Skip to main content

Table 2 Multilevel logistic regression models for individual and household/community factors associated with girl child marriage

From: A multilevel analysis of prevalence and factors associated with female child marriage in Nigeria using the 2018 Nigeria Demographic and Health Survey data

Variables

Model 0

Model I

Model II

Model III

aOR[95% CI]

aOR[95% CI]

aOR[95% CI]

Fixed effects results

    

Individual-level variables

    

Educational level

    

 No education

 

RC

 

RC

 Primary

 

0.63***[0.49–0.81]

 

0.75*[0.58–0.97]

 Secondary and above

 

0.24***[0.20–0.30]

 

0.36***[0.29–0.46]

Currently working

    

 Unemployed

 

RC

 

RC

 Employed

 

1.23*[1.04–1.45]

 

1.16 [0.98–1.37]

Religious affiliation

    

 Christianity

 

RC

 

RC

 Islam

 

1.44**[1.15–1.80]

 

1.40**[1.09–1.81]

 Traditionalist and others

 

1.65 [0.70–3.88]

 

1.52 [0.63–3.70]

Ethnicity

    

 Hausa

 

RC

 

RC

 Yoruba

 

0.17***[0.13–0.25]

 

0.37***[0.22–0.62]

 Igbo

 

0.20***[0.14–0.29]

 

0.41**[0.23–0.74]

 Others

 

0.46***[0.37–0.57]

 

0.53***[0.41–0.68]

Ever circumcised

    

 No

 

RC

 

RC

 Yes

 

0.75 [0.49–1.16]

 

0.80 [0.52–1.24]

Parity

    

 None

 

RC

 

RC

 1–3

 

5.94***[4.74–7.46]

 

5.96***[4.73–7.52]

 4–6

 

50.54***[25.13–101.66]

 

52.39***[25.91–105.92]

Media exposure

    

 No

 

RC

 

RC

 Yes

 

0.78**[0.84–1.71]

 

0.96 [0.80–1.16]

Household/community level variables

   

Place of residence

    

 Urban

  

RC

RC

 Rural

  

0.90 [0.73–1.11]

1.00 [0.80–1.24]

Wealth index

    

 Poorest

  

RC

RC

 Poorer

  

0.92 [0.73–1.16]

1.14 [0.90–1.46]

 Middle

  

0.62***[0.49–0.80]

0.86 [0.66–1.12]

 Richer

  

0.38***[0.29–0.50]

0.59**[0.43–0.80]

 Richest

  

0.18***[0.12–0.27]

0.35***[0.23–0.54]

Sex of household head

    

 Male

  

RC

RC

 Female

  

0.71*[0.54–0.92]

0.84 [0.64–1.11]

Region

    

 North Central

  

RC

RC

 North East

  

2.40***[1.89–3.06]

1.55**[1.19–2.10]

 North West

  

3.35***[2.66–4.21]

1.59**[1.18–2.16]

 South East

  

0.48***[0.34–0.68]

0.73 [0.40–1.33]

 South South

  

1.01 [0.74–1.38]

1.31 [0.94–1.82]

 South West

  

0.60**[0.44–0.82]

0.80 [0.50–1.27]

Community literacy level

    

 Low

  

RC

RC

 Medium

  

0.67**[0.52–0.86]

0.74*[0.57–0.95]

 High

  

0.48***[0.39–0.59]

0.74**[0.59–0.92]

Community socioeconomic status

   

 Low

  

RC

RC

 High

  

1.21 [0.92–1.58]

1.04 [0.79–1.37]

Random effects results

    

 PSU variance (95% CI)

1.88 [1.50–2.34]

0.20 [0.09–0.45]

0.28 [0.15–0.49]

0.16 [0.06–0.43]

 ICC

0.36

0.06

0.08

0.05

 LR test

χ2 = 346.66, p < 0.001

χ2 = 8.06, p < 0.05

χ2 = 17.60, p < 0.001

χ2 = 5.06, p < 0.05

 Wald χ2

Reference

852.30***

649.36***

838.61***

Model fitness

    

 Log-likelihood

− 2759.72

− 2192.48

− 2396.81

− 2141.45

 AIC

5523.45

4412.95

4825.62

4338.91

 BIC

5536.28

4502.77

4928.26

4518.53

 Number of clusters

1191

1191

1191

1191

  1. Weighted NDHS, 2018
  2. Exponentiated coefficients; 95% confidence intervals in brackets; AOR: adjusted Odds Ratios; CI: Confidence Interval; RC: Reference Category
  3. *p < 0.05; **p < 0.01; ***p < 0.001
  4. AIC: Akaike’s Information Criterion; BIC: Schwarz’s Bayesian Information Criteria; PSU: Primary Sampling Unit; ICC: Intra-Class Correlation; LR Test: Likelihood ratio Test
  5. Model 0 is the null model, a baseline model without any determinant variable
  6. The model I is adjusted for individual-level variables (Educational level, currently working, Religious, Ethnicity ever circumcised, parity and media exposure)
  7. Model II is adjusted for household/community level variables (Place of residence, region, wealth index and sex of household head)
  8. Model III is the final model adjusted for both individual and household/community level variables
Back to article page

BMC Women's Health

ISSN: 1472-6874

Contact us