Skip to main content

Table 2 Association between HIV self-test kit sales price and observed customer characteristics

From: Sales and pricing decisions for HIV self-test kits among local drug shops in Tanzania: a prospective cohort study

Linear regression models of log price of test kits sold (Coefficient, standard error)

Model

1

2

3

4

5

Predictors

Age

Sex

Age & Sex

All

All (age-sex)

Customer age

 

   

 - 15–19

ref

ref

ref

ref

 - 20–24

0.617** (0.168)

0.367* (0.151)

0.150** (0.044)

0.082 (0.046)

 - 25–34

0.817*** (0.193)

0.252 (0.133)

0.231** (0.062)

0.107** (0.037)

 - 35–44

0.884*** (0.207)

0.338* (0.151)

0.198** (0.056)

0.065 (0.057)

 - 45–54

1.069*** (0.204)

0.535** (0.157)

0.219* (0.086)

0.120* (0.054)

Customer sex (Male)

−0.049 (0.075)

−0.608** (0.202)

−0.002 (0.026)

−0.145* (0.059)

P-values from F-tests of joint significance

 Age

< 0.001

0.031

0.013

0.029

 Sex

0.518

0.007

0.927

0.023

 Age x Sex interaction

0.079

0.233

 Shop

< 0.001

< 0.001

 Week

0.023

0.042

Control variables included in model

 - Treatment status

X

X

X

X

X

 - Shop fixed effects

X

X

 - Week fixed effects

X

X

R2

0.166

0.0147

0.195

0.802

0.803

N

506

499

492

492

492

  1. * p < 0.05, **p < 0.01, ***p < 0.001. Standard errors are clustered at the shop level. Only coefficients for customer age and sex are reported for brevity