Textbook Notes (362,768)
Canada (158,052)
SOAN 2120 (129)
D Walters (55)
Chapter 6

SOAN chapter 6, 7, 9.docx

11 Pages
Unlock Document

University of Guelph
Sociology and Anthropology
SOAN 2120
D Walters

SOAN 2120- Introductory Methods Chapter Six – Survey Research - Survey is most widely used data-gathering technique used in sociology, developed within a positivist approach Research Questions Appropriate for a Survey 1. Behaviour (how frequently do you brush your teeth? Did you vote?) 2. Attitudes/Beliefs/Opinions (do you think mayor is doing good job?) 3. Characteristics (age, marital status, do you belong to a union?) 4. Expectations (Do you think pop’n of town will grow, shrink, stay same) 5. Self-Classification (Consider self liberal? Social Class? Highly religious) 6. Knowledge (Who was elected mayor in last election?) - Survey responses have very little connection to the actual behaviours exhibited by employers who said they would hire criminal offenders  An association among variables is measured with statistical techniques; researchers think of alternative explanations, measure variables that represent alternative explanations, then examine effects to rule out alternatives Principles of Good Question Writing  Keep it clear, keep it simple, keep respondent’s perspective in mind 1. Avoid jargon, slang and abbreviations (NATO has may meanings) 2. Avoid Ambiguity, confusion and vagueness  “What is your income” – weekly, monthly, annually?, before or after taxes, this year or last year, salary or from all sources. 3. Avoid emotional language  “What do you think about a policy to pay murderous terrorists who steal the freedoms of peace-loving people” 4. Avoid prestige bias  Titles in society (president, expert) carry prestige or status, people will answer if they want to agree/disagree with what President or doctors say 5. Avoid double-barreled questions (2 or more questions joined together)  Does this company have pension and health-insurance benefits? 6. Do not confuse beliefs with reality  “Do you rate a teacher higher if they tell many jokes”- implies that the students rating of the teacher includes joke telling 7. Avoid leading questions  “You don’t smoke, do you?” 8. Avoid asking questions that are beyond respondents’ capabilities  “How did you feel about your brother when you were 6?” 9. Avoid False Premises  “The post office is open too many hours. Do you want it to open four hours later or close four hours earlier each day?” 10. Avoid asking about intentions in the distant future 11. Avoid double negatives  “students should not be required to take a comprehensive exam to graduate”- disagree with a statement is a double negative 12. Avoid overlapping or unbalanced response categories  Mutually exclusive (numerical 5-10,-10-20 can be fixed 5-9, 10-19)  Exhaustive, every respondent has a choice (working/unemployed?)  Balanced (What kind of job is mayor doing? Good, excellent, satisfactory, outstanding) Threatening Questions: presentation of self & ego protection from questions about sexual behaviour, drug/alcohol use, mental health, deviant behaviour  Develop trust w/ respondent and phrase the question in an enhanced way “In past surveys may men have reported that in some point in live gay experience…” Socially Desirable Questions: distort answers to make reports conform to social norms (reporting donating to charities when really haven’t) Knowledge Questions: majority of the public cannot correctly answer elementary questions or identify important political documents  “How much, if anything, have you heard about…” Skip/Contingency Questions: gives ability to avoid asking questions that are irrelevant for a respondent. Nonattitudes and the Middle Positions (not sure, don’t know, no opinion) Standard-format question does not offer “don’t know” Quasi-filter question does offer “don’t know” alternative Full-Filter question first asks if respondent has an opinion, then asks for it Wording Issues  More people are willing to “not allow” something than to “forbid” it  Less educated people most influenced by minor wording differences Questionnaire Design Issues 1. Length of Survey/Questionnaire  Longer = cost effective, but responses drop significantly 2. Question Order or Sequence  Organization of Questionnaire (pleasant, not boring, say thank you)  Order effects: the order questions are presented may influence answers (strongest for people who lack strong views, less educated and older respondents) First question asks if abortion is okay if woman is unmarried and fetus has serious birth defects  Context effects: difference in topic order (talking about crime before drunk driving) o Funnel sequence: general questions before specific 3. Format and Layout  Questionnaire Layout and Question format (horizontal, matrix) 4. Nonresponse  People refuse to participate in surveys because they feel there are too many, fear of crime and strangers, a more hectic lifestyle, loss of privacy & rising distrust of authority or gov’t  Misuse of surveys to sell products or persuade, poorly designed, inadequate explanations also increase refusals  Decrease refusals: send