한국인의 난민에 대한 태도 예측요인 탐색을 위한 다중회귀분석 및 상호작용효과 분석
- 아래는 디아스포라연구(KCI 등재)에 게재한 논문을 요약한 내용이다. 분석에 활용한 코드도 함께 제시하였다.
- 전재은, 박기나. (2023). 한국인의 난민신청자(asylum seeker)에 대한 태도: 통합위협이론과 접촉이론을 중심으로. 디아스포라연구, 17(1), 145-195.
- 2018년 제주 예멘 난민 사태로 인해 한국에서 난민 수용에 대한 찬반 논쟁이 불거진 가운데,
Abstract
1. Objective
- The purpose of this study is to explore factors which predict Korean public’s attitude toward asylum seekers, using integrated threat theory and contact theory as theoretical frameworks.
2. Methods
- Data were collected in February 2021 via online survey with 800 adults in Korea.
3. Results
- The findings showed that integrated threat theory and contact theory are useful frameworks in explaining Koreans’ attitude towards asylum seekers.
- As the results of multiple regression analyses, realistic, symbolic, and safety and cohesion threats were shown to predict attitudes toward refugees. Most of the participants in this study were found to contact with refugees (asylum seekers) through mass media or the Internet and SNS. Analyzing by contact types also confirmed the importance of contact quality in predicting attitudes toward refugees.
- In addition, the analysis of moderating effect showed the significant moderating effect of contact quality on the relationship between the frequency of mass media contact and positive behavioral intention toward asylum seekers.
- In other words, increase in the frequency of contact through mass media is negatively associated with positive behavioral intention toward asylum seekers, but as the positive evaluation of contact through mass media increases, positive behavioral intention toward asylum seekers also increases.
4. Conclusion & Implications
- Accordingly, this study suggested considering ways to facilitate the formation of public’s positive attitude toward refugees by promoting understanding refugees and positive contact experiences with them through mass media and the Internet.
Code
from google.colab import drive
drive.mount('/content/drive')
!pip install seaborn
!pip install statsmodels
!pip install pyprocessmacro
!pip install scikit-learn
!pip install pyreadstat
!sudo apt-get install -y fonts-nanum
!sudo fc-cache -fv
!rm ~/.cache/matplotlib -rf
import warnings
warnings.filterwarnings('ignore')
import pandas
import pyreadstat
import matplotlib.pyplot as plt
import seaborn as sns
from statsmodels.formula.api import ols
import statsmodels.api as sm
from pyprocessmacro import Process
dtafile = '/content/drive/MyDrive/Analysis_0602/refugee_0426.2023.dta'
df, meta = pyreadstat.read_dta(dtafile)
df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 800 entries, 0 to 799
Columns: 287 entries, no to _Ireligion2_2
dtypes: float64(279), int64(6), object(2)
memory usage: 1.8+ MB
# 데이터 값 변경
df.replace({"q92":{1:0, 2:1, 3:2, 4:3, 5:4},
"q96":{1:0, 2:1, 3:2, 4:3, 5:4},
"q97":{1:0, 2:1, 3:2, 4:3, 5:4},
"q175":{2:0}}, inplace=True)
# 데이터 컬럼명 변경
df.rename(columns = {'q1r_gender' : 'gender',
'q2_1':'age',
'q175':'child',
'_Iq5_region_1':'region_1',
'_Iq5_region_2':'region_2',
'q180edu2':'edulevel',
'q181':'income',
'_Ireligion2_1':'religion_1',
'_Ireligion2_2':'religion_2',
'q92':'dir_contact',
'q96':'massmedia',
'q97':'sns',
'q99':'dir_cont_qual',
'q103':'mass_media_qual',
'q104':'sns_qual',
'q43':'poli_pers',
'q44':'relig_pers',
'q45':'econo',
'q46':'fake',
'q153':'pol_orien'}, inplace = True)
# 다양한 데이터셋 구성
# N = 800
columns = ["gender", "age", "child", "region_1", "region_2", "edulevel", "income", "religion_1", "religion_2",
"T_realf", "T_symbf", "T_safecoh2", "T_health", "dir_contact", "massmedia", "sns",
"poli_pers", "relig_pers", "econo", "fake",
"humani", "nat_eth2", "nat_cit2", "SDO_D", "SDO_E", "rightw", "pol_orien", "islamo", "socd_p", "socd_n", "beh_p", "beh_n"]
df_800 = df[columns]
# N = 410 (직접 접촉 경험 o)
df_410 = df[df["dir_cont_qual"] != 6.0]
columns = ["gender", "age", "child", "region_1", "region_2", "edulevel", "income", "religion_1", "religion_2",
"T_realf", "T_symbf", "T_safecoh2", "T_health", "dir_contact", "massmedia", "sns", 'dir_cont_qual',
"poli_pers", "relig_pers", "econo", "fake",
"humani", "nat_eth2", "nat_cit2", "SDO_D", "SDO_E", "rightw", "pol_orien", "islamo", "socd_p", "socd_n", "beh_p", "beh_n"]
df_410 = df_410[columns]
# N = 715 (대중매체 접촉 경험 o)
df_715 = df[df["mass_media_qual"] != 6.0]
columns = ["gender", "age", "child", "region_1", "region_2", "edulevel", "income", "religion_1", "religion_2",
"T_realf", "T_symbf", "T_safecoh2", "T_health", "dir_contact", "massmedia", "sns", 'mass_media_qual',
"poli_pers", "relig_pers", "econo", "fake",
"humani", "nat_eth2", "nat_cit2", "SDO_D", "SDO_E", "rightw", "pol_orien", "islamo", "socd_p", "socd_n", "beh_p", "beh_n"]
df_715 = df_715[columns]
# N = 591 (SNS 접촉 경험 o)
df_591 = df[df["sns_qual"] != 6.0]
columns = ["gender", "age", "child", "region_1", "region_2", "edulevel", "income", "religion_1", "religion_2",
"T_realf", "T_symbf", "T_safecoh2", "T_health", "dir_contact", "massmedia", "sns", 'sns_qual',
"poli_pers", "relig_pers", "econo", "fake",
"humani", "nat_eth2", "nat_cit2", "SDO_D", "SDO_E", "rightw", "pol_orien", "islamo", "socd_p", "socd_n", "beh_p", "beh_n"]
df_591 = df_591[columns]
# 다중회귀분석
def regression_model(df, y_data):
x_data = df.drop(["beh_n", "beh_p"], axis=1)
x_data_add = sm.add_constant(x_data, has_constant = "add")
y_data = y_data
reg_model = sm.OLS(y_data, x_data_add)
fitted_reg_model = reg_model.fit()
return fitted_reg_model.summary()
# 조절효과분석
def moderation_model(df, x_var, m_var, y_var):
control_var = df.drop(["beh_n", "beh_p"], axis=1).columns.tolist()
control_var.remove(x_var)
control_var.remove(m_var)
p = Process(data=df,
model=1,
x=x_var,
m=m_var,
y=y_var,
controls=control_var)
print(f"******************** 종속변수 {y_var}에 대한 {x_var} x {m_var}의 조절효과 분석 결과********************")
return p.summary()
# 회귀분석 예시
y_list = ["beh_p", "beh_n"]
df_list = [df_800, df_410, df_715, df_591]
n_list = [800, 410, 715, 591]
for (df, n) in zip(df_list, n_list):
for y in y_list:
print(f" n = {n}일때, 종속변수 {y}에 대한 회귀분석 결과")
print(regression_model(df, df[y]))
print()
print()
n = 800일때, 종속변수 beh_p에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_p R-squared: 0.628
Model: OLS Adj. R-squared: 0.613
Method: Least Squares F-statistic: 43.24
Date: Sun, 04 Jun 2023 Prob (F-statistic): 3.11e-143
Time: 07:33:41 Log-Likelihood: -615.98
No. Observations: 800 AIC: 1294.
Df Residuals: 769 BIC: 1439.
