Question 1

Which technique would you recommend to a researcher who is interested in obtaining a good sample of a large population?

Accepted Answer

Questionnaires are effective for obtaining a good sample from a large population as they can be distributed widely and collected efficiently, allowing for a larger and more diverse sample size.

Question 2

The main objection to using anonymous questionnaires is that they

Accepted Answer

Anonymous questionnaires make it difficult to conduct follow-ups because the respondents' identities are not known, preventing researchers from asking for clarifications or additional information.

Question 3

The cover letter for a questionnaire survey should

Accepted Answer

The cover letter should emphasize the importance of the study to encourage recipients to complete the questionnaire. It should also provide a brief overview of the purpose of the survey without revealing too much information about the anticipated findings. Avoiding information about the way the recipient was selected is not necessary in the cover letter.

Question 4

In a survey of n = 100 teachers, 50 preferred year-around school.Set the .95 confidence interval around this sample proportion.

Accepted Answer

The confidence interval for a proportion is calculated using the formula $p \pm Z \sqrt{\frac{p(1-p)}{n}}$, where $p$ is the sample proportion, $Z$ is the Z-score corresponding to the confidence level (for 95% confidence, $Z \approx 1.96$), and $n$ is the sample size. Here, $p = 0.50$ and $n = 100$, so the margin of error is $1.96 \sqrt{\frac{0.50 \times 0.50}{100}} = 0.098$. Adding and subtracting this from the sample proportion gives an interval of $0.50 \pm 0.098$, which is $0.402$ to $0.598$, rounded to $0.40$ to $0.60$.

Question 5

The structured questionnaire has an advantage over the unstructured interview in all of the following except in

Accepted Answer

The structured questionnaire does not offer freedom to respond from an individual frame of reference, as the questions are pre-determined and standardized. The other options listed are all advantages of the structured questionnaire over the unstructured interview.

Question 6

A researcher may choose to use open-ended rather than closed questions in a questionnaire when the

Accepted Answer

Open-ended questions allow the researcher to identify the possible alternative responses to use later in closed questions. Closed questions limit the range of responses that participants can provide, whereas open-ended questions allow participants to provide detailed and diverse responses to a given question. This can help the researcher to identify possible responses and incorporate them into structured questions for future surveys.

Question 7

The main reason for the inappropriateness of an interview study designed to gather parents' opinions of the validity of the new statewide achievement test for elementary students is that

Accepted Answer

The main reason for the inappropriateness of an interview study designed to gather parents' opinions of the validity of the new statewide achievement test for elementary students is that most parents do not have sufficient knowledge about test validity. Therefore, conducting an interview study may not yield accurate results. A survey or questionnaire may be more appropriate to gather parents' opinions.

Question 8

What is wrong with the following Likert-type scale item? "The university needs to provide additional computer clusters, and the student technology fee should be increased to provide the necessary funds."

Accepted Answer

The item is double-barreled because it is asking about two different things (the need for additional computer clusters and the increase of the student technology fee) in one statement, making it unclear what the respondent's true attitude is towards each individual element.

Question 9

In a survey, statistical precision in estimating population parameters from sample statistics can be increased by

Accepted Answer

Increasing sample size will decrease the margin of error and increase the accuracy of the estimate, as the sample mean will be a better representative of the population mean. Increasing variability of data, limiting the number of subjects surveyed, and using a stricter confidence interval will not improve the precision of the estimate.

Question 10

Unstructured interviews are most appropriate for

Accepted Answer

Unstructured interviews are designed for preliminary exploration of relevant questions in a research study, allowing for an open-ended conversation between the researcher and participant. This type of interview is ideal for collecting data in the initial stages of a study, before developing more specific research questions or hypotheses.

Question 11

The first step in conducting survey research is to

Accepted Answer

Determining the purpose of the survey is essential as it guides the overall approach, including the design of the questionnaire, the selection of the population, and the choice of sample. It sets the foundation for all subsequent steps in the survey process.

Question 12

A sample of 500 voters was surveyed; 400 said they favored a bond issue and 100 said they opposed it.Based on these sample data, estimate at the 95 percent confidence level the proportion in the population that likely favors the bond issue.

