Because the effect of deworming is small and becomes negative over time, our best guess finds that the overall cost-effectiveness of deworming is negative. Even under more generous assumptions (but still plausible according to this data), deworming is less cost-effective than StrongMinds. Therefore, we do not recommend any deworming charities at this time. To overturn this conclusion, proponents of deworming would either need to (1) appeal to different SWB data (we’re not aware of any) or (2) appeal to a non-SWB method of comparison which concludes that deworming is more cost-effective than StrongMinds.
Couldn’t you estimate the causal effects of deworming on consumption (or income or wealth) and combine with the causal effects of consumption on SWB? (Did HLI do this before?) My best guess would still be that the effect is positive, but the study here was underpowered.
Maybe an alternative analysis would be to estimate the smallest positive effect size on SWB that should have been detectable with a study of this size and design (with a given p-value cutoff), and use that to get an approximate upper bound estimate on the cost-effectiveness of deworming on SWB.
Couldn’t you estimate the causal effects of deworming on consumption (or income or wealth) and combine with the causal effects of consumption on SWB? (Did HLI do this before?) My best guess would still be that the effect is positive, but the study here was underpowered.
We did this in section 3.4 of the Appendix. Converting from economic benefits to SWB came up with 28 WELLBYs per $1000. This is better than GiveDirectly (8 WELLBYs per $1000) but less than StrongMinds (77 WELLBYs per $1000).
Maybe an alternative analysis would be to estimate the smallest positive effect size on SWB that should have been detectable with a study of this size (with a given p-value cutoff, say 5% or lower), and use that as an approximate upper bound for the cost-effectiveness of deworming.
We did a variant of this in a previous version of the report (it didn’t seem to make it into the final cut), where we estimated whether we could detect if deworming in the KLPS had a large enough effect to be more cost-effective than StrongMinds. Assuming (generously) that the effects of deworming couldn’t become negative, then the initial would need to be 0.08 SDs for deworming to beat StrongMinds, which we had 98% power to detect.
Couldn’t you estimate the causal effects of deworming on consumption (or income or wealth) and combine with the causal effects of consumption on SWB? (Did HLI do this before?) My best guess would still be that the effect is positive, but the study here was underpowered.
Maybe an alternative analysis would be to estimate the smallest positive effect size on SWB that should have been detectable with a study of this size and design (with a given p-value cutoff), and use that to get an approximate upper bound estimate on the cost-effectiveness of deworming on SWB.
We did this in section 3.4 of the Appendix. Converting from economic benefits to SWB came up with 28 WELLBYs per $1000. This is better than GiveDirectly (8 WELLBYs per $1000) but less than StrongMinds (77 WELLBYs per $1000).
We did a variant of this in a previous version of the report (it didn’t seem to make it into the final cut), where we estimated whether we could detect if deworming in the KLPS had a large enough effect to be more cost-effective than StrongMinds. Assuming (generously) that the effects of deworming couldn’t become negative, then the initial would need to be 0.08 SDs for deworming to beat StrongMinds, which we had 98% power to detect.