Df Model: 30
Covariance Type: nonrobust
===============================================================================
coef std err t P>|t| [0.025 0.975]
-------------------------------------------------------------------------------
const 2.2106 0.343 6.450 0.000 1.538 2.883
gender -0.0426 0.041 -1.033 0.302 -0.123 0.038
age 0.0018 0.002 0.849 0.396 -0.002 0.006
child 0.1341 0.056 2.409 0.016 0.025 0.243
region_1 0.0792 0.040 1.977 0.048 0.001 0.158
region_2 -0.1348 0.174 -0.775 0.439 -0.476 0.207
edulevel 0.0206 0.013 1.569 0.117 -0.005 0.046
income -0.0086 0.010 -0.835 0.404 -0.029 0.012
religion_1 -0.0466 0.058 -0.808 0.419 -0.160 0.067
religion_2 -0.0715 0.053 -1.358 0.175 -0.175 0.032
T_realf -0.2353 0.033 -7.047 0.000 -0.301 -0.170
T_symbf -0.1059 0.028 -3.818 0.000 -0.160 -0.051
T_safecoh2 -0.0992 0.037 -2.689 0.007 -0.172 -0.027
T_health 0.0167 0.025 0.668 0.505 -0.033 0.066
dir_contact 0.1036 0.029 3.587 0.000 0.047 0.160
massmedia -0.0312 0.022 -1.419 0.156 -0.074 0.012
sns 0.0597 0.020 3.009 0.003 0.021 0.099
poli_pers -0.0041 0.027 -0.150 0.881 -0.057 0.049
relig_pers 0.1573 0.025 6.187 0.000 0.107 0.207
econo 0.1363 0.023 5.879 0.000 0.091 0.182
fake 0.0321 0.021 1.559 0.119 -0.008 0.073
humani 0.1994 0.030 6.645 0.000 0.140 0.258
nat_eth2 0.0257 0.029 0.872 0.383 -0.032 0.084
nat_cit2 -0.0521 0.035 -1.475 0.141 -0.121 0.017
SDO_D 0.0348 0.033 1.069 0.285 -0.029 0.099
SDO_E -0.0638 0.027 -2.387 0.017 -0.116 -0.011
rightw -0.0372 0.062 -0.596 0.551 -0.160 0.085
pol_orien -0.0546 0.025 -2.220 0.027 -0.103 -0.006
islamo -0.0045 0.022 -0.205 0.837 -0.047 0.038
socd_p 0.0291 0.036 0.801 0.423 -0.042 0.100
socd_n 0.0593 0.030 1.975 0.049 0.000 0.118
==============================================================================
Omnibus: 10.391 Durbin-Watson: 1.989
Prob(Omnibus): 0.006 Jarque-Bera (JB): 13.156
Skew: 0.156 Prob(JB): 0.00139
Kurtosis: 3.545 Cond. No. 897.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
n = 800일때, 종속변수 beh_n에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_n R-squared: 0.287
Model: OLS Adj. R-squared: 0.260
Method: Least Squares F-statistic: 10.34
Date: Sun, 04 Jun 2023 Prob (F-statistic): 1.72e-39
Time: 07:33:41 Log-Likelihood: -989.51
No. Observations: 800 AIC: 2041.
Df Residuals: 769 BIC: 2186.
Df Model: 30
Covariance Type: nonrobust
===============================================================================
coef std err t P>|t| [0.025 0.975]
-------------------------------------------------------------------------------
const 0.6277 0.547 1.148 0.251 -0.445 1.701
gender -0.0556 0.066 -0.846 0.398 -0.185 0.073
age -0.0045 0.003 -1.307 0.192 -0.011 0.002
child 0.1500 0.089 1.690 0.091 -0.024 0.324
region_1 0.0377 0.064 0.590 0.555 -0.088 0.163
region_2 0.2390 0.277 0.861 0.389 -0.306 0.784
edulevel 0.0350 0.021 1.674 0.094 -0.006 0.076
income -0.0142 0.016 -0.866 0.386 -0.046 0.018
religion_1 0.1169 0.092 1.270 0.205 -0.064 0.298
religion_2 -0.1270 0.084 -1.513 0.131 -0.292 0.038
T_realf -0.0269 0.053 -0.504 0.614 -0.131 0.078
T_symbf -0.0058 0.044 -0.131 0.896 -0.093 0.081
T_safecoh2 0.2529 0.059 4.299 0.000 0.137 0.368
T_health 0.0254 0.040 0.635 0.525 -0.053 0.104
dir_contact 0.1578 0.046 3.426 0.001 0.067 0.248
massmedia -0.0008 0.035 -0.022 0.983 -0.070 0.068
sns 0.0122 0.032 0.385 0.700 -0.050 0.074
poli_pers -0.0821 0.043 -1.894 0.059 -0.167 0.003
relig_pers 0.0511 0.041 1.261 0.208 -0.028 0.131
econo 0.0537 0.037 1.451 0.147 -0.019 0.126
fake 0.1633 0.033 4.969 0.000 0.099 0.228
humani -0.1283 0.048 -2.681 0.008 -0.222 -0.034
nat_eth2 0.0731 0.047 1.553 0.121 -0.019 0.165
nat_cit2 -0.0822 0.056 -1.458 0.145 -0.193 0.028
SDO_D 0.1119 0.052 2.158 0.031 0.010 0.214
SDO_E -0.0323 0.043 -0.758 0.449 -0.116 0.051
rightw 0.0907 0.100 0.911 0.363 -0.105 0.286
pol_orien -0.0295 0.039 -0.752 0.452 -0.107 0.048
islamo 0.0544 0.035 1.565 0.118 -0.014 0.123
socd_p 0.0612 0.058 1.057 0.291 -0.052 0.175
socd_n 0.0036 0.048 0.075 0.940 -0.090 0.098
==============================================================================
Omnibus: 2.935 Durbin-Watson: 2.199
Prob(Omnibus): 0.230 Jarque-Bera (JB): 2.986
Skew: 0.128 Prob(JB): 0.225
Kurtosis: 2.847 Cond. No. 897.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
n = 410일때, 종속변수 beh_p에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_p R-squared: 0.700
Model: OLS Adj. R-squared: 0.675
Method: Least Squares F-statistic: 28.44
Date: Sun, 04 Jun 2023 Prob (F-statistic): 4.98e-80
Time: 07:33:41 Log-Likelihood: -284.06
No. Observations: 410 AIC: 632.1
Df Residuals: 378 BIC: 760.6
Df Model: 31
Covariance Type: nonrobust
=================================================================================
coef std err t P>|t| [0.025 0.975]
---------------------------------------------------------------------------------
const 1.8867 0.472 3.993 0.000 0.958 2.816
gender -0.0094 0.056 -0.170 0.865 -0.119 0.100
age 0.0062 0.003 2.029 0.043 0.000 0.012
child 0.0332 0.078 0.426 0.670 -0.120 0.187
region_1 0.0912 0.055 1.663 0.097 -0.017 0.199
region_2 -0.1818 0.263 -0.691 0.490 -0.699 0.336
edulevel 0.0014 0.018 0.078 0.938 -0.034 0.037
income -0.0034 0.014 -0.243 0.808 -0.030 0.024
religion_1 0.0050 0.076 0.066 0.948 -0.145 0.155
religion_2 -0.0063 0.071 -0.090 0.929 -0.145 0.132
T_realf -0.3206 0.045 -7.057 0.000 -0.410 -0.231
T_symbf -0.1099 0.038 -2.872 0.004 -0.185 -0.035
T_safecoh2 0.0037 0.050 0.076 0.940 -0.094 0.101
T_health -0.0199 0.035 -0.574 0.566 -0.088 0.048
dir_contact 0.0277 0.032 0.855 0.393 -0.036 0.091
massmedia -0.0950 0.029 -3.263 0.001 -0.152 -0.038
sns 0.0927 0.028 3.299 0.001 0.037 0.148
dir_cont_qual 0.1016 0.031 3.277 0.001 0.041 0.163
poli_pers 0.0112 0.038 0.292 0.771 -0.064 0.086
relig_pers 0.2035 0.036 5.710 0.000 0.133 0.274
econo 0.1195 0.033 3.567 0.000 0.054 0.185
fake 0.0301 0.027 1.120 0.264 -0.023 0.083
humani 0.1254 0.042 3.020 0.003 0.044 0.207
nat_eth2 -0.0446 0.043 -1.029 0.304 -0.130 0.041
nat_cit2 -0.0148 0.047 -0.312 0.755 -0.108 0.078
SDO_D 0.0697 0.045 1.545 0.123 -0.019 0.158
SDO_E -0.0664 0.033 -2.040 0.042 -0.130 -0.002
rightw 0.0286 0.096 0.300 0.765 -0.159 0.216
pol_orien -0.0191 0.033 -0.585 0.559 -0.083 0.045
islamo 0.0104 0.029 0.356 0.722 -0.047 0.068
socd_p -0.0018 0.051 -0.036 0.971 -0.101 0.098
socd_n 0.0791 0.041 1.928 0.055 -0.002 0.160
==============================================================================
Omnibus: 2.074 Durbin-Watson: 1.936
Prob(Omnibus): 0.355 Jarque-Bera (JB): 2.026
Skew: 0.037 Prob(JB): 0.363
Kurtosis: 3.336 Cond. No. 941.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
n = 410일때, 종속변수 beh_n에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_n R-squared: 0.300
Model: OLS Adj. R-squared: 0.242
Method: Least Squares F-statistic: 5.216
Date: Sun, 04 Jun 2023 Prob (F-statistic): 9.96e-16
Time: 07:33:41 Log-Likelihood: -504.78
No. Observations: 410 AIC: 1074.
