Task 4:使用Amos制作CFA模型

 使用Amos制作CFA模型

数据

《结构方程模型:方法与应用》-王济川,2011,chapter2

https://pan.baidu.com/s/1cxglKa

视频地址:https://pan.baidu.com/s/1jI6tuJG

高清无水印版视频:https://pan.baidu.com/s/1i4W34xB

完整带截图版报告见网盘分享word文件:

word文档:https://pan.baidu.com/s/1c2BpttA

数据原始格式为.dat,转换整理成为.sav格式后进行使用。

数据结构及变量释义

该数据源自一项对美国农村非法药物使用者无干预随访观察研究,考虑到该人群由于非法使用药物所导致的不良影响,包括罹患各种疾病的高风险,其心理疾患的发生率必定较高,高人群是测试BSI-18量表的重点人群。BSI-18量表是《简明症状问卷53》的缩减版,BSI量表广泛用于临床和一般人群心理疾患的评估,Derogatis编制了缩减版的问卷BSI-18用作常见的心理疾患的筛检工具。这些疾患包括:躯体化,抑郁和焦虑。BSI-18表中共有18个测量条目或观察指标,分别负载在三个因子SOM, DEP和ANM上,每个因子各有6各条目。如下表。

最新研究通过运用CFA模型确认BSI量表具有最初设计的三维结构,虽然被选的4因子模型(SOM, DEP, AGI和PAN)也能你和数据。尽管3因子和4因子模型均你和数据向好,但3因子模型比4因子模型更简约,结果更易理解。

本次操作通过使用实际数据用CFA评估BSI-18因子结构。其模型如下图。

该数据中包括BSI-18量表中的18个条目,分别为V1-V18,次序对应上图所示的18个变量。

使用Amos对该数据进行验证性因子分析,采用固定方差法。

结果

从校正指数(M.I.)表中可以看出,只有V2ßF1的校正指数是大于4的,因此可以考虑V2可能是属于F1的,本次操纵只关心因子与项目之间的关系而不关心其他无理论意义的修正参数。

上表显示因子到项目的载荷,标准误,C.R.以及显著性,本例中可以看到V17<-F2的点估计较小,但是其p值显著,说明我们应当保留此项,其他项目的点估计均大于.4。继续看下表的标准化载荷。

在标准化载荷中可以看到V17<-F2的标准化载荷为.368,接近传统认为较好的数值.4,并且因为其显著性较好,故保留V17<-F2。

在以上两个表格中可以看到模型的拟合优度,可以看到该模型RMSEA系数在.068,气90%置信区间为(.057, .079),在.05到.08之间,说明模型拟合较好。NFI和CFI分别为.862和.920,在这两个系数上说明模型拟合尚可。

模型修正

根据以上结果,考虑添加F1→V2单向箭头作为竞争模型。重新拟合结果如下:

从上表可以看出V2<-F1的点估计为.299,虽然数值较小但是其显著性为非常显著,继续看下图标准化后的载荷。

从上图可看出V2<-F1标准化后的载荷为.231,小于.4。

对照原始模型,竞争模型RMSEA指数基本没有变化。

对照原始模型,竞争模型NFI以及CFI变化不大,只是略微改善。

综合考虑到V2<-F1标准化后的点估计并不大,因此对原模型不做修改,即不增加F1→V2的单向箭头,模型表示为:

Task 3:使用MLwiN比较两组独立样本均值

使用MLwiN比较两组独立样本均值

数据:北京市老年人焦虑症状水平调查:

http://www.cnsda.org/index.php?r=projects/view&id=60493698

视频地址:https://pan.baidu.com/s/1slPHUJ7

高清无水印视频(文件较大):https://pan.baidu.com/s/1c18Xmjm

注:完整带截图版报告见网盘分享word文件!

