Jump to content

Talk:Internal consistency

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

This article should be merged into the article on Cronbach's alpha because the concept of internal consistency cannot be discussed without reference to the formula contained in that article. As used in this article, the examples are misleading by trying to show responses to items with sematic similarity, when internal consistency actually refers to the degree of correlation among items of a test. The examples would more effective if added to the Cronbach's alpha article, which already has a short (one-sentence) discussion of how spurious correlations can arise from redundant semantic/syntactic similarity in the items of a scale. Amead (talk) 10:35, 29 November 2009 (UTC)[reply]

In gaming

[edit]

Does this REALLY belong in the 'Role-playing video games' category? It's not a game and has no more to do with games than with, say, novels or other art forms.

Arkaaito 16:11, 23 February 2007 (UTC)[reply]

This article was obviously written by someone who thinks any idea that's been around for 2000 years or so was first invented when an occasion arose to apply it to computer games. Michael Hardy 01:37, 28 October 2007 (UTC)[reply]

The online references to "internal consistency" in gaming tend to use a rather different notion of what the phrase means; see for example this column. In the context of gaming "internal consistency" is a concept without a well-defined and generally accepted meaning. Wikipedia should not randomly make up a definition for this and present it as "the" meaning.  --Lambiam 08:01, 28 October 2007 (UTC)[reply]
You have now completely eliminated the original purpose for which Kate (talk · contribs) created this article by removing the last mention of gaming without (as far as I know) putting it into another article such as Internal consistency (video games). Would it not have been more proper to do an Afd rather than completely changing the contents? JRSpriggs 17:44, 28 October 2007 (UTC)[reply]
That's not exactly what happened. I added a section on the statistical concept a few months ago, and later on a different editor removed the reference to video games. I have no ability to comment on the gaming concept's notability or the accuracy of the version presented here. Inhumandecency 15:11, 30 October 2007 (UTC)[reply]
As for my removing the gaming part, I have to confess that I did not look at the early history of the article; nevertheless, my criticism of that part stands. More precisely, I see no evidence that the combination "internal consistency" has a specific technical meaning when applied to games, just like "juvenile humor" has no specific technical meaning when applied to movies. In contrast, the term "internal consistency" does have a specific technical meaning in the context of tests, so I see no reason to take the article to AfD.  --Lambiam 19:16, 30 October 2007 (UTC)[reply]
I looked up this article trying to find what it meant in terms of the Fantasy genre, which uses "internal consistency" partially to explain its concept. I didn't find a good or understandable answer here at all, but I did find the definition Lambiam gave somewhat helpful. I do think it should have a place somewhere. Spirit Stiff (talk) 06:24, 12 January 2008 (UTC)[reply]

Example of statistical concept is misleading

[edit]

I think that to be consistent the example in the first paragraph should read more like this:

For example, if a respondent expressed agreement with the statements, "I like to ride bicycles" and "I've enjoyed riding bicycles in the past," but disagreed with the statement "I hate bicycles", our test would have good internal consistency.

Even if the statement "I hate bicycles" is reverse-scored, this still would not imply agreement with it.


--Ashoumarov (talk) 02:21, 23 February 2008 (UTC)[reply]

Table measuring alpha inconsistent

[edit]

The table showing how to read a Cronbach's alpha differs from the one in the wiki for Cronbach's alpha, an inconsistency that can be misleading. Please find the correct one and use it for both. (Mostly that 0.7 > α ≥ 0.6 is Questionable in one and Acceptable in the other) — Preceding unsigned comment added by 136.167.212.73 (talk) 17:39, 20 February 2014 (UTC)[reply]