An executive recruitment firm used a lengthy mail survey to gather information on the job mobility of mid-level managers. A sample of 500 eligible middle managers was selected, using a sample random sample procedure. The firm used 3 waves of mailings. Sixteen of the questionnaires were returned due bad addresses, all but two of the returned surveys were usable. After the third mailing, each of the non-responding (NR) sample units was contacted by phone and asked to answer only four questions regarding variables that the recruitment firm thought were particularly important, given the objective of the study. The table below gives mean values for these variables.
Years in Total Years of
Wave # of responses Age Income Current Position Management Experience
1 125 30 $22,000 1.2 5.1
2 100 37 $27,000 4.0 9.4
3 75 42 $32,500 5.1 15.1
NR 200 50 $31,250 10.2 24.2
a) What is the response rate for the completed questionnaire?
b)Which variables, if any, see, to be the most affected by potential non-response bias? Does this tell you anything about the sample selection procedure?
Explain answer and show calculations