Table 2.

Coefficient estimates, linear regression model predicting weight-for-height z-score (Model 1) and wasting (Model 2)

Variable	Model 1 (WHZ)		Model 2 (Wasting)
Climate anomalies, 12 months before survey
Temperature	−0.0226	^**	0.0105	^***
Temperature²	0.0070		−0.0031	^**
Precipitation	0.0328	^***	0.0035	^**
Precipitation²	−0.0188	^**	0.0016

Control variables
Child’s age (months)	0.0078	^***	−0.0039	^***
Child’s age² (months)	0.0000		0.0000	^***
Child’s sex = female	0.0356	^***	−0.0147	^***
Child’s birth order	−0.0088	^**	0.0011	^†
Coresident under-5 children	0.0057		−0.0010
Mother’s age (years)	0.0002		0.0002
Mother’s education = primary+	0.1023	^***	−0.0178	^***
Residence = urban	0.0725	^***	−0.0181	^***
Historical monthly temperature, mean	−0.0516	^***	0.0053	^***
Historical monthly temperature, SD	−0.4675	^***	0.0900	^**
Historical monthly precipitation, mean	0.0002	^***	0.0000	^**
Historical monthly precipitation, SD	0.0002		0.0000

Season of interview month fixed effects	Yes		Yes
Interview decade fixed effects	Yes		Yes
Province fixed effects	Yes		Yes

R²	0.0944		0.0612
Joint test, temperature	^†		^***
Joint test, precipitation	^***		^**
Joint test, all climate variables	^***		^***

Sample size	182,272		182,272

^†

p<0.10

^**

p<0.05

^***

p<0.001

Standard errors clustered at the community level. Historical climate variables measured at the cluster level. Interview month (ordinal) ranked by historical monthly precipitation within cluster. WHZ = weight-for-height z-score.