![]() |
|
|
| Subject:
Industrial/Organization Psychology
Category: Miscellaneous Asked by: graduate-ga List Price: $2.50 |
Posted:
07 Jun 2005 14:27 PDT
Expires: 07 Jul 2005 14:27 PDT Question ID: 530501 |
What does it mean to say that a test can be no more valid than it is reliable? |
|
| There is no answer at this time. |
|
| Subject:
Re: Industrial/Organization Psychology
From: jba2905-ga on 07 Jun 2005 21:47 PDT |
Hi, My specializtion in Psychology is IO so I thought that I would have a crack at this question. Firstly I think that it is important to distinguish between validity and reliability. Reliablity basically means that a test or experiment can be administered over and over again and you will reliably be able to measure something. For example you may have adminstered a test to five patients that have a condition and the test will show that they have the condition, it does not mean that the test actually worked just that it was reliable. In general Validity refers to weather the test actually meaures what it claims to measure. So for example we took 10 patients, and we know that five have a condition and five do not, then does the test when administered blindly does the test pick the five that have the condition and the five that do not have the condition. One of the ways that I think about it is to restate it in another way. A test can be reliable without being valid, but you cannot have a test that is valid without being reliable. Let me know if this answers your question. |
| Subject:
Re: Industrial/Organization Psychology
From: needsomeinfo-ga on 07 Jun 2005 21:57 PDT |
Here's another example to build from jba2905... Reliability is similar to consistency and validity is similar to accuracy. I like the dart board example. If you throw darts reliably, they will always land in the same area...it might not be at the bulls-eye, but it will be in the same place. Validity is actually hitting the bulls-eye. So even if you can hit the bulls-eye (validity), if you can't do it consistently (reliability) there's not much point. So bottom line...even if you come up with a test to measure something accurately (validity), if you can't measure it consistently (reliability) the test won't do much for you. |
| Subject:
Re: Industrial/Organization Psychology
From: graduate-ga on 08 Jun 2005 03:53 PDT |
to: jba22905 and needsomeinfo: These are very explanatory. I have a difficult time explaining what I know in writing even though it's in my head. Thesea re great. Thanks much. |
| Subject:
Re: Industrial/Organization Psychology
From: jba2905-ga on 12 Jun 2005 05:07 PDT |
Did it answer your question |
| Subject:
Re: Industrial/Organization Psychology
From: cynthia-ga on 12 Jun 2005 05:19 PDT |
Personally, I think graduate-ga has indeed said his question is answered. Good Job!!! ~~Cynthia |
If you feel that you have found inappropriate content, please let us know by emailing us at answers-support@google.com with the question ID listed above. Thank you. |
| Search Google Answers for |
| Google Home - Answers FAQ - Terms of Service - Privacy Policy |