Df Residuals: 378 BIC: 1202.
Df Model: 31
Covariance Type: nonrobust
=================================================================================
coef std err t P>|t| [0.025 0.975]
---------------------------------------------------------------------------------
const 1.3572 0.809 1.677 0.094 -0.234 2.949
gender -0.1189 0.095 -1.249 0.212 -0.306 0.068
age -0.0055 0.005 -1.046 0.296 -0.016 0.005
child 0.1937 0.134 1.448 0.148 -0.069 0.457
region_1 0.1630 0.094 1.735 0.084 -0.022 0.348
region_2 0.1942 0.451 0.431 0.667 -0.692 1.081
edulevel 0.0370 0.031 1.194 0.233 -0.024 0.098
income -0.0055 0.024 -0.231 0.817 -0.052 0.041
religion_1 -0.0077 0.130 -0.059 0.953 -0.264 0.249
religion_2 -0.2058 0.121 -1.702 0.090 -0.444 0.032
T_realf -0.1317 0.078 -1.692 0.091 -0.285 0.021
T_symbf 0.0167 0.066 0.254 0.799 -0.112 0.146
T_safecoh2 0.2559 0.085 3.013 0.003 0.089 0.423
T_health 0.0797 0.059 1.345 0.179 -0.037 0.196
dir_contact 0.1305 0.056 2.350 0.019 0.021 0.240
massmedia 0.0051 0.050 0.102 0.919 -0.093 0.103
sns 0.0012 0.048 0.024 0.981 -0.094 0.096
dir_cont_qual -0.0210 0.053 -0.395 0.693 -0.125 0.083
poli_pers -0.0660 0.066 -1.007 0.315 -0.195 0.063
relig_pers 0.0966 0.061 1.583 0.114 -0.023 0.217
econo 0.0301 0.057 0.525 0.600 -0.083 0.143
fake 0.1779 0.046 3.865 0.000 0.087 0.268
humani -0.1210 0.071 -1.700 0.090 -0.261 0.019
nat_eth2 0.0055 0.074 0.074 0.941 -0.141 0.152
nat_cit2 -0.0291 0.081 -0.360 0.719 -0.188 0.130
SDO_D 0.1657 0.077 2.143 0.033 0.014 0.318
SDO_E -0.0748 0.056 -1.343 0.180 -0.184 0.035
rightw -0.1557 0.164 -0.952 0.342 -0.477 0.166
pol_orien -0.0305 0.056 -0.546 0.586 -0.140 0.079
islamo 0.0678 0.050 1.358 0.175 -0.030 0.166
socd_p 0.0840 0.087 0.967 0.334 -0.087 0.255
socd_n -0.0026 0.070 -0.037 0.970 -0.141 0.136
==============================================================================
Omnibus: 0.660 Durbin-Watson: 1.989
Prob(Omnibus): 0.719 Jarque-Bera (JB): 0.767
Skew: 0.048 Prob(JB): 0.681
Kurtosis: 2.811 Cond. No. 941.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
n = 715일때, 종속변수 beh_p에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_p R-squared: 0.640
Model: OLS Adj. R-squared: 0.624
Method: Least Squares F-statistic: 39.15
Date: Sun, 04 Jun 2023 Prob (F-statistic): 1.20e-129
Time: 07:33:41 Log-Likelihood: -546.93
No. Observations: 715 AIC: 1158.
Df Residuals: 683 BIC: 1304.
Df Model: 31
Covariance Type: nonrobust
===================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------
const 1.9990 0.364 5.495 0.000 1.285 2.713
gender -0.0214 0.043 -0.493 0.622 -0.106 0.064
age 0.0026 0.002 1.102 0.271 -0.002 0.007
child 0.0912 0.060 1.530 0.126 -0.026 0.208
region_1 0.0917 0.042 2.169 0.030 0.009 0.175
region_2 -0.1127 0.184 -0.614 0.540 -0.474 0.248
edulevel 0.0204 0.014 1.472 0.141 -0.007 0.048
income -0.0075 0.011 -0.681 0.496 -0.029 0.014
religion_1 -0.0668 0.061 -1.093 0.275 -0.187 0.053
religion_2 -0.0726 0.055 -1.309 0.191 -0.182 0.036
T_realf -0.2497 0.036 -6.968 0.000 -0.320 -0.179
T_symbf -0.0991 0.029 -3.388 0.001 -0.157 -0.042
T_safecoh2 -0.0582 0.040 -1.452 0.147 -0.137 0.020
T_health 0.0108 0.026 0.408 0.683 -0.041 0.063
dir_contact 0.0958 0.030 3.203 0.001 0.037 0.154
massmedia -0.0349 0.025 -1.411 0.159 -0.084 0.014
sns 0.0568 0.020 2.774 0.006 0.017 0.097
mass_media_qual 0.0718 0.025 2.898 0.004 0.023 0.120
poli_pers -0.0137 0.029 -0.481 0.631 -0.070 0.042
relig_pers 0.1643 0.027 6.096 0.000 0.111 0.217
econo 0.1428 0.024 5.879 0.000 0.095 0.191
fake 0.0462 0.022 2.109 0.035 0.003 0.089
humani 0.1845 0.033 5.579 0.000 0.120 0.249
nat_eth2 0.0220 0.032 0.696 0.487 -0.040 0.084
nat_cit2 -0.0643 0.037 -1.722 0.086 -0.138 0.009
SDO_D 0.0198 0.035 0.571 0.568 -0.048 0.088
SDO_E -0.0622 0.028 -2.193 0.029 -0.118 -0.007
rightw -0.0374 0.066 -0.570 0.569 -0.166 0.091
pol_orien -0.0373 0.026 -1.409 0.159 -0.089 0.015
islamo -0.0091 0.023 -0.393 0.695 -0.055 0.036
socd_p 0.0243 0.038 0.632 0.528 -0.051 0.100
socd_n 0.0523 0.032 1.647 0.100 -0.010 0.115
==============================================================================
Omnibus: 12.012 Durbin-Watson: 2.059
Prob(Omnibus): 0.002 Jarque-Bera (JB): 15.077
Skew: 0.200 Prob(JB): 0.000532
Kurtosis: 3.588 Cond. No. 904.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
n = 715일때, 종속변수 beh_n에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_n R-squared: 0.307
Model: OLS Adj. R-squared: 0.275
Method: Least Squares F-statistic: 9.751
Date: Sun, 04 Jun 2023 Prob (F-statistic): 4.06e-37
Time: 07:33:41 Log-Likelihood: -876.59
No. Observations: 715 AIC: 1817.
