Textbook Notes (363,212)
Canada (158,271)
Psychology (9,573)
PSYB01H3 (585)
Anna Nagy (283)
Chapter 7

Chapter 7 – Asking People About Themselves.docx

13 Pages
Unlock Document

University of Toronto Scarborough
Anna Nagy

Chapter 7 Asking People About Themselves: Survey Research - survey research employs questionnaires and interviews to ask people to provide info about themselves attitudes/beliefs, demographics and other factors, and past or intended future behaviours Why Conduct Surveys? - surveys are a common and important method of studying behaviour - provide us a methodology for asking people to tell us about themselves - become important as society demands data about issues rather than only intuition and anecdotes - in basic research many important variables, including attitudes current emotional states, and self-reports of behaviors are most easily studied using questionnaires or interviews - not only important for how ppl behave at a given point in time but also in order to study relationships among variables and ways that attitudes and behaviors change over time - survey research is important as a complement to experimental research - an assumption that underlies the use of questionnaires/interviews is that peple are willing and able to provide truthful/accurate answers - response set: a tendency to respond to all questions from a particular perspective rather than to provide answers that are directly related to the questions [social desirability set > answer in the most socially accepted way] - socially desirable answers most likely when dealing w/ sensitive topic [sex] - ppl only likely to lie when they do not trust the researcher Constructing Questions to Ask - most important factors to consider when constructing a questionnaire Defining the Research Objectives - first thing is to explicitly determine the research objectives - requires deciding what type of questions to ask Attitudes and Beliefs: focus on the ways that people eval and think about issues Facts and Demographics: what people know! - demographic info necessary to describe sample - age, sex, marital status [no need to ask questions that have no real reason] Behaviours: focus on past behaviours or intended future behaviours Question Wording - issues with question wording: unfamiliar technical terms, vagues or imprecise, ungrammatical sentence structure, phrasing that overloads working memory, embedding the question with misleading info - things to consider when writing questions: o Simplicity: easily understandable and easy to respond to Avoid jargon and technical terms, sometimes must make it more complex to make it easier to understand o Double-Barreled Questions: avoid questions that ask 2 things at once o Loaded Questions: avoid questions that lead ppl to respond in one way question w/ emotionally charged words can lead to biased ans o Negative Wording: avoid phrasing questions w/ negatives o Yea-Sayinng and Nay-Saying: a tendency for a respondent to employ a response set to agree or disagree w/ all the questions true agreement vs agreeing w/ you [word qs so consistent agreement is unlikely] QUAID: Quesiton understating Aid Responses to Questions Closed vs Open-Ended Questions - closed-ended: a limited number of response alternatives are given o more structured approach [useful: variable dimensions well-defined] - open-ended: respondents are free to respond in any way they like o require time to categorize and code the responses = more costly o more useful when: need to know what they are thinking - can lead to different conclusions Number of Response Alternatives - closed-ended > fixed number of responses: sufficient [5-7 scale preferable] Rating Scales - how much judgments on any # of dimensions - Graphic Scale: requires a mark along a continuous 100-milimeter line that is anchored w/ descriptions at each end o ruler then placed on the line to obtain a scale that ranges form 0-100 - Semantic Differential Scale: a measure of the meaning of concepts that o rate any concept on a series of bipolar adjectives [7 pt scale] o virtually anything can be measured [people, ideas, things places] o rated along 3 basic dimensions: evaluation [good/bad], activity [active/passive], potency [weak/strong] - Nonverbal Scale for Children: young children may not understand previous o ie. point to the face tat shows how you feel about the toy Labeling Response Alternatives - more clearly define the meaning of each alternative - some perfectly balanced some not [disagree > agree or lower 50 > upper 5] - labeling alternatives is interesting for frequency of beh [H/L freq scales] Finalizing the Questionnaire Formatting the Questionnaire - should be attractive and professional - be consistent with scale formats [space out questions so not confusing] - sequence of questions: generally most interesting ones so ppl motivated to finish the questionnaire and demographic questions last - group questions that address a similar topic or theme Refining Quesitons - before actually administering give it to small group and have them think aloud while answering them [tell you how they interpret the quesitons] Administering Surveys - 2 ways ro administer: 1. written questionnaire 2. interview format Questionnaires - questions are presented in written format and the respondents write their answers [less costly than interviews, can be anonymous - but ppl must be able to read/comprehend the questions [boring motivatio] - can be given to groups = captive audience that is likely to complete it - also instructor is available to ask questions - mailed surveys = inexpensive but low response rates and no help if confused - internet survey = easy to design open or closed surveys, sent immediately to researchers when completed [samples are from search engines] o easy to obtain samples of people with specific characteristics - internet results are comparable to traditional methods [no extensive studies] - prob of ppl misrepresenting themselves online [age/gender/ethnicity] - for most research it is unlikely ppl will misrep more online than in person - other technologies: assist w/ the collection of data [sampling beh over a period of time easier w/ cell phone/pagers] - can contact ppl at get info on their current activities and emotional reactions o computerized sampling Interviews - more likely to agree to answer questions to a real person than in the mail - response rates higher, rapport w/ researcher helps respondent complete it - will leave less questions unanswered - questions and answers can be clarified by either party - interviewer bias: all of the biases that can arise from the fact that the interviewer is a unique human being interacting with another human - interviewer could subtly/inadvertently influence answers o interviewers expectations can also influence the ask they look for - face-to-face interviews: interviewer travels to the persons home/office or vice versa/expensive and time consuming [when sample size small and slear benefit of face to face interaction] - telephone interview: most large scale surveys are done this way [less expensive/data collection quick/may people can collect at same time/CATI Computer-assisted telephone interview] - focus group interviews: interview w/ a group of 6-10 ppl for 2/3 hrs [ppl are selected for their knowledge or interest in a topic/usually a monetary incentive to participate/questions are open and to whole group] o advantage of group interaction [interviewer must mediate hostility] o recorded and may be transcribed > analyzed to find areas of group consensus and disagreement [at least 2/3 groups so reflects populat] Survey Designs to Study Changes over time - panel study: the same people are surveyed at two or more points in time - two-wave: surveyed 2 points in time/3-wave there are 3 surveys - important when comparing the relationship between 1 variable at time one and another variable at some other time two [couples attitude similarity] Sampling form a Population - sampling: participants forma population of interest - population: composed of all individuals of interest to the researcher - studying entire population is unrealistic in most cases so we use samples to draw conclusions about the population - Confidence Intervals: survey are accurate w/i 3% using a 95% level of confidence = actual pop value is probably between 58-64% [not always %] - have a 95% confidence that the true population value lies w/i this interval around the obtained sample result - gives u info about the likely error aka sampling error - Sample Size: larger sample size reduces the confidence interval o the needed sample size does not change much as pop size increases o a sample size fo 150 will describe a pop of 1500 or 15 mil w/ pretty much the same degree of accuracy Sampling Techniques - 2 techniques for sampling individuals from a population
More Less

Related notes for PSYB01H3

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.