letters in advance, offering reschedule, small incentives/gifts, expressing sincerity, explaining Mail and Self-Administered Questionnaires  ADVANTAGES: cheapest, conducted by a single researcher, can be sent to wide geographical area, convenient, anonymity and avoid interviewer bias  DISADVANTAGES: low response rate, reminder letters add to time/cost, no one to clarify questions or probe more info, someone else might complete it, serious errors can go undetected, not suited for the illiterate Web Surveys  ADVANTAGES: fast, cheap, visual, audio, video, flexible design  DISADVANTAGES: lack of coverage: (older, poor, less-educated less likely to have good internet access, multiple email addresses), privacy, verification that survey done only once per individual, compatibility of various web software Face-To-Face Interviews  ADVANTAGES: highest response rate, longest, observe non-verbal communication, use visual aids, complex questions, extensive probes  DISADVANTAGES: high cost, training of interviewer, interviewer bias, tone of voice/appearance of interviewer may affect response Interviewer  Not always paid well or provided regular employment  Must be experienced, able to talk to all demographics  Interviewer bias: answers may differ if others are around, race/gender of the interviewer may influence respondent answer Computer-Assisted Telephone Interviewing (CATI)  Fast, reduces errors, if you put in an invalid code computer will request another answer Ethical Issues Privacy: people will answer when it is asked in a comfortable context w/ mutual trust, when they believe serious answers are needed for research, and when they believe answers will remain confidential Voluntary participation, exploitation of surveys and pseudosurveys: suggle “sell in the guise of a survey”, sloppy reporting of survey results permits abuse SOAN 2120 – Introductory Methods Chapter Seven- Experimental Research Experiment 1. Begin with a hypothesis 2. Modify something in a situation 3. Compare outcomes with and without the modification - Experimental research is the strongest for testing causal relationships because 3 conditions for causality (temporal order, association and no alternative explanations are best met in experimental designs Research Questions Appropriate for an Experiment  Experiment is usually limited to questions in which a researcher is able to manipulate conditions  Cannot answer “do people who complete college increase their annual income more than people who do not” because you cannot randomly assign people to attend college and prevent others  Experiments best suited for micro-level studies (individual or small group phenomena) Random Assignment: the method for assigning cases (individuals, organizations) to groups for purpose of making comparisons  Groups must be similar except in terms of variable isolated (studying people who took a specific course with those who  Random means a case has an exactly equal chance of ending up in one or the other group  Random samples aren’t very common, most experimenters begin with a convenience sample then randomly assign  Matching is an alternative, but presents a problem: what are the relevant characteristics to match on, can one locate exact matches? Experimental Design Logic Parts of the experiment 1. Treatment/Independent Variable : stimulus/manipulation, what the researcher modifies 2. Dependent Variable: outcomes in experimental research are the physical conditions, social behaviours, attitudes, feelings, beliefs of subjects that change in response to treatment. 3. Pretest: the measurement of the dependent variable prior to introduction of the treatment. 4. Posttest: the measurement of the dependent variable after the treatment has been introduced into the experimental situation 5. Experimental Group: the group that receives the treatment 6. Control Group: the group that does not receive the treatment. 7. Random Assignment Control in experiments  A researcher wants to control all aspects of the experimental situation to isolate the effects of the treatment and eliminate alternative explanations  Use deception to control setting: occurs when the researcher intentionally misleads subjects, use confederates or stooges- people who pretend to be other subjects but actually work with the researcher Types of Design  All designs are variations of the classical experimental design: has random assignment, pretest, posttest, experimental group and a control group Preexperimental Designs: lack random assignment and are compromises or shortcuts, make inferring a causal relationship more difficult  One-Shot Case Study Design (one-group posttest-only): has only one group, a treatment and a posttest. Because there is only one group, no random assignment.  One-group Pretest-Posttest Design: one group, a pretest, a treatment and a post-test. It lacks a control group and random assignment.  Static group comparison (posttest-only nonequivalent group design): has 2 groups, a posttest and treatment. Lacks random assignment and a pretest. Quasi-Experime
More Less

Related notes for SOAN 2120

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.