Chapter 6 Surveys Surveys and Polls a sample of people being asked to answer questions via mail in person interview phone interview or online Types of Questions Yes No Questions respondent answers yes or no to the items is it possible to believe in both God and evolution do you approve of the job the president is doing Forced alternative questions the respondent must select between 2 alternative responses often agree or disagree no opinion or not sure are not options is the position of women better or worse than it way 10 years ago the respondent must select the most suitable response from among several a question is asked to which the respondent must construct his or her own Multiple Choice alternatives Open Ended questions short answer advantages information more complete disadvantages participant may not understand what you are looking for some answers committed inadvertently summarizing data dif cult interrater reliability partially open ended multiple house with last choice being Other usually good to include Likert designated scale is to represent neutral The individual answers a question by selecting a response alternative from a anchors the labeled opposite extremes and sometimes middle especially if it no set rules for labeling but exact point labeling reduces ambiguity Response set tendency for participants to answer all or most of the questions the same way whenever questions in a series all have the same choices for responding example participant uses the nutria answer for everything Fence sitting maybe the question is confusing or unclear maybe the question is controversial and the respondent is playing safe maybe they really do not feel strongly about any of the items so they really are neutral maybe they want to nish quickly What can a researcher do Take away the neutral option dont anchor 3 make it a 4 point scale yay sayers AKA acquiescence bias what can a researcher do use some statements that require agree and others that require disagree to address the same construct Avoiding the response set Items should include a mixture of positive and negative statements including some alternate phrasing of the same item forces respondents to move back and forth between opposite ends of the scale so that they cannot fall into a single response set for answering the questions double barreled my instructor was enthusiastic and knowledgable the phone representative was helpful and courteous i am always running last and feeling disorganized both have to do with conscientiousness but its still a double barreled item Writing well worded questions Leading questions Do you think that relations between Blacks and Whites will always be a problem or that a solution will eventually worked out 45 responded optimistically Do you think that relations between Blacks and Whites are as good as they re going to get or will they eventually get better 73 respond optimistically How do I know if wording matters test different wording if the results are the same no matter the wording it clearly doesn t matter if results differ may need to report results differently for each question Double negatives Ordering effects cause confusion decrease construct validity Prepare different versions of the survey with different questions in different orders so you can look for order effects ALSO present demographic items at the end of the survey not the beginning Problems with accurate responses socially desirable responding ensuring anonymity use weed out items My table manners at home are as good as when I eat out at a restaurant I am always a good listener no matter who I am talking to I never hesitate to go out of the way to help someone in trouble How do researchers avoid socially desirable responding mask the true purpose of the survey with various unrelated items interested about racial attitudes but also ask about politics gender rolls and include ller items education use implicit measures computerized measures to evaluate people s implicit opinions about sensitive topics Social desirability isn t the only issue self reporting more than they can know you know your opinion better than anyone else so self report can be useful but self report can be inaccurate when asking some questions why do you think what you do why do you behave the way you dp why do you feel the way you do memories for attitudes are notoriously inaccurate and sometimes memory isn t the real issue Problems with accurate responses self reporting reasons for behavior women given samples of stockings to examine Wilson 1977 asked which they preferred and why all elected their favorite reason was quality but they were all exactly the same Analysis of Behavioral Observation The potential problems an be subjective Can re ect researchers bias Might be inaccurate or incompltete portrayal of the situation might not generalize beyond particular So how do researchers evaluate quality of observational results Assessing quality of behavioral observation con rmability indicate procedures used to check recheck data solicit input from colleagues regarding potential inconsistencies contradictions or instances of bias use more than one investigator to record same data and compare dependability similar to test retest reliability in numerical data can a study and its results be replicated credibility the accuracy of the identi cation and description of the subject of stuudy the best judge is the subject him herself if the subject reads the report he she can judge its accuracy as long as he she is able to do this and is being honest also accounted for during the Peer Review Process transferability generalizability can the results can be generalized to other setting or groups the more thorough and clear the report the easier it will be for the other to judge this Observation vs Self report how much do you talk per day is your happiness related to the types of conversations you have do parents behave well at their children s athletic events are kids happy to see their parents at the end of a work day in these cases at least observation is probably more reliable Construct validity of behavioral observation What is the variable of interest How is it being operationalized Observer bias observers might see what they expect to see Lots of examples intellectual bloomers versus normal kids futures tried not to show favoritism Rosenthal effect maze bright versus maze dull rats told teachers that half the students were bloomers really responses great participants
View Full Document
Unlocking...