RM - Chp. 6 - survey & observation: construct validity - Exam 2

Rating products

researchers found that consumer ratings were correlated with the cost of the product and the prestige of its brand. it suggests that people may not always be able to accurately report on the quality of products they buy

Encouraging Accurate Responses, Part 2

socially desirable responding/faking good: the idea is that bc respondents are embarrassed, shy, or worried about giving an unpopular opinion, they will not tell the truth on a survey or other self report measure. - anonymity may not be the perfect solution. so researchers try to minimize it by including special survey items that identify socially desirable respondents like, "my table manners at home are as good as when I eat out in a restaurant". If ppl agree then the researchers will discard that individuals data from the final set - researchers use IAT to evaluate peoples implicit opinions about sensitive topics.

Observations Can Be Better Than Self-Reports, Part 1

- Making reliable and valid observations - the construct validity can be threatened by three problems: observers bias, observer effects, and reactivity. observations have good construct validity

Reactivity: When Participants React to Being Watched

- Solution 1: blend in -Unobtrusive observations: unchaining environment, make yourself less noticeable - Solution 2: wait it out - Solution 3: measure the behavior's results: researchers measure the traces a particular beh. leaves behind. (e.g. in a museum, wear and tear on the flooring can indicate which areas of the museum ar most popular. empty liquor bottles in parking garages can show how much alcohol is being consumed in the community)

Encouraging Accurate Responses, Part 1

- People can give meaningful responses - self reports are often ideal & provide the most meaningful information you can get

Construct Validity of Surveys and Polls: choosing question formats: open-ended questions

- A survey question format that allows respondents to answer any way they like. - Ex. ask people to name the public figure they like the most, or ask a sample of ppl to describe their view on immigration. - Ex. What do you think of this class?

Writing Well-Worded Questions: leading question

- Ex. opponents of voter ID laws argue that they will prevent people who are eligible to vote from voting. What is your opinion? Do you strongly favor...mostly appose? - Ex. opponents of voters ID laws argue that they will prevent people who are eligible to vote from voting, and that the laws will affect African American voters espieclly hard. what is your opinion?

Writing Well-Worded Questions: double-barreled question

- asking two questions in one - Do you enjoy swimming and wearing sunscreen?

Writing Well-Worded Questions: negatively worded questions

- does it seem possible or does it seem impossible to you that the nazi extermination of the jews never happened? - People who do not drive with a suspended license should never be punished. Disagree 1 2 3 4 5 Agree

Preventing Observer Bias and Observer Effects: masked design

- don't tell you the purpose, that way you won't have a bias. - if the rat study students had not known which rats were expected to be bright and dull, the students would have not evoked diff. beh. in their charges. when cleavers Hans observers did not know the right answer to the question they were asking, the horse acted differently; he looked much less intelligent.

Encouraging Accurate Response: acquiescence

- giving answers you think the researchers want to hear - how to tell if they are actually yea saying? to include reverse worded items: "if I had my life to live over, I would change almost everything". one drawback is that they can turn into negatively worded questions.

Construct Validity of Surveys and Polls: choosing question formats: semantic differential format

- instead of degree of agreement, respondents may be asked to rate a target object using a numerical scale - Ex. ratemyprofessors.com - 5 = a real gem, 1 = professor got an F too. - Ex. Yelp review - 1 star = poor, 5 star = outstanding

observing people ethically

- is it ethical for researchers to observe the behaviors of others? It depends - most psychologists believe it is ethical to watch people in museums, classrooms, hockey games, public restroom sinks, they do not specifically identify any of the ppl being observed. - psychologists doing research must obtain permission in advanced to watch or record peoples private beh.

Encouraging Accurate Response: fence sitting

- researchers may try to jostile people out of this tendency by taking away the neutral option.

Preventing Observer Bias and Observer Effects

- researchers train their observers well. they develop clear rating instructions called codebooks, so the observers can make reliable judgments with less bias. - codebooks are precise statements of how the variables are operationalized, the more precise and clear the codebooks are, the more valid operationalizations will be.

Self-reporting "more than they can know"

- self reports can be inaccurate, especially when ppl are asked to describe why they are thinking, behaving, or feeling the way they do. ppl give unintentional inaccurate responses to the researcher - a study was done where 6 socks were layed out and women were told to pick the sock they liked the best even though the socks were the same. they all picked the far right sock & the researchers suggested that they picked It bc it was on the far right of the table, the women gave an explanation that it was bc of the quality. for this reason researchers can't assume the reasons ppl give for their own beh. are their actual reasons.

Encouraging Accurate Response: response sets

- sometimes ppl use shortcuts - rather than thinking carefully about each answer, ppl might answer all of them positive, negatively, or neutral - Christmas treeing - weakens construct validity

Construct Validity of Surveys and Polls: choosing question formats: forced-choice questions

- specific way to ask survey questions - true, false/ yes, no - Ex. used in political polls, such as asking which of the two or three candidates they plan to vote for. - Ex. narcisitic personality test - picking one statement out of two items paired. - Ex. Do you like this class so far? Yes or No

Construct Validity of Surveys and Polls: choosing question formats: likert scale

- strongly agree, agree, neither agree nor disagree, disagree, and strongly disagree - How difficult is this class? - Easy 1 2 3 4 5 Hard

Construct Validity of Behavioral Observations

- survey and poll results are among the most common types of data used to support a frequency claim - the kind you read most often in newspapers of on websites - some scientists believe observing behavior is better than collecting self-reports though surveys, bc ppl can't always report on their beh. or past events accurately - observational research is not just for frequency claims: it can also be used to operationalize variables in associations & casual claims - important to have good construct validity

observers bias

- when observers see what they expect to see - Ex. phsychoanayic therapists were told to watch a video of a man talking to a professor about his feelings & work expereince. one group was told he was a applicant & the other group wast old he was a patient & both groups described him differently even though they watched the same video.

observer effects

- when participants confirm observer expectations - Ex. bright & dull rats: students were either given a "dull" rat or a "bright" rat, even though the rats came from the same liter, those that were believed "bright" completed the maze a little faster each day with fewer mistakes & the "dull" rats did not improve their performance. shows that observers not only see what they expect to see; sometimes they even cause the beh. of those they are observing to confirm their expectations. - Ex. clever Hans - accepted what he was saying without evidence

Self-reporting memories of events

flashbulb memories are not usually accurate