Df Residuals: 683 BIC: 1963.
Df Model: 31
Covariance Type: nonrobust
===================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------
const 0.6718 0.577 1.165 0.245 -0.461 1.805
gender -0.0553 0.069 -0.805 0.421 -0.190 0.080
age -0.0026 0.004 -0.715 0.475 -0.010 0.005
child 0.1659 0.094 1.755 0.080 -0.020 0.351
region_1 0.0467 0.067 0.696 0.487 -0.085 0.178
region_2 -0.0140 0.291 -0.048 0.962 -0.586 0.558
edulevel 0.0428 0.022 1.953 0.051 -0.000 0.086
income -0.0118 0.017 -0.679 0.497 -0.046 0.022
religion_1 0.0729 0.097 0.752 0.453 -0.118 0.263
religion_2 -0.1377 0.088 -1.565 0.118 -0.311 0.035
T_realf -0.0595 0.057 -1.047 0.296 -0.171 0.052
T_symbf -0.0112 0.046 -0.241 0.810 -0.102 0.080
T_safecoh2 0.2592 0.063 4.083 0.000 0.135 0.384
T_health 0.0146 0.042 0.347 0.729 -0.068 0.097
dir_contact 0.1683 0.047 3.550 0.000 0.075 0.261
massmedia 0.0101 0.039 0.257 0.798 -0.067 0.087
sns 0.0184 0.032 0.567 0.571 -0.045 0.082
mass_media_qual -0.0913 0.039 -2.325 0.020 -0.168 -0.014
poli_pers -0.0990 0.045 -2.191 0.029 -0.188 -0.010
relig_pers 0.0674 0.043 1.577 0.115 -0.017 0.151
econo 0.0599 0.039 1.554 0.121 -0.016 0.136
fake 0.1512 0.035 4.349 0.000 0.083 0.220
humani -0.1642 0.052 -3.130 0.002 -0.267 -0.061
nat_eth2 0.0509 0.050 1.016 0.310 -0.047 0.149
nat_cit2 -0.0540 0.059 -0.912 0.362 -0.170 0.062
SDO_D 0.1324 0.055 2.413 0.016 0.025 0.240
SDO_E -0.0211 0.045 -0.470 0.639 -0.110 0.067
rightw 0.1263 0.104 1.216 0.225 -0.078 0.330
pol_orien -0.0292 0.042 -0.696 0.486 -0.112 0.053
islamo 0.0326 0.037 0.887 0.375 -0.040 0.105
socd_p 0.0917 0.061 1.504 0.133 -0.028 0.211
socd_n 0.0244 0.050 0.485 0.628 -0.074 0.123
==============================================================================
Omnibus: 3.601 Durbin-Watson: 2.164
Prob(Omnibus): 0.165 Jarque-Bera (JB): 3.536
Skew: 0.136 Prob(JB): 0.171
Kurtosis: 2.789 Cond. No. 904.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
n = 591일때, 종속변수 beh_p에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_p R-squared: 0.659
Model: OLS Adj. R-squared: 0.640
Method: Least Squares F-statistic: 34.88
Date: Sun, 04 Jun 2023 Prob (F-statistic): 6.75e-110
Time: 07:33:41 Log-Likelihood: -444.90
No. Observations: 591 AIC: 953.8
Df Residuals: 559 BIC: 1094.
Df Model: 31
Covariance Type: nonrobust
===============================================================================
coef std err t P>|t| [0.025 0.975]
-------------------------------------------------------------------------------
const 1.9320 0.397 4.866 0.000 1.152 2.712
gender -0.0004 0.048 -0.009 0.993 -0.094 0.093
age 0.0029 0.003 1.128 0.260 -0.002 0.008
child 0.1257 0.067 1.880 0.061 -0.006 0.257
region_1 0.0854 0.047 1.833 0.067 -0.006 0.177
region_2 -0.1450 0.206 -0.703 0.482 -0.550 0.260
edulevel 0.0227 0.015 1.502 0.134 -0.007 0.052
income -0.0068 0.012 -0.566 0.571 -0.030 0.017
religion_1 -0.0187 0.067 -0.277 0.782 -0.151 0.114
religion_2 -0.0289 0.060 -0.478 0.633 -0.148 0.090
T_realf -0.2671 0.039 -6.839 0.000 -0.344 -0.190
T_symbf -0.0930 0.033 -2.853 0.004 -0.157 -0.029
T_safecoh2 -0.0352 0.044 -0.805 0.421 -0.121 0.051
T_health -0.0116 0.030 -0.391 0.696 -0.070 0.047
dir_contact 0.0883 0.030 2.898 0.004 0.028 0.148
massmedia -0.0471 0.028 -1.682 0.093 -0.102 0.008
sns 0.0514 0.025 2.063 0.040 0.002 0.100
sns_qual 0.0595 0.027 2.204 0.028 0.006 0.112
poli_pers -0.0208 0.031 -0.667 0.505 -0.082 0.040
relig_pers 0.1711 0.029 5.827 0.000 0.113 0.229
econo 0.1532 0.027 5.627 0.000 0.100 0.207
fake 0.0421 0.024 1.749 0.081 -0.005 0.089
humani 0.1834 0.036 5.143 0.000 0.113 0.253
nat_eth2 0.0085 0.036 0.237 0.813 -0.062 0.079
nat_cit2 -0.0591 0.041 -1.433 0.152 -0.140 0.022
SDO_D 0.0522 0.039 1.351 0.177 -0.024 0.128
SDO_E -0.0506 0.030 -1.663 0.097 -0.110 0.009
rightw -0.0355 0.074 -0.479 0.632 -0.181 0.110
pol_orien -0.0402 0.029 -1.374 0.170 -0.098 0.017
islamo -0.0076 0.025 -0.309 0.757 -0.056 0.041
socd_p 0.0197 0.042 0.463 0.643 -0.064 0.103
socd_n 0.0487 0.035 1.379 0.168 -0.021 0.118
==============================================================================
Omnibus: 10.618 Durbin-Watson: 2.026
Prob(Omnibus): 0.005 Jarque-Bera (JB): 13.991
Skew: 0.183 Prob(JB): 0.000916
Kurtosis: 3.658 Cond. No. 896.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
n = 591일때, 종속변수 beh_n에 대한 회귀분석 결과
OLS Regression Results
==============================================================================
Dep. Variable: beh_n R-squared: 0.315
Model: OLS Adj. R-squared: 0.277
Method: Least Squares F-statistic: 8.309
Date: Sun, 04 Jun 2023 Prob (F-statistic): 7.97e-30
Time: 07:33:41 Log-Likelihood: -728.22
No. Observations: 591 AIC: 1520.
Df Residuals: 559 BIC: 1661.
