Methoden der empirischen Kommunikations- und Medienforschung
Freie Universität Berlin
| Parameter | Coefficient | 95% CI | t(990) | p | Std. Coef. | Std. Coef. 95% CI | Fit |
|---|---|---|---|---|---|---|---|
| (Intercept) | 0.99 | (0.67, 1.31) | 6.07 | < .001 | 0.00 | (-0.06, 0.06) | |
| Age | 0.02 | (0.02, 0.03) | 8.24 | < .001 | 0.24 | (0.19, 0.30) | |
| Newspapers | 0.22 | (0.18, 0.27) | 9.31 | < .001 | 0.28 | (0.22, 0.33) | |
| R2 (adj.) | 0.16 |
Text auf der nächsten Folie
| Parameter | Coefficient | 95% CI | t(990) | p | Std. Coef. | Std. Coef. 95% CI | Fit |
|---|---|---|---|---|---|---|---|
| (Intercept) | 0.99 | (0.67, 1.31) | 6.07 | < .001 | 0.00 | (-0.06, 0.06) | |
| Age | 0.02 | (0.02, 0.03) | 8.24 | < .001 | 0.24 | (0.19, 0.30) | |
| Newspapers | 0.22 | (0.18, 0.27) | 9.31 | < .001 | 0.28 | (0.22, 0.33) | |
| R2 (adj.) | 0.16 |
| Parameter | Coefficient | 95% CI | t(990) | p | Std. Coef. | Std. Coef. 95% CI | Fit |
|---|---|---|---|---|---|---|---|
| (Intercept) | 0.99 | (0.67, 1.31) | 6.07 | < .001 | 0.00 | (-0.06, 0.06) | |
| Age | 0.02 | (0.02, 0.03) | 8.24 | < .001 | 0.24 | (0.19, 0.30) | |
| Newspapers | 0.22 | (0.18, 0.27) | 9.31 | < .001 | 0.28 | (0.22, 0.33) | |
| R2 (adj.) | 0.16 |
Beispiel für vier Befragte:
| Age | y | yhat | e | e2 | e_M | e_M2 |
|---|---|---|---|---|---|---|
| 23 | 1 | 2.2 | -1.2 | 1.3 | -2 | 4.2 |
| 37 | 4 | 2.6 | 1.4 | 2.0 | 1 | 0.9 |
| 55 | 2 | 3.1 | -1.1 | 1.2 | -1 | 1.1 |
| 62 | 5 | 3.3 | 1.7 | 2.9 | 2 | 3.8 |

\(R^2 = \frac{\sum(y_i - \bar{y})^2 - \sum(y_i - \hat y)^2}{\sum(y_i - \bar{y})^2}\)
| Parameter | Coefficient | 95% CI | t(990) | p | Std. Coef. | Std. Coef. 95% CI | Fit |
|---|---|---|---|---|---|---|---|
| (Intercept) | 0.99 | (0.67, 1.31) | 6.07 | < .001 | 0.00 | (-0.06, 0.06) | |
| Age | 0.02 | (0.02, 0.03) | 8.24 | < .001 | 0.24 | (0.19, 0.30) | |
| Newspapers | 0.22 | (0.18, 0.27) | 9.31 | < .001 | 0.28 | (0.22, 0.33) | |
| R2 (adj.) | 0.16 |
Vergleiche Abschnitt Regression und Mittelwertvergleich in Einheit zur bivariaten linearen Regresssion
| Parameter | Coefficient | 95% CI | t(991) | p | Fit |
|---|---|---|---|---|---|
| (Intercept) | 3.44 | (3.33, 3.55) | 60.48 | < .001 | |
| Gender (female) | -0.84 | (-1.00, -0.67) | -10.14 | < .001 | |
| R2 | 0.09 |
Dummy-Variablen
| Zugehörigkeit | Middle | High |
|---|---|---|
| Lower | 0 | 0 |
| Middle | 1 | 0 |
| High | 0 | 1 |
Mittelwerte
| Lower | Middle | High |
|---|---|---|
| 2.58 | 2.97 | 3.25 |
Regression
| Parameter | Coefficient | 95% CI | t(990) | p | Fit |
|---|---|---|---|---|---|
| (Intercept) | 2.58 | (2.35, 2.81) | 22.39 | < .001 | |
| Education (Middle) | 0.39 | (0.13, 0.65) | 2.92 | 0.004 | |
| Education (High) | 0.67 | (0.41, 0.93) | 5.09 | < .001 | |
| R2 (adj.) | 0.03 |
Dummy-Variablen
| Zugehörigkeit | Lower | High |
|---|---|---|
| Middle | 0 | 0 |
| Lower | 1 | 0 |
| High | 0 | 1 |
Mittelwerte
