Mean of computer experience for Males: 10.21818 Standard Deviation of computer e
ID: 374349 • Letter: M
Question
Mean of computer experience for Males: 10.21818
Standard Deviation of computer experience for Males: 2.973282
Mean of computer experience for Females: 9.296296
Standard Deviation of computer experience for Females: 2.839486
The 95% confidence interval for the difference in the two means is:
Since this interval contains the point zero, we can conclude that the differences in the 2 means of computer experiance of males and females is not significant i.e. the difference in the mean can be statistically approximated to zero.
The 95% confidence interval is obtained by t test using the statistical software R. The above figure shows the result. The R code to obtain this is:
d1 = read.csv("data1.csv")
d1_gender_male = subset(d1, d1$Gender=="Male")
d1_gender_female = subset(d1, d1$Gender=="Female")
mean(d1_gender_male$Years.Experience)
mean(d1_gender_female$Years.Experience)
sd(d1_gender_male$Years.Experience)
sd(d1_gender_female$Years.Experience)
t.test(d1_gender_male$Years.Experience, d1_gender_female$Years.Experience)
(b)
Mean of computer experience of people with PS's: 8.685714
Standard Deviation of computer experience of people with PS's: 2.446744
Mean of computer experience of people without PS's: 10.82979
Standard Deviation of computer experience of people without PS's: 2.973208
The 95% confidence interval for the difference in the two means is:
Since this interval does not contain the point zero, we can conclude that the differences in the 2 means of computer experiance of people with and wothout PC's is significant i.e. the difference in the mean cannot be statistically approximated to zero. Thus there is difference in the mean of the years of experiance of people with and wothout PC's and since the confidence interval shows negative value it means that the years of experiance of people without PC's is more than people with PC's.
The 95% confidence interval is obtained by t test using the statistical software R. The above figure shows the result. The R code to obtain this is:
d1 = read.csv("data1.csv")
d1_pc_yes = subset(d1, d1$Own.PC == "Yes")
d1_pc_no = subset(d1, d1$Own.PC == "No")
mean(d1_pc_yes$Years.Experience)
mean(d1_pc_no$Years.Experience)
sd(d1_pc_yes$Years.Experience)
sd(d1_pc_no$Years.Experience)
t.test(d1_pc_yes$Years.Experience, d1_pc_no$Years.Experience)
(c)
Since the PC ownership is statistically significant than the Gender, the PC ownership is likely to be a better predictor than the Gender.
Explanation / Answer
Consider the washroom data above and construct a logistic regression model for whether users washed their hands based on the regressors whether users carried a backpack and whether other were present. Estimate the parameter vector by three steps of Newton-Raphson approximation of the maximum likelihood estimator. Obtain the initial values of the parameters from the odds-ratios.
Gender BkPk Male Male Male Male no no yes alone 20 yes others7 Others WashYes WashNo alone 31 others 5 54 10 15 8