1. Take look at the job data shown below with the outcome variable being an over
ID: 3226959 • Letter: 1
Question
1. Take look at the job data shown below with the outcome variable being an overall rating of job satisfaction. We suspect that variables such as median income, job stress, hiring prospects, and work environment quality contribute to the overall rating of job satisfaction. Enter the data in SPss and respond to all the questions following the table. (Attach your relevant spss output to the back of your exam. Median Income Job Stress Hiring Work Overall Job Prospects Environment Satisfaction Quality 101360 19 63 68970 73 75560 12 96 93680 103 69720 31 110 70210 51 93350 56 125Explanation / Answer
a) R = 0.9455
R sqr = 0.8940
F = 31.625
p = 3.77 E -07 (much less than 0 i.e 0.0000000038)
b) Best predictors by rank are
Hiring prospects - coeff 1.02, p-value 0.0000049
Work environment - coeff 1.08 p-value 0.0000005
Job stress - coeff 0.9463, p-value 0.000075
because p-value is significant for all these variables, less than 0.01. Lower the p-value, higher their significance
c) y = 115.9378 -0.0010*55000+0.9463*10+1.0174*20+1.08*2
overall job satisfaction = 92.99
d) Multicollinearity happens when there is high correlation between some predictors.
there is only one pair median income and job stress which are correlated with 0.64 value. Multicollinearity might not be a problem if we take a cutoff of 0.8
e) If we have to exclude then we can exclude median income. We will do it if we keep the cutoff value of correlation as 0.6. Then we will be left with choosing between median income and job stress (correlation of 0.64), but since job stress has significant p-value we will exclude median income.
Median income Job stress Hiring prospects work environment overall job satisfaction Median income 1 Job stress 0.64 1 Hiring prospects 0.06 (0.15) 1 work environment 0.13 0.27 (0.36) 1 overall job satisfaction 0.10 0.45 0.24 0.65 1Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.