Income Deprivation Affecting Children Index ( IDACI ). 48 However , area-based measures such as IMD and IDACI were deliberately avoided due to the existing area-based comparison being made between learners from Uni Connect target areas and learners from other areas . Nonetheless , as described in our sensitivity analysis , when IMD and IDACI were included as alternative matching criteria to free school meals status , the conclusions of this analysis were unchanged ; there was no reduction in the application rate gap between 2016 and 2021 after differences in matched characteristics were taken into account . f .
The purpose of this report is to evaluate the long-term impact of the Uni Connect programme in achieving one of its stated aims : to reduce the gap in higher education participation between the most and least represented groups of learners . However , beyond comparing the headline application rates of the most and least represented areas in England ( as defined by POLAR3 quintiles ), this report also explores whether the Uni Connect programme appears to be associated with a reduction in this gap among learners for whom we would expect the impact of the programme ( if any ) to be most evident . In the previous analysis , while we made clear that learners from target areas had not necessarily been engaged by the Uni Connect programme , we presented no attempt to quantify the extent of this engagement in target areas . This update includes a new section on this topic , finding that less than one in five learners in Uni Connect target areas in the most recent cohort received the full amount of engagement intended in the programme design . This is important , because it suggests that the impact of the Uni Connect programme may be limited by its scale .
g . We received feedback that the decision to apply to higher education was not the only outcome of interest in the Uni Connect programme , although it was most closely related to the programme aims after the placed rate itself . In light of this feedback , we have now conducted analysis of applications to more selective – or ‘ high tariff ’– providers , in addition to the four existing application outcomes which were analysed in the previous report .
h . We updated the criteria for excluding learners from non-Uni Connect areas from the matched counterfactual analysis , in order to take account of the timing of the schoolbased engagement received by each cohort . i .
Although we did not receive any external feedback on our statistical model , we have reconsidered this part of the analysis . Given the findings of two statistical approaches continue to be extremely similar , in the interests of interpretability , we opted to present only the findings from the matching analysis in this update . As described in our sensitivity analysis , we did nonetheless run the statistical model described in our previous report for quality assurance purposes , but the overall conclusions remained unchanged . The results from the statistical model also happen to relate to a broader population of learners , including some who would otherwise have been discarded from the matching analysis because they were too dissimilar from the group of learners living in Uni Connect areas . We felt that the matching analysis was more closely aligned with the aims of this evaluation for this reason .
48
See opendatacommunities . org / def / concept / general-concepts / imd / idaci .
67