for DV, higher numbers = more green choices
summary(aov(DV ~ default, data=data))
## Df Sum Sq Mean Sq F value Pr(>F)
## default 2 58.6 29.31 6.881 0.00121 **
## Residuals 279 1188.5 4.26
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
br<-subset(data, default=="brown")
gr<-subset(data, default=="green")
no<-subset(data, default=="none")
#default manipulation worked
t.test(br$DV, no$DV)
##
## Welch Two Sample t-test
##
## data: br$DV and no$DV
## t = -1.8631, df = 177, p-value = 0.0641
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.22822966 0.03532606
## sample estimates:
## mean of x mean of y
## 5.347368 5.943820
t.test(gr$DV, no$DV)
##
## Welch Two Sample t-test
##
## data: gr$DV and no$DV
## t = 1.8517, df = 180.97, p-value = 0.0657
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.03314852 1.04346725
## sample estimates:
## mean of x mean of y
## 6.44898 5.94382
t.test(gr$DV, br$DV)
##
## Welch Two Sample t-test
##
## data: gr$DV and br$DV
## t = 3.5738, df = 174.15, p-value = 0.0004553
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.4932291 1.7099932
## sample estimates:
## mean of x mean of y
## 6.448980 5.347368
and the lay theory manipulations worked - the manipulation checks were affected as expected
change<-subset(data, changecheck==2)
stable<-subset(data, stablecheck==1)
t.test(change$MC, stable$MC)
##
## Welch Two Sample t-test
##
## data: change$MC and stable$MC
## t = -7.8373, df = 277.45, p-value = 9.893e-14
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.6336167 -0.3792173
## sample estimates:
## mean of x mean of y
## 3.119737 3.626154
t.test(change$certain, stable$certain)
##
## Welch Two Sample t-test
##
## data: change$certain and stable$certain
## t = -2.0067, df = 280, p-value = 0.04574
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.46479033 -0.00447418
## sample estimates:
## mean of x mean of y
## 5.674342 5.908974
default condition and lay theory manipulation interact in the opposite way we expected
summary(aov(DV ~ lay*default, data=data))
## Df Sum Sq Mean Sq F value Pr(>F)
## lay 1 6.1 6.058 1.444 0.23053
## default 2 61.9 30.930 7.373 0.00076 ***
## lay:default 2 21.3 10.662 2.542 0.08058 .
## Residuals 276 1157.9 4.195
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#default condition was effective for people who read that preferences are stable
summary(aov(DV ~ default, data=stable))
## Df Sum Sq Mean Sq F value Pr(>F)
## default 2 75.6 37.81 8.429 0.000365 ***
## Residuals 127 569.6 4.49
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#but not for people who read that preferences are malleable
summary(aov(DV ~ default, data=change))
## Df Sum Sq Mean Sq F value Pr(>F)
## default 2 7.6 3.787 0.959 0.386
## Residuals 149 588.3 3.948
but if we use measured preference certainty, then we get the pattern that we expected
summary(aov(DV ~ certain*default, data=data))
## Df Sum Sq Mean Sq F value Pr(>F)
## certain 1 58.5 58.48 14.672 0.000158 ***
## default 2 55.3 27.66 6.939 0.001147 **
## certain:default 2 33.3 16.66 4.181 0.016266 *
## Residuals 276 1100.0 3.99
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(data$certain)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.833 5.167 5.833 5.783 6.667 7.000
certain<-subset(data, certain>5.8)
uncertain<-subset(data, certain<=5.8)
#no default effect when preferences are certain
summary(aov(DV~default, data=certain))
## Df Sum Sq Mean Sq F value Pr(>F)
## default 2 15.8 7.891 1.755 0.176
## Residuals 158 710.4 4.496
#but there is an effect when they are uncertain
summary(aov(DV~default, data=uncertain))
## Df Sum Sq Mean Sq F value Pr(>F)
## default 2 64.6 32.28 9.789 0.000117 ***
## Residuals 118 389.2 3.30
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1