Desktop version

Home arrow Economics arrow Children and Forced Migration: Durable Solutions During Transient Years

Results

The empirical results of all three models are presented here, depending on the outcome of interest. For the sake of parsimony, the focus is solely on the variable of interest, whether the household is internally displaced or not. Still, it also is indicated if the previously explained household-level covariates and location fixed effects are included in the model specification. The results of all full models can be found in the Appendix.

Table 2 presents the estimates for the effect of displacement on school attendance for those school-age children of the household, reporting the

Table 2 School Attendance

Base

category:

none

All school-age children

Only school-age boys

Internally

displaced

  • 0.4937***
  • (0.0740)
  • 0.4838***
  • (0.0849)
  • 0.7608
  • (0.1628)
  • 0.6013***
  • (0.1099)
  • 0.6008***
  • (0.1175)
  • 0.8363
  • (0.1976)

Household-

level

covariates

No

Yes

Yes

No

Yes

Yes

Location- based FE

No

No

Yes

No

No

Yes

Adj.

R-squared

0.0102

0.1158

0.1742

0.0102

0.1158

0.1742

N

1020

1020

1020

1020

1020

1020

Notes: *p < 0.10, **p < 0.05, ***p < 0.01 Full results are presented in the Appendix.

relative risk ratio[1] in comparison to the base category of no school attendance and robust standard errors in parentheses. Across both response categories, being a child within an internally displaced household yields a statistically significant and negative result for school attendance when not controlling for location fixed effects. None of the estimates, however, are robust to the inclusion of location fixed effects, including whether the household is located in a rural setting along with the specific province of the current location. As such, even though there is indication that internal displacement plays a role in school attendance, the relationship is not uniform across all local settings. It seems that is possible to deduce that the dynamics within the communities of displacement are more likely to influence educational outcomes regardless of whether the household is displaced. This may be because of the lack local services within the community (e.g., schools), which is not uncommon throughout Afghanistan; or perhaps areas receiving the displaced are impoverished in general, which results in a substitution of schooling for income-generating activities by all children.

Table 3 provides the results for the effect of displacement on food insecurity, again reporting the relative risk ratio in comparison to the base category of never having problems satisfying food needs, along with the standard errors in parentheses. Contrary to the former model, robust results are found here across the model specification that a displaced household has a greater likelihood of being food insecure at both levels, rarely having problems satisfying food needs (1-2 times/month), and often having problems satisfying food needs (> 3 times/month). Additionally, when comparing across these two distinct levels of food insecurity, displaced households are seen to have greater odds of being more food insecure, although this result does not hold after controlling for location. Regardless, the findings show a clear picture that displaced households are more food insecure than those that never migrated.

To finish, Table 4 illustrates the effect of displacement on dietary diversity by taking into consideration the number of times the household had eaten meat in the week prior. Because a zero-censored tobit model is employed, the marginal effects are reported for both the probability of having eaten meat as well as the number of times, if the former is greater

Table В Food Insecurity

Base category: never

Rarely (1-2 times/month)

Often (> 3 times/month)

Internally displaced

  • 1.9005***
  • (0.3522)
  • 1.9340***
  • (0.3836)
  • 2.0554***
  • (0.4628)
  • 2.0895***
  • (0.3964)
  • 2.1239***
  • (0.4434)
  • 2.0358***
  • (0.4717)

Household-level

covariates

No

Yes

Yes

No

Yes

Yes

Location-based FE

No

No

Yes

No

No

Yes

Adj. R-squared

0.0080

0.0533

0.0823

0.0080

0.0533

0.0823

N

1020

1020

1020

1020

1020

1020

Table 4 Dietary Diversity

Probability of eating meat

Number of times eating meat (if > 0)

Internally displaced

  • -0.1186***
  • (0.0270)
  • -0.1161***
  • (0.0286)
  • -0.1700***
  • (0.0331)
  • -0.2419***
  • (0.0558)
  • -0.2189***
  • (0.0556)
  • -0.2986***
  • (0.0599)

Household-level

covariates

No

Yes

Yes

No

Yes

Yes

Location-based FE

No

No

Yes

No

No

Yes

Adj. R-squared

0.0058

0.0518

0.0900

0.0058

0.0518

0.0900

N

1020

1020

1020

1020

1020

1020

than zero. Again, the estimates are robust across all model specifications with a displaced household being less likely to have eaten meat and, if so, the occupants ate less. In the full model, which takes into account household-level covariates and location fixed effects as indicated in columns three and six, a displaced household is on average 17 % less likely to have eaten meat in the week prior to the survey. For those displaced households that had eaten meat, it was eaten 30 % less in comparison to their nondisplaced counterparts.

  • [1] Keep in mind, when reporting “relative risk ratios,” a result greater than one indicates a positiverelationship while a result less than one indicates a negative relationship.
 
Source
< Prev   CONTENTS   Source   Next >

Related topics