Df Model: 31
Covariance Type: nonrobust
===============================================================================
coef std err t P>|t| [0.025 0.975]
-------------------------------------------------------------------------------
const 0.4566 0.641 0.712 0.477 -0.803 1.716
gender -0.0784 0.077 -1.021 0.308 -0.229 0.072
age -0.0065 0.004 -1.538 0.125 -0.015 0.002
child 0.2586 0.108 2.394 0.017 0.046 0.471
region_1 0.1337 0.075 1.776 0.076 -0.014 0.282
region_2 0.3559 0.333 1.069 0.286 -0.298 1.010
edulevel 0.0326 0.024 1.334 0.183 -0.015 0.080
income 0.0040 0.019 0.210 0.834 -0.034 0.042
religion_1 0.0429 0.109 0.395 0.693 -0.171 0.256
religion_2 -0.1176 0.098 -1.204 0.229 -0.309 0.074
T_realf -0.0777 0.063 -1.231 0.219 -0.202 0.046
T_symbf -0.0262 0.053 -0.498 0.619 -0.130 0.077
T_safecoh2 0.2912 0.071 4.121 0.000 0.152 0.430
T_health 0.0528 0.048 1.099 0.272 -0.042 0.147
dir_contact 0.1698 0.049 3.451 0.001 0.073 0.266
massmedia 0.0187 0.045 0.412 0.680 -0.070 0.108
sns -0.0059 0.040 -0.146 0.884 -0.085 0.073
sns_qual -0.0443 0.044 -1.015 0.310 -0.130 0.041
poli_pers -0.0928 0.050 -1.844 0.066 -0.192 0.006
relig_pers 0.0746 0.047 1.573 0.116 -0.019 0.168
econo 0.0631 0.044 1.435 0.152 -0.023 0.149
fake 0.1592 0.039 4.092 0.000 0.083 0.236
humani -0.1364 0.058 -2.368 0.018 -0.250 -0.023
nat_eth2 0.1163 0.058 2.006 0.045 0.002 0.230
nat_cit2 -0.1263 0.067 -1.896 0.059 -0.257 0.005
SDO_D 0.1272 0.062 2.037 0.042 0.005 0.250
SDO_E -0.0257 0.049 -0.523 0.601 -0.122 0.071
rightw 0.1307 0.120 1.091 0.276 -0.105 0.366
pol_orien -0.0356 0.047 -0.754 0.451 -0.128 0.057
islamo 0.0415 0.040 1.040 0.299 -0.037 0.120
socd_p 0.0848 0.069 1.236 0.217 -0.050 0.219
socd_n 0.0074 0.057 0.129 0.897 -0.105 0.119
==============================================================================
Omnibus: 1.088 Durbin-Watson: 2.114
Prob(Omnibus): 0.580 Jarque-Bera (JB): 1.143
Skew: 0.050 Prob(JB): 0.565
Kurtosis: 2.810 Cond. No. 896.
==============================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
# 조절효과
y_list = ["beh_p", "beh_n"]
for y in y_list:
moderation_model(df = df_410, x_var="dir_contact", m_var="dir_cont_qual", y_var=y)
for y in y_list:
moderation_model(df = df_715, x_var="massmedia", m_var="mass_media_qual", y_var=y)
for y in y_list:
moderation_model(df = df_591, x_var="sns", m_var="sns_qual", y_var=y)
Process successfully initialized.
Based on the Process Macro by Andrew F. Hayes, Ph.D. (www.afhayes.com)
****************************** SPECIFICATION ****************************
Model = 1
Variables:
Cons = Cons
x = dir_contact
m = dir_cont_qual
y = beh_p
Statistical Controls:
gender, age, child, region_1, region_2, edulevel, income, religion_1, religion_2, T_realf, T_symbf, T_safecoh2, T_health, massmedia, sns, poli_pers, relig_pers, econo, fake, humani, nat_eth2, nat_cit2, SDO_D, SDO_E, rightw, pol_orien, islamo, socd_p, socd_n
Sample size:
410
******************** 종속변수 beh_p에 대한 dir_contact x dir_cont_qual의 조절효과 분석 결과********************
***************************** OUTCOME MODELS ****************************
Outcome = beh_p
OLS Regression Summary
R² Adj. R² MSE F df1 df2 p-value
0.7011 0.6748 0.2536 27.6276 32 377 0.0000
Coefficients
coeff se t p LLCI ULCI
Cons 1.9986 0.4815 4.1505 0.0000 1.0548 2.9424
dir_contact -0.0596 0.0804 -0.7415 0.4588 -0.2172 0.0980
dir_cont_qual 0.0826 0.0349 2.3671 0.0184 0.0142 0.1509
dir_contact*dir_cont_qual 0.0277 0.0233 1.1872 0.2359 -0.0180 0.0734
gender -0.0080 0.0556 -0.1436 0.8859 -0.1169 0.1009
age 0.0063 0.0031 2.0659 0.0395 0.0003 0.0123
child 0.0259 0.0783 0.3314 0.7405 -0.1274 0.1793
region_1 0.0941 0.0549 1.7162 0.0870 -0.0134 0.2017
region_2 -0.1890 0.2631 -0.7183 0.4730 -0.7047 0.3267
edulevel -0.0020 0.0183 -0.1108 0.9118 -0.0379 0.0338
income -0.0014 0.0139 -0.0980 0.9220 -0.0286 0.0259
religion_1 0.0056 0.0761 0.0736 0.9413 -0.1435 0.1547
religion_2 -0.0124 0.0707 -0.1754 0.8609 -0.1510 0.1262
T_realf -0.3205 0.0454 -7.0593 0.0000 -0.4095 -0.2315
T_symbf -0.1133 0.0384 -2.9546 0.0033 -0.1885 -0.0381
T_safecoh2 0.0057 0.0496 0.1159 0.9078 -0.0914 0.1029
T_health -0.0233 0.0347 -0.6708 0.5028 -0.0913 0.0447
massmedia -0.0938 0.0291 -3.2199 0.0014 -0.1509 -0.0367
sns 0.0933 0.0281 3.3225 0.0010 0.0383 0.1484
poli_pers 0.0108 0.0382 0.2829 0.7774 -0.0641 0.0858
relig_pers 0.2044 0.0356 5.7371 0.0000 0.1346 0.2742
econo 0.1161 0.0336 3.4568 0.0006 0.0503 0.1820
fake 0.0330 0.0270 1.2230 0.2221 -0.0199 0.0858
humani 0.1264 0.0415 3.0447 0.0025 0.0450 0.2078
nat_eth2 -0.0463 0.0434 -1.0668 0.2867 -0.1313 0.0388
nat_cit2 -0.0105 0.0474 -0.2206 0.8255 -0.1034 0.0825
SDO_D 0.0665 0.0452 1.4716 0.1420 -0.0221 0.1550
SDO_E -0.0682 0.0325 -2.0949 0.0368 -0.1320 -0.0044
rightw 0.0294 0.0955 0.3083 0.7580 -0.1577 0.2166
pol_orien -0.0173 0.0326 -0.5315 0.5954 -0.0813 0.0466
islamo 0.0077 0.0292 0.2642 0.7918 -0.0496 0.0650
socd_p -0.0099 0.0511 -0.1930 0.8470 -0.1100 0.0903
socd_n 0.0735 0.0413 1.7813 0.0757 -0.0074 0.1544
-------------------------------------------------------------------------
********************** CONDITIONAL EFFECTS **********************
Conditional effect(s) of dir_contact on beh_p at values of the moderator(s):
dir_cont_qual Effect SE t p LLCI ULCI
1.0000 -0.0319 0.0598 -0.5340 0.5936 -0.1492 0.0853
2.0000 -0.0043 0.0421 -0.1010 0.9196 -0.0869 0.0783
3.0000 0.0234 0.0326 0.7184 0.4730 -0.0405 0.0873
4.0000 0.0511 0.0379 1.3478 0.1785 -0.0232 0.1254
5.0000 0.0788 0.0539 1.4631 0.1443 -0.0268 0.1844
Process successfully initialized.
