You are here

Group Exercise: Model Specification

Each group will build a causal model to explain one DV. Compute the linear model and interpret the results, as appropriate. Paste your R code beneath your interpretation.

A. political view
B. feminism
C. religiosity
D. party affiliation

Comments

DV: RELIGIOSITY (V161241) - Is Religion important part of R life?
IV: Social Class (V162132) - R social class
IV: Working now: had reduction in work hrs or pay cut (V161299)
IV: Highest Level of Education (V161270)

Social Class
Reflection: This variable is appropriate because social class and class identity encompasses variances in cultures, including differences in day-to-day life and work occupations. These differences may influence how important religion is in ones life depending on the emphasis that religion provides as well as the opportunity to pursue religion in one’s “free time”. Differences in culture allows for differences in schedules that allows for more “free time” from work, and other demands.

Reduction in Hours
Reflection: This variable is appropriate because a reduction in work hours may be meaningful of one’s lowered economic status. Therefore, one may turn to religion to explain why this social status was lowered.

Highest Level of Education
Reflection: This variable is appropriate because higher education attainment may bring about more education knowledge that contradicts with religious thought and beliefs.

> summary(lm(V161241~socialclass+V161270+V161299))

Call:
lm(formula = V161241 ~ socialclass + V161270 + V161299)

Residuals:
Min 1Q Median 3Q Max
-0.6593 -0.3961 -0.3453 0.5946 0.7572

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.103059 0.106482 10.359 <2e-16 ***
socialclass 0.044559 0.025310 1.761 0.0787 .
V161270 0.003097 0.003093 1.001 0.3170
V161299 0.064195 0.049226 1.304 0.1925
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.4824 on 883 degrees of freedom
(3384 observations deleted due to missingness)
Multiple R-squared: 0.008608, Adjusted R-squared: 0.00524
F-statistic: 2.556 on 3 and 883 DF, p-value: 0.05411

We have no evidence to support the null hypothesis.

no significance.

> Gaymar<-0;Gaymar[V161231==1]<-1;Gaymar[V161231==3]<-3
> abort<-0;abort[V161232==1 || V161232==2]<-1;abort[V161232==4]<-4
> regbus<-ifelse(V162186<=3,1,0)
> tradfam<-ifelse(V162210>=3,1,0)
> summary(lm(V161128~abort+Gaymar+V162106+V162313+regbus+tradfam))

Call:
lm(formula = V161128 ~ abort + Gaymar + V162106 + V162313 + regbus +
tradfam)

Residuals:
Min 1Q Median 3Q Max
-2.3606 -1.0230 -0.1538 0.9415 4.4096

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.3322654 1.8453597 -0.180 0.85714
abort 0.6512318 0.4586526 1.420 0.15587
Gaymar 0.1150116 0.0710630 1.618 0.10580
V162106 -0.0004966 0.0007665 -0.648 0.51720
V162313 0.0058125 0.0014651 3.967 7.65e-05 ***
regbus 0.3434475 0.1042864 3.293 0.00102 **
tradfam 0.0277319 0.0776494 0.357 0.72104
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.368 on 1375 degrees of freedom
(2889 observations deleted due to missingness)
Multiple R-squared: 0.02432, Adjusted R-squared: 0.02006
F-statistic: 5.712 on 6 and 1375 DF, p-value: 7.03e-06

We reject the null hypothesis and we can see that illegal immigration favorability and opinion on regulation of businesses affect political view.

> ANES2016<-read.csv("http://www.shortell.nyc/online/files/anes_timeseries_2016.csv")
> attach(ANES2016)
>
> Gender[V161342==1]<-1;
Error in Gender[V161342 == 1] <- 1 : object 'Gender' not found
> Gender[V161342==1]<-1;Gender[V161342==2]<-2;
Error in Gender[V161342 == 1] <- 1 : object 'Gender' not found
> Gender<-0;Gender[V161342==1]<-1;Gender[V161342==2]<-2;
> SocClass<-ifelse(V161307<=2,1,0)
> summary(lm(V161128~V161241+Gender+V161310a+SocClass+V161277+V161271))

Call:
lm(formula = V161128 ~ V161241 + Gender + V161310a + SocClass +
V161277 + V161271)

Residuals:
Min 1Q Median 3Q Max
-3.2952 -1.3345 -0.3155 1.1188 4.5147

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.50075 0.50271 4.975 9.33e-07 ***
V161241 0.32442 0.17117 1.895 0.05869 .
Gender 0.10462 0.16485 0.635 0.52601
V161310a -1.08003 0.19918 -5.422 9.62e-08 ***
SocClass 0.53439 0.17363 3.078 0.00221 **
V161277 0.02308 0.03577 0.645 0.51901
V161271 0.36937 0.21928 1.684 0.09278 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.721 on 449 degrees of freedom
(3815 observations deleted due to missingness)
Multiple R-squared: 0.1041, Adjusted R-squared: 0.09212
F-statistic: 8.695 on 6 and 449 DF, p-value: 5.731e-09

We reject the null hypothesis. For every step there is a -1 decreased favorability towards less liberal. Higher social class are more favorable towards liberalism.

