Question

1-A researcher sends out 2,000 questionnaires via e-mail. Fifty are returned because the addresses are inaccurate. Of the 1,950 delivered questionnaires, 100 are completed and e-mailed back. However, 40 of these respondents wrote that they did not want to participate in the survey. The researcher indicates the response rate was 5.0 percent. Is this the right thing to do?

2-What potential sources of error might be associated with the following situations?

a. In a survey of frequent fliers age 50 and older, researchers concluded that price does not play a significant role in air- line travel because only 25 percent of the respondents check off price as the most important consideration in determining where and how they travel.

b- A survey comes with a water hardness packet to test the hardness of the water in a respondent’s home. The packet includes a color chart and a plastic strip to dip into hot water. The respondent is given instructions in six steps on how to compare the color of the plastic strip with the color chart that indicates water hardness.

3-A researcher investigating public health issues goes into a junior high school classroom and asks the students if they have ever smoked a cigarette. The students are asked to respond orally in the presence of other students. What types of error might enter into this process? What might be a better approach?

Answer 1

(1-A)

This is not the right thing to do.

REASON:

Response Rate of 5.0 was obtained from:the following:

100/2000 = 0.05 = 5 %.

This is not correct because :

50 are returned because the addresses are inaccurate. So, it is 200 - 50 1950 instead of 2000.

40 of the respondendents were not partricipationg. So, it is 100 - 40 = 60 instead of 100.

(2)

(a)

Potential source of bias: Generalization bias:

REASON:

The survey covered not all people, but only are 50 and older. So, these people will concentrate more for facilities than low cost. So, generalization to all persons is wrong.

(b)

Potential source of error: Non response bias

REASON: The respondant will not be interested in troubling strictly following the six steps. The respondents will not be willing to participate in the survey.

3. Type of error that might enter into this process: Student, even if he smokes, when asked orally in the presence of other students, will respond that he has not smoked at all so that he will be a "good" boy by others. The better approach is to conduct secret ballot without identity of the student, so that the student can respond that he has smoked in case he has smoked.