Key results of regression (1) and (2)
| Variable | Longo et al8 | New model (2) | ||||
| 2017/2018 | 2018/2019 | 2019/2020 | 2017/2018 | 2018/2019 | 2019/2020 | |
| Public adult social care expenditure per user | 0.003*** | 0.002** | 0.002** | 0.006*** | 0.005*** | 0.009*** |
| (0.001) | (0.001) | (0.001) | (0.002) | (0.002) | (0.003) | |
| Public adult social care expenditure per user squared | 0.0004** | 0.0003* | 0.0004** | |||
| (0.0002) | (0.0002) | (0.0002) | ||||
| Observations | 52 602 | 55 570 | 50 441 | 52 602 | 55 570 | 50 441 |
| F-test of expenditure and its square’s p value | – | – | – | 0.007 | 0.004 | 0.002 |
| First stage Kleibergen-Paap rk Wald F statistic | 434.3 | 398.6 | 408.9 | 17.7 | 10.9 | 7.4 |
| Over-identification test’s p value | 0.595 | 0.715 | 0.777 | 0.191 | 0.871 | 0.334 |
Longo et al8=regression (1) as proposed by Longo et al,8 new model=regression (2) in this paper.
***P value <0.01, **p value <0.05, *p value <0.10.
The dependent variable is the user care-related quality of life measured at the individual level. Results on control variables are not reported in this table. The instrumental variables are the council tax base per user for the model proposed by Longo et al,8 and the council tax base per user, its square and LA type dummies for the new model (2). Following Longo et al,8 the over-identification test in their model is run by using the business rates tax base per user and the area cost adjustment index as additional instruments. All regressions are weighted using the survey weight. SEs are clustered within LAs and strata, and they are reported in parenthesis.