#############################

# 調整媒介分析（R: mediationパッケージ）

# Output

#############################

# 「回避」を調整変数,「自己活用接近対処」を媒介変数とする調整媒介分析

## 「回避」が平均値の時のの偏回帰係数

> summary(a.int)

Call:

lm(formula = 自己活用接近対処_c ~ 性別_c + 現浪_c +

学年_c + 自己活用回避対処_c * 回避_c, data = dat)

Residuals:

Min 1Q Median 3Q Max

-3.5915 -0.4702 0.0227 0.5091 2.2638

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 0.046621 0.037451 1.245 0.213794

性別_c 0.092035 0.074244 1.240 0.215726

現浪_c 0.356854 0.096679 3.691 0.000249 ***

学年_c 0.001785 0.035089 0.051 0.959461

自己活用回避対処_c 0.533513 0.039907 13.369 < 2e-16 ***

回避_c -0.157060 0.039414 -3.985 7.81e-05 ***

自己活用回避対処_c:回避_c -0.173131 0.029571 -5.855 8.90e-09 ***

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.8051 on 477 degrees of freedom

Multiple R-squared: 0.3623, Adjusted R-squared: 0.3543

F-statistic: 45.16 on 6 and 477 DF, p-value: < 2.2e-16

> summary(b.int)

Call:

lm(formula = 情動知能の成長感 ~ 性別_c + 現浪_c +

学年_c + 自己活用接近対処_c * 回避_c + 自己活用回避対処_c *

回避_c, data = dat)

Residuals:

Min 1Q Median 3Q Max

-3.5087 -0.3338 0.0484 0.4787 1.9595

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 3.150662 0.034390 91.616 <2e-16 ***

性別_c 0.069475 0.067945 1.023 0.3071

現浪_c -0.007414 0.089477 -0.083 0.9340

学年_c -0.058930 0.032053 -1.838 0.0666 .

自己活用接近対処_c 0.485062 0.041804 11.603 <2e-16 ***

回避_c 0.066103 0.036611 1.806 0.0716 .

自己活用回避対処_c 0.088382 0.043547 2.030 0.0430 *

自己活用接近対処_c:回避_c -0.029926 0.031820 -0.940 0.3475

回避_c:自己活用回避対処_c -0.025532 0.032490 -0.786 0.4324

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.7347 on 475 degrees of freedom

Multiple R-squared: 0.3882, Adjusted R-squared: 0.3779

F-statistic: 37.67 on 8 and 475 DF, p-value: < 2.2e-16

## 「回避」が+1SDの時の偏回帰係数

> summary(a.int.plus1SD)

Call:

lm(formula = 自己活用接近対処_c ~ 性別_c + 現浪_c +

学年_c + 自己活用回避対処_c * 回避_high, data = dat)

Residuals:

Min 1Q Median 3Q Max

-3.5915 -0.4702 0.0227 0.5091 2.2638

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) -0.108769 0.053372 -2.038 0.042108 *

性別_c 0.092035 0.074244 1.240 0.215726

現浪_c 0.356854 0.096679 3.691 0.000249 ***

学年_c 0.001785 0.035089 0.051 0.959461

自己活用回避対処_c 0.362222 0.052207 6.938 1.30e-11 ***

回避_high -0.157060 0.039414 -3.985 7.81e-05 ***

自己活用回避対処_c:回避_high -0.173131 0.029571 -5.855 8.90e-09 ***

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.8051 on 477 degrees of freedom

Multiple R-squared: 0.3623, Adjusted R-squared: 0.3543

F-statistic: 45.16 on 6 and 477 DF, p-value: < 2.2e-16

> summary(b.int.plus1SD)

Call:

lm(formula = 情動知能の成長感 ~ 性別_c + 現浪_c +

学年_c + 自己活用接近対処_c * 回避_high + 自己活用回避対処_c *

回避_high, data = dat)

Residuals:

Min 1Q Median 3Q Max

-3.5087 -0.3338 0.0484 0.4787 1.9595

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 3.216063 0.049190 65.381 <2e-16 ***

性別_c 0.069475 0.067945 1.023 0.3071

現浪_c -0.007414 0.089477 -0.083 0.9340

学年_c -0.058930 0.032053 -1.838 0.0666 .

