【文章內(nèi)容簡(jiǎn)介】
Qdf)/Q)*100% Q is the chisquared statistic。 df is its degrees of freedom(自由度). This describes the percentage of the variability in effect estimates that is due to heterogeneity rather than sampling error (chance).,?,異質(zhì)性檢驗(yàn)(heterogeneity test),A rough guide to interpretation is as follows: 0% to 40%: might not be important。 30% to 60%: may represent moderate heterogeneity。 50% to 90%: may represent substantial heterogeneity。 75% to 100%: considerable heterogeneity.,?,異質(zhì)性檢驗(yàn),?,?,異質(zhì)性分析與處理方法,無(wú)異質(zhì)性或I2小于50%(即異質(zhì)性較小,可以忽略) 固定效應(yīng)模型(fixed effects model),?,固定效應(yīng)模型(fixed effects model),?,關(guān)于隨機(jī)效應(yīng)模型與固定效應(yīng)模型,固定效應(yīng)模型(fixed effects model) 前提是假定全部研究結(jié)果的方向與效應(yīng)大小基本相同,即各獨(dú)立研究的結(jié)果趨于一致,一致性檢驗(yàn)差異無(wú)顯著性。因此固定效應(yīng)模型適用于各獨(dú)立研究間無(wú)差異,或差異較小的研究。,?,關(guān)于隨機(jī)效應(yīng)模型與固定效應(yīng)模型,隨機(jī)效應(yīng)模型(random effects model) 隨機(jī)效應(yīng)模型