| Lower | Middle | High |
|---|---|---|
| 2.58 | 2.97 | 3.25 |
Regression
| Parameter | Coefficient | 95% CI | t(990) | p | Fit |
|---|---|---|---|---|---|
| (Intercept) | 2.97 | (2.84, 3.10) | 44.41 | < .001 | |
| Education (Lower) | -0.39 | (-0.65, -0.13) | -2.92 | 0.004 | |
| Education (High) | 0.28 | (0.10, 0.46) | 3.03 | 0.002 | |
| R2 (adj.) | 0.03 |
| Contrast | Estimate | Std. Error | z | Pr(>|z|) | S | 2.5 % | 97.5 % |
|---|---|---|---|---|---|---|---|
| Type: response | |||||||
| High - Lower | 0.669 | 0.131 | 5.09 | < 0.001 | 21.4 | 0.4109 | 0.926 |
| High - Middle | 0.279 | 0.092 | 3.03 | 0.00241 | 8.7 | 0.0988 | 0.459 |
| Middle - Lower | 0.389 | 0.133 | 2.92 | 0.00348 | 8.2 | 0.1282 | 0.651 |
| Estimate | Std. Error | z | Pr(>|z|) | S | 2.5 % | 97.5 % |
|---|---|---|---|---|---|---|
| 0.669 | 0.131 | 5.09 | < 0.001 | 19.8 | 0.3618 | 0.975 |
| 0.279 | 0.092 | 3.03 | 0.00724 | 7.1 | 0.0644 | 0.494 |
| 0.389 | 0.133 | 2.92 | 0.01045 | 6.6 | 0.0784 | 0.700 |
m1 <- lm(Political_knowledge ~ Radio + Television + Newspapers + Online_news_sites + Twitter +
Facebook + Gender + Age + Education + Political_interest, data = d)
m4 <- lm(Political_knowledge ~ Radio + Television + Newspapers + Online_news_sites + Twitter +
Facebook + Gender + Age + Education + Political_interest + Information_overload, data = d)| Model 1 | Model 4 | |
|---|---|---|
| (Intercept) | 0.46 (0.22)* | 0.67 (0.23)** |
| Radio | -0.01 (0.02) | -0.01 (0.02) |
| Television | 0.08 (0.03)** | 0.09 (0.03)** |
| Newspapers | 0.08 (0.02)*** | 0.08 (0.02)*** |
| Online_news_sites | 0.06 (0.02)** | 0.06 (0.02)** |
| -0.06 (0.04) | -0.05 (0.04) | |
| -0.07 (0.02)*** | -0.07 (0.02)*** | |
| Genderfemale | -0.48 (0.07)*** | -0.46 (0.07)*** |
| Age | 0.02 (0.00)*** | 0.02 (0.00)*** |
| EducationMiddle | 0.28 (0.11)* | 0.27 (0.11)* |
| EducationHigh | 0.48 (0.11)*** | 0.48 (0.11)*** |
| Political_interest | 0.18 (0.01)*** | 0.17 (0.01)*** |
| Information_overload | -0.03 (0.01)** | |
| Num.Obs. | 993 | 993 |
| R2 Adj. | 0.371 | 0.376 |
| Parameter | Coefficient | 95% CI | t(980) | p | Std. Coef. | Std. Coef. 95% CI | Fit |
|---|---|---|---|---|---|---|---|
| (Intercept) | 0.67 | (0.21, 1.12) | 2.90 | 0.004 | -0.08 | (-0.22, 0.06) | |
| Radio | -0.01 | (-0.05, 0.04) | -0.34 | 0.736 | -0.01 | (-0.06, 0.05) | |
| Television | 0.09 | (0.03, 0.15) | 2.80 | 0.005 | 0.08 | (0.03, 0.14) | |
| Newspapers | 0.08 | (0.04, 0.13) | 3.46 | < .001 | 0.10 | (0.04, 0.16) | |
| Online news sites | 0.06 | (0.02, 0.11) | 2.73 | 0.006 | 0.08 | (0.02, 0.14) | |
| -0.05 | (-0.13, 0.02) | -1.37 | 0.171 | -0.04 | (-0.09, 0.02) | ||
| -0.07 | (-0.11, -0.03) | -3.38 | < .001 | -0.10 | (-0.16, -0.04) | ||
| Gender (female) | -0.46 | (-0.60, -0.32) | -6.34 | < .001 | -0.34 | (-0.44, -0.23) | |
| Age | 0.02 | (0.01, 0.02) | 6.03 | < .001 | 0.17 | (0.12, 0.23) | |
