Annex E : Summary of feedback from previous release and changes to the analysis
13 . This annex presents some of the feedback received for our previous analysis and describes the places where we have attempted to incorporate this feedback into this report .
a . Some users felt that the comparison of POLAR3 quintiles 1 and 5 might be missing the impact of the spillover effect of the Uni Connect programme on learners from underrepresented areas on the margins ( such as those in POLAR3 quintile 2 ). We incorporated this suggestion by presenting a new chart ( Figure 3 ) comparing quintiles 1 and 2 combined against quintile 5 .
b . It was also argued that the attempt to match learners in the previous analysis was missing an important interaction between school type and application rates , and the fact that this relationship differs between Uni Connect areas and other areas . For example , it might have been that changes in the application gap over time were driven by the fact that school type was unbalanced between the two groups , rather than one group having lived in a Uni Connect target area while the other did not . In response to this , as set out in our sensitivity analysis , we tried matching on school type in addition to the other matching criteria . But this made no difference to the overall conclusions of the analysis ; there was no reduction in the application gap between 2016 and 2021 .
c . Another suggestion we received was to conduct separate analyses for each English region . We decided against this because it is beyond the scope of this national evaluation ; other parts of the Uni Connect evaluation are designed to evaluate the programme at a local level .
d . Some users were unclear whether learners applying to higher education courses provided by further education colleges ( FECs ) were within scope of the UCAS Undergraduate scheme ( and therefore this analysis ). We want to clarify that applications to these higher education providers are included in the UCAS applications data . However , there is a possibility that learners studying Level 3 qualifications at an FEC , who then move on to study qualifications at Level 4 or above at the same college , will bypass the UCAS undergraduate scheme when entering higher education . These learners will therefore not be captured in our analysis of application outcomes . They will instead be captured in the new analysis of higher education entrant data , discussed in Annex C .
e . One limitation with the matching analysis is that our ability to account for differences in socioeconomic background between learners from Uni Connect areas and other areas is based solely on free school meals status . While this represents a reliable indicator of learner disadvantage , it lacks granularity because it can only divide the population of learners in two . It is unable to distinguish between different levels of disadvantage , unlike other measures such as the Index for Multiple Deprivation ( IMD ) 47 and the
47
See www . gov . uk / government / collections / english-indices-of-deprivation .
66