Based on the Process Macro by Andrew F. Hayes, Ph.D. (www.afhayes.com)
****************************** SPECIFICATION ****************************
Model = 1
Variables:
Cons = Cons
x = dir_contact
m = dir_cont_qual
y = beh_n
Statistical Controls:
gender, age, child, region_1, region_2, edulevel, income, religion_1, religion_2, T_realf, T_symbf, T_safecoh2, T_health, massmedia, sns, poli_pers, relig_pers, econo, fake, humani, nat_eth2, nat_cit2, SDO_D, SDO_E, rightw, pol_orien, islamo, socd_p, socd_n
Sample size:
410
******************** 종속변수 beh_n에 대한 dir_contact x dir_cont_qual의 조절효과 분석 결과********************
***************************** OUTCOME MODELS ****************************
Outcome = beh_n
OLS Regression Summary
R² Adj. R² MSE F df1 df2 p-value
0.3020 0.2407 0.7445 5.0970 32 377 0.0000
Coefficients
coeff se t p LLCI ULCI
Cons 1.5400 0.8251 1.8665 0.0628 -0.0771 3.1571
dir_contact -0.0121 0.1378 -0.0880 0.9299 -0.2821 0.2579
dir_cont_qual -0.0520 0.0598 -0.8707 0.3845 -0.1692 0.0651
dir_contact*dir_cont_qual 0.0452 0.0400 1.1314 0.2586 -0.0331 0.1235
gender -0.1165 0.0952 -1.2244 0.2216 -0.3031 0.0700
age -0.0053 0.0052 -1.0106 0.3129 -0.0155 0.0050
child 0.1818 0.1341 1.3555 0.1761 -0.0810 0.4446
region_1 0.1678 0.0940 1.7850 0.0751 -0.0164 0.3520
region_2 0.1825 0.4508 0.4047 0.6859 -0.7012 1.0661
edulevel 0.0314 0.0314 1.0008 0.3176 -0.0301 0.0928
income -0.0022 0.0238 -0.0929 0.9260 -0.0488 0.0444
religion_1 -0.0067 0.1303 -0.0513 0.9591 -0.2621 0.2488
religion_2 -0.2157 0.1212 -1.7799 0.0759 -0.4533 0.0218
T_realf -0.1316 0.0778 -1.6918 0.0915 -0.2841 0.0209
T_symbf 0.0111 0.0657 0.1692 0.8658 -0.1177 0.1399
T_safecoh2 0.2591 0.0850 3.0504 0.0024 0.0926 0.4256
T_health 0.0741 0.0595 1.2470 0.2132 -0.0424 0.1907
massmedia 0.0071 0.0499 0.1422 0.8870 -0.0907 0.1049
sns 0.0022 0.0481 0.0457 0.9636 -0.0922 0.0966
poli_pers -0.0665 0.0655 -1.0152 0.3107 -0.1950 0.0619
relig_pers 0.0981 0.0610 1.6073 0.1088 -0.0215 0.2178
econo 0.0246 0.0576 0.4282 0.6688 -0.0882 0.1375
fake 0.1826 0.0462 3.9530 0.0001 0.0921 0.2731
humani -0.1194 0.0711 -1.6781 0.0942 -0.2588 0.0201
nat_eth2 0.0028 0.0743 0.0380 0.9697 -0.1429 0.1485
nat_cit2 -0.0221 0.0812 -0.2719 0.7858 -0.1813 0.1371
SDO_D 0.1604 0.0774 2.0723 0.0389 0.0087 0.3121
SDO_E -0.0778 0.0558 -1.3953 0.1637 -0.1871 0.0315
rightw -0.1544 0.1636 -0.9440 0.3458 -0.4750 0.1662
pol_orien -0.0276 0.0559 -0.4942 0.6215 -0.1372 0.0819
islamo 0.0635 0.0501 1.2684 0.2054 -0.0346 0.1617
socd_p 0.0709 0.0875 0.8096 0.4187 -0.1007 0.2424
socd_n -0.0117 0.0707 -0.1655 0.8686 -0.1503 0.1269
-------------------------------------------------------------------------
********************** CONDITIONAL EFFECTS **********************
Conditional effect(s) of dir_contact on beh_n at values of the moderator(s):
dir_cont_qual Effect SE t p LLCI ULCI
1.0000 0.0331 0.1025 0.3228 0.7470 -0.1678 0.2339
2.0000 0.0783 0.0722 1.0841 0.2790 -0.0633 0.2198
3.0000 0.1235 0.0559 2.2101 0.0277 0.0140 0.2330
4.0000 0.1687 0.0650 2.5962 0.0098 0.0413 0.2961
5.0000 0.2139 0.0923 2.3181 0.0210 0.0330 0.3948
Process successfully initialized.
Based on the Process Macro by Andrew F. Hayes, Ph.D. (www.afhayes.com)
****************************** SPECIFICATION ****************************
Model = 1
Variables:
Cons = Cons
x = massmedia
m = mass_media_qual
y = beh_p
Statistical Controls:
gender, age, child, region_1, region_2, edulevel, income, religion_1, religion_2, T_realf, T_symbf, T_safecoh2, T_health, dir_contact, sns, poli_pers, relig_pers, econo, fake, humani, nat_eth2, nat_cit2, SDO_D, SDO_E, rightw, pol_orien, islamo, socd_p, socd_n
Sample size:
715
******************** 종속변수 beh_p에 대한 massmedia x mass_media_qual의 조절효과 분석 결과********************
***************************** OUTCOME MODELS ****************************
Outcome = beh_p
OLS Regression Summary
R² Adj. R² MSE F df1 df2 p-value
0.6428 0.6255 0.2811 38.3611 32 682 0.0000
Coefficients
coeff se t p LLCI ULCI
Cons 2.2558 0.3782 5.9640 0.0000 1.5144 2.9971
massmedia -0.1550 0.0561 -2.7632 0.0059 -0.2650 -0.0451
mass_media_qual -0.0288 0.0489 -0.5892 0.5559 -0.1247 0.0670
massmedia*mass_media_qual 0.0435 0.0182 2.3834 0.0174 0.0077 0.0792
gender -0.0221 0.0432 -0.5124 0.6085 -0.1068 0.0625
age 0.0026 0.0023 1.1437 0.2532 -0.0019 0.0072
child 0.0876 0.0594 1.4754 0.1406 -0.0288 0.2041
region_1 0.0849 0.0423 2.0079 0.0450 0.0020 0.1677
region_2 -0.1258 0.1832 -0.6867 0.4925 -0.4849 0.2333
edulevel 0.0200 0.0138 1.4481 0.1480 -0.0071 0.0470
income -0.0081 0.0109 -0.7434 0.4575 -0.0296 0.0133
religion_1 -0.0684 0.0610 -1.1223 0.2621 -0.1879 0.0511
religion_2 -0.0771 0.0553 -1.3935 0.1639 -0.1856 0.0313
T_realf -0.2452 0.0358 -6.8557 0.0000 -0.3153 -0.1751
T_symbf -0.0959 0.0292 -3.2874 0.0011 -0.1531 -0.0387
T_safecoh2 -0.0601 0.0399 -1.5051 0.1328 -0.1383 0.0182
T_health 0.0142 0.0264 0.5375 0.5911 -0.0376 0.0660
dir_contact 0.0915 0.0299 3.0649 0.0023 0.0330 0.1500
sns 0.0589 0.0204 2.8866 0.0040 0.0189 0.0990
poli_pers -0.0097 0.0285 -0.3402 0.7338 -0.0655 0.0461
relig_pers 0.1643 0.0269 6.1164 0.0000 0.1117 0.2170
econo 0.1445 0.0242 5.9637 0.0000 0.0970 0.1919
fake 0.0515 0.0220 2.3428 0.0194 0.0084 0.0945
humani 0.1785 0.0331 5.3991 0.0000 0.1137 0.2433
nat_eth2 0.0152 0.0316 0.4798 0.6315 -0.0468 0.0771
nat_cit2 -0.0598 0.0373 -1.6052 0.1089 -0.1328 0.0132
SDO_D 0.0184 0.0345 0.5335 0.5938 -0.0492 0.0860
SDO_E -0.0656 0.0283 -2.3156 0.0209 -0.1211 -0.0101
rightw -0.0386 0.0653 -0.5914 0.5545 -0.1666 0.0894
pol_orien -0.0335 0.0264 -1.2667 0.2057 -0.0853 0.0183
islamo -0.0122 0.0231 -0.5277 0.5979 -0.0576 0.0331
socd_p 0.0249 0.0383 0.6497 0.5161 -0.0502 0.1000
socd_n 0.0545 0.0317 1.7222 0.0855 -0.0075 0.1166
-------------------------------------------------------------------------
********************** CONDITIONAL EFFECTS **********************
Conditional effect(s) of massmedia on beh_p at values of the moderator(s):
mass_media_qual Effect SE t p LLCI ULCI
1.0000 -0.1115 0.0405 -2.7526 0.0061 -0.1910 -0.0321
2.0000 -0.0681 0.0283 -2.4030 0.0165 -0.1236 -0.0126
3.0000 -0.0246 0.0251 -0.9818 0.3266 -0.0737 0.0245
4.0000 0.0189 0.0335 0.5638 0.5731 -0.0467 0.0844
5.0000 0.0623 0.0477 1.3067 0.1918 -0.0312 0.1558
Process successfully initialized.
