Landreys Class Notes

# Run this code to load the required packages suppressMessages(suppressWarnings(suppressPackageStartupMessages({ library(coursekata) })))

Run to view results

## Read in the data moralemotions <- read.csv("moralemotions.csv")

Run to view results

gf_histogram (~ WC, data = moralemotions, color = "red")

Run to view results

data$tWC <- sqrt(data$WC)

Run to view results

** I thought this was assinging the new varrible and doing the sqrt but coninousley got error symbol **

empty_model<- lm(WC ~ NULL, data = moralemotions) empty_model

Run to view results

## dat$shame_withdrawl<- rowMeans(data.frame(dat$gasp_4, dat$gasp_7, dat$gasp_8, dat$gasp_12),na.rm=T) tosca$guilt <- rowMeans(data.frame(tosca_01_3, tosca_02_1,tosca_03_3,tosca_04_4,tosca_05_4,tosca_06_4,tosca_07_3, tosca_08_4,tosca_09_3,tosca_10_3,tosca_11_2),

Run to view results

Tried moving all the comas but not sure why the error continued to show up because when I am going and looking at my R studio notes you have to put the varible that is going to be measured and added in followed by the , at the end to seperate them to get the ultimate goal of the row means function of this command.

tosca$guilt <- lm( tosca_guilt ~ tWc, data = moralemotions)

Run to view results

confint(tosca_guilt)

Run to view results

If my varible were to be found which I think is where something went wrong in the first few question it would show the confidence interval that when interperting we can yes yes statsically significant but have to be careful to say that 95 out of 100 intervals would contain the true population mean

as.factor (empathy)

Run to view results

gf_jitter(tosca_guilt ~ empathy, data = moralemotions, color = "red") ## This should be the right code

Run to view results

num_empathy_1 <- sum(participants$empathy_score == 1)

Run to view results

tWc= tosca_guilt + empathy + error

model <- lm(tWC ~ tosca_guilt + empathy, data = moralemotions )

Run to view results

I would predict this running would should based off the backround information and looking at my notes that the survey and code would predict that they are both good predictors but the only solifited way to test this is if the v vlaue = less than .05

Run to view results

Both

b0= twc which is accounting for the word count in the disclouser

b1= guilt score that they recorded and the the b1 incremident would be added on for thenext group mena which includes empthaty

It should explain more varation than a empty model becuase it is accounting for two more things that effect guilt and shame scores that if there p vlaues are greater thatn .05 is signiifcant and needs to be reported and acknoleded as reduced error and more explanantion!

Run to view results

It is easy to get lost in all the statiscal lingo reggrefssions and all but the important take aways at a basic level are that the implemtation of testing empthy and tosca guilt is that it allows us to explain more of our answers and varations in the reponses that were collected. The key way to spot importance and impact is going to be a P vlaue that is less than .05. Based on this you can asses outputs that adding somethign else in when looking at guilt scores of particpants is helpful.

Apology=+β 1 ×tosca_guilt+ϵ

apology= tWC + tosca_guilt + error

logistic_dat <- filter(dat, apology != 99)

Run to view results

model_logistic_dat <- glm(apology ~ tosca_guilt + , data = moralemotions,) model_logistic_dat

Run to view results

Could not get output but based on the filter I would say you cannot predict apology singificantly

They are slightly different espically at then end when we use exp(model) to exponaite them and the explanantioon that logistic regression is a method for fitting a regression curve,

shame_items <- data.frame (moralemotions$gasp_3 moralemotions$gasp_4, moralemotions$gasp_51, moralemotions$gasp_7 moralemotions$gasp_8, moralemotions$gasp_10, moralemotions$gasp_12, moralemotions$gasp_13

Run to view results

fa(Shame_Item, nfactors= 2)

Run to view results

Using factor anaylisis the output below represents that we can conclude that Factor 1 is primarily characterized by dat.gasp_4, dat.gasp_7, dat.gasp_8, dat.gasp_10, and dat.gasp_12, while Factor 2 is primarily characterized by dat.gasp_3, dat.gasp_5, dat.gasp_10, and dat.gasp_13.

I am so sorry that a lot of my code boxes wouldn't run again-- I have no idea what happened because I felt so prepared for this and had been doing well in class but something small must have slipped through and I kind of panicied and just tried to finish the rest out and show my ability of understanding the content even if the code did not process. Thank you for a great quarter and your willingness to help me-- at the start of the class I was very concered if I could make it through and you made it very acessible to me. Thank you.