先放个果子：
父母控制量表的探索性因素分析结果
探索性因素分析 Exploratory Factor Analysis
-
判断提取的因子个数
最常见的是基于特征值的方法，每个因子都与相关系数矩阵的特征值关联，第一个因子与最大的特征值相关联，第二个因子与第二大的特征值相关联，依此类推。 在R中，有许多包／函数可以实现这个功能。在这个case中，我们也会使用到。
library(ggplot2) library(psych) library(GPArotation) fa.parallel() -
判断提取的因子个数的检验指标
- Adequacy Test
- The root mean square of residuals (RMSR) should be closer to 0. （RMSR -> 小一点比较好）
- RMSEA (root mean square error of approximation) index should be lower than 0.05（RMSEA -> 小于0.05）
- Tucker-Lewis Index (TLI) should be over 0.9（TLI -> 应该要大于0.9）
为了更方便地一次性得到所有的这些指标，我编写了一个名为EFA_factor的函数，帮助大家更好更快地确定因素分析的结果。
～～函数在此～～
EFA_factor <- function(data)
{
library(ggplot2)
library(psych)
library(GPArotation)
# Next we’ll find out the number of factors that we’ll be selecting for factor analysis.
# Parallel Analysis / eigenvalue（特征值）
parallel <- fa.parallel(data, fm = 'minres', fa = 'fa')
n_fact <- parallel$nfact
# Factor Analysis
# r ??? Raw data or correlation or covariance matrix
# nfactors ??? Number of factors to extract
# rotate ??? Although there are various types rotations, `Varimax` and `Oblimin` are most popular
# fm ??? One of the factor extraction techniques like `Minimum Residual (OLS)`, `Maximum Liklihood`, `Principal Axis` etc.
# In this case, we will select oblique rotation (rotate = “oblimin”) as we believe that there is correlation in the
# factors. Note that Varimax rotation is used under the assumption that the factors are completely uncorrelated.
# We will use `Ordinary Least Squared/Minres` factoring (fm = “minres”), as it is known to provide results similar to
# `Maximum Likelihood` without assuming multivariate normal distribution and derives solutions through iterative
# eigendecomposition like principal axis.
RMSR <- c()
RMSEA <- c()
TLI <- c()
my_list <- list()
# 创建一系列变量名
code <- as.character(sprintf("%01d",2:(n_fact+1)))
# 这部分是统一命名一个方便你提取的变量名
varname <- paste("factor_",code,sep="")
n <- 2
while(n <= (n_fact+1))
{
factor_opt <- fa(data,nfactors = n,rotate = "oblimin",fm="minres")
factor_opt_cut <- print(factor_opt$loadings,cutoff = 0.3)
fa.diagram(factor_opt)
my_list[[n-1]] <- factor_opt_cut
RMSR <- c(RMSR,factor_opt$rms)
RMSEA <- c(RMSEA,factor_opt$RMSEA[1])
TLI <- c(TLI,factor_opt$TLI)
n <- n+1
}
data_output <- data.frame(cbind(varname,RMSR, RMSEA,TLI))
output_list <- list("factors_compare" = data_output, "loadings" = my_list)
return(output_list)
}
这里哇这里需要补充介绍一下这个函数
对父母控制量表进行探索性因素分析
父母控制量表
这里的心里控制量表是一个由三部分量表拼成的新量表，还没有经过检验。参考文献如下：
Item 1 – 8: parents’ autonomy support
Adapted from measures used in prior research (Cheung, Pomerantz, Wang, & Qu, 2016; McPartland & Epstein, 1977; Robbins, 1994; Steinberg, Lamborn, Dornbusch, & Darling, 1992).
Reference:
Cheung, C. S., Pomerantz, E. M., Wang, M. & Qu, Y. (2016). Controlling and Autonomy-Supportive Parenting in the United States and China: Beyond Children’s Reports. Child Development, 87, 1992-2007.
McPartland, J. M., & Epstein, J. L. (1977). Open schools and achievement: Extended tests of a finding of no relationship. Sociology of Education, 50(2), 133-144. doi: 10.2307/2112375
Robbins, R. J. (1994). An assessment of perceptions of parental autonomy support and control: Child and parent correlates. Unpublished doctoral dissertation, University of Rochester.
Steinberg, L., Lamborn, S. D., Dornbusch, S. M., & Darling, N. (1992). Impact of parenting practices on adolescent achievement: Authoritative parenting, school involvement, and encouragement to succeed. Child Development, 63(5), 1266-1281. doi: 10.2307/1131532
Item 9 – 18: parents’ psychological control
Adapted from measures used in prior research (Barber, 1996; Cheung, Pomerantz, Wang, & Qu, 2016; Silk et al., 2003)
Reference:
Barber, B. K. (1996). Parental psychological control: Revisiting a neglected construct. Child Development, 67(6), 3296-3319. doi:10.2307/1131780
Cheung, C. S., Pomerantz, E. M., Wang, M. & Qu, Y. (2016). Controlling and Autonomy-Supportive Parenting in the United States and China: Beyond Children’s Reports. Child Development, 87, 1992-2007.
Silk, J. S., Morris, A. S., Kanaya, T., & Steinberg, L. (2003). Psychological control and autonomy granting: Opposite ends of a continuum or distinct constructs? Journal of Research on Adolescence, 13(1), 113-128. doi:10.1111/1532-7795.1301004
Item 19 – 26: Parents’ Warmth and Rejection
Adapted from Rohner’s Parental Acceptance-Rejection Questionnaire
Reference:
Rohner, R. P. & Khaleque, A. (2005). Handbook for the Study of Parental Acceptance and Rejection 4th edition. E-book. Storrs, CT: Rohner Research Publications.
本帖希望对这个量表进行探索性因素分析，看看26个条目可以分为多少个维度。
这里使用的数据集是435个初中学生填写的真实数据。
有兴趣的小伙伴可以点击进入数据下载页面（密码:48ry）
使用函数进行分析
使用函数对数据集进行了分析后，我们得到一个可以看到各项指标的表格。
varname RMSR RMSEA TLI
RMSEA factor_2 0.059 0.13 0.776
RMSEA.1 factor_3 0.042 0.111 0.837
RMSEA.2 factor_4 0.027 0.094 0.884
RMSEA.3 factor_5 0.019 0.07 0.936
得到Scree Plot如下：
最终得到合适的因子分析结果如下：
结合因子分析结果对原量表进行分析
根据因素分析的结果，对各个维度进行了归纳和命名，得到的结果已经在文章开头展示啦。 分为了5个维度，分别是：
* 父母的自主性支持
* 父母的心理控制
* 父母控制
* 父母支持
* 父母温暖
