The topic I am researching is school Mindset and I am interested to do a survey validation of the instrument. Any information on this topic and survey would be greatly appreciated.
Did you see AAPOR and Michigan State University collections of texts? I think you could find a lot in the publications of academics who affiliate in this institutions.
I appreciate the references. I am also looking for a reference to a specific text or study that you may have found especially useful in a validation study you have done.
It sounds as if you already have an instrument, but no information about its validity?
The classic article on this topic is 60 years old:
CONSTRUCT VALIDITY IN PSYCHOLOGICAL TESTS, Lee J. Cronbach and Paul E. Meehl (1955), First published in Psychological Bulletin, 52, 281-302.
Basically, Cronbach and Meehl say that you need to demonstrate validity by showing that your measure performs as predicted vis a vis other well validated measures. So, you need to have both existing measures with adequate validity and hypotheses about how your own measure will relate to those measures.
Instrument validation studies are usually reported by researchers in psychology, such as social psychology. These studies may be on a different topic but they are valuable as they can show a direction in which to proceed with the validation (and construction):
I am so grateful for your response. Yes, I have the instrument and my plan is to run another, validated instrument with similar constructs, along with the survey instrument under study. Your reference to Cronbach's article is perfect.
The enclosed article is very elaborate, check and customise
Survey Instruments and Scales
The following surveys and scales have been designed and tested by CAPS scientists and are made available free of charge for use by HIV researchers, evaluators, prevention program planners, and designers. Keep in mind that survey needs vary from population to population and project to project. Please consider carefully the needs of your particular intervention before using one of these instruments.
These are all great responses. Don't forget that there can be many differences between the "essential survey conditions" of the validation studies and your implementation, such as target/eligible population, mode of data collection, etc. No validation can say that a question is "good" for all places and times. It's always a good idea to go do your on question testing (at least qualitative testing like cognitive interviewing, e.g., http://www.amazon.com/Cognitive-Interviewing-Improving-Questionnaire-Design/dp/0761928049) to make sure the questions work well in your situation.
Also, keep in mind that the "validation" paradigm can be conceptualized differently for latent measures (e.g., attitudes) v. behaviors and facts. For the latter, records or alternative data collection can be used for validation. Essentially, "validity" in this context is the degree to which the self report matches the record (or alternative report), but there is no modeling of a latent construct. This type of measurement error may be best explained in Hansen, Hurwitz, and Maddow (http://www.amazon.com/Sample-Survey-Methods-Theory/dp/0471309664) but see Tourangeau et al for a more contemporary review (http://www.amazon.com/Psychology-Survey-Response-Roger-Tourangeau/dp/0521576296/ref=sr_1_2?s=books&ie=UTF8&qid=1421907142&sr=1-2&keywords=Tourangeau)
My dissertation has many of the references, including cronbach's for the methodology of validating and instrument that was not validated. The instrument that I validated was created by me for use in the secondary population and was relative to alcohol and drug use, however, the methodology is similar regardless of topic.