Accepted Answer

The correct answer is A. The confidence interval is 0.77 to 0.83.

Question 13

If all other factors are held constant, increasing the level of confidence will cause the width of a confidence interval to

Accepted Answer

Increasing the level of confidence will increase the width of a confidence interval. This is because as the level of confidence increases, the range of values that the true population parameter may fall within also increases, leading to a wider range of values being included within the confidence interval.

Question 14

The main advantage of the questionnaire over the interview is that the questionnaire

Accepted Answer

Questionnaires are generally less expensive in terms of time and money compared to interviews, as they do not require the interviewer to meet with the respondent in person or spend time conducting the interview.

Question 15

According to the text, ____ percent returns should be a reasonable expectation from a mailed questionnaire.

Accepted Answer

The answer of According to the text, ____ percent returns...

Question 16

Which of the following is $\underline{not}$ a rule for a questionnaire format?

Accepted Answer

Asking specific questions before general questions on a related topic can be confusing for respondents and may decrease the accuracy of their answers. Therefore, it is not a rule for a questionnaire format. A, C, and D are all valid rules for questionnaire formatting.

Question 17

The cover letter accompanying a questionnaire typically contains all but one of the following.Which one is not included?

Accepted Answer

An explanation of the anticipated findings of the survey is not typically included in the cover letter accompanying a questionnaire. It may be included in the introduction section of the questionnaire itself. A suggested date for the return of the questionnaire, an assurance of confidentiality to the respondent, and an explanation of the purpose of the survey are all commonly included in the cover letter.

Question 18

A university surveyed 36 randomly selected graduate students in a pilot study and found that 27 said they favored the new university health insurance plan and 9 said they opposed it.How many students must the university survey in order to make an estimate of how all graduate students feel about the plan with a margin of error to be no more than 5 percentage points? Use the 95 percent confidence level.

Accepted Answer

To estimate the sample size needed for a proportion with a specified margin of error (E) at a certain confidence level, we use the formula: $n = \frac{Z^2 \cdot p \cdot (1-p)}{E^2}$, where $Z$ is the Z-score corresponding to the confidence level (for 95%, $Z = 1.96$), $p$ is the sample proportion (in this case, 27/36 favoring the plan, so $p = 0.75$), and $E$ is the margin of error expressed as a decimal (for 5%, $E = 0.05$). Plugging in the values: $n = \frac{1.96^2 \cdot 0.75 \cdot (1-0.75)}{0.05^2}$, which simplifies to approximately 288. Therefore, the university needs to survey 288 students to achieve a margin of error of no more than 5 percentage points at the 95% confidence level.

Question 19

The .95 confidence interval for a population proportion based on a sample of n cases ranged from .40 to .56.If the sample proportion was unchanged, but the sample size was increased to 4n, the .95 confidence interval would be about

Accepted Answer

The formula for calculating the confidence interval for a population proportion is:

CI = sample proportion +/- margin of error

where the margin of error can be calculated as:

margin of error = z* * sqrt[(sample proportion * (1 - sample proportion))/n]

where z* is the z-score corresponding to the desired level of confidence.

Given that the original confidence interval had a range of 0.56 - 0.40 = 0.16 and the sample size is increased to 4n, we can calculate the new margin of error as:

margin of error = 1.96 * sqrt[(0.48 * 0.52)/(4n)] = 0.049

where we have used the fact that the original sample proportion was (0.40 + 0.56)/2 = 0.48.

Thus, the new confidence interval will be:

CI = 0.48 +/- 0.049

CI = (0.431, 0.529)

which rounds to (0.44, 0.52) at the .95 confidence level. Therefore, the best choice is C.

Question 20

If all other factors are held constant, increasing the number of subjects in a survey will cause the width of a confidence interval to

Accepted Answer

Increasing the number of subjects in a survey will cause the width of a confidence interval to decrease. This is because as the sample size increases, the standard error of the mean decreases, which leads to a narrower confidence interval.

Quiz 14: Survey Research