Skip to main content

Table 5 Logit regression estimates for the probability of owning a bed net at household level and using a bed net at individual level

From: How equitable is bed net ownership and utilisation in Tanzania? A practical application of the principles of horizontal and vertical equity

 

Panel A: Horizontal equity analysis: Household level

(N = 1603; F < 0.0001)

Panel B: Vertical equity analysis: Individual level

(N = 8663; F < 0.0001)

Level of Analysis

All Nets

Treated nets

All Nets

Treated nets

 

Marginal/Average effects†

Marginal/Average effects

Marginal/Average effects

Marginal/Average effects

Explanatory variable

dF/dx

p-value

dF/dx

p-value

dF/dx

p-value

dF/dx

p-value

Family size

0.001

0.973

-0.002

0.656

-0.274

<0.001*

-0.008

<0.001*

Male

-0.032

0.555

-0.01

0.859

-0.056

<0.001*

-0.02

0.023*

Urban

0.228

<0.001*

0.205

<0.001*

0.315

<0.001*

0.192

<0.001*

Married

0.007

0.887

0.043

0.408

0.031

0.051*

0.04

<0.001*

Sambaa

0.072

0.214

0.031

0.496

-0.087

0.003*

0.011

0.546

Digo

-0.085

0.108

-0.102

0.012*

-0.176

0.001*

-0.086

0.001*

Bondei

0.079

0.248

-0.051

0.314

-0.05

0.15

-0.056

0.005*

Other ethnic group

0.006

0.908

-0.078

0.048*

-0.09

0.001*

-0.056

<0.001*

Age of household head

-0.002

0.097

-0.001

0.133

n/a

n/a

n/a

n/a

Under-five

n/a

n/a

n/a

n/a

0.157

<0.001*

0.06

<0.001*

Using other prevention measures

-0.232

<0.001*

-0.006

0.81

-0.182

<0.001*

-0.004

0.684

Education

0.013

0.008*

0.02

<0.001*

0.018

<0.001*

0.02

<0.001*

Wealth

0.11

<0.001*

0.053

<0.001*

0.091

0.001*

0.043

<0.001*

Education-squared

-0.002

0.028*

-0.001

0.073

n/a

n/a

n/a

n/a

Wealth-squared

-0.008

<0.001*

-0.004

<0.001*

-0.007

0.001*

-0.004

<0.001*

Poor Road

0.012

0.698

-0.07

0.085

n/a

n/a

n/a

n/a

Market centre

-0.016

0.764

-0.008

0.906

n/a

n/a

n/a

n/a

Constant

-

-

-

-

-

-

-

-

  1. * Coefficient significant at 5% level
  2. †marginal (average) effects are computed using the dprobit command and reflect the effect of a unity change in the continuous (dummy) variable on the dependent variable.