Textbook Notes (368,316)
Canada (161,798)
Sociology (1,053)
SOCB05H3 (26)
Chapter 9

Chapter 9

8 Pages
115 Views
Unlock Document

Department
Sociology
Course
SOCB05H3
Professor
Katherine Stuart
Semester
Fall

Description
• Chapter.9 – Survey Research Topics Appropriate to Survey Research • Respondent: a person who provides data for analysis by responding to a survey questionnaire • “Push polls” – phony telemarketers pushing you to vote against opposing candidates Guidelines for asking questions • Sometimes asked by interviewer or written down to be answered Choose appropriate question forms • Strongly agree, agree, disagree etc…using such text gives you more flexibility and interesting (Rensis Likert/Likert Scale) Open-Ended and Closed-Ended Questions • Open-ended: the respondent gives his own answer to the question • Closed-ended: the respondent has to choose from given options • both have their drawbacks (open-ended hard to analyze, closed-ended not enough options) Structuring Questions • Make them Clear • Avoid Double-Barreled Questions (do not use the word “AND”) • Respondents must be competent to answer (people don’t remember what they wore last week) • Respondent must be willing to answer (some people are not comfortable telling their salary) • Questions should be relevant (only if people have insight can they respond to given questions) • Short Items are best (should be able to read it quick, understand it and answer it) • Avoid Negative Items (make sure you don’t use “NOT”) • Avoid Biased Items and Terms (“Don’t you agree…..”?) Questionnaire Construction General Questionnaire Format • Spread out questions so people are not forced to write in tiny spaces or have no room to answer Formats for Respondents • Make boxes that they can check off, keep it professional if survey is on the computer ex. • Use numbers on the sides for respondent choice 1. Yes 2. No 3. Don’t know Contingency Questions • Survey applies ONLY to people who answer “yes” to first questions assuring their involvement in survey (cannot ask me how they feel during pregnancy) • If “yes” comes more than once it is known as a complex contingency table Ex. Are you a woman? If yes, did u ever take birth control pills? If yes, did it impact health? Matrix Questions • Also known as “Likert Response” benefits: space efficient, easier to answer, quickly review answers Disadvantages: might have to structure question to fit format, may agree with everything (response-set/ may think all question are about liberal view and miss-read questions in hurry) Ordering Items in a Questionnaire • It is important to ask open-ended questions first, otherwise they will make their answer consistent to previous answers • If it’s not continuous they switch their attention from one topic to another • Begin questionnaire with most interesting set of items • Try to start off with demographic data (age, gender etc…) in a face-to- face situation to have a formal approach BUT avoid in a written survey otherwise it may seem dull and boring Questionnaire Instructions • Clear instructions/ introductory comments ex. “In this portion we would like to know….” • Rank-ordering only used when no way of getting data; (1, 2, 3, 4, 5…) Pretesting the Questionnaire • Give it to a diverse set of 10 people before full distribution (look for problems/difficulties etc…) • Make sure the options are easy for data entry Self-Administered Questionnaires • Surveys administered through telephone, face-to-face and by mail • Other techniques used are door to door/mailed and then picked up/ dropped off and picked up Mail Distribution and Return • Send a questionnaire with instructions and a stamped envelope for return • Some people think it’s’ trouble keeping the envelope or might lose it so researchers construct surveys in such a way where they can be sent as is when folded, such methods improve study • You pay for peoples stamps whether they reply or not (cheapest way) next to business-reply permit (pay for those that are used and a nickel surcharge) Monitoring Returns • Record varying rate of return on a return rate graph which includes days and # of graphs returned • Respondent dates can show biases (if GPA asked, and kids with lower GPA taking longer to respond) Follow-up Mailings • Send a letter for encouragement to participate in survey, with a new cover letter and survey in case they lost/misplaced the original one therefore having no response initially • An original and 2 follow-ups in the most efficient method to increase response rates • Respondents get a postcard attached which is described in the cover letter describing anonymity, after sending in the survey they send the post-card, this way they do not get follow-up after they respond Acceptable Response Rates • Response rate: # of people participating divided by # selected in the sample • 50% response rate is adequate for analysis and reporting, 60% Is good, 70% very good • Some people enclose money in order to get people to respond • Telep
More Less

Related notes for SOCB05H3

Log In


OR

Join OneClass

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

Sign up

Join to view


OR

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.


Submit