构建招收飞行员基本认知能力预测学业成绩的模型

Constructing a model of basic cognitive ability predicting academic performance for pilots' selection

目的:通过空军招收飞行员心理品质检测第一平台的基本认知能力测验,建立学生学业成绩的预测模型,提前淘汰高考成绩不良的学生,以节省开支。方法:①实验1:采用完全随机设计的方法,于2005-12选取西安市某部队院校入校3个月的854名大学生,进行加法计算、比较刻度、找特殊图形、组合图形、填写数据、识符检数、判别方向、选词配对、找规律填数等9项基本能力测试,利用Pearson相关分析,选出与学业成绩显著相关的4项测验。②实验2:2006-12选取与实验1不同部队院校的2006级入校3个月的200名大学生,进行4项选定的测验,通过SPSS12.0统计软件包进行多元回归分析最终建立预测学业成绩的基本认知能力模型,并分析其预测效度。结果:①实验1结果:选定的4项能力测试与学业成绩的相关分别是:加法计算X1(r=0.233)、找特殊图形X2(r=0.226)、填写数据X3(r=0.100)、选词配对X4(r=0.176)。②实验2结果:最终建立的预测方程是Y=532.75+1.001X1+1.247X3+1.257X4,它能解释学业成绩中13.0%的变异。结论:基本认知能力测验能预测学业成绩,并且能达到较好的效度,可以应用到飞行员选拔的实际中。

AIM： To establish a model of predicting academic performance by using basic cognitive ability test on psychological detection subsystem Ⅰ for pilots＇ selection, so as to eliminate the learning disability students in advance and save expenses for pilots＇ selection. METHODS： ①Test 1： Using completely random design, 854 three-month college students, who were newly admitted in troop university of Xi＇an in December 2005, ware investigated by nine basic ability tests： addition calculation, comparing simulated scale, detecting special graph, assembling figures, simulating keyboard input, recognition regulation, distinguishing direction, selection paired word and regulanty array. Four significant related tests were selected with Pearson correlation. ②Test 2： In December 2006, 200 three-month college students of 2006 grade from another university ware tested by four selected basic ability tests. Using multiple linear regression analysis, we constructed a model of basic cognitive ability predicting academic performance by SPSS12.0 software, and its predictive validity was analyzed. RESULTS： ①Test 1： The four selected ability tests were addition calculation X1 （r =0.233）, detecting special graph X2 （r =0.226）, simulating keyboard input X3 （r =0.100） and selecting paired word X4 （r =0.176）.②Test 2： The regression equation was Y=532.75＋1.001X1＋1.247X3＋1.257X4, which could account for 13.0% variance on the academic record. CONCLUSION： The basic cognitive ability tests have a significant influence on predicting academic performance. It can be applied for the pilots＇ selection.