Does Performing the Chinese Eye Exercises Help Protect Children’s Vision? – New Evidence from Primary Schools in Rural Northwestern China

Figures & data

Table 1 Summary Statistics, by Eye-Exercise Performance Frequency

	(1)	(2)	(3)	(4)
	Full Sample	Students Performing Eye Exercises Regularly	Students Not Performing Eye Exercises Regularly	Difference = (2) – (3)
A.Visual outcomes
Uncorrected visual acuity (raw)	0.730	0.721	0.745	−0.024***
	(0.275)	(0.277)	(0.271)	[0.006]
LogMAR (defined in Equation 1(1) )	0.188	0.195	0.179	0.016***
	(0.251)	(0.250)	(0.252)	[0.005]
Vision impairment (Vimpaired, defined in Equation 2(2) )	0.212	0.224	0.194	0.030***
	(0.409)	(0.417)	(0.396)	[0.008]
Myopia (Vmyopia, defined in Equation 3(3) )	0.154	0.167	0.133	0.034***
	(0.361)	(0.373)	(0.340)	[0.008]
B.Student/family characteristics
Male (dummy, =1 if yes)	0.524	0.515	0.536	−0.021**
	(0.499)	(0.500)	(0.499)	[0.010]
Age (years)	10.062	10.065	10.060	0.005
	(1.044)	(1.009)	(1.093)	[0.216]
Father’s education (years)	7.824	7.838	7.802	0.036
	(3.321)	(3.446)	(3.127)	[0.069]
Mother’s education (years)	6.232	6.250	6.205	0.045
	(3.966)	(4.119)	(3.730)	[0.082]
Father is a migrant worker (dummy, =1 if yes)	0.454	0.440	0.476	−0.036***
	(0.498)	(0.496)	(0.500)	[0.010]
Mother is a migrant worker (dummy, =1 if yes)	0.183	0.172	0.200	−0.029***
	(0.387)	(0.377)	(0.400)	[0.008]
Household wealth (RMB, in log)	9.652	9.771	9.481	0.290***
	(0.973)	(1.019)	(0.873)	[0.021]
C.School/teacher characteristics
Teacher-student ratio	0.075	0.078	0.071	0.007***
	(0.036)	(0.035)	(0.037)	[0.001]
Proportion of classmates with eyeglasses	0.032	0.033	0.031	0.002**
	(0.036)	(0.037)	(0.035)	[0.001]
School playing eye-exercise music (dummy, =1 if yes)	0.779	0.923	0.562	0.360***
	(0.415)	(0.267)	(0.496)	[0.008]
N	9737	5805	3932

Notes: Source: Author’s survey. 1. Columns 1–3 report means and standard deviations (in parentheses) of variables; column 4 reports t-test results for differences in means between columns 2 and 3 (standard errors in brackets). 2. One US dollar = 6.81 RMB in 2012. 3. ***p<0.01, **p<0.05.

Table 2 Estimated Impacts of Regularly Performing the Chinese Eye Exercises on 4th Graders’ Visual Outcomes

Outcome Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	Uncorrected Visual Acuity (LogMAR)		Visual Impairment (Vimpaired)			Myopia (Vmyopia)
Estimator	OLS	2SLS	OLS	2SLS	Bivariate Probit	OLS	2SLS	Bivariate Probit
Reported Estimates	Coef.	Coef.	Coef.	Coef.	APE	Coef.	Coef.	APE
Performed eye exercises regularly (PCEE)	−0.005	0.033**	−0.006	0.062***	0.079*	−0.006	0.076***	0.051
	(0.008)	(0.015)	(0.011)	(0.023)	(0.043)	(0.009)	(0.020)	(0.114)
Male (dummy, =1 if yes)	−0.028***	−0.027***	−0.032***	−0.032***	−0.021***	−0.029***	−0.028***	−0.019***
	(0.006)	(0.006)	(0.010)	(0.010)	(0.005)	(0.009)	(0.009)	(0.007)
Father’s education (years)	−0.001	−0.001	−0.001	−0.001	0.000	0.000	0.000	0.000
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Mother’s education (years)	0.001*	0.001	0.001	0.001	0.001	0.001	0.000	0.000
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Father is a migrant worker (dummy, =1 if yes)	0.005	0.004	0.021**	0.018*	0.011**	0.019**	0.017**	0.012**
	(0.006)	(0.006)	(0.009)	(0.009)	(0.005)	(0.008)	(0.008)	(0.006)
Mother is a migrant worker (dummy, = 1 if yes)	−0.005	−0.012	−0.022**	−0.036***	−0.017**	−0.018*	−0.032***	−0.015*
	(0.008)	(0.008)	(0.011)	(0.011)	(0.007)	(0.010)	(0.010)	(0.008)
Household wealth (RMB, in log)	0.004	0.008**	0.010*	0.020***	0.007**	0.008	0.019***	0.005*
	(0.003)	(0.004)	(0.005)	(0.005)	(0.003)	(0.005)	(0.005)	(0.003)
Teacher-student ratio	0.171	0.275**	0.047	0.405**	0.035	0.071	0.357**	0.038
	(0.106)	(0.110)	(0.173)	(0.195)	(0.096)	(0.171)	(0.176)	(0.091)
Proportion of classmates with eyeglasses	0.398***	0.902***	0.689***	1.478***	0.304***	0.608***	1.355***	0.267***
	(0.143)	(0.136)	(0.213)	(0.227)	(0.085)	(0.199)	(0.226)	(0.095)
Constant	0.222***	0.056	0.230***	−0.078		0.178***	−0.131***
	(0.038)	(0.035)	(0.062)	(0.054)		(0.058)	(0.048)
Log (likelihood)					−8734.322			−7401.669
N	8813	8673	8772	8632	8772	8330	8197	8197
R^2(Pseudo R^2)	0.053	0.024	0.056	0.027	0.164	0.062	0.027	0.182