> ANES2016<-read.csv("http://www.shortell.nyc/online/files/anes_timeseries_2016.csv")
> attach(ANES2016)
>
> Gender[V161342==1]<-1;
Error in Gender[V161342 == 1] <- 1 : object 'Gender' not found
> Gender[V161342==1]<-1;Gender[V161342==2]<-2;
Error in Gender[V161342 == 1] <- 1 : object 'Gender' not found
> Gender<-0;Gender[V161342==1]<-1;Gender[V161342==2]<-2;
> SocClass<-ifelse(V161307<=2,1,0)
> summary(lm(V161128~V161241+Gender+V161310a+SocClass+V161277+V161271))

Call:
lm(formula = V161128 ~ V161241 + Gender + V161310a + SocClass +
V161277 + V161271)

Residuals:
Min 1Q Median 3Q Max
-3.2952 -1.3345 -0.3155 1.1188 4.5147

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.50075 0.50271 4.975 9.33e-07 ***
V161241 0.32442 0.17117 1.895 0.05869 .
Gender 0.10462 0.16485 0.635 0.52601
V161310a -1.08003 0.19918 -5.422 9.62e-08 ***
SocClass 0.53439 0.17363 3.078 0.00221 **
V161277 0.02308 0.03577 0.645 0.51901
V161271 0.36937 0.21928 1.684 0.09278 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.721 on 449 degrees of freedom
(3815 observations deleted due to missingness)
Multiple R-squared: 0.1041, Adjusted R-squared: 0.09212
F-statistic: 8.695 on 6 and 449 DF, p-value: 5.731e-09

We reject the null hypothesis. For every step there is a -1 decreased favorability towards less liberal. Higher social class are more favorable towards liberalism.

Dependent Variable: Favorability towards feminism
Independent Variables: Party affiliation (V161019), Gender (V161342), Sexuality (V161511), Abortion Legal or Illegal (V161232) 1=Abortion should never be permitted -> 5= Abortion is choice, Liberal to Conservative Scale (V161126) 1= Liberal -> 7= Conservative, How much more attention should media pay to discrimination against women (V162231a), R's interest in politics (V162256) 1 = interested -> 4 = very interested

> summary(lm(V162096~Gender+Party+V161511+V161232+V161126+V162367+V162256))

This chart is reliable. Gender, political party, the legality of abortion, and interest in politics have a unique influence on the favorability towards feminists. Favorability towards feminists on average increases 6.5 degrees from men to women. Favorability towards feminists on average decreases 18.017849 degrees from democrats to republicans. Favorability towards feminists on average increases by 6.05 degrees as perception as abortion as a legal choice increases. Favorability towards feminists on average decreases by 2.9 degrees as interest in politics decreases. This model accounts for 28% of the variation among favorability towards feminists.

I TYPED IN THE WRONG NAME LMAOOOOOO

Call:
lm(formula = V161088 ~ Gender + Abortion + SocialClass + V161230 +
V161246 + V161228)

Residuals:
Min 1Q Median 3Q Max
-52.585 -11.174 3.771 12.178 53.105

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 49.1198 12.7883 3.841 0.000154 ***
Gender 0.3312 2.5929 0.128 0.898458
Abortion 3.3520 3.5181 0.953 0.341588
SocialClass 5.3894 2.7076 1.990 0.047600 *
V161230 -8.3280 4.6066 -1.808 0.071801 .
V161246 -1.9052 2.6121 -0.729 0.466432
V161228 -4.1158 3.2526 -1.265 0.206872
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 20.87 on 257 degrees of freedom
(4007 observations deleted due to missingness)
Multiple R-squared: 0.03619, Adjusted R-squared: 0.01369
F-statistic: 1.608 on 6 and 257 DF, p-value: 0.1451

We fail to reject the null hypothesis. The overall model is not reliable.

Cat, Alessia, Kenton

Call:
lm(formula = V161128 ~ Gender + Abortion + SocialClass + V161230 +
V161246 + V161228)

Residuals:
Min 1Q Median 3Q Max
-2.5844 -1.0022 -0.0161 0.8898 4.9833

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.05420 0.75141 1.403 0.161595
Gender 0.10629 0.15428 0.689 0.491350
Abortion -0.01787 0.20574 -0.087 0.930854
SocialClass 0.55042 0.15912 3.459 0.000615 ***
V161230 -0.27595 0.27030 -1.021 0.308075
V161246 0.01395 0.15471 0.090 0.928234
V161228 0.74135 0.19075 3.886 0.000124 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.388 on 320 degrees of freedom
(3944 observations deleted due to missingness)
Multiple R-squared: 0.09626, Adjusted R-squared: 0.07931
F-statistic: 5.681 on 6 and 320 DF, p-value: 1.258e-05

The overall model is significant. This model accounts for 9% of the variation between Liberals and Conservatives.