Do now – identify the sampling method A sample of 2,000 was sought to estimate the average achievement in science of fifth graders in a city’s public schools. The average fifth grade enrollment in the city’s elementary schools is 100 students. Thus, 20 schools were randomly selected and within each of those schools all fifth graders were tested. A sociologist conducts an opinion survey in a major city. Part of the research plan calls for describing and comparing the opinions of four different ethnic groups: African Americans, Asian Americans, European Americans, and Native Americans. For a total sample of 300, the researcher selects 75 participants from each of the four predetermined subgroups. A researcher has a population of 100 third grade children from a local school district from which a sample of 25 children is to be selected. Each child’s name is put on a list, and each child is assigned a number from 1 to 100. Then the numbers 1 to 100 are written on separate pieces of paper and shuffled. Finally, the researcher picks 25 slips of paper and the numbers on the paper determine the 25 participants.

Problems with Sample surveys Lesson 4.1 Day 3 Problems with Sample surveys

Objectives Explain how undercoverage, nonresponse, question wording, and other aspects of a sample survey can lead to bias.

Definitions Undercoverage: some people can’t be chosen or are less likely to be chosen (ex. Landlines) Nonresponse: people can’t be reached or refuse (ex. No answer, hang up) Response bias: Interviewer influences answer, wording of questions, people lie These all lead to bias!

*Voluntary response bias is when an invitation to be part of a survey is given and people decided whether or not to participate.  They were NOT specifically chosen by the researchers. *Nonresponse is when people ARE chosen by the researchers for the survey but the people do not complete the survey.

Introduce Project The Project:   You and your partner (or you by yourself) will design and conduct an experiment to investigate the effects of response bias in surveys.  You may choose the topic for your surveys, but you must design your experiment so that it can answer at least one of the following questions: Can the wording of a question create response bias? Does providing additional information create response bias?   Do the characteristics of the interviewer create response bias? Does anonymity change the responses to sensitive questions? Does manipulating the answer choices/order of answer choices change the response? Can revealing other peoples’ answers to a question create response bias?

proposal Proposal (25 points): The proposal is due: _______. The proposal will be worth 25% of the grade, so don’t treat it casually.   If the proposal isn’t approved the first time, you will need to resubmit it for a reduced grade.  You must attach the original proposal to any resubmissions. In your proposal, you should: Describe your topic and state which type of response bias you are investigating Describe how you will obtain your subjects in an unbiased manner (minimum sample size is 50). This must be practical!!  Your population does not need to be from school nor should you interrupt any classes. Describe what your questions will be and how they will be asked, including how you will incorporate the principles of a good experiment and avoid potentially confounding variables.  You should also indicate what your hypotheses are. Convince me that you have a good design!

Pick your partner You have the rest of class to think about your project proposal

Do Now You are on the staff of a member of Congress who is considering a bill that would provide government-sponsored insurance for nursing- home care. You report that 1128 letters have been received on the issue, of which 871 oppose the legislation. “I’m surprised that most of my constituents oppose the bill. I thought it would be quite popular,” says the congresswoman. Are you convinced that a majority of the voters oppose the bill? How would you explain the statistical issue to the congresswoman?

Do Now A researcher was interested in studying the effect of number of hours spent watching tv on children's aggression level. The children were randomly split in two groups varying ages 5-10. After a month of time has passed then researcher returned to their homes to record the number of hours spent watching tv and any changes in aggression level. The results were monitored by the families of the children. What could be a confounding variable?