Notes: Source: Author’s survey. 1. The 2SLS models (columns 2, 4 and 7) use the indicator of whether a school plays the official eye-exercise music during class breaks (M_CEE) as an instrumental variable for the indicator of whether a student performed the Chinese eye exercises regularly (P_CEE). 2. Columns 6–8 exclude observations with “high myopia” (spherical equivalent refractive error ≤ −6 D)³³ or with spherical equivalent refractive error > 0. 3. Robust standard errors in parentheses, adjusted for intra-school clustering. 4. APE (average partial effects) is defined as the sample average of the individual partial effect estimated for each of the observations.²² 5. ***p<0.01, **p<0.05, *p<0.1.

Table 3 First-Stage Regression Results

Outcome Variables	(1)	(2)
	Whether Performed the Chinese Eye Exercises Regularly
Estimator	OLS	Bivariate Probit
Reported Parameters	Coef.	Coef.
IV: School plays official eye-exercise music (MCEE)	0.272***
	(0.050)
Male (dummy, =1 if yes)	−0.023***	−0.086***
	(0.009)	(0.031)
Father’s education (years)	0.003**	0.012**
	(0.001)	(0.005)
Mother’s education (years)	−0.000	0.003
	(0.001)	(0.004)
Father is a migrant worker (dummy, =1 if yes)	−0.010	−0.012
	(0.012)	(0.032)
Mother is a migrant worker (dummy, =1 if yes)	−0.016	−0.078*
	(0.013)	(0.041)
Household wealth index (RMB, in log)	0.015**	0.046***
	(0.006)	(0.018)
Teacher-student ratio	−0.122	−0.152
	(0.512)	(0.622)
Proportion of classmates wearing eyeglasses	−0.459	−0.725
	(0.600)	(0.543)
Constant	0.248*	−0.204
	(0.139)	(0.214)
N	8673	8772
R^2(Pseudo R^2)	0.337	0.126

Notes: Source: Author’s survey. 1. The F-statistic of testing the null hypothesis that the coefficient of the instrument variable (M_CEE) in column 1 is zero (ie H₀: ) is 75.13 (P-value = 0.0000). 2. One US dollar = 6.81 RMB in 2012. 3. Robust standard errors in parentheses, adjusted for intra-school clustering. 4. ***p<0.01, **p<0.05, *p<0.1.

Table 4 Coefficient Estimates of Bivariate-Probit Models

Outcome Variables	(1)	(2)
	Visual Impairment (Vimpaered)	Myopia (Vmyopia)
Reported Estimates	Coefficient	Coefficient
Performed eye exercises regularly (PCEE)	0.514*	0.396
	(0.304)	(0.944)
Male (dummy, =1 if yes)	−0.098***	−0.112**
	(0.033)	(0.049)
Father’s education (years)	−0.005	−0.001
	(0.005)	(0.006)
Mother’s education (years)	0.004	0.003
	(0.004)	(0.005)
Father is a migrant worker (dummy, =1 if yes)	0.079**	0.096***
	(0.032)	(0.037)
Mother is a migrant worker (dummy, =1 if yes)	−0.075*	−0.085
	(0.045)	(0.057)
Household wealth (RMB, in log)	0.024	0.024
	(0.018)	(0.024)
Teacher-student ratio	0.292	0.444
	(0.604)	(0.681)
Proportion of classmates with eyeglasses	2.285***	2.318***
	(0.529)	(0.581)
Constant	−1.026***	−1.183***
	(0.240)	(0.434)
	−0.324*	−0.264
	(0.197)	(0.593)
N	8772	8197
Pseudo R^2	0.164	0.182

Notes: Source: Author’s survey. 1. One US dollar = 6.81 RMB in 2012. 2. ***p<0.01, **p<0.05, *p<0.1.

