Methodological research on moderation effects in China’s mainland

doi:10.3724/SP.J.1042.2022.01703

Abstract

Abstract:

The analysis of moderation effects has become an important statistical method in multivariate studies. Methodological research on moderation effects in China’s mainland covers the following topics: moderation effects of observed variables, latent variables, multi-level data and longitudinal data; the single-level moderation effect analysis based on a two-level regression model; the integration model of moderation and mediation (see Wen et al. 2022).

Methodological research on the moderation effect of observed variables includes three aspects: standardized resolution, simple slope test, and the moderation effect of category variables. The research on latent moderation includes three aspects too: standardized resolution, model simplification, and comparison of analytical methods. Under the normal condition, latent moderated structural equations (LMS) are recommended to estimate the moderation effect of latent variables. Otherwise, after centralizing all indicators, the unconstrained product indicator method is recommended to establish a latent moderation model; Bayesian method is an alternative, especially in the case of a small sample.

The model development of multilevel moderation effect involves the conflated multilevel model, unconflated multilevel model (UMM), and multilevel structural equation model (MSEM). All independent variables at Level-1 are not centered in the conflated multilevel model, whereas in the UMM all independent variables at level-1 are centered using group-mean, and the group mean is included at Level-2. If the group-mean was treated as a latent variable, MSEM is recommended. Further, two ways are adopted to test multilevel moderation in the multilevel structural equation model: random coefficient prediction (RCP) for cross-level moderations, and LMS for same-level moderations.

The moderation effect analysis of longitudinal data is divided into three types. The first type is moderation analysis in two-instance repeated measures designs, in which only the dependent variable is repeated measurement. In the second type, there isn’t any moderator, while both the independent and dependent variables are repeated measurement (e.g., the cross-lagged model, and the contextual moderation model). In the third type, all variables are repeated measurement, such as the latent growth model and multilevel model.

Two-level regression model is recommended to analyze the moderation effect of single-level data. It can be employed to analyze the moderation effect of both observed variables and latent variables.

Some international frontiers of methodological research on moderation analysis are briefly introduced: the combination of LMS and Bayesian method, moderation analysis of multiple moderators; moderation analysis of longitudinal data.

Key words: moderating effect, latent variable, categorical variable, multilevel data, longitudinal data

CLC Number:

B841

FANG Jie, WEN Zhonglin, OUYANG Jinying, CAI Baozhen. Methodological research on moderation effects in China’s mainland[J]. Advances in Psychological Science, 2022, 30(8): 1703-1714.

