Skip to main content

Table 4 Ordinary least squares regression models of treatment effects on handwashing outcomes among children (as reported by caregivers)

From: Washing with hope: evidence of improved handwashing among children in South Africa from a pilot study of a novel soap technology

Panel A: Handwashing after toilet

A1

A2

A3

A4

A5

Handwashing score [1–10]

Handwashing score [1–10]

Handwashing score [1–10]

Handwashing score [1–10]

≥8 on scale [0/1]

Treatment: HOPE SOAP©

− 0.01a

0.11

0.20

1.29

0.08

 

(−0.69–0.67)

(−0.62–0.84)

(−0.49–0.88)

(−0.33–2.91)

(−0.06–0.23)

bBaseline handwashing measure

  

0.22***

0.32***

0.29***

   

(0.10–0.34)

(0.15–0.50)

(0.15–0.43)

Treatment*Baseline handwashing measure

   

−0.19*

 
    

(−0.41–0.03)

 

Controls

No

Yes

Yes

Yes

Yes

Observations

249

247

242

242

242

R-squared

0.00

0.08

0.16

0.17

0.16

Panel B: Handwashing before meal

B1

B2

B3

B4

B5

Score 1–10

Score 1–10

Score 1–10

Score 1–10

≥8 on scale

Treatment: HOPE SOAP©

−0.10

0.27

0.28

1.14

0.05

 

(−0.86–0.65)

(− 0.49–1.03)

(−0.46–1.02)

(− 0.46–2.73)

(− 0.08–0.19)

bBaseline handwashing measure

  

0.13**

0.22**

0.10

   

(0.01–0.25)

(0.04–0.39)

(− 0.05–0.26)

Treatment*Baseline handwashing measure

   

− 0.16

 
    

(−0.40–0.09)

 

Controls

No

Yes

Yes

Yes

Yes

Observations

249

247

244

244

244

R-squared

0.00

0.09

0.11

0.11

0.12

Panel C: General soap usage when handwashing

C1

C2

C3

C4

C5

 

Score 1–10

Score 1–10

Score 1–10

Score 1–10

≥8 on scale

Treatment: HOPE SOAP©

0.31

0.42*

0.41

0.91

0.14**

 

(−0.27–0.89)

(−0.07–0.91)

(− 0.09–0.90)

(− 0.89–2.70)

(0.03–0.24)

bBaseline handwashing measure

  

0.13**

0.17*

0.15**

   

(0.01–0.26)

(−0.02–0.36)

(0.02–0.28)

Treatment*Baseline handwashing measure

   

− 0.07

 
    

(−0.31–0.17)

 

Controls

No

Yes

Yes

Yes

Yes

Observations

249

247

236

236

236

R-squared

0.01

0.07

0.10

0.10

0.13

  1. aBeta coefficient presented followed by 95% confidence intervals in parentheses
  2. bBaseline equivalent of dependent variable
  3. *** p < 0.01, ** p < 0.05, * p < 0.1
  4. Additional controls included but not reported: Female, age, household size, number of children in HH, asset ownership, piped water available in HH, HH limits water use; soap observed in HH; HH received hygiene training; caregiver depressed/anxious; child had difficulty opening tap; child cannot reach taps; child’s hands too small for soap. The full models, with coefficients for all control variables, are presented in Additional file 1: Tables S2, S3 and S4