Steps in validating a new scale

[K Shahid]

I appreciate if someone can provide me a list of relevant steps required to validate the development of a new management scale based on a likert scale. its a farily simple scale with 60 short items.

 

i am hoping to divide this into two steps:

 

1) sample size issues

2) statistical issues like doing EFA and CFA etc

 

please help.

 

thanks

 

shahid

[Carlos Avalos]

Although it´s not a factor analysis book, I recommend you take a look in this book. Summated rating scale construction. It will be very helpful in your task.

Regards.

[Vanessa Arcilla]

Since your talking about validation and scale development, face and content validity could be checked using ratings/comments from at least 3 experts about your scale prior to actual testing on your sample/participants. Factor validity is more about the steps you have identified. You can also check books about test development for more step-by-step guide. Hope this helps.

[Ingo Rohlfing]

I am not sure what you mean by sample size issues. But in addition to face validation and content validation (on the latter see the classic Carmines, Edward G. and Richard A. Zeller (1979): Reliability and Validity Assessment. Beverly Hills, Calif.: Sage Publications), you could do a principal component analysis or exploratory factor analysis in order to see on how many components the items are loading in what way (on the difference between PCA and EFA, see the chapter in Tabachnik and Fidell’s book on statistics). A confirmatory FA could be done when you have clear expectations about the number of factors and loadings. The Carmines and Zeller book also gives you advice on reliability measurement, which is relevant as well when constructing a new scale.

[Author]

Posts