Talk:Level of measurement

What is a yardstick?

I am confused as what "yardstick" advanced operation for intervals stand for? The linked page does not help, and some clarification would be much appreciated.

Thanks, User:boramalper

Direct links to terms

I modified the source so you can link to a measurement type using a HTML label.

Just put in Nominal Measurement

Thanks User:dmccreary

Definition of Levels

This article should clarify the meaning of the word level at the very beginning, and disambiguate it from a related meaning. The word level is used two different ways in this article. Besides its primary meaning (a type of measurement), here is a sentence fragment from this article where it is used differently:

...any difference between the levels of an attribute can be multiplied by...

The same use of level occurs in the Wikipedia article on Theory of Conjoint Measurement in this sentence fragment:

...Via specific relations between the levels of P...

In these cases, it appears that the word level means a particular value of a nominal (or categorical) variable (or measurement). The programming language R has a type of variable called a factor which is used to represent categorical variables, and its individual values are called levels, which seems to again be the same meaning of that word as the above two. This use of the word level seems to be quite common and used freely as if any educated reader will understand it, but that understanding of level is not that widespread.

Likert scale as example of interval level?

I was surprised to see the Likert scale given unreservedly as an example of interval-level data. Surely this assertion is a bit controversial and should at least have a caveat.

Also, it should have a hyperlink to the Wikipedia entry for Likert scale, (where, incidentally, the issue of whether it is interval or ordinal is teased out a bit more). ===Likert Scale chettan I could not find anything mentioned on the page regarding Likert scales. Given how often these are used, as well as some important influences about how you label the possible discrete & ordered choices, I think that they should at least put in a brief cameo appearance.

Please indicate whether or not you agree with the assertion that Likert scales should be included as part of this entry:

• Strongly agree
• Agree somewhat
• Neutral, neither agree nor disagree
• Disagree somewhat
• Strongly disagree

Cheers, Kevin (forgot my log on info for WP) 2014-02-19

Dichotomous data as example of ordinal level?

Similar to the comment above, I was surprised to see dichotomous data given as an example of ordinal level of measurement. Dichotomous data doesn't have a natural order and should be listed as nominal. Its mathematical operation is only equal or not equal. It doesn't become ordinal until a metric is applied (i.e. positive is better than negative when the metric is blood type match, while negative is better than positive when the metric is being tested for an incurable disease). This is also true for data that is 'trichotomous' or 'n-chotomous' (if you'll excuse my abuse of English). Additionally, a median value for 'n-chotomous' data doesn't naturally occur (i.e. if a survey asks a participant for their gender, and the answers are 'male', 'female', and 'other').

I'm reluctant to make an edit on the page since I'm not an active WP editor, but I propose changing the second sentence under ordinal data to:

"Examples include nominal data whose values are naturally transitive, such as 'completely agree', 'mostly agree', 'mostly disagree', 'completely disagree' when measuring opinion, or 'No Education', 'High School Diploma', 'Bachelor's Degree', 'Master's Degree', 'PhD' when assessing level of education."

BAusmus, May 2019 — Preceding unsigned comment added by BAusmus (talk • contribs) 17:55, 31 May 2019 (UTC)