How to Fit Discrete-Time Multilevel Multiple-spell & Multiple-state Survival Analysis Model Using Stata?

[smith Jason]

The current study would like to model the relationship between on-time grade progression (state) and student English proficiency reclassification time period (reclat) using the attached dataset, which is created for the expected discrete-time multilevel multiple-spell & multiple-state survival analysis. More specifically, the variables include,

id (student id),
schid (school id),
male (1-male;0-female),
esl_pgm (Time-varying variable:1-student attended ESL program in the current year;0-No),
bil_pgm (Time-varying variable:1-student attended bilingual program in the current year;0-No),
everspec (1-student ever received special education;0-No),
hispanic (1-student ethnicity is Hispanic;0-No),
grade (student current grade-and it would repeat if a student had grade retention)
reclat (student English proficiency reclassification time period;1-primary school;2-middle school;3-high school;4-never reclassified)
state (if a student had grade retention;0-No;1-grade retention Note: student can have multiple times grade retentions)
spell
time.

I am totally new to survival analysis and really don't know how to do this using Stata.
Can anybody provide Stata code with explanations?

Thank you!

[smith Jason]

Can anybody help in any aspect? Thank you!