Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
reliability test
Message posted by Lokman (via 194.168.45.250) on November 16, 2001 at 9:35 AM (ET)
What are the steps to perform reliability test?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: reliability test
Message posted by beverley (via 193.62.3.247) on November 23, 2001 at 8:52 AM (ET)
i am studying the measurement of body composition using skinfold calipers, we are comparing the two studies of jackson(1980), and Durnin and Womersley(1974), can you help us to find out the reliability of these tests,we really appreciate your time,
lv, bevxx
Re: reliability test
Message posted by Phil (via 66.32.156.145) on November 23, 2001 at 12:16 PM (ET)
It all depends on what you are referring to. If you mean the reliability of a survey instrument (that is, will the instrument produce consistent results if repeated), then there are two basic ways to assess reliability: One is to repeat the test with the same subjects and compare, and the other is internal reliability. The "split half" method is popular and also "Cronbach's Alpha" (which is an average of all possible split half measures for a test). You generally improve the reliability by eliminating or fixing the questions that have the least consistency.If you are talking about the reilability of a product or process, then you track such things as "Mean Time Between Failures (MTBF)", or "Mean Time Between Repairs (MTBR)". You see if you can increase these numbers by improving preventative maintenance, better repairs, better parts, etc.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.