Then the time has come to switch to R :)))

Trust me, it is easy to use....See lavaan.org for a great tutorial.

Best,

Holger

Thanks Safa

Hi Ceren,

mean substitution is - given that there more valid and easy approaches - considered as a "no go". If you use R/lavaan package or Mplus, you can use the FIML estimator (full information maximum likelihood) to consider missing data.

See

Newman, D. A. (2014). Missing data: Five practical guidelines. Organizational Research Methods, 17(4), 372-411. doi:10.1177/1094428114548590

Schlomer, G., Bauman, S., & Card, N. A. (2010). Best practices for missing data management in counseling psychology. Journal of Counseling Psychology, 57(1), 1-10.

Graham, J. W. (2009). Missing data analysis: Making it work in the real world. Annual Review of Psychology, 60, 549-576.

Best,

Holger

Hi Holger, I use AMOS and AMOS does not accept missing data when you click ML.

Then the time has come to switch to R :)))

Trust me, it is easy to use....See lavaan.org for a great tutorial.

Best,

Holger

Thanks Holger

Please, be careful if imputing mean will provide you logical results. Sometimes this is not the best imputation method. Good luck.

MISSING DATA means no data. No matter what software you use, the fact remains, there is no data, i.e. missing data. If you takes courses in your Ph.D. degree program, there are 25 courses you should take, but you did not take 5 courses, should the university award you the degree by imputing 5 courses that you did not take as....fixing the missing data? Any row of observation that contains a missing data, if hbat data is relevent and material to the calculation or determinative of the result, it must be treated as a defective data point and, therefore, removed. To do otherwise, no matter how innovative the method may be: correcting, imputation, etc., is not fixing but faking data that is not there. What is the integrity of the data? What is the integrity of the researcher? This is an issue in methodology as much as it is an issue of ethics.