Based on the Process Macro by Andrew F. Hayes, Ph.D. (www.afhayes.com)
****************************** SPECIFICATION ****************************
Model = 1
Variables:
Cons = Cons
x = massmedia
m = mass_media_qual
y = beh_n
Statistical Controls:
gender, age, child, region_1, region_2, edulevel, income, religion_1, religion_2, T_realf, T_symbf, T_safecoh2, T_health, dir_contact, sns, poli_pers, relig_pers, econo, fake, humani, nat_eth2, nat_cit2, SDO_D, SDO_E, rightw, pol_orien, islamo, socd_p, socd_n
Sample size:
715
******************** 종속변수 beh_n에 대한 massmedia x mass_media_qual의 조절효과 분석 결과********************
***************************** OUTCOME MODELS ****************************
Outcome = beh_n
OLS Regression Summary
R² Adj. R² MSE F df1 df2 p-value
0.3074 0.2738 0.7121 9.4590 32 682 0.0000
Coefficients
coeff se t p LLCI ULCI
Cons 0.5392 0.6020 0.8956 0.3708 -0.6407 1.7191
massmedia 0.0721 0.0893 0.8075 0.4197 -0.1029 0.2471
mass_media_qual -0.0394 0.0778 -0.5055 0.6134 -0.1919 0.1132
massmedia*mass_media_qual -0.0225 0.0290 -0.7735 0.4395 -0.0793 0.0344
gender -0.0549 0.0687 -0.7990 0.4246 -0.1896 0.0798
age -0.0027 0.0037 -0.7274 0.4672 -0.0099 0.0045
child 0.1677 0.0946 1.7736 0.0766 -0.0176 0.3530
region_1 0.0502 0.0673 0.7467 0.4555 -0.0816 0.1820
region_2 -0.0072 0.2916 -0.0248 0.9802 -0.5788 0.5643
edulevel 0.0430 0.0219 1.9620 0.0502 0.0000 0.0860
income -0.0115 0.0174 -0.6589 0.5102 -0.0456 0.0226
religion_1 0.0737 0.0970 0.7597 0.4477 -0.1165 0.2639
religion_2 -0.1354 0.0881 -1.5374 0.1247 -0.3081 0.0372
T_realf -0.0618 0.0569 -1.0862 0.2778 -0.1734 0.0497
T_symbf -0.0128 0.0464 -0.2754 0.7831 -0.1038 0.0782
T_safecoh2 0.2602 0.0635 4.0968 0.0000 0.1357 0.3847
T_health 0.0128 0.0421 0.3044 0.7609 -0.0696 0.0952
dir_contact 0.1706 0.0475 3.5893 0.0004 0.0774 0.2637
sns 0.0173 0.0325 0.5316 0.5952 -0.0464 0.0810
poli_pers -0.1011 0.0453 -2.2324 0.0259 -0.1899 -0.0123
relig_pers 0.0674 0.0428 1.5766 0.1154 -0.0164 0.1512
econo 0.0590 0.0386 1.5315 0.1261 -0.0165 0.1346
fake 0.1485 0.0350 4.2491 0.0000 0.0800 0.2171
humani -0.1610 0.0526 -3.0601 0.0023 -0.2642 -0.0579
nat_eth2 0.0544 0.0503 1.0813 0.2800 -0.0442 0.1530
nat_cit2 -0.0563 0.0593 -0.9499 0.3425 -0.1725 0.0599
SDO_D 0.1331 0.0549 2.4251 0.0156 0.0255 0.2406
SDO_E -0.0194 0.0451 -0.4309 0.6667 -0.1078 0.0689
rightw 0.1269 0.1039 1.2215 0.2223 -0.0767 0.3306
pol_orien -0.0312 0.0421 -0.7420 0.4584 -0.1137 0.0513
islamo 0.0342 0.0368 0.9288 0.3533 -0.0380 0.1064
socd_p 0.0913 0.0610 1.4981 0.1346 -0.0282 0.2109
socd_n 0.0233 0.0504 0.4616 0.6445 -0.0755 0.1221
-------------------------------------------------------------------------
********************** CONDITIONAL EFFECTS **********************
Conditional effect(s) of massmedia on beh_n at values of the moderator(s):
mass_media_qual Effect SE t p LLCI ULCI
1.0000 0.0496 0.0645 0.7697 0.4418 -0.0768 0.1761
2.0000 0.0272 0.0451 0.6030 0.5467 -0.0612 0.1156
3.0000 0.0047 0.0399 0.1187 0.9055 -0.0735 0.0829
4.0000 -0.0177 0.0532 -0.3328 0.7394 -0.1221 0.0866
5.0000 -0.0402 0.0759 -0.5291 0.5969 -0.1890 0.1086
Process successfully initialized.
Based on the Process Macro by Andrew F. Hayes, Ph.D. (www.afhayes.com)
****************************** SPECIFICATION ****************************
Model = 1
Variables:
Cons = Cons
x = sns
m = sns_qual
y = beh_p
Statistical Controls:
gender, age, child, region_1, region_2, edulevel, income, religion_1, religion_2, T_realf, T_symbf, T_safecoh2, T_health, dir_contact, massmedia, poli_pers, relig_pers, econo, fake, humani, nat_eth2, nat_cit2, SDO_D, SDO_E, rightw, pol_orien, islamo, socd_p, socd_n
Sample size:
591
******************** 종속변수 beh_p에 대한 sns x sns_qual의 조절효과 분석 결과********************
***************************** OUTCOME MODELS ****************************
Outcome = beh_p
OLS Regression Summary
R² Adj. R² MSE F df1 df2 p-value
0.6592 0.6391 0.2795 33.7347 32 558 0.0000
Coefficients
coeff se t p LLCI ULCI
Cons 1.9209 0.4096 4.6900 0.0000 1.1181 2.7236
sns 0.0568 0.0548 1.0366 0.3004 -0.0506 0.1643
sns_qual 0.0628 0.0403 1.5581 0.1198 -0.0162 0.1418
sns*sns_qual -0.0019 0.0171 -0.1117 0.9111 -0.0355 0.0316
gender -0.0006 0.0476 -0.0124 0.9901 -0.0939 0.0928
age 0.0030 0.0026 1.1303 0.2588 -0.0022 0.0081
child 0.1260 0.0670 1.8818 0.0604 -0.0052 0.2573
region_1 0.0855 0.0467 1.8332 0.0673 -0.0059 0.1770
region_2 -0.1451 0.2063 -0.7032 0.4822 -0.5495 0.2593
edulevel 0.0228 0.0151 1.5049 0.1329 -0.0069 0.0525
income -0.0068 0.0119 -0.5663 0.5714 -0.0302 0.0166
religion_1 -0.0187 0.0674 -0.2775 0.7815 -0.1507 0.1133
religion_2 -0.0286 0.0606 -0.4719 0.6372 -0.1474 0.0902
T_realf -0.2671 0.0391 -6.8324 0.0000 -0.3437 -0.1905
T_symbf -0.0930 0.0326 -2.8525 0.0045 -0.1570 -0.0291
T_safecoh2 -0.0349 0.0439 -0.7964 0.4261 -0.1209 0.0510
T_health -0.0117 0.0298 -0.3934 0.6941 -0.0702 0.0467
dir_contact 0.0886 0.0306 2.8943 0.0039 0.0286 0.1485
massmedia -0.0475 0.0283 -1.6817 0.0932 -0.1029 0.0079
poli_pers -0.0210 0.0313 -0.6719 0.5019 -0.0823 0.0403
relig_pers 0.1713 0.0295 5.8171 0.0000 0.1136 0.2290
econo 0.1532 0.0273 5.6191 0.0000 0.0997 0.2066
fake 0.0420 0.0242 1.7381 0.0827 -0.0054 0.0893
humani 0.1837 0.0358 5.1302 0.0000 0.1135 0.2539
nat_eth2 0.0087 0.0359 0.2407 0.8099 -0.0618 0.0791
nat_cit2 -0.0592 0.0413 -1.4334 0.1523 -0.1401 0.0217
SDO_D 0.0523 0.0387 1.3516 0.1770 -0.0236 0.1282
SDO_E -0.0503 0.0305 -1.6468 0.1002 -0.1101 0.0096
rightw -0.0352 0.0743 -0.4743 0.6355 -0.1808 0.1104
pol_orien -0.0405 0.0294 -1.3768 0.1691 -0.0982 0.0172
islamo -0.0077 0.0247 -0.3128 0.7545 -0.0562 0.0407
socd_p 0.0195 0.0425 0.4592 0.6462 -0.0638 0.1029
socd_n 0.0487 0.0353 1.3787 0.1686 -0.0205 0.1179
-------------------------------------------------------------------------
********************** CONDITIONAL EFFECTS **********************
Conditional effect(s) of sns on beh_p at values of the moderator(s):
sns_qual Effect SE t p LLCI ULCI
1.0000 0.0549 0.0403 1.3616 0.1739 -0.0241 0.1340
2.0000 0.0530 0.0289 1.8354 0.0670 -0.0036 0.1096
3.0000 0.0511 0.0251 2.0398 0.0418 0.0020 0.1002
4.0000 0.0492 0.0317 1.5498 0.1218 -0.0130 0.1114
5.0000 0.0473 0.0444 1.0642 0.2877 -0.0398 0.1344
Process successfully initialized.
