Statistics: Section 1.5
Nonsampling error
results from undercoverage, nonresponse bias, response bias, or data-entry error
Surveys tend to suffer from low response rates. Based on past experience, a researcher determines that the typical response rate for an e-mail survey is 20%. She wishes to obtain a sample of 200 respondents, so she e-mails the survey to 2000 randomly selected e-mail addresses. Assuming the response rate for her survey is 20%, will the respondents form an unbiased sample? Explain.
No. The survey still suffers from undercoverage (sampling bias), nonresponse bias, and potentially response bias.
Response bias
exists when the answers on a survey do not reflect the true feelings of the respondent. Example: Asking someone how many push ups they can do and them giving a response vs. how many push ups they can actually do when told.
Consider the two questions shown below. (a) Who is your favorite actor? (b) What is the best movie you have seen in the past year?Will the order in which the questions are asked affect the survey results? If so, what can the pollster do to alleviate this response bias?
Yes,question order will affect the survey results. The pollster should alternate the order of the questions given in the questionnaire so that different respondents receive questionnaires with the same questions but different question orderings.
The following survey has bias. (a) Determine the type of bias. (b) Suggest a remedy. A magazine is conducting a study on the effects of infidelity in a marriage. The editors randomly select 400 women whose husbands were unfaithful and ask "Do you believe a marriage can survive when the husband destroys the trust that must exist between husband and wife? "
(a) What is the type of bias? Response bias (b) What is a possible remedy? Reword the question
The following survey has bias. (a) Determine the type of bias. (b) Suggest a remedy. Suppose a survey regarding the sleeping habits of students is being conducted. From a list of registered students, a researcher obtains a simple random sample of 150 students. One survey question is "How much sleep do you get? "
(a) What is the type of bias? Response bias (b) What is a possible remedy? Reword the question.
The survey has bias. (a) Determine the type of bias. (b) Suggest a remedy. A polling organization conducts a study to estimate the percentage of households that have high- speed Internet access. It mails a questionnaire to 1778 randomly selected households across the country and asks the head of each household if he or she has high dash speed Internet access. Of the 1778 households selected, 49 responded.
(a) Which of these best describes the bias in the survey? Nonresponse bias b) How can the bias be remedied? The polling organization should try contacting households that do not respond by phone or face-to-face.
Bias
If the results of a sample are not representative of the population then the sample is bias.
Suppose a random phone survey is being conducted to find the population's feelings on a new stamp design. If the randomly selected person is not home they are called again later in the day.What tactic is being used to increase the response rate for this survey?
The company is using callbacks.
Suppose a surveyor wants to conduct a phone survey about a new car. She plans to take a simple random sample. However, some people do not want to participate. Do you believe this can affect the ability of the surveyor to obtain accurate polling results? If so, how?
Yes, especially if the people who do not want to participate have a trait that is not accurately represented by the remaining people in the sample.
The survey has bias. (a) Determine the type of bias. (b) Suggest a remedy. Suppose you are conducting a survey regarding activity level of children in a school district. You obtain a cluster sample of 25 schools within a school district and sample all sophomore students in the randomly selected schools. The survey is administered by the parents.
a) Which of these best describes the bias in the survey? Response bias b) How can the bias be remedied? The survey should be administered by an impartial party.
The survey has bias. (a) Determine the type of bias. (b) Suggest a remedy. A pro dash evolution advocate wants to estimate the percentage of people who favor teaching only evolution in schools. He conducts a nationwide survey of 1540 randomly selected adults 18 years and older. The interviewer asks the respondents, "Do you favor promoting science by teaching only evolution in schools?
(a) Which of these best describes the bias in the survey? Response bias (b) How can the bias be remedied? The interviewer should reword the question.
The owner of a shopping mall wishes to expand the number of shops available in the food court. She has a market researcher survey the first 80 customers who come into the food court during weekday evenings to determine what types of food the shoppers would like to see added to the food court. Complete parts (a) and (b) below.
(a) The survey has bias. Determine whether the flaw is due to the sampling method or the survey itself. For biased surveys, identify the cause of the error. What is the cause of the bias? Sampling bias (b) Suggest a remedy to the problem. Which of the following is the best way to remedy this problem? Ask customers throughout the day on both weekdays and weekends.
Nonresponse bias
when individuals selected to be in the sample who do not respond to the survey have different opinions from those who do. Example: Honda sent me an email survey but I did not complete the survey so the receptionist called me to take the survey.
Sampling error
results from using a sample to estimate information about a population. This type of error occurs because a sample gives incomplete information about a population.
Sampling Bias
means that the technique used to obtain the individuals to be in the sample tends to favor one part of the population over another. Any convenience sampling has bias because the individuals are not chosen through random sampling. Example: Roosevelt and Landon's election stats about whose going to win. There was bias in this sample because most of the car owners telephone owners, etc. where republican's.
