Is there a possibility that the data are valid and realiable if you use a self -made qustionaire.
Requesting references
If you create your own question, then you need to demonstrate the reliability and validity of those items. This is definitely the case when you are creating a new multi-item scale. If so, you would begin by assessing the reliability of the scale, typically through assessing Cronbach's alpha and often Exploratory Factor Analysis. You start there because a measure that has low reliability will, by definition have low correlations with anything. If you do have sufficient reliability, then you can check validity be examining the pattern of correlations between your new measure and other established measures, in a process known as "construct validity."
Necesariamente, debe hacer una evaluación, con un alfa de Cronbach, midiendo la fiabilidad y asimismo la validez. Todo conlleva a tener una encuesta robustecida que le permitirá ejecutarla o aplicarla, con la seguridad, que pueda medir confiablemente el comportamiento de la variable de su investigación.
1. Self-made questionnaire where the measurement items (items of question) are not linked to existing relevant literature (or questions not developed relevant literature) appears to violate quantitative (deductive approach) because in quantitative approach the researcher test and prove a theory or theories.
2. Self-made questionnaire also indicates inadequate review of relevant literature.
3. There is the tendency that consistency between the measurement items may not be achieved thus poor or no reliability in the questions, when checked using Cronbach;s Alpha. Thanks
While i strongly agree with Morgan, i would like to say that there are various ways of testing the validity and reliability of a self-designed questionnaire. One simple way for validity is to rely on best practice instruments in pevious related research or ensure that the items on your instrument actually capture the objectives of your study. This can be done through expert opinion (face validity) Reliability in most cases can be achieved through repeated tests before the main study and measured as per Morgan's recommendatiom above.
I disagree with Lasisi Raimi on one point, which is the use of repeated measures for reliability. This is an expensive and time consuming process, which Cronbach's alpha is specifically designed to replace.
- Badges
- Science topic
- Similar topics
- Engineering
- Chemical Engineering