Figures/Tables 4

References 66

[1]	董维维, 庄贵军, 王鹏. (2012). 调节变量在中国管理学研究中的应用. 管理学报, 9(12), 1735-1743.
[2]	方杰, 温忠麟. (2018). 基于结构方程模型的有调节的中介效应. 心理科学, 41(2), 453-458.
[3]	方杰, 温忠麟. (2022a). 基于两水平回归模型的调节效应分析及其效应量. 心理科学进展, 30(5), 1183-1190.
[4]	方杰, 温忠麟. (2022b). 纵向数据的调节效应分析. 心理科学进展, 30.
[5]	方杰, 温忠麟. (印刷中a). 两类常见的类别变量调节效应分析. 心理科学.
[6]	方杰, 温忠麟. (印刷中b). 有调节的多层中介效应分析. 心理科学.
[7]	方杰, 温忠麟, 梁东梅, 李霓霓. (2015). 基于多元回归的调节效应分析. 心理科学, 38(3), 715-720.
[8]	方杰, 温忠麟, 吴艳. (2018). 基于结构方程模型的多层调节效应. 心理科学进展, 26(5), 781-788.
[9]	方杰, 张敏强, 顾红磊, 梁东梅. (2014). 基于不对称区间估计的有调节的中介模型检验. 心理科学进展, 22(10), 1660-1668.
[10]	付会斌, 孔丹莉, 潘海燕, 丁元林. (2010). 非线性结构方程模型的研究进展. 中国卫生统计, 27(1), 101-103.
[11]	侯杰泰, 温忠麟, 成子娟. (2004). 结构方程模型及其应用. 北京: 教育科学出版社.
[12]	李艾, 李君文. (2008). 调节变量(moderator)辨析:类型、表述和识别. 数理统计与管理, 27(2), 257-264.
[13]	李建明, 曲成毅. (2007). 隐变量交互作用分析建模及其在SAS上的实现. 数学的实践与认识, 37(1), 31-37.
[14]	李建明, 曲成毅. (2011). 隐变量交互作用分析建模方法——两阶段最小二乘法(2SLS) 及其在SAS软件上的实现. 数理医药学杂志, 24(6), 631-634.
[15]	廖卉, 庄瑗嘉. (2012). 多层次理论模型的建立及研究方法. 见陈晓萍, 徐淑英, 樊景立 (编).组织与管理研究的实证方法(第二版, pp. 442-476). 北京: 北京大学出版社.
[16]	刘红云. (2019). 高级心理统计. 北京: 中国人民大学出版社.
[17]	刘红云, 袁克海, 甘凯宇. (2021). 有中介的调节模型的拓展及其效应量. 心理学报, 53(3), 322-338.
[18]	刘源. (2021). 多变量追踪研究的模型整合与拓展:考察往复式影响与增长趋势. 心理科学进展, 29(10), 1755-1772.
[19]	卢谢峰, 韩立敏. (2007). 中介变量、调节变量与协变量——概念、统计检验及其比较. 心理科学, 30(4), 934-936.
[20]	罗畏畏, 方杰, 孙雅文. (2018). 网络被欺负与大学生自杀意念的关系及生命意义感的调节作用. 心理技术与应用, 6(4), 211-217.
[21]	乔元波, 王砚羽, 邹仁余. (2017). 非线性模型连续调节变量检验的实现方法. 统计与决策, (9), 9-14.
[22]	王阳, 温忠麟, 王惠惠, 管芳. (2022). 第二类有中介的调节模型. 心理科学进展, 30(9), 2131-2142.
[23]	温忠麟, 方杰, 沈嘉琦, 谭倚天, 李定欣, 马益铭. (2021). 新世纪20年国内心理统计方法研究回顾. 心理科学进展, 29(8), 1331-1344.
[24]	温忠麟, 方杰, 谢晋艳, 欧阳劲樱. (2022). 国内中介效应的方法学研究. 心理科学进展, 30(8), 1692-1702.
[25]	温忠麟, 侯杰泰. (2004). 隐变量交互效应分析方法的比较与评价. 数理统计与管理, 23(3), 37-42.
[26]	温忠麟, 侯杰泰, Marsh, H. W. (2003). 潜变量交互效应分析方法. 心理科学进展, 11(5), 593-599.
[27]	温忠麟, 侯杰泰, Marsh, H. W. (2008). 结构方程模型中调节效应的标准化估计. 心理学报, 40(6), 729-736.
[28]	温忠麟, 侯杰泰, 张雷. (2005). 调节效应与中介效应的比较和应用. 心理学报, 37(2), 268-274.
[29]	温忠麟, 刘红云. (2020). 中介效应和调节效应:方法及应用. 北京: 教育科学出版社.
[30]	温忠麟, 刘红云, 侯杰泰. (2012). 调节效应和中介效应分析. 北京: 教育科学出版社.
[31]	温忠麟, 欧阳劲樱, 方俊燕. (2022). 潜变量交互效应标准化估计: 方法比较与选用策略. 心理学报, 54(1), 91-107.
[32]	温忠麟, 吴艳. (2010). 潜变量交互效应建模方法演变与简化. 心理科学进展, 18(8), 1306-1313.
[33]	温忠麟, 吴艳, 侯杰泰. (2013). 潜变量交互效应结构方程: 分布分析方法. 心理学探新, 33(5), 409-414.
[34]	温忠麟, 叶宝娟. (2014). 有调节的中介模型检验方法: 竞争还是替补? 心理学报, 46(5), 714-726.
[35]	温忠麟, 张雷, 侯杰泰. (2006). 有中介的调节变量和有调节的中介变量. 心理学报, 38(3), 448-452.
[36]	吴冰, 王重鸣. (2010). 带有乘积项的结构方程潜变量交互模型研究. 统计与决策, (16), 26-28.
[37]	吴艳, 温忠麟, 侯杰泰, Marsh, H. W. (2011). 无均值结构的潜变量交互效应模型的标准化估计. 心理学报, 23(10), 1219-1228.
[38]	吴艳, 温忠麟, 李碧. (2014). 潜变量交互效应模型标准化估计中的检验问题. 心理学探新, 34(3), 260-264.
[39]	吴艳, 温忠麟, 林冠群. (2009). 潜变量交互效应建模:告别均值结构. 心理学报, 41(12), 1252-1259.
[40]	叶宝娟, 温忠麟. (2013). 有中介的调节模型检验方法: 甄别和整合. 心理学报, 45(9), 1050-1060.
[41]	张莉, Wan, F, 林与川, Qiu, P. (2011). 实验研究中的调节变量和中介变量. 管理科学, 24(1), 108-116.
[42]	郑舒方, 张沥今, 乔欣宇, 潘俊豪. (2021). 密集追踪数据分析: 模型及其应用. 心理科学进展, 29(11), 1948-1969.
[43]	Aiken L. S., & West S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.
[44]	Asparouhov T., & Muthén B. (2019). Latent variable interactions using maximum-likelihood and Bayesian estimation for single- and two-level models. Mplus Web Notes, No. 23. Muthén & Muthén.
[45]	Asparouhov T., & Muthén B. (2021). Bayesian estimation of single and multilevel models with latent variable interactions. Structural Equation Modeling: A Multidisciplinary Journal, 28(2), 314-328. doi: 10.1080/10705511.2020.1761808 URL
[46]	Baron R. M., & Kenny D. A. (1986). The moderator- mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182. doi: 10.1037//0022-3514.51.6.1173 pmid: 3806354
[47]	Carden S. W., Holtzman N. S., & Strube M. J. (2017). CAHOST: An excel workbook for facilitating the Johnson- Neyman technique for two-way interactions in multiple regression. Frontiers in Psychology, 8, 1-7.
[48]	Cham H., West S. G., Ma Y., & Aiken L. S. (2012). Estimating latent variable interactions with nonnormal observed data: A comparison of four approaches. Multivariate Behavioral Research, 47(6), 840-876. doi: 10.1080/00273171.2012.732901 URL
[49]	Hayes A. F. (2018). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (2nd ed.). New York: Guilford Press.
[50]	Jackman M. G. A., Leite W. L., & Cochrane D. J. (2011). Estimating latent variable interactions with the unconstrained approach: A comparison of methods to form product indicators for large, unequal numbers of items. Structural Equation Modeling: A Multidisciplinary Journal, 18(2), 274-288. doi: 10.1080/10705511.2011.557342 URL
[51]	Kelava A., & Nagengast B. (2012). A Bayesian model for the estimation of latent interaction and quadratic effects when latent variables are non-normally distributed. Multivariate Behavioral Research, 47(5), 717-742. doi: 10.1080/00273171.2012.715560 URL
[52]	Klein A., & Moosbrugger H. (2000). Maximum likelihood estimation of latent interaction effects with the LMS method. Psychometrika, 65(4), 457-474. doi: 10.1007/BF02296338 URL
[53]	Klein A. G., & Muthén B. O. (2007). Quasi-maximum likelihood estimation of structural equation models with multiple interaction and quadratic effects. Multivariate Behavioral Research, 42(4), 647-673. doi: 10.1080/00273170701710205 URL
[54]	Lee S. Y., Song X. Y., & Tang N. S. (2007). Bayesian methods for analyzing structural equation models with covariates, interaction, and quadratic latent variables. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 404-434. doi: 10.1080/10705510701301511 URL
[55]	Lin G. C., Wen Z., Marsh H. W., & Lin H. S. (2010). Structural equation models of latent interactions: Clarification of orthogonalizing and double-mean-centering strategies. Structural Equation Modeling: A Multidisciplinary Journal, 17(3), 374-391. doi: 10.1080/10705511.2010.488999 URL
[56]	Liu H., & Yuan K.-H. (2021). New measures of effect size in moderation analysis. Psychological Methods, 26(6), 680-700. doi: 10.1037/met0000371 URL
[57]	Liu H., Yuan K.-H., & Liu F. (2020). A two-level moderated latent variable model with single level data. Multivariate Behavioral Research, 55(6), 873-893. doi: 10.1080/00273171.2019.1689350 URL
[58]	Liu H., Yuan K.-H., Wen Z. (in press). Two-level moderated mediation models with single level data and new measures of effect sizes. Behavior Research Methods.
[59]	Marsh H. W., Wen Z., & Hau K. -T. (2004). Structural equation models of latent interactions: Evaluation of alternative estimation strategies and indicator construction. Psychological Methods, 9(3), 275-300. doi: 10.1037/1082-989X.9.3.275 URL
[60]	McNeish D., & Hamaker E. L. (2020). A primer on two- level dynamic structural equation models for intensive longitudinal data in Mplus. Psychological Methods, 25(5), 610-635. doi: 10.1037/met0000250 URL
[61]	Muthén L. K., & Muthén B. O. (1998/2017). Mplus User’s Guide. Muthén & Muthén.
[62]	Ozkok O., Vaulont M. J., Zyphur M. J., Zhang Z., Preacher K. J., Koval P., & Zheng Y. (in press). Interaction effects in cross-lagged panel models: SEM with latent interactions applied to work-family conflict, job satisfaction, and gender. Organizational Research Methods.
[63]	Serang S., Jacobucci R., Brimhall K. C., & Grimm K. J. (2017). Exploratory mediation analysis via regularization. Structural Equation Modeling: A Multidisciplinary Journal, 24(5), 733-744. doi: 10.1080/10705511.2017.1311775 URL
[64]	Wen Z., Marsh H. W., & Hau K. -T. (2010). Structural equation models of latent interactions: An appropriate standardized solution and its scale-free properties. Structural Equation Modeling: A Multidisciplinary Journal, 17(1), 1-22. doi: 10.1080/10705510903438872 URL
[65]	Wen Z., Marsh H. W., Hau K.-T., Wu Y., Liu H., & Morin A. J. S. (2014). Interaction effects in latent growth models: Evaluation of alternative estimation approaches. Structural Equation Modeling: A Multidisciplinary Journal, 21(3), 361-374. doi: 10.1080/10705511.2014.915205 URL
[66]	Wu Y., Wen Z., Marsh H. W., & Hau K. -T. (2013). A comparison of strategies for forming product indicators for unequal numbers of items in structural equation models of latent interactions. Structural Equation Modeling: A Multidisciplinary Journal, 20(4), 551-567. doi: 10.1080/10705511.2013.824772 URL

