Table 13 KHB analysis for mediation effect.

From: Preschool education and non-cognitive skills: evidence from rural China’s junior high students

 

Monetary, time investment and family dynamics

Peer level

 

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

Variables

A lot of books at home

Exercise with students

Visit museums, zoos, science museums with parents

Going out to the movies, shows, sports games with parents

Parent-students relationship

Talk about school things with students

Talk about student’s mood with students

Visit museums, zoos, science museums with classmates

Going out to the movies, shows, sports games with classmates

Reduced model

0.382***

0.358***

0.391***

0.384***

0.382***

0.358***

0.360***

0.353***

0.357***

 

(0.079)

(0.083)

(0.081)

(0.080)

(0.078)

(0.082)

(0.081)

(0.081)

(0.081)

Full model

0.243***

0.299***

0.303***

0.305***

0.335***

0.264***

0.321***

0.260***

0.276***

 

(0.080)

(0.083)

(0.081)

(0.080)

(0.078)

(0.082)

(0.081)

(0.081)

(0.082)

Diff

0.139***

0.059***

0.088***

0.078***

0.047**

0.094***

0.039*

0.093***

0.081***

 

(0.020)

(0.015)

(0.015)

(0.016)

(0.023)

(0.020)

(0.023)

(0.016)

(0.015)

Controls

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

N

9428

8827

9361

9388

9384

8695

8651

9223

9195

  1. This table reports the results of the Karlson–Holm–Breen (KHB) decomposition analysis to examine how family investments and peer interactions mediate the relationship between preschool attendance and non-cognitive outcomes. The dependent variable in all models is non-cognitive outcomes. The reduced models estimate the total effect of preschool attendance without including mediators, while the full models include each mediator separately. The “Diff” rows show the indirect effects, representing the portion of the total effect transmitted through each mediator. All mediators show statistically significant indirect effects, with the largest reductions observed for “A lot of books at home” (indirect effect = 0.139, p < 0.01) and “Visit museums, zoos, science museums with parents” (indirect effect = 0.088, p < 0.01). These results indicate that both monetary and time investments by families, as well as peer-related activities, partially mediate the positive effect of preschool attendance on non-cognitive development. However, significant direct effects remain, suggesting that preschool has additional benefits beyond these family and peer mechanisms. Standard errors are in parentheses. *p < 0.10, **p < 0.05, ***p < 0.01.
Back to article page