Skip to main content

Table 4 Logistic regression results, age and gender subsamples

From: Using the Uganda National Panel Survey to analyze the effect of staple food consumption on undernourishment in Ugandan children

Stunting

 

Full Sample

12mos or less

Greater than 12mos

Males only

Females only

Number of obs (n)

6101

955

5146

3066

3035

Number of children

3427

936

2993

1679

1751

 

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

 SBS

0.007*

(0.002)

1.007

−0.003

(0.006)

0.997

0.009*

(0.003)

1.009

0.006*

(0.003)

1.006

0.008*

(0.004)

1.008

 Spending

−0.005*

(0.002)

0.995

−0.005

(0.004)

0.995

−0.005*

(0.002)

0.995

−0.008*

(0.002)

0.992

−0.003

(0.002)

0.997

 Average age

−0.008

(0.008)

0.992

−0.015

(0.018)

0.985

−0.011

(0.009)

0.989

0.009

(0.009)

1.009

−0.034*

(0.014)

0.967

 Urban

−0.709*

(0.169)

0.492

−0.481

(0.389)

0.618

−0.782*

(0.196)

0.457

−0.849*

(0.228)

0.428

−0.578*

(0.248)

0.561

 Mother present

−0.325*

(0.162)

0.723

0.110

(0.773)

1.116

−0.205

(0.182)

0.815

−0.129

(0.218)

0.879

−0.558*

(0.244)

0.572

 Male

0.404*

(0.098)

1.498

0.275

(0.249)

1.317

0.431*

(0.114)

1.539

---

---

---

---

Log pseudolikelihood

−3053.290

−333.994

−2658.276

−1621.735

−1420.928

Wald X2 =

88.13

8.46

86.76

48.18

42.14

Prob > X2 =

0.000

0.956

0.000

0.0004

0.000

Wasting

 

Full Sample

12mos or less

Greater than 12mos

Males only

Females only

Number of obs (n)

6101

955

5146

3066

3035

Number of children

3427

936

2993

1679

1751

 

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

Coef.

(std. error)

Odds Ratio

 SBS

0.011*

(0.005)

1.011

0.015

(0.012)

1.015

0.008

(0.006)

1.008

0.006

(0.007)

1.006

0.017*

(0.008)

1.017

 Food insecurity

0.332

(0.198)

1.393

0.291

(0.451)

1.338

0.351

(0.225)

1.421

0.148

(0.261)

1.159

0.595*

(0.298)

1.813

 Household head educated

−0.402*

(0.205)

0.669

−0.385

(0.440)

0.680

−0.357

(0.226)

0.700

−0.203

(0.268)

0.816

−0.660*

(0.316)

0.517

 Male

0.445*

(0.184)

1.560

0.568

(0.448)

1.765

0.436*

(0.206)

1.547

---

---

---

---

Log pseudolikelihood

−807.927

−200.675

−596.030

−463.902

−336.692

Wald X2 =

56.79

6.12

45.08

34.9

30.55

Prob > X2 =

0.000

0.992

0.000

0.004

0.015

  1. *indicates p ≤ 0.05. Each regression includes the same covariates as in Table 3 with the exception of the males only and females only regressions which did not include the gender indicator. For ease of exposition, only those covariates with significant coefficients in at least one regression are reported here. Complete results are available upon request. Heteroskedasticity-robust standard errors are clustered at the person-level. OR = ecoef = ratio of the odds of stunting/wasting after a unit increase in the corresponding covariate to the odds before the increase