类别		文献	推荐方法
单层数据的调节效应	显变量	温忠麟等(2003, 2005, 2008); 卢谢峰等(2007); 李艾等(2008); 董维维等(2012); 方杰等(2015, 印刷中a); 乔元波等(2017)	回归分析法(连续和类别变量都适用)
单层数据的调节效应	潜变量	温忠麟等(2003, 2004, 2008, 2010, 2013); 温忠麟和欧阳劲樱等(2022); 李建明等(2007, 2011); 吴艳等(2009, 2011, 2014); 吴冰等(2010); 付会斌等(2010)	多组SEM(类别调节变量) 无约束的乘积指标法 LMS法
多层数据的调节效应	显变量	方杰、温忠麟和吴艳(2018)	无混淆的MLM 多层SEM(RCP和LMS法)
多层数据的调节效应	潜变量	方杰、温忠麟和吴艳(2018)	无混淆的MLM 多层SEM(RCP和LMS法)
基于两层回归模型的单层调节分析	显变量	方杰等(2022a)	两层回归模型分析法 LMS法
基于两层回归模型的单层调节分析	潜变量	方杰等(2022a)	两层回归模型分析法 LMS法
纵向数据的调节效应	显变量	方杰等(2022b, 印刷中a)	无混淆的MLM 多层SEM、CLM和LGM
纵向数据的调节效应	潜变量	方杰等(2022b, 印刷中a)	无混淆的MLM 多层SEM、CLM和LGM
调节和中介的整合模型	显变量	温忠麟等(2006, 2014); 张莉等(2011); 叶宝娟等(2013); 方杰等(2014, 2018, 印刷中b); 刘红云等(2021); 王阳等(2022)	回归分析法 LMS法
调节和中介的整合模型	潜变量		回归分析法 LMS法

