4.90 Judging sampling design In each of the following situations, summarize nega

Question

4.90 Judging sampling design In each of the following situations, summarize negative aspects of the sample design. a. A newspaper asks readers to vote at its Internet site to determine if they believe government expenditures should be reduced by cutting social programs Based on 1434 votes, the newspaper reports that 93% of the city's residents believe that social programs should be reduced eports that letters to her office are running 3 to 1 in opposition to the passage of stricter gun control laws. She concludes that approximately 75% of her constituents oppose stricter gun control laws c. An anthropology professor wants to compare attitudes toward premarital sex of physical science majors and social science majors. She administers a questionnaire to her Anthropology 437, Comparative Human Sexuality class. She finds no appreciable difference in attitudes between the two majors, so she concludes that the two student groups are about the same in their views about premarital sex d. A questionnaire is mailed to a simple random sample of 500 household addresses in a city. Ten are returmed as bad addresses, 63 are returned completed, and the rest are not returned The researcher analyzes the 63 cases and reports that they represent a "simple random sample of city households."

Explanation / Answer

Ans:

a. Here the sample is biased and not representative. Reason: As the survey was conduscted by a newspaper group, its reach was restricted to its reader and not the whole population. So the sample taken is not a random sample and inherently carries a bias. (Also, the survey was conducted on internet so that too implement a bias, although small, as those will respond who have easy acess to internet.)

b. Here again, the sample is not representative. Reason: As all the potential identity in the population is not considered for the sample. Only those who take an effort to write the letter to the particular congresswomen got to have their say. So the sample is not random and carries a bias.

c. Here, all the division of physical sciences and social sciences are not considered. So the sample suffers from undercoverage. The correct way should have been to identify the whole population of respective class and then do the sampling by P.P.S. (probability proportional to size)

d. First, the sample size is too small to what was planned prior, so the inference is not reliable. Moreover, The unresponsiveness must have something in common, so the sample might contain a possible bias.

Related Questions

Navigate

• Browse (All)
• Browse 4
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.