1、 SZ 2012, Vol. 20, No. 6, 825833 Advances in Psychological Science DOI: 10.3724/SP.J.1042.2012.00825 825 研究方法(Research Method) 二次响应面回归方法及其在个体 环境匹配研究中的使用*f 1,2,3fy1,2 1,2(1S S r B L i , 100101) (2S S , 100101) (3S S 3 , 100039) K 1 二次响应面回归分析是指二次多项式回归与响应面分析相结合的一种研究方法, 该方法在工业与组织心理学领域起源于个体环境(P-E)匹配研究。它使
2、用对个体环境进行间接匹配测量的策略, 考察二者间的最佳匹配以及与结果变量的复杂关系, 并在三维空间对其进行检验和解释。该文在分析以往P-E匹配研究方法(例如差异分数)缺陷性的基础上, 阐述了二次响应面回归方法的产生与发展, 逻辑思想, 统计分析方法, 并以实例演示其操作步骤与结果解释, 以及相对于其他方法的优势与不足。 1oM 二次多项式回归; 响应面分析; 个体环境(P-E)匹配; 差异分数 s | B841 1 引言=Q TB (quadratic polynomial regression)Y s (response surface analysis; Box , Tr1 Tr H ,
3、Tg rd9il A (a3=0.44, p0.05)b m 2 1 (X)- (Y)Tr Tg (Z)Y 4 二次响应面回归方法的优势与不足 4.1 二次响应面回归方法的贡献 =QY Bs P-E 4 Bd9s d O , TBZ=Q“ V$z d , 7 O ?_t“ Y + ; KV 1“ , 7 1“/Y9 V_bV7P-E M1 4 F r m b H , =QY BsM.dW ZE rb Edwards Parry (1993) - Qs , ?CF K=Q TBZE V dZMs (vss ZE ; i Oss s KMB“ , =Q TB$ 4 b , French (1982
4、)ss ZE?C L= XT T i1“b7Edwards Parry (1993) TB , ?C L= XT T i1“ db L s U , 1 - Tr B f / , = ( ( M1 , 8Tg , .dss s f b t M , =QY BsZE9 l b 8v 1“ , vM SB v =8s 7 , , TBL , V7 P8ZsF S (Jansen Hecht V j A , =Q TBYVY ZEM , Z P-E ?_L !S , P-E sL !_4 b , =QY Bs K.d d9s O , VCd9s/ A , =QY BZE P-E B 5 -b 6 ,
5、cKV P-E =QY BZE , L= , OX 5b , 5 5 , I“d L=/ PY (Venkatesh g 5 , sg MGig Y (Kim H “ S u ; K “ xZ ub | q w q_ Q Y | q w q9 u , 9 T Q5 = (F6 + F7); R | q w q S9 u , 9 T R5 = SQRT (H6*H6) + (H7*H7) + (2*K5); S T | q w q t _ u , 9 T S5 = Q5/R5(t ), T5 = TDIST (ABS (S5), F12,2) (P ); U A_ u , 9 T U5 =IF(
6、T50.05,“Sig!“,“ “)b m | -2 2 Tm , 9 T P16 = (F5 + (F6*P15) + (F7*O16) + (F8*P15*P15) + (F9*P15*O16) + (F10*O16*O16)b The Quadratic Polynomial Regression with Response Surface Analysis and Its Use in Person-Environment Fit Research ZHANG Shan-Shan1,2; ZHANG Jian-Xin1; ZHOU Ming-Jie1(1Key Laboratory o
7、f Mental Health, Institute of Psychology, Chinese Academy of Sciences, Beijing 100101, China) (2 Graduate school, the Chinese Academy of Sciences, Beijing 100039, China) Abstract: Quadratic polynomial regression with response surface analysis is a methodology and statistical approach that has become
8、 increasingly popular in Industrial and Organizational Psychology, and it originated in the person-environment fit research. The approach is an indirect measurement of person- environment fit, it allows researchers to study the best match between person and environment and the relations with outcome
9、s, and further examination and interpretation in the three-dimensional space. First of all, this article presented the numerous substantive and methodological problems of traditional methods (i.e., difference scores); secondly, we elaborated the emergence and development, logical thinking, methods a
10、nd process of statistical analysis, and also gave an example to show how to conduct this approach and analyzed the results; Finally, we summarized the advantages and disadvantages of quadratic polynomial regression with response surface analysis. Key words: quadratic polynomial regression; response surface analysis; person-environment fit; difference scores
