The remittance behaviour of Kenyan sibling migrants

Table 7 The probability of remitting and the volume of remittances (control variables)

Variables	Probability of remitting	Remittance level
=1 if employed	0.415***	49,555***
=1 if employed	(0.0298)	(12,471)
=1 if self-employed	0.268***	24,387*
=1 if self-employed	(0.0563)	(14,351)
=1 if lives with relatives and/or friends	−0.0829**	21,568
=1 if lives with relatives and/or friends	(0.0388)	(22,115)
=1 if migration length is 1 to 3 years	0.0213**	9515***
=1 if migration length is 1 to 3 years	(0.00968)	(3314)
=1 if receives transfers from the household of origin	−0.186***	−9110
=1 if receives transfers from the household of origin	(0.0433)	(16,186)
=1 if located in external destination	0.0335	79,129***
=1 if located in external destination	(0.0324)	(13,914)
=1 if the head is unemployed	−0.100*	−12,951
=1 if the head is unemployed	(0.0543)	(34,623)
Proportion of children in the household of origin	−0.357***	−63,352
Proportion of children in the household of origin	(0.137)	(38,669)
=1 if head is male^Ϯ	−0.0776*
=1 if head is male^Ϯ	(0.0351)
Size of the household of origin^Ϯ	0.0321***
Size of the household of origin^Ϯ	(0.0087)
Birth rank	−0.0437***	−724
Birth rank	(0.0105)	(4467)
N	1092	542
R-squared		0.235
Pseudo R-squared	0.324

Notes to the table: (i) *, **, and *** denote statistical significance from zero at the 10, 5, and 1 % levels, respectively using two-tailed tests. (ii) Robust standard errors clustered by the household are reported in parentheses. (iii) ^Ϯ These variables are not included in the OLS regression model to prevent over-parameterizing the model. The variables were originally used as identification variables in the Heckman selection model and were found to have statistically insignificant effects on the level of remittances sent. (iv) Other variables are included in these models but not shown here in order to conserve space (see text for details)