Based on the Process Macro by Andrew F. Hayes, Ph.D. (www.afhayes.com)
****************************** SPECIFICATION ****************************
Model = 1
Variables:
Cons = Cons
x = sns
m = sns_qual
y = beh_n
Statistical Controls:
gender, age, child, region_1, region_2, edulevel, income, religion_1, religion_2, T_realf, T_symbf, T_safecoh2, T_health, dir_contact, massmedia, poli_pers, relig_pers, econo, fake, humani, nat_eth2, nat_cit2, SDO_D, SDO_E, rightw, pol_orien, islamo, socd_p, socd_n
Sample size:
591
******************** 종속변수 beh_n에 대한 sns x sns_qual의 조절효과 분석 결과********************
***************************** OUTCOME MODELS ****************************
Outcome = beh_n
OLS Regression Summary
R² Adj. R² MSE F df1 df2 p-value
0.3180 0.2776 0.7263 8.1317 32 558 0.0000
Coefficients
coeff se t p LLCI ULCI
Cons 0.2240 0.6603 0.3392 0.7346 -1.0701 1.5180
sns 0.1085 0.0884 1.2275 0.2201 -0.0647 0.2818
sns_qual 0.0259 0.0650 0.3978 0.6909 -0.1015 0.1532
sns*sns_qual -0.0401 0.0276 -1.4529 0.1468 -0.0942 0.0140
gender -0.0821 0.0768 -1.0698 0.2852 -0.2326 0.0683
age -0.0063 0.0042 -1.4907 0.1366 -0.0145 0.0020
child 0.2647 0.1080 2.4519 0.0145 0.0531 0.4763
region_1 0.1358 0.0752 1.8059 0.0715 -0.0116 0.2833
region_2 0.3537 0.3326 1.0632 0.2882 -0.2983 1.0056
edulevel 0.0345 0.0244 1.4135 0.1581 -0.0133 0.0824
income 0.0039 0.0193 0.2036 0.8387 -0.0338 0.0417
religion_1 0.0423 0.1086 0.3899 0.6967 -0.1705 0.2552
religion_2 -0.1105 0.0977 -1.1316 0.2583 -0.3020 0.0809
T_realf -0.0772 0.0630 -1.2254 0.2210 -0.2007 0.0463
T_symbf -0.0281 0.0526 -0.5345 0.5932 -0.1312 0.0750
T_safecoh2 0.2971 0.0707 4.2021 0.0000 0.1585 0.4356
T_health 0.0508 0.0481 1.0563 0.2913 -0.0434 0.1450
dir_contact 0.1760 0.0493 3.5676 0.0004 0.0793 0.2727
massmedia 0.0109 0.0455 0.2401 0.8103 -0.0783 0.1002
poli_pers -0.0973 0.0504 -1.9316 0.0539 -0.1961 0.0014
relig_pers 0.0790 0.0475 1.6650 0.0965 -0.0140 0.1721
econo 0.0618 0.0439 1.4071 0.1600 -0.0243 0.1480
fake 0.1560 0.0389 4.0057 0.0001 0.0797 0.2323
humani -0.1295 0.0577 -2.2431 0.0253 -0.2427 -0.0163
nat_eth2 0.1192 0.0579 2.0564 0.0402 0.0056 0.2327
nat_cit2 -0.1279 0.0666 -1.9217 0.0552 -0.2584 0.0025
SDO_D 0.1294 0.0624 2.0732 0.0386 0.0071 0.2517
SDO_E -0.0199 0.0492 -0.4035 0.6867 -0.1163 0.0766
rightw 0.1369 0.1197 1.1430 0.2535 -0.0978 0.3716
pol_orien -0.0434 0.0474 -0.9144 0.3609 -0.1364 0.0496
islamo 0.0392 0.0399 0.9834 0.3258 -0.0389 0.1173
socd_p 0.0816 0.0685 1.1900 0.2346 -0.0528 0.2159
socd_n 0.0081 0.0569 0.1430 0.8864 -0.1035 0.1197
-------------------------------------------------------------------------
********************** CONDITIONAL EFFECTS **********************
Conditional effect(s) of sns on beh_n at values of the moderator(s):
sns_qual Effect SE t p LLCI ULCI
1.0000 0.0684 0.0650 1.0519 0.2933 -0.0591 0.1959
2.0000 0.0283 0.0466 0.6077 0.5436 -0.0630 0.1196
3.0000 -0.0118 0.0404 -0.2924 0.7701 -0.0910 0.0674
4.0000 -0.0519 0.0512 -1.0146 0.3107 -0.1522 0.0484
5.0000 -0.0920 0.0716 -1.2849 0.1994 -0.2324 0.0483
# Scatter plot 그리기
scatter = plt.scatter(massmedi, beh_p, color='black') # massmedi를 x축으로, beh_p를 y축으로 설정하고, 점 색상을 검은색으로 표시
# 선 그래프 그리기
lines_dict = {}
for i in range(len(massmedi) - 1):
if mass_qua[i] == 2 and mass_qua[i + 1] == 2:
line, = plt.plot([massmedi[i], massmedi[i + 1]], [beh_p[i], beh_p[i + 1]], linewidth=2, color='black')
lines_dict.setdefault('대중매체 접촉의 질: 2', line)
elif mass_qua[i] == 3 and mass_qua[i + 1] == 3:
line, = plt.plot([massmedi[i], massmedi[i + 1]], [beh_p[i], beh_p[i + 1]], linewidth=2, linestyle='--', color='black')
lines_dict.setdefault('대중매체 접촉의 질: 3', line)
elif mass_qua[i] == 4 and mass_qua[i + 1] == 4:
line, = plt.plot([massmedi[i], massmedi[i + 1]], [beh_p[i], beh_p[i + 1]], linewidth=2, linestyle='-.', color='black')
lines_dict.setdefault('대중매체 접촉의 질: 4', line)
# 축과 제목 설정
plt.xlabel('대중매체 접촉 빈도')
plt.ylabel('긍정적 행동의사')
# plt.title('대중매체 접촉 빈도 X 대중매체 접촉의 질')
# 범례 설정
lines = list(lines_dict.values())
labels = list(lines_dict.keys())
plt.legend(list(reversed(lines)), list(reversed(labels)))
# 그래프 출력
plt.show()