自己活用接近対処_c 0.455455 0.051521 8.840 <2e-16 ***

回避_high 0.066103 0.036611 1.806 0.0716 .

自己活用回避対処_c 0.063121 0.050601 1.247 0.2129

自己活用接近対処_c:回避_high -0.029926 0.031820 -0.940 0.3475

回避_high:自己活用回避対処_c -0.025532 0.032490 -0.786 0.4324

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.7347 on 475 degrees of freedom

Multiple R-squared: 0.3882, Adjusted R-squared: 0.3779

F-statistic: 37.67 on 8 and 475 DF, p-value: < 2.2e-16

## 「回避」が-1SDの時の偏回帰係数

> summary(a.int.minus1SD)

Call:

lm(formula = 自己活用接近対処_c ~ 性別_c + 現浪_c +

学年_c + 自己活用回避対処_c * 回避_low, data = dat)

Residuals:

Min 1Q Median 3Q Max

-3.5915 -0.4702 0.0227 0.5091 2.2638

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 0.202012 0.054752 3.690 0.000251 ***

性別_c 0.092035 0.074244 1.240 0.215726

現浪_c 0.356854 0.096679 3.691 0.000249 ***

学年_c 0.001785 0.035089 0.051 0.959461

自己活用回避対処_c 0.704803 0.046599 15.125 < 2e-16 ***

回避_low -0.157060 0.039414 -3.985 7.81e-05 ***

自己活用回避対処_c:回避_low -0.173131 0.029571 -5.855 8.90e-09 ***

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.8051 on 477 degrees of freedom

Multiple R-squared: 0.3623, Adjusted R-squared: 0.3543

F-statistic: 45.16 on 6 and 477 DF, p-value: < 2.2e-16

> summary(b.int.minus1SD)

Call:

lm(formula = 情動知能の成長感 ~ 性別_c + 現浪_c +

学年_c + 自己活用接近対処_c * 回避_low + 自己活用回避対処_c *

回避_low, data = dat)

Residuals:

Min 1Q Median 3Q Max

-3.5087 -0.3338 0.0484 0.4787 1.9595

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 3.085262 0.050692 60.863 <2e-16 ***

性別_c 0.069475 0.067945 1.023 0.3071

現浪_c -0.007414 0.089477 -0.083 0.9340

学年_c -0.058930 0.032053 -1.838 0.0666 .

自己活用接近対処_c 0.514670 0.053132 9.687 <2e-16 ***

回避_low 0.066103 0.036611 1.806 0.0716 .

自己活用回避対処_c 0.113642 0.057435 1.979 0.0484 *

自己活用接近対処_c:回避_low -0.029926 0.031820 -0.940 0.3475

回避_low:自己活用回避対処_c -0.025532 0.032490 -0.786 0.4324

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.7347 on 475 degrees of freedom

Multiple R-squared: 0.3882, Adjusted R-squared: 0.3779

F-statistic: 37.67 on 8 and 475 DF, p-value: < 2.2e-16

## 間接効果の検定

> summary(result.plus1SD)

Causal Mediation Analysis

Nonparametric Bootstrap Confidence Intervals with the BCa Method

(Inference Conditional on the Covariate Values Specified in `covariates')

Estimate 95% CI Lower 95% CI Upper p-value

ACME 0.1650 0.0889 0.2872 0.00

ADE 0.0631 -0.1293 0.2037 0.50

Total Effect 0.2281 0.0608 0.3666 0.01

Prop. Mediated 0.7233 0.4870 11.7464 0.01

Sample Size Used: 484

Simulations: 5000

> summary(result.minus1SD)

Causal Mediation Analysis

Nonparametric Bootstrap Confidence Intervals with the BCa Method

(Inference Conditional on the Covariate Values Specified in `covariates')

Estimate 95% CI Lower 95% CI Upper p-value

ACME 0.3627 0.2691 0.4813 0.00

ADE 0.1136 -0.0302 0.2463 0.12

Total Effect 0.4764 0.3738 0.5710 0.00

Prop. Mediated 0.7614 0.5550 1.1034 0.00

Sample Size Used: 484

Simulations: 5000

# ACMEが間接効果を指します。