Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Construct validity
Message posted by arien (via 161.142.15.157) on June 30, 2001 at 10:42 AM (ET)
Anyone can explain how to know validity a questionnaire ??
Can statistical test assess the validity ?? If yes, what test we can use for assess it ??
Thanks for helping me with clear answers ..
Arien
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: Construct validity
Message posted by JG (via 128.8.22.77) on June 30, 2001 at 2:37 PM (ET)
Validity means that the qiestionnair studies the right subject and the other thing we look for is reliability. There are many levels of validity. The weakest form of validity is face validity - it looks OK.
Re: Construct validity
Message posted by JG (via 128.8.22.45) on July 1, 2001 at 8:08 AM (ET)
One way to think of validity is to consider how would you check that a ruler really measures distance, a thermometer really measures temperature, a scale really measures weight, a gas pump really gives a gallon of gasoline, etc.
Re: Construct validity
Message posted by Jen (via 155.42.42.91) on July 4, 2001 at 2:10 PM (ET)
One way to examine the validity of a questionnaire is to compare it to other questionnaires that measure similar constructs (concurrent validity) or dissimilar constructs (discriminant validity). You want there to be correlations between your instrument and other similar instruments, and you don't want a correlation between your instrument and other questionnaires that measure other constructs
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.
