Skip to main content

Table 3 Logistic regression results for the model including interaction terms with province, area, and gender

From: Cigarette smoking and reasons for leaving school among school dropouts in South Africa

   

95% Confidence Interval

 
 

B

S.E.

Lower

Exp (B)

Upper

p-value

Kwazulu Natal (ref Gauteng)

−1.082*

.328

0.178

.339

0.644

.001

Mpumalanga

0.595

.360

0.896

1.813

3.667

.098

Western Cape

−0.786*

.343

0.233

.456

0.893

.022

Urban (ref rural)

0.406

.312

0.815

1.501

2.765

.193

Peri-urban

0.511

.337

0.861

1.667

3.229

.130

Timing of the dropout

−0.089*

.025

0.872

.915

0.960

.000

Coloured (ref black African)

1.020*

.127

2.163

2.772

3.553

.000

Indian

0.245

.332

0.667

1.277

2.447

.461

White

0.388

.337

0.761

1.474

2.856

.250

Other

0.205

.269

0.724

1.227

2.080

.447

Boys versus Girls

−0.903*

.416

0.179

.405

0.917

.030

No reason for leaving school

−0.376

.565

0.227

.687

2.079

.506

Being pregnant or made someone pregnant

0.593

.552

0.614

1.810

5.336

.282

Working to help the family

0.033

.520

0.373

1.034

2.863

.949

Not enough money to pay for school fees

−0.065

.631

0.272

.937

3.227

.918

Had to help with looking after the house and siblings

0.072

.654

0.298

1.074

3.870

.913

Problems with your school work, teachers or the learners

0.159

.571

0.383

1.172

3.591

.781

The school was too far

−0.356

.672

0.188

.701

2.615

.596

Other

−0.198

.602

0.252

.820

2.671

.742

Gender * No reason for leaving school

0.159

.411

0.524

1.173

2.625

.699

Gender * Being pregnant or made someone pregnant

−0.288

.390

0.349

.750

1.611

.461

Gender * Working to help the family

−0.079

.378

0.441

.924

1.936

.834

Gender * Not enough money to pay for school fees

−0.116

.446

0.371

.890

2.135

.795

Gender * Had to help with looking after the house and siblings

−0.049

.462

0.385

.952

2.355

.916

Gender * Problems with your school work, teachers or the learners

−0.329

.422

0.315

.720

1.647

.437

Gender * The school was too far

0.064

.485

0.412

1.066

2.759

.895

Gender * Other

−0.326

.409

0.323

.721

1.610

.425

Being pregnant or made someone pregnant *Urban (rural ref)

−0.613*

.240

0.338

.542

0.867

.011

Being pregnant or made someone pregnant * Peri-urban

−0.246

.262

0.468

.782

1.308

.349

Not enough money to pay for school fees * Urban (rural ref)

1.449

.734

1.011

4.259

17.942

.048

Not enough money to pay for school fees *Peri-urban

−0.221

.747

0.185

.801

3.464

.767

Other * Kwazulu Natal (ref Gauteng)

0.595

.336

0.939

1.814

3.505

.077

Other * Mpumalanga

1.325*

.360

1.858

3.761

7.616

.000

Other * Western Cape

0.353

.341

0.729

1.423

2.778

.302

Gender * Kwazulu Natal (ref Gauteng)

0.125

.225

0.728

1.133

1.761

.580

Gender * Mpumalanga

−0.996*

.255

0.224

.369

0.609

.000

Gender * Western Cape

0.532*

.231

1.083

1.703

2.676

.021

Gender * Urban (rural ref)

0.059

.212

0.701

1.060

1.605

.781

Gender * Peri-urban

−0.219

.231

0.511

.803

1.263

.342

Gender * Not enough money to pay for school fees *Urban (rural ref)

−1.098*

.511

0.123

.334

0.907

.032

Gender * Not enough money to pay for school fees * Peri-urban

0.283

.500

0.498

1.328

3.541

.571

Constant

1.401

.569

 

4.057

 

.014

  1. Multivariate logistic regression used to generate p-values, *p < .05 indicates significance, Beta (B), Standard error (S.E)
Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us