Skip to main content
. 2017 Aug 29;7:9813. doi: 10.1038/s41598-017-10230-3

Figure 5.

Figure 5

Subgroup meta-analysis for the effect of different sample size. Note: The sample size of the included studies is different. In our study, a sample size of lower than 1000 juveniles was considered as small sample size. Accordingly, a sample size of higher than 1000 juveniles was thought as large sample. In spite of different sample size, the meta-analysis still showed that juveniles with sleep problems still hold a higher risk of being injured than juveniles without sleep problems. For the study with small sample size, the heterogeneity would be lower.