完整报告:https://pan.baidu.com/s/1c3lPl8

本次操作旨在使用已有数据对老年人焦虑水平是否存在性别差异进行分析。

对象及方法:调查总体为海淀区行政区域内60周岁以上的户籍人口。以典型抽样方式选取海淀区13个街道(镇)中的45个村/居委会,社区类型包括新建社区、老旧小区、城乡结合部社区、大院型社区和村。在社区内以系统抽样方式选取个人调查对象,每个社区选取30名,实际调查人数为1350人。机构老人来自北京市四季青养老院和嘉德养老公寓,通过方便取样方式获得,实际调查101人。

变量及变量释义:

老年人焦虑水平:对澳大利亚学者Pachana2007年编制的英文版老年焦虑量表(Geriatric Anxiety Inventory, GAI)进行中文翻译并施测,在本次操作中,GAI(焦虑)变量共20道题目,每道题目均为得分均为0或1,对样本得到的GAI总分进行分析,得分越高,被试焦虑水平越高。

分类变量:性别,“0”=“女”;“1”=“男”。

APA Section

An independent-samples t test was conducted to evaluate the hypothesis that old females are more anxious than old males. The test was significant, t(1316) =-4.94, p < .01 , and the results were confirm the research hypothesis. Old females (M =2.66, SD =4.57) are more anxious than old males (M =1.52, SD =3.43). The 95% confidence interval for the difference in means was quite wide compare to the mean of the scores of the old man`s anxiety index, ranging from .69 to 1.60. Figure 1 shows the distributions for the two group.

Figure 1. Error bars(two standard deviations above and below the mean) for anxiety index for each sex group

%e5%9b%be%e7%89%87-1

 

Task 2:使用LMATRIX语句进行简单主效应检验及两两比较

使用LMATRIX语句进行简单主效应检验及两两比较

数据下载:https://pan.baidu.com/s/1c1YFB1I

视频观看:https://pan.baidu.com/s/1jHOZWsm

高清无水印版(文件较大):https://pan.baidu.com/s/1jHXN2tg

注:完整带截图版报告见网盘分享word文件!

完整报告:https://pan.baidu.com/s/1nuXbtvr

数据及变量释义

使用《SPSS统计分析第五版》-卢纹岱data09-07。

该数据为教育心理学实验,数据是心理运动测验分数(score)与被试者必须瞄准的目标大小关系的资料,数据中共有4个大小不同的目标(target):T1、T2、T3、T4,两种不同明暗程度的照明环境(Light):L1、L2,数据中还包括一个不同设备变量,由于三因素方差分析过程太过复杂且为重复劳动意义不大,本次操作中对设备变量不予分析,本次操作选取device = 1的数据,即挑选使用设备1的被试,共剩下40个得分数据。

即本次操作使用的是一个4x2的析因实验设计,共40个得分数据。

首先进行Two-Way ANOVA分析,结果如下:

在方差分析表中可以看出light和target的交互作用显著,需要继续进行简单主效应分析,对简单主效应的分析使用LMATRIX语句。

首先分析不同明暗程度的照明条件(L1,L2)在不同target大小不同水平(T1,T2,T3,T4)上的差异。

LMATRIX语句如下:

UNIANOVA

score BY light target

/METHOD=SSTYPE(3)

/lmatrix 'L1 vs L2 within T1' light*target 1 0 0 0 -1 0 0 0 light 1 -1

/lmatrix 'L1 vs L2 within T2' light*target 0 1 0 0 0 -1 0 0 light 1 -1

/lmatrix 'L1 vs L2 within T3' light*target 0 0 1 0 0 0 -1 0 light 1 -1

/lmatrix 'L1 vs L2 within T4' light*target 0 0 0 1 0 0 0 -1 light 1 -1

 

而后对不同target大小(T1,T2,T3,T4)条件在不同照明程度(L1,L2)两个水平上的差异进行分析。

LMATRIX语句如下:

UNIANOVA

score BY light target

/METHOD=SSTYPE(3)

/lmatrix 'Target within Light1'

light*target 1 -1 0 0 0 0 0 0 target 1 -1 0 0;

light*target 0 0 1 -1 0 0 0 0 target 0 0 1 -1

/lmatrix 'Target within Light2'

light*target 0 0 0 0 1 -1 0 0 target 1 -1 0 0;

light*target 0 0 0 0 0 0 1 -1 target 0 0 1 -1

 

运行后得到结果如下:

Custom Hypothesis Tests #1

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate .400
Hypothesized Value 0
Difference (Estimate - Hypothesized) .400
Std. Error .524
Sig. .451
95% Confidence Interval for Difference Lower Bound -.668
Upper Bound 1.468
a. Based on the user-specified contrast coefficients (L') matrix: L1 vs L2 within T1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast .400 1 .400 .582 .451
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #2

