. Agreement with such statements is not informed by factual or moral questions, but is a function of definitions of their terms and can’t be evaluated when the terms remain undefined. I have no preference on definition of ill-defined politicized words (my preference is to use better tools, not to use bad tools in a particular way, especially when there is no principled way of improving them), so I can’t supply these definitions.
I read the article as a piece of self-help/personal development writing. Part of the corresponding reading protocol is that I interpret all questions to the reader at the beginning as being intended to prime a response contrasting with the article’s message.
Anyway, those three questions are just the short form quoted from the slides in reference [4]. The full version used as a psychological scale is linked in reference [1]. It includes the scoring key. I achieved a perfect “internal locus” score, even though I left two questions unanswered. They and several others turned out to be unscored. I guess they are there to camouflage the purpose of the test. (I presume that when the test is actually administered, it isn’t labelled in big letters, “Rotter’s Locus of Control Scale”.)
The article asks the questions of the reader, it doesn’t just cite the poll as a tool used in standard experiments. I find the article’s suggestion to answer such questions deplorable, as it requires the reader to at least play along with the confusion in the questions, instead of fixing or sidestepping it. Even as a tool, the problems with the questions suggest looking for better tools, although that’s probably outside the scope of the article.
Saying that language is ill-defined dodges the issue. People do have beliefs and those beliefs are made up of language.
Maybe it’s helpful to think about the question as checking aliefs. Which of the two resonates more? Which produces a bigger feeling of agreement?
Working with people who are in denial about their beliefs they have because they focus on a intellectual answer to such a question can be hard.
Agreement with such statements is not informed by factual or moral questions, but is a function of definitions of their terms and can’t be evaluated when the terms remain undefined.
That’s also not the point of the exercise. The exercise is about taking account for the beliefs that the reader holds
Which of the two resonates more? Which produces a bigger feeling of agreement?
Such things can be easily and dramatically changed by minor adjustments to expressions used—that has been demonstrated repeatedly and pollsters know that very well.
The article leads with a poll that asks the reader about agreement with statements whose meaning depends on ill-defined politicized words. I’ll cite the first question:
A poll like this can be seen as either a poll on respondents’ definitions of the words, or as being intended to use the meaning assumed by poll’s author (we are not told which one it should be; I think the respondents are not supposed to notice the distinction). The second option is too misleading for this particular poll, as there is a wide distribution of people’s preferences about these words and we don’t know very much about the author. The first option is particularly annoying when you notice how some words don’t have settled meanings, forming statements that don’t have well-defined truth value. Agreement with such statements is not informed by factual or moral questions, but is a function of definitions of their terms and can’t be evaluated when the terms remain undefined. I have no preference on definition of ill-defined politicized words (my preference is to use better tools, not to use bad tools in a particular way, especially when there is no principled way of improving them), so I can’t supply these definitions.
I read the article as a piece of self-help/personal development writing. Part of the corresponding reading protocol is that I interpret all questions to the reader at the beginning as being intended to prime a response contrasting with the article’s message.
Anyway, those three questions are just the short form quoted from the slides in reference [4]. The full version used as a psychological scale is linked in reference [1]. It includes the scoring key. I achieved a perfect “internal locus” score, even though I left two questions unanswered. They and several others turned out to be unscored. I guess they are there to camouflage the purpose of the test. (I presume that when the test is actually administered, it isn’t labelled in big letters, “Rotter’s Locus of Control Scale”.)
The article asks the questions of the reader, it doesn’t just cite the poll as a tool used in standard experiments. I find the article’s suggestion to answer such questions deplorable, as it requires the reader to at least play along with the confusion in the questions, instead of fixing or sidestepping it. Even as a tool, the problems with the questions suggest looking for better tools, although that’s probably outside the scope of the article.
Saying that language is ill-defined dodges the issue. People do have beliefs and those beliefs are made up of language.
Maybe it’s helpful to think about the question as checking aliefs. Which of the two resonates more? Which produces a bigger feeling of agreement?
Working with people who are in denial about their beliefs they have because they focus on a intellectual answer to such a question can be hard.
That’s also not the point of the exercise. The exercise is about taking account for the beliefs that the reader holds
Such things can be easily and dramatically changed by minor adjustments to expressions used—that has been demonstrated repeatedly and pollsters know that very well.
The point of a psychological test instead getting polling results. It doesn’t make sense to treat the test the same way as an opinion poll.