The length of the LTC RAP programs is an important issue for any evaluation because of its implications for the cost of data collection. An evaluation that involves measuring the apprentice at the beginning and the end of the program (and perhaps during the course of the program and afterwards) would need to consider how long it is necessary to follow an entrant to the program. The longer the program, the more difficult it is to gather information, the more likely that some apprentices will not complete the program, and the more likely that apprentices will be lost to follow-up. The programs visited on the site visits have a wide range in time for completion, with the shortest program being 232 hours and the longest program being 3,000 hours (approximately 1.5 years) (Kuehn et al., 2011). The two remaining programs were approximately 2,000 hours, which is a full year.
Length of the intervention is also important because longer evaluations are usually more expensive than shorter ones, particularly if they involve multiple waves of data collection. In order to follow entrants to the LTC RAP, data would have to be collected on new entrants to the program for at least a year, if not longer, to obtain a large enough sample size, and then followed on a flow basis for another year. Because of the small number of people in the LTC RAP, it would take a long period of data collection to build up an adequate sample size, which would be expensive.
Finally, the longer the program, the more likely it will be that apprentices will not complete the program. Some apprentices will find it too difficult to continue; others may leave the employment of the LTC RAP sponsor, either to go to another long-term care provider or to leave the field. As a result, to the extent possible, analyses will need to be done on several groups -- all persons who ever participated in the program, current apprentices, and direct care workers who have completed the apprenticeship program. Sample sizes will dictate how many of these different analyses can be actually conducted.