In my DOE result, there is no lack of fit computed. I found a piece of information, it mentioned that " There will be no Lack of Fit statistics if there are no replicates and/or more unique design points than model coefficients".
what does it mean? what should I do to get the lack of fit to prove that the model is seen as a fit model? can I see only from pred R-square to say that my model is well?
Try to see some answers in this post:
https://www.researchgate.net/post/Design_expert_regression_model_lack_of_fit-can_anyone_help
Thank you so much
I find another one where it's close to my case. I obtained pure of error zero
https://www.researchgate.net/post/Is_it_neccessary_to_get_F_value_and_p_value_of_Lack_of_Fit
Your question cannot be answered. The details are needed e.g the model you are measuring is adopted and modified or a new one? How this model has been constructed? or the data you have used to meet reliability and accuracy measures?
Arjumand Bano Soomro
please see the file attach on how I constructed my model
The images you have shared, I believe are the extraction of strong factors and definitely lead you to the model fit. If you are worried for goodness of fit then refer
Hinkin, T. . R., Tracey, J. . B., & Enz, C. A. (1997). Scale construction: Developing reliable and valid measurement instruments. Journal of Hospitality and Tourism Research, 21(1), 100–120. http://doi.org/10.1177/1096348097021001 08
It is recommended to have smaller Chi-Square values for a better fitted model (goodness-of-fit) (Hinkin et al., 1997).
- Badges
- Science topic
- Similar topics
- Engineering
- Computer Engineering
- Computer