Skip to main content

Table 3 Results of multinomial logistic regression for factors affecting the choice among three clinical specialties (with “basic science’ as the base outcome)

From: Factors affecting the choice of medical specialties in Turkiye: an analysis based on cross-sectional survey of medical graduates

Clinical specialization selected and factors affecting choice of specialty

Relative risk ratio

Standard error

Z

P > z

Basic medical sciences (base outcome)

Internal medicine

Male

1.406

0.451

1.060

0.288

Graduated more than a year ago

0.496**

0.159

-2.180

0.029

Self-perceived competency is important

2.308**

0.808

2.390

0.017

Ability to serve the community by specialty selection is important

2.924**

0.966

3.250

0.001

Family expectation on selection of specialty is important

2.009*

0.828

1.690

0.091

Income earning potential is important

1.164

0.368

0.480

0.631

Potential risk of violence against healthcare providers important

0.127**

0.058

-4.550

0.000

High level of performance in medical school

0.835

0.267

-0.560

0.572

Risk of malpractice and other legal issues important

0.753

0.308

-0.690

0.487

Constant term (baseline Relative Risk Ratio)

2.892**

1.373

2.240

0.025

Surgery

Male

2.886**

1.107

2.760

0.006

Graduated more than a year ago

0.377**

0.148

-2.480

0.013

Self-perceived competency is important

5.782**

2.879

3.520

0.000

Ability to serve the community by specialty selection is important

2.338**

0.919

2.160

0.031

Family expectation on selection of specialty is important

2.695**

1.297

2.060

0.039

Income earning potential is important

2.385**

0.920

2.250

0.024

Potential risk of violence against healthcare providers important

0.110**

0.056

-4.300

0.000

High level of performance in medical school

0.341**

0.148

-2.470

0.013

Risk of malpractice and other legal issues important

0.254**

0.115

-3.020

0.003

Constant term (baseline Relative Risk Ratio)

0.902

0.546

-0.170

0.865

  1. Multinomial logistic regression model: Number of observations = 333, LR χ2(18) = 160.93, Prob > χ2 = 0.0000, Log likelihood = -278.047, Pseudo R2 = 0.2244
  2. ** Significant at 5% level or better
  3. * Significant at 10% level