I am working at a school who have asked me to pop in their designed timetable into CMIS using scheduler. (no problem here done it several times). However they want to begin this new timetable at the start of the last half of the summer term.
This would mean importing the schedule into the current data set, otherwise i'm sure that they would have attendance and Slasc issues. Then once the change has been made rolling the timetable into the newly created dataset.
The real pain here being that it would all have to be done on thier live database in one massive intergration just before they wanted to make the switch.
However assuming this is the case, I was wondering if anyone else has done this before. What effect would it have on the current teaching groups? Would changes mess with current data especially assessment wise.
Can it be done vice versa? create the timetable incrementally in the new dataset and then import back into the previous one?
Last edited by GeorgeFaux; 24th March 2009 at 03:34 PM.
I'm not a timetabling expert, but i think you need to look at weeks. Events have weeks assigned, and you can set them to occur for any range. For example, on a one week timetable, your current events will all (unless some are in rotations) span weeks 1-52. If you're on a 2 week timetable they will span 1 - 26.
Assuming it's a 1 week timetable and you're running on 1 - 52. You need to work out where the current ones need to stop, and change them accordingly. You can do this in bulk.
You can then set your new events to start the week after the first set finish. Off you go
Also, i'd recommend trying all this out on a standalone system first to make sure it's going to work ok.
If you don't know how to do that PM me your email address and i'll send you the docs (it's an easy job to do if you can find 30 mins to an hour and a spare standard desktop machine with M$ Access on it). You can also get the docs from the firstline site if you have access to that.
I'm fine with both the bulk changes and the stand alone machine but yes I see how that can work. Just create it from the existing dataset and then at the end of year process roll over the timetable in full and bulk change the weeks again for next year.
No - a 2 week timetable will be fine, if you follow the timetable guidance you do the following:
A 26 week year of 14 days, ie: you have a Monday A, Monday B etc.
This works because the both the first weeks are still Week 1, i.e: monday a and monday b will still be week 1 - the issue we're having relates to Events existing in one week and not another, but cover rotas not accommodating it.
I'm afraid that things are rather more complex than you all imagine.
You cannot successfuly integrate a timetable into a different dataset. Rolling timetables forward from one dataset to another is a nightmare and not to be recommended.
You do use 'weeks' to allow the new timetable to be started in the current dataset but effectively, the new timetable has to be entered twice. Once in the current dataset and again in the new dataset.
Timetabling a different timetable in a dataset that already has a timetable is difficult as both sets of events will often 'pop up' in the screens that are usually empty. You also have to cope with an entirely new set of teaching groups for the new start but you can't get rid of the old ones for a number of reasons.
Several schools in Birmingham have done this in the last couple of years. One did it 'manually' two years ago and found it quite difficult. Since then the schools that have needed this have done it with the support of Link2ICT. They did their new timetable in the new dataset as per normal and then Link2ICT SQLed the new timetable back into the current dataset for the weeks at the end of the year. This keeps the timetabling process as simple as possible and allows a relatively stress free switch over.
(Oh by the way, I should mention that I work for Link2ICT for part of the week)
Last edited by jmc; 26th March 2009 at 11:08 AM.
Reason: Forgot to declare an 'interest'
We've been having a bit of a discussion about this here in the office over the last few days and we've come to the conclusion that if you are effectively rolling your students up to start next year's courses in June then you should start a new data set providing that you have finished all of your reporting home to parents by that point.
This will create restrictions on reporting on certain data, mainly attendance data (for both students and staff) as it is only possible to report on the current data set. So to report on attendance for the academic year from September to August you would need to run reports in 2 data sets.
There is no particularly elegant solution if you are not happy to start a new data set, using week ranges will give you a new timetable, but may cause issues in Cover as already stated (just getting that one double checked and I'll post back when I have something), and as people have already said you will need to re-do the timetable in next years data set.
Thank you all for your responses. I understand that the advice is to start a new Dataset. As a technical solution this seems sound.
I just wanted to play devils advocate to ensure that this will have no second order repercussions down the line.
Week Ranges:- In terms of creating the new data set I should be looking at a 52 week year running June to June (assuming they want to do this again next year.) This will probably create a calendar with too many or two few teaching days (isn't 194 statutory). Will this cause issues?
What about the current dataset, should I change the week ranges to reflect when the year is ending, again this will create a year with too few days. Or should I just leave it.
Attendance:- As stated this will screw their 2008/9 attendance figures. Is this reported to the DFES anywhere other than SLASC. What other issues might there be on that.
SLASC:- As I understand it the final SLASC date for this year is May 21st. Since I will be starting the new timetable after this date this should not be a problem. However what about the Autumn term Slasc for 2009/10, will this be sent haywire by having the first term start in June? Is CMIS smart enough to just take the data from September 1st.
Examinations:- I assume we will just import results back into the old database and this should not be effected by Timetabling or shortening the current dataset.
Student and staff Records:- Any changes would just have to take place in the new Dataset unless they corresponded to Examinations I would imagine.
What have I missed?
Last edited by GeorgeFaux; 27th March 2009 at 11:18 AM.
Here in Birmingham, we did start with the idea of starting the new dataset early. This would be the neatest solution as it actually reflects the reality of what was happening in a school. If, indeed, it is possible, this would be good news all round.
However, we dismissed this solution because of complications with stats in general and Census in particular. I can't remember all of the details but I'm sure that there were problems with attendance and exclusions. In particular, as I understand it, the Autumn census reports on all exclusions from the Summer term and an early end to the dataset might affect this number.
There maybe other subtle issues. For instance, if there's a sixth form, I'm not sure how QAMS will be affected.
When a timetable is rolled forward, it seems to lose the connection between events and virtual groups which makes later maintenance and certain printouts (such as Master Class) a nightmare. When you look at a typical matrix, after rolling forward, all of the events are 'piled on top of each other' in an unassigned column at the right of the matrix! I seem to recall problems with the naming of teaching groups as well.