2020年中国高等教育规模预测偏差成因分析及重估

On the Sources of Bias in the Prediction of China's Higher Education Scale in 2020 and Its Re-evaluation

2010年公布的《国家中长期教育改革和发展规划纲要(2010-2020年)》明确提出:"到2020年,高等教育毛入学率达到40%。"但是,教育部最新统计数据显示,我国2015年的高等教育毛入学率已经提前5年超前实现了40%的目标。为什么会出现如此大的偏差?结合对各学者预测内容的梳理,笔者认为原因主要有三:一是信息更新约束与模型差异;二是理性偏误;三是不可预测的政策因素。在连续毛入学率预测数据中,预测误差较小的学者使用了Logistic模型、GM(1,1)灰色预测模型及高等教育规模滚动回归模型。本文使用方差-协方差组合预测法,辅以改进后的三种模型来重新估计2020年高等教育毛入学率。结果显示,2020年我国高等教育毛入学率将达到且超过50%。

China＇s National Plan for Medium and Long-term Education Reform and Development （2010- 2020） proposes a strategic goal that China＇s gross enrollment rate of higher education will reach 40% by 2020. However, new statistics show that China has achieved the long-term development goals of higher education in 2015, five years in advance. Based on the literature review, the paper identifies three main reasons for the bias of predications, i.e. the constraints of information updating and model difference, the rational bias and the unpredictable policy factors. In some forecast variables of the continuous gross enrollment rate, more accurate predictions are achieved by employing Logistic model, GM （1, 1） model and rolling regression model. By a combination of variance-covariance prediction method and the aforementioned three models, the paper shows that China＇s higher education gross enrollment rate will reach 50% in 2020.