| Education (Middle) | 0.27 | (0.06, 0.48) | 2.48 | 0.013 | 0.20 | (0.04, 0.35) | |
| Education (High) | 0.48 | (0.26, 0.69) | 4.27 | < .001 | 0.35 | (0.19, 0.51) | |
| Political interest | 0.17 | (0.14, 0.20) | 11.79 | < .001 | 0.34 | (0.28, 0.39) | |
| Information overload | -0.03 | (-0.05, -0.01) | -3.03 | 0.003 | -0.08 | (-0.13, -0.03) | |
| R2 (adj.) | 0.38 |

| Estimate | Std. Error | z | Pr(>|z|) | S | 2.5 % | 97.5 % | |
|---|---|---|---|---|---|---|---|
| Type: response | |||||||
| 1 | 3.53 | 0.0687 | 51.4 | <0.001 | Inf | 3.40 | 3.66 |
| 2 | 3.46 | 0.0640 | 54.1 | <0.001 | Inf | 3.34 | 3.59 |
| 3 | 3.39 | 0.0657 | 51.6 | <0.001 | Inf | 3.26 | 3.52 |
| 4 | 3.32 | 0.0734 | 45.2 | <0.001 | Inf | 3.18 | 3.47 |
| 5 | 3.25 | 0.0855 | 38.1 | <0.001 | 1050.3 | 3.08 | 3.42 |
| 6 | 3.18 | 0.1003 | 31.7 | <0.001 | 731.9 | 2.99 | 3.38 |
| Term | Contrast | Estimate | Std. Error | z | Pr(>|z|) | S | 2.5 % | 97.5 % |
|---|---|---|---|---|---|---|---|---|
| Type: response | ||||||||
| Education | High - Lower | 0.476 | 0.1114 | 4.27 | < 0.001 | 15.7 | 0.2575 | 0.694 |
| Education | High - Middle | 0.206 | 0.0757 | 2.72 | 0.00654 | 7.3 | 0.0575 | 0.354 |
| Education | Middle - Lower | 0.270 | 0.1086 | 2.48 | 0.01297 | 6.3 | 0.0570 | 0.483 |
| (x + sd) - (x - sd) | -0.270 | 0.0799 | -3.38 | < 0.001 | 10.5 | -0.4271 | -0.114 | |
| Gender | female - male | -0.461 | 0.0727 | -6.34 | < 0.001 | 32.0 | -0.6030 | -0.318 |
| Newspapers | (x + sd) - (x - sd) | 0.276 | 0.0796 | 3.46 | < 0.001 | 10.9 | 0.1198 | 0.432 |
Siehe Einheit zur bivariaten linearen Regression für Details
$NCV
| Parameter | Political_knowledge | Radio | Television | Newspapers | Online_news_sites | Age | Political_interest | Information_overload | ||
|---|---|---|---|---|---|---|---|---|---|---|
| Political_knowledge | 0.14*** | 0.26*** | 0.33*** | 0.22*** | -0.04 | -0.13*** | 0.30*** | 0.49*** | -0.09* | |
| Radio | 0.14*** | 0.39*** | 0.28*** | 0.22*** | 0.09 | 0.18*** | 0.05 | 0.20*** | 0.05 | |
| Television | 0.26*** | 0.39*** | 0.31*** | 0.28*** | 0.04 | 0.19*** | 0.24*** | 0.30*** | 0.08 | |
| Newspapers | 0.33*** | 0.28*** | 0.31*** | 0.33*** | 0.09 | 0.06 | 0.21*** | 0.33*** | 0.02 | |
| Online_news_sites | 0.22*** | 0.22*** | 0.28*** | 0.33*** | 0.23*** | 0.27*** | -0.08 | 0.32*** | 0.04 | |
| -0.04 | 0.09 | 0.04 | 0.09 | 0.23*** | 0.33*** | -0.15*** | 0.05 | 0.09 | ||
| -0.13*** | 0.18*** | 0.19*** | 0.06 | 0.27*** | 0.33*** | -0.25*** | 0.06 | 0.12** | ||
| Age | 0.30*** | 0.05 | 0.24*** | 0.21*** | -0.08 | -0.15*** | -0.25*** | 0.14*** | 0.05 | |
| Political_interest | 0.49*** | 0.20*** | 0.30*** | 0.33*** | 0.32*** | 0.05 | 0.06 | 0.14*** | -0.02 | |
| Information_overload | -0.09* | 0.05 | 0.08 | 0.02 | 0.04 | 0.09 | 0.12** | 0.05 | -0.02 |
p-value adjustment method: Holm (1979)
$VIF
Digitale Verhaltensdaten und Webtracking
Multiple lineare Regression III: Moderation
Marko Bachl