类别		文献	推荐方法
单层数据的调节效应	显变量	温忠麟等(2003, 2005, 2008); 卢谢峰等(2007); 李艾等(2008); 董维维等(2012); 方杰等(2015, 印刷中a); 乔元波等(2017)	回归分析法(连续和类别变量都适用)
单层数据的调节效应	潜变量	温忠麟等(2003, 2004, 2008, 2010, 2013); 温忠麟和欧阳劲樱等(2022); 李建明等(2007, 2011); 吴艳等(2009, 2011, 2014); 吴冰等(2010); 付会斌等(2010)	多组SEM(类别调节变量) 无约束的乘积指标法 LMS法
多层数据的调节效应	显变量	方杰、温忠麟和吴艳(2018)	无混淆的MLM 多层SEM(RCP和LMS法)
多层数据的调节效应	潜变量	方杰、温忠麟和吴艳(2018)	无混淆的MLM 多层SEM(RCP和LMS法)
基于两层回归模型的单层调节分析	显变量	方杰等(2022a)	两层回归模型分析法 LMS法
基于两层回归模型的单层调节分析	潜变量	方杰等(2022a)	两层回归模型分析法 LMS法
纵向数据的调节效应	显变量	方杰等(2022b, 印刷中a)	无混淆的MLM 多层SEM、CLM和LGM
纵向数据的调节效应	潜变量	方杰等(2022b, 印刷中a)	无混淆的MLM 多层SEM、CLM和LGM
调节和中介的整合模型	显变量	温忠麟等(2006, 2014); 张莉等(2011); 叶宝娟等(2013); 方杰等(2014, 2018, 印刷中b); 刘红云等(2021); 王阳等(2022)	回归分析法 LMS法
调节和中介的整合模型	潜变量		回归分析法 LMS法