This study tested Marguerite Foxon’s 1993 proposal of a Transfer of Training Process. The transfer of supervisor/management training was evaluated for municipal workers in two southwestern metropolitan areas. The measurement of transfer was based upon self-reporting by the training participants who completed two surveys: one at the conclusion of training and another 30 days after training. Analysis of the data included calculations of instrument reliability, tests for normality, and multicoliniarity. Path analysis of the initial and proposed models for the transfer process was based on variable correlation, stepwise regression, and mediation tests.

      Regression tests of the transfer model showed general support for Foxon’s proposal.  However, the relationships between the stages of transfer indicated mediation that was not included in the original model.  The relationship between the intention to transfer variable and the maintained transfer variable was very strong with intention to transfer mediating the effects of all of the other variables on maintained transfer.  The relationship between initiation of transfer variable and partial transfer variable was also quite strong with partial transfer mediating the effects of the other variables on initiation of transfer.

      Based on the results of data analysis, a new model of the transfer process is proposed.  The effects of organizational support and reaction to training are included along with the stages of transfer in the new proposal.  Additionally, the partial transfer variable is redefined and placed into the process in three places: partial intention to transfer, partial initiation of transfer, and extent of transfer.