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate 3.600
Hypothesized Value 0
Difference (Estimate - Hypothesized) 3.600
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound 2.532
Upper Bound 4.668
a. Based on the user-specified contrast coefficients (L') matrix: L1 vs L2 within T2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 32.400 1 32.400 47.127 .000
Error 22.000 32 .687

 

 

Custom Hypothesis Tests #3

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate 3.200
Hypothesized Value 0
Difference (Estimate - Hypothesized) 3.200
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound 2.132
Upper Bound 4.268
a. Based on the user-specified contrast coefficients (L') matrix: L1 vs L2 within T3

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 25.600 1 25.600 37.236 .000
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #4

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate .400
Hypothesized Value 0
Difference (Estimate - Hypothesized) .400
Std. Error .524
Sig. .451
95% Confidence Interval for Difference Lower Bound -.668
Upper Bound 1.468
a. Based on the user-specified contrast coefficients (L') matrix: L1 vs L2 within T4

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast .400 1 .400 .582 .451
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #5

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -6.000
Hypothesized Value 0
Difference (Estimate - Hypothesized) -6.000
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -7.068
Upper Bound -4.932
L2 Contrast Estimate 1.600
Hypothesized Value 0
Difference (Estimate - Hypothesized) 1.600
Std. Error .524
Sig. .005
95% Confidence Interval for Difference Lower Bound .532
Upper Bound 2.668
a. Based on the user-specified contrast coefficients (L') matrix: Target within Light1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 96.400 2 48.200 70.109 .000
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #6

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -2.800
Hypothesized Value 0
Difference (Estimate - Hypothesized) -2.800
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -3.868
Upper Bound -1.732
L2 Contrast Estimate -1.200
Hypothesized Value 0
Difference (Estimate - Hypothesized) -1.200
Std. Error .524
Sig. .029
95% Confidence Interval for Difference Lower Bound -2.268
Upper Bound -.132
a. Based on the user-specified contrast coefficients (L') matrix: Target within Light2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 23.200 2 11.600 16.873 .000
Error 22.000 32 .687

 

 

上表中可以看出:

L1 VS L2 within T2以及L1 VS L2 within T3都是显著的,也就是说不同照明条件(L1,L2)在T2以及T3这两个目标大小条件下差异是显著的;

从Test#5 和 Test#6中可以看出,不同目标大小变量(T1,T2,T3,T4)在不同照明条件(L1,L2)两个水平上差异是显著的,由于目标大小变量有四个水平,故需要继续进行两两比较,共12对需要比较的均值。

两两比较继续使用LMATRIX语句,语句如下:

UNIANOVA

score BY light target

/METHOD=SSTYPE(3)

/lmatrix ‘T1 vs T2 within L1’

light*target 1 -1 0 0 0 0 0 0 target 1 -1 0 0

/lmatrix ‘T1 vs T3 within L1’

light*target 1 0 -1 0 0 0 0 0 target 1 0 -1 0

/lmatrix ‘T1 vs T4 within L1’

light*target 1 0 0 -1 0 0 0 0 target 1 0 0 -1

/lmatrix ‘T2 vs T3 within L1’

light*target 0 1 -1 0 0 0 0 0 target 0 1 -1 0

/lmatrix ‘T2 vs T4 within L1’

light*target 0 1 0 -1 0 0 0 0 target 0 1 0 -1

/lmatrix ‘T3 vs T4 within L1’

light*target 0 0 1 -1 0 0 0 0 target 0 0 1 -1

/lmatrix ‘T1 vs T2 within L2’

light*target 0 0 0 0 1 -1 0 0 target 1 -1 0 0

/lmatrix ‘T1 vs T3 within L2’

light*target 0 0 0 0 1 0 -1 0 target 1 0 -1 0

/lmatrix ‘T1 vs T4 within L2’

light*target 0 0 0 0 1 0 0 -1 target 1 0 0 -1

/lmatrix ‘T2 vs T3 within L2’

light*target 0 0 0 0 0 1 -1 0 target 0 1 -1 0

/lmatrix ‘T2 vs T4 within L2’

light*target 0 0 0 0 0 1 0 -1 target 0 1 0 -1

/lmatrix ‘T3 vs T4 within L2’

light*target 0 0 0 0 0 0 1 -1 target 0 0 1 -1.

