Skip to main content

Table 5 Marginal effects from probit regression of household and household head characteristics accounting for type I attrition

From: The Egypt labor market panel survey: introducing the 2012 round

Base probability

0.149

Household composition in 2006

 

    No. of children 0 to 5 in household

-0.003

    No. of children 6 to 14 in household

-0.023***

    No. of working age males in household

-0.031***

    No. of working age females in household

-0.020**

    No. of elderly males in household

-0.040*

    No. of elderly females in household

-0.011

Region of residence in 2006 (Greater Cairo omitted)

    Alex & Suez Canal

-0.037**

    Urban Lower

-0.077***

    Urban Upper

-0.086***

    Rural Lower

-0.127***

    Rural Upper

-0.117***

Home ownership status in 2006 (Own omitted)

 

    Old rental law

0.065***

    New rental law

0.292***

HH head age group in 2006 (35+ omitted)

 

    15-24

0.017

    25-34

0.034*

HH head gender (Male omitted)

 

    Female

0.019

HH head marital status in 2006 (Single omitted)

 

    Married

0.069*

    Divorced

0.105

    Widowed

0.038

HH head education in 2006 (Illiterate or read & write omitted)

 

    Primary/Preparatory

0.002

    Secondary

0.026*

    University

0.084***

HH head status/sector of employment in 2006 (Government omitted)

 

    Formal private regular wage

0.067***

    Informal private regular wage

0.069***

    Informal temporary

-0.036

    HH enterprise worker

0.008

    Out of labor force/manpower

0.009

    Unemployed

0.043

No. of households

8,164

  1. Dependent variable: probability that the original household attrited (was not found).
  2. Asterisks indicate significance of marginal effects: *p < 0.05; **p < 0.01; ***p < 0.001.
  3. Marginal effects are estimated at the omitted category for binary and categorical variables, and at the mean for continuous variables.