Table 5 Robustness Checks

Outcome Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
	Visual Acuity (LogMAR)			Visual Impairment (Vimpaired)			Myopia (Vmyopia)
Estimator	2SLS			2SLS			2SLS
Performed eye exercises regularly (PCEE)	0.034**	0.035**	0.033**	0.063***	0.065***	0.064***	0.077***	0.081***	0.076***
	(0.015)	(0.015)	(0.015)	(0.023)	(0.023)	(0.022)	(0.019)	(0.020)	(0.019)
Standardized math score	0.004			0.010*			0.021***
	(0.003)			(0.005)			(0.005)
Classmates’ average standardized math score		0.017*			0.028*			0.044***
		(0.010)			(0.016)			(0.014)
School organized vision screening			−0.001			−0.006			−0.001
			(0.009)			(0.013)			(0.011)
Constant	0.058*	0.060*	0.056	−0.073	−0.071	−0.077	−0.121**	−0.121**	−0.131***
	(0.035)	(0.035)	(0.035)	(0.054)	(0.054)	(0.055)	(0.048)	(0.048)	(0.049)
Other control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	8672	8672	8673	8631	8631	8632	8196	8196	8197
R^2	0.024	0.025	0.024	0.027	0.027	0.027	0.030	0.029	0.027

Notes: Source: Author’s survey. 1. “Other control variables” include a male dummy, parents’ years of schooling, parents’ migration status (dummies), family wealth, teacher-to-student ratio, and the proportion of classmates wearing eyeglasses. 2. Robust standard errors in parentheses, adjusted for intra-school clustering. 3. ***p<0.01, **p<0.05, *p<0.1.

Table 6 Heterogeneity Analysis

Outcome Variables		(1)	(2)	(3)	(4)
		Performing Eye Exercises Regularly (PCEE)	Visual Acuity (LogMAR)	Visual Impairment (Vimpaired)	Myopia (Vmyopia)
Estimator			2SLS	2SLS	2SLS
Subsamples defined by different values of:
Gender	Boys [N = 5099]	0.587	0.031*	0.045	0.052**
		[0.492]	(0.018)	(0.028)	(0.024)
	Girls [N = 4638]	0.607	0.036**	0.077***	0.095***
		[0.489]	(0.018)	(0.029)	(0.024)
Father’s education	≥ 9 years [N = 5282]	0.600	0.024	0.054*	0.055**
		[0.490]	(0.018)	(0.029)	(0.025)
	< 9 years [N = 4455]	0.592	0.039**	0.068**	0.097***
		[0.492]	(0.018)	(0.027)	(0.025)
Mother’s education	> 6 years [N = 3565]	0.616	0.010	0.051	0.070**
		[0.486]	(0.021)	(0.035)	(0.030)
	≤ 6 years [N = 6172]	0.585	0.040***	0.058***	0.071***
		[0.493]	(0.015)	(0.022)	(0.020)
Is father a migrant worker?	Yes [N = 4395]	0.577	0.038**	0.071***	0.059**
		[0.494]	(0.017)	(0.026)	(0.023)
	No [N = 5276]	0.613	0.026	0.048*	0.082***
		[0.487]	(0.017)	(0.027)	(0.024)
Is mother a migrant worker?	Yes [N = 1761]	0.557	0.022	0.074*	0.078**
		[0.497]	(0.027)	(0.040)	(0.032)
	No [N = 7854]	0.605	0.036**	0.061***	0.075***
		[0.489]	(0.015)	(0.023)	(0.020)
Household wealth	> median [N = 4955]	0.650	0.033	0.057*	0.076***
		[0.477]	(0.021)	(0.032)	(0.027)
	≤ median [N = 4782]	0.540	0.033**	0.062**	0.070***
		[0.498]	(0.017)	(0.026)	(0.023)

Notes: Source: Author’s survey. 1. Column 1 reports means and standard deviations of the indicator of whether a student performed the Chinese eye exercises regularly (P_CEE) ; columns 2–4 report estimated coefficients and standard errors, adjusted for intra-school clustering . 2. ***p<0.01, **p<0.05, *p<0.1.

Tong L, Saw SM, Tan D, et al. Sensitivity and specificity of visual acuity screening for refractive errors in school children. Optom Vis Sci. 2002;79(10):650–657. doi:10.1097/00006324-200210000-0001112395920

 PubMed Web of Science ®Google Scholar

Greene W. Econometric Analysis. 8th ed. London: Pearson Education; 2018.

 Google Scholar