# Alejandro1 comments on 2014 Less Wrong Census/​Survey—Call For Critiques/​Questions

• Is the “Birth Month” bonus ques­tion just to sort peo­ple ar­bi­trar­ily into groups to do statis­tics, or to find cor­re­la­tions be­tween birth month and other traits? If the lat­ter, since the causal mechanism is al­most cer­tainly sea­sonal weather, the ques­tion should ask di­rectly for sea­sonal weather at birth to avoid South Hemi­sphere con­founders.

• I have no idea what the sea­sonal weather was like when I was born—but I know which hemi­sphere it hap­pened in.

• the causal mechanism is al­most cer­tainly sea­sonal weather

Not the only pos­si­ble one. If Alice was born in Jan­uary and Bob was born in De­cem­ber, she will be 11 months older than him when they start go­ing to school (and their class­mates will be in av­er­age 5.5 months younger than her and 5.5 months older than him), which I hear can make a differ­ence.

avoid South Hemi­sphere confounders

The sur­vey already asks for coun­try. Sure, some peo­ple will have been born and grown up in a coun­try in a hemi­sphere other than that they “most iden­tify with” to­day, but they’ll prob­a­bly be a small enough minor­ity that they wouldn’t screw up the statis­tics too much.

• If Alice was born in Jan­uary and Bob was born in De­cem­ber, she will be 11 months older than him when they start go­ing to school (and their class­mates will be in av­er­age 5.5 months younger than her and 5.5 months older than him), which I hear can make a differ­ence.

I think this way of sort­ing classes by cal­en­dar year of birth might also be six months shifted in differ­ent hemi­spheres (or per­haps vary with coun­try in more capri­cious ways). IIRC, in Ar­gentina my classes had peo­ple born from one July to the fol­low­ing June, not from one Jan­uary to the fol­low­ing De­cem­ber.

• In Mas­sachusetts, USA, the classes of peo­ple were from one Septem­ber to the fol­low­ing Au­gust, be­ing based on the day classes started.