Exam 2 Thursday 3 6 in class The exam will consist of 25 multiple choice questions 2 pts each 50 pts 15 points short answer 10 points short essays Total 75 points Pay close attention to recommendations regarding research methodology or the pros and cons of various approaches discussed in class This is what you need to know for the exam 2 25 14 2 25 14 Define survey research and understand the benefits of this method Collecting information from a sample through responses to standardized questions o Standardized questions are asking everyone the same questions the same way Benefits of survey research o Versatile can use in a lot of different situations and context o Efficient o Generalizable because surveys can be given to a lot of people Know the types of surveys benefits drawbacks to each I m not sure if we need to know solutions so I included them just in case Mailed Self administered survey o Benefits Cheaper Large and generalizable samples o Drawbacks Poor response rate Household members can share responses No control of who in the household fills out the survey o Solutions ways to control the and quantity you get back Group administered survey given at the same time to a group of people I e class Coupon included 2 bills etc Good cover letter Clear instructions Follow up mailings Token incentives o Benefits Very cheap High response rate o Drawbacks Even though you don t have to do it awkward to say no what else would you be doing Must have a captive audience Researchers need to avoid comments that can bias the group because it will effect a lot more of the data May be worried about confidentiality i e filling out survey about your job with co workers and boss all in the room o Solutions Read a standard statement of instructions and if possible have the same person give the instructions to all groups Insure confidentiality and voluntary partipcaton Make it clear that the research is separate from the insitiuton i e on the job you may be worried that your boss will read your survey and be able to tell it s you Phone survey o Benefits Most people have a phone Large generalizable samples can be used by using random digit dialing i e computer dials random numbers until someone picks up High response rate once you actually get someone on the phone o Drawbacks Not reaching proper sampling units Because cell phones are not listed so therefore not used for random digit dialing May not get enough complete responses Telemarketing call screening or people just aren t home Need to make a lot of attempts Ask questions initially to ensure you re talking to the right person i e Are you the head of the house o Solutions In person interview o Benefits High response rate Surveys can be longer and more complex You can control the order of the questions Essential for some groups especially children o Drawbacks Interviewer effects bias Expensive Embarassement of interviewee so may not be honest Lower response rates in central cities Demographic bias may occur or may be difficult in central cities o Solutions Interviewer ranting Advance contact Send a postcard ahead of time i e letting them know someone this is legit and someone will be coming to their door to interview them about Multiple interview attempts Electronic web surveys o Benefits In general very low cost I e qualatrics at OSU questions Attractive pull down menus and skip patters participants cannot skip Data entry time and errors are eliminated Because can automatically put data in spreadsheet o Drawbacks Nonrepresentative samples Some people don t have internet access Almost anyone with internet access can participate and sometimes they can participate 10x o Solutions Use when the population of interest has internet access I e OSU student and not the elderly in rural areas Understand the difference between anonymity and confidentiality Anonymity anonymous o No identifying information is ever recorded to link participants with their response No risk Confidentiality This cannot work for panel research o Surveys and data are both labeled with linking numbers o Close tabs are kept on the key that links the numbers with individuals identities involved However researcher can know what you said so there is some risk Be able to identify double negatives and double barreled questions Double negatives o Do you disagree that there should not be a tax increase Yes agree that there should be a tax increase o The responsibilities of fatherhood never overshadows the joys A better phrase would be the responsibilities of fatherhood is worth the joys Double barreled questions 2 parts o Most women enjoy caring for their homes AND men just don t enjoy that stuff You can agree with one part and disagree with the other so provides inaccurate answers Understand the logic of skip patterns Skip patterns have fileter questions o I e are you currently working Followed by contingent questions only answer if yes in filter o I e If yes how many hours per week do you work o Online you can just not see the questions your not suppose to answer Know what floaters and fence sitters are and what we have to do in surveys to accommodate them o Floaters o People who choose an answer even though they do not know the answer We accommodate them by including a don t know option o Fence sitters o Truly neutral people We accommodate them by including neither agree nor disagree or neutral options Understand tips for designing surveys including cognitive interviews and context effects Make response categories exhaustive o At the very least include an other option Make response categories mutually exclusive o No overlap should be possible between answer choices Build on existing reliable and valid instruments o Don t reinvent the wheel Refine and pretest questions o Do a cognitive interview Give questions to a few people and ask them to think out loud so you can get a sense of problems with questions and describe how they understand the questions o Pilot study Test out the survey with a small group Maintain consistent focus o Be thorough but you will always miss something o Avoid very long surveys with redundant or irrelevant questions to prevent subject fatigue participant attrition Order the questions o In a logical order organize the sections Easy online or in person o Watch for context effects I e beliefs and fathers If you ask a father how important spending time with your child is they are going to say important then they will probably not also say later that they don t