 

 

得到如下结果:

 

Custom Hypothesis Tests #1

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -6.000
Hypothesized Value 0
Difference (Estimate - Hypothesized) -6.000
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -7.068
Upper Bound -4.932
a. Based on the user-specified contrast coefficients (L') matrix: T1 vs T2 within L1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 90.000 1 90.000 130.909 .000
Error 22.000 32 .687

 

 

Custom Hypothesis Tests #2

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -6.800
Hypothesized Value 0
Difference (Estimate - Hypothesized) -6.800
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -7.868
Upper Bound -5.732
a. Based on the user-specified contrast coefficients (L') matrix: T1 vs T3 within L1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 115.600 1 115.600 168.145 .000
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #3

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -5.200
Hypothesized Value 0
Difference (Estimate - Hypothesized) -5.200
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -6.268
Upper Bound -4.132
a. Based on the user-specified contrast coefficients (L') matrix: T1 vs T4 within L1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 67.600 1 67.600 98.327 .000
Error 22.000 32 .687

 

 

Custom Hypothesis Tests #4

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -.800
Hypothesized Value 0
Difference (Estimate - Hypothesized) -.800
Std. Error .524
Sig. .137
95% Confidence Interval for Difference Lower Bound -1.868
Upper Bound .268
a. Based on the user-specified contrast coefficients (L') matrix: T2 vs T3 within L1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 1.600 1 1.600 2.327 .137
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #5

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate .800
Hypothesized Value 0
Difference (Estimate - Hypothesized) .800
Std. Error .524
Sig. .137
95% Confidence Interval for Difference Lower Bound -.268
Upper Bound 1.868
a. Based on the user-specified contrast coefficients (L') matrix: T2 vs T4 within L1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 1.600 1 1.600 2.327 .137
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #6

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate 1.600
Hypothesized Value 0
Difference (Estimate - Hypothesized) 1.600
Std. Error .524
Sig. .005
95% Confidence Interval for Difference Lower Bound .532
Upper Bound 2.668
a. Based on the user-specified contrast coefficients (L') matrix: T3 vs T4 within L1

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 6.400 1 6.400 9.309 .005
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #7

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -2.800
Hypothesized Value 0
Difference (Estimate - Hypothesized) -2.800
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -3.868
Upper Bound -1.732
a. Based on the user-specified contrast coefficients (L') matrix: T1 vs T2 within L2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 19.600 1 19.600 28.509 .000
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #8

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -4.000
Hypothesized Value 0
Difference (Estimate - Hypothesized) -4.000
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -5.068
Upper Bound -2.932
a. Based on the user-specified contrast coefficients (L') matrix: T1 vs T3 within L2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 40.000 1 40.000 58.182 .000
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #9

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -5.200
Hypothesized Value 0
Difference (Estimate - Hypothesized) -5.200
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -6.268
Upper Bound -4.132
a. Based on the user-specified contrast coefficients (L') matrix: T1 vs T4 within L2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 67.600 1 67.600 98.327 .000
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #10

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -1.200
Hypothesized Value 0
Difference (Estimate - Hypothesized) -1.200
Std. Error .524
Sig. .029
95% Confidence Interval for Difference Lower Bound -2.268
Upper Bound -.132
a. Based on the user-specified contrast coefficients (L') matrix: T2 vs T3 within L2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 3.600 1 3.600 5.236 .029
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #11

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -2.400
Hypothesized Value 0
Difference (Estimate - Hypothesized) -2.400
Std. Error .524
Sig. .000
95% Confidence Interval for Difference Lower Bound -3.468
Upper Bound -1.332
a. Based on the user-specified contrast coefficients (L') matrix: T2 vs T4 within L2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 14.400 1 14.400 20.945 .000
Error 22.000 32 .687

 

 

 

Custom Hypothesis Tests #12

Contrast Results (K Matrix)a
Contrast Dependent Variable
得分
L1 Contrast Estimate -1.200
Hypothesized Value 0
Difference (Estimate - Hypothesized) -1.200
Std. Error .524
Sig. .029
95% Confidence Interval for Difference Lower Bound -2.268
Upper Bound -.132
a. Based on the user-specified contrast coefficients (L') matrix: T3 vs T4 within L2

