• Exact, physical measurement of user actions (traces, real entries and exits): Use Stats block (“Activity measurement”) and Training sessions report (TrainingSessions), based on Moodle's logs.
• A flat-rate pedagogical contract, with times allotted to activities and resources: LearningTimeCheck (“Pedagogical progress and time” in French) and its accessory block.

It requires users to have a measurable and formal presence on the interfaces (via the standard logs) to accumulate “laptimes” (presence intervals) in their trace, whether they are cooking during that time or not. It matches well the former requirements of the OPCA/OPCO funding bodies for a transposition of “time in the training room”.

The accuracy of this measure depends on the heuristic for detecting the ends of sessions: see the calculation model of the Use Stats block.

It offers a model more compatible with online practices, based on the validation of knowledge and skills, with a transposition into time of these validations. It is intended for cases where it is possible to negotiate with the funding bodies a training service contract based on this calculation, allowing a more flexible use (more or less) of the training material. From there, the notions of time are no longer arithmetically tied to presence times: they represent “fundable time equivalents” based on pedagogical validations.

Reconciling the two models is therefore very complicated, since the times displayed by the second have no physical reality in the traces. They can come close to them for a person who consumes the material in a way close to the approach planned by the trainer, if the material is well calibrated.

In all cases, it is the “session exits” that pose a consistency problem (in the sense of auditing) with respect to the calculation of the flat-rate credits, and only if the real consumption is lower than the flat-rate allowance.

Conversely, flat rates that are too underestimated, which stand for “fundable quantity”, pose an acceptability problem: learners feel they are working far too much for the quantity paid.

In this practice, the “real measured” time is integrated into the LTC as a complementary measure, which should only be used to audit and correct the calibration of the contents. The coupling settings are described in the module's Installation and configuration page. What is expected from this comparison:

• Is the programme not globally under-calibrated?
• Is the programme not globally over-calibrated?
• Do the variations between positive and negative gaps balance out?
• Is the distribution of learners between under-calibrated, over-calibrated and average “normal” (normal distribution)?

1. The funder does not want to leave the principle of time observed “in real terms” and requires a transposition of attendance sheets.

In principle, the flat-rate time technique cannot be applied. Only the recorded events can generate entry and exit reports. The only room for manoeuvre is to decide whether the “outside course” times (home, personal spaces, blog, profile, etc.) are included or not in the calculation: see the Show “Outside course” time setting and the additional times of the report columns.

2. The funder accepts a principle of pre-negotiation of the pedagogical contract with the organisation and agrees on a pre-calibration of the contents.

The flat-rate technique can be applied, but must be explicitly documented for the participant. The participant must be warned that their working time cannot be held against the validation of the funding of their training, which they can take advantage of in either direction. The negative point of this model for funders is that it leads to funding knowledge and skills already acquired, where they would like to fund only “necessary and useful training”. Our answer to this argument is that the analysis of the real training needs (of a modular training) of a participant will have had to be done before their registration to the module.

Flat-rate times allow “fast” participants to validate a training time by spending less time on it than the normal pedagogical forecast, but they create a difference between the actual entries and exits and the measure of the capitalised time.

Providing all the data from both systems to an unaware inspector necessarily leads to a conclusion of inconsistency.

Trying to correct the exit dates, by adding the difference of LTC times, could only work over the whole learning duration, without the risk of producing inexplicable overlaps of ranges. In any case, it is a manipulation of data whose legality may be challenged.

It is therefore structurally not possible to find an algorithm that accounts for the flat-rate times on each entry and exit session, while taking into account the “reality of uses” of the material (split or interrupted sessions, sessions straddling midnight, etc.). Except by presenting “anomalies” in the result, mainly a list of sessions with a non-zero probability of two work sessions overlapping for the same person. A more radical correction of the course session list, to avoid overlaps, would be even more dangerous: the divergence between the real trace data, the participant's known schedules and the sessions presented to funders would be even greater, and impossible to justify.

Back to the LTC module index - Back to plugins index - Back to catalogue