 

Test Results
Dependent Variable: 得分
Source Sum of Squares df Mean Square F Sig.
Contrast 3.600 1 3.600 5.236 .029
Error 22.000 32 .687

 

从上表可看出,两两比较显著的有:

p***(T1 vs T2 within L1);

p***(T1 vs T3 within L1);

p***(T1 vs T4 within L1);

p***(T3 vs T4 within L1);

p***(T1 vs T2 within L2);

p***(T1 vs T3 within L2);

p***(T1 vs T4 within L2);

p*(T2 vs T3 within L2);

p***(T2 vs T4 within L2);

p*(T3 vs T4 within L2)。

 

 

 

Task 1:多自变量回归

使用SPSS进行回归模型操作

注:完整带截图版报告见网盘分享word文件!

数据来源:科技人才工作压力调查问卷:

http://www.cnsda.org/index.php?r=projects/view&id=34848626

视频地址:

https://pan.baidu.com/s/1i5LUy5f

高清无水印版:

https://pan.baidu.com/s/1eR4jyb4

word版报告及结果屏幕截图见:

https://pan.baidu.com/s/1c221OaO

变量

IV:挑战性压力指数;阻碍性压力指数

DV:员工工作满意度

变量释义:

Cavanaugh等人(2000)明确提出了挑战性—阻碍性压力源(challenge and hindrance stressor)概念,按压力的属性对不同压力源进行了区分。挑战性压力源所带来的压力个体认为能够克服,对自己的工作绩效与成长具有积极意义,例如工作负荷、时间紧迫性、工作范围与职责、工作复 杂性等;阻碍性压力源所带来的压力个体认为难 以克服,对自己工作目标的实现与职业生涯的发展具有阻碍作用,例如组织政治、角色模糊与冲突、官僚程序、工作不安全感等(Cavanaugh等, 2000;LePine等, 2005;Webster 等, 2011)。先前研究表明挑战性压力源与阻碍性压力源对个体工作态度与行为的影响存在差异,主要表现为挑战性压力源能够对工作态度产生正向影响,而阻碍性压力源则相反。Cavanaugh等人研究发现,挑战性压力源与管理者的工作满意度正相关,而阻碍性压力源则与工作满意度负相关。

本次操作选用数据调查对象为来自两个国家级科研院所,一所北京市属高层次科研人员服务机构和一所著名央企的研发部的科研人员。共计发放问卷1000份,回收问卷479份,其中有效问卷364份,问卷回收率为48.90%,有效问卷回收率为36.40%。

变量挑战性压力指数包括11道题目,采用五点计分法,分数越高表示压力越大;变量阻碍性压力指数包括15道题目,采用七点计分法,分数越高表示压力越大。工作满意度共八道题,采用七点计分。

 

APA Section:

A multiple regression analysis was conducted to evaluate how well the challenge and hindrance stressor predicted Job Satisfaction. The predictors were the two stressor, which the criterion variable was the Job Satisfaction. The linear combination of stressor was significantly related to the Job Satisfaction, F (2, 361 ) = 92.92, p<.001. The sample correlation coefficient was .58 indicating that approximately 34% of the variance of the Job Satisfaction in the sample can be accounted for by the linear combination of the two kind of stressors.

In Table 1, we present indices to indicate the relative strength of the individual predictors. The bivariate correlations reveal that the bivariate between the challenge stressor and Job Satisfaction was positive while the bivariate between the hindrance stressor and Job Satisfaction was negative, as expected, all of the two indices were statistically significant (p < .01). Also, all the partial correlation between the two stressor and Job Satisfaction was significant. On the basis of these correlation analyses, we can conclude that the two stressors are the useful predictors. The challenge stressor counted for 29%(-.538 = .29) of the variance of the Job Satisfaction, while the hindrance stressor contributed an additional 5%(34% - 29%). And the correlation between the two stressors is 0.01 which is not is statistically significant.

 

Table 1
The Bivariate and Partial Correlations of the Predictor with Job Satisfaction
Predictors Correlation between each predictor and the Job Satisfaction Correlation between each predictor and the Job Satisfaction controlling for all other predictors
阻碍性压力源 -0.54** -0.55*
挑战性压力源 0.232** 0.267*
p < .05, ** < .01