Chapter 7: surveys

Post on 23-Feb-2016

40 views 2 download

description

Chapter 7: surveys. Survey Research: The Hite Report. 84% of respondents are not satisfied emotionally with their relationships 95% report "emotional and psychological harassment" from their men 98% desire more communication - PowerPoint PPT Presentation

Transcript of Chapter 7: surveys

CHAPTER 7: SURVEYS

SURVEY RESEARCH: THE HITE REPORT

84% of respondents are not satisfied emotionally with their relationships

95% report "emotional and psychological harassment" from their men

98% desire more communication

13% of women married more than two years still "in love“ with husbands

THE HITE REPORT

1972 behavioral researcher, Shere Hite, conducted large study of female sexuality

Sample size = 4,500 womenQuestionnaires distributed through many

women's groups and other sourcesNOWMs. MagazineVillage VoiceChurch groupsPolitical organizations

Survey--127 essay questions

THE STUDY

100,000 surveys distributed

4,500 returnedResponse rate= 4.5 %

Response rate of 70 to 80% to generalize to population

Motivation of respondents?

What about the 95.5% who did not respond?

PROBLEMS

SURVEY RESEARCH

Structured way to collect standardized information from individuals using a questionnaire.

Conducted once or at repeated intervals Concurrently with multiple samples From few people or many people

WHAT DO YOU THINK? ANSWER YES OR NO TO

EACH OF THE FOLLOWING1. A survey is always appropriate 2. Surveys are one of the most

popular ways to collect information3. An email or online survey is better

than the old mail or telephone surveys

4. Careful planning is necessary 5. Advance notice to potential

respondents helps increase response rate

6. A low response rate increases the likelihood of biased results

CHECK YOUR ANSWERS

1. A survey is always appropriate - NO2. Surveys are one of the most

popular ways to collect information - YES

3. An email or online survey is better than the old mail or telephone surveys – NO, not necessarily

4. Careful planning is necessary - YES5. Advance notice to potential

respondents helps increase response rate - YES

6. A low response rate increases the likelihood of biased results - YES

To collect information from individuals (vs. a group or collective)

Want standardized information

Respondents can read and write

1. SURVEYS ARE APPROPRIATE…

Want information from many people

Privacy is important or independent opinions and responses are needed

Have resources to send, track, analyze and interpret questionnaires

SURVEYS ARE USED WHEN…

PROS AND CONS OF SURVEY

PROSCollect information from many people

Anonymous

Standardized

Easy tabulation

CONSResults easily biased

Miss important information

Question and answer choices predetermined

Literacy skills

Literacy levelTradition of reading, writing SettingTranslationSequence of questionsPretesting questionnaire (Intrusive?)Computer access and use of electronic survey

CULTURALLY APPROPRIATE?

TYPES OF SURVEYS

1. Hand-out2. Mail3. Telephone4. Face-to-face 5. Email6. Web survey – Online survey7. Mixed mode: uses two or more of

above

Use mix of modes to ensure everyone can and does respond

Proportion of people who respond:

Example: If you distribute 50 questionnaires and get back 25, your response rate is 50%.

RESPONSE RATE

# that answered = response rate# you contacted

LOW RESPONSE RATE ?

Determine how respondents differ from non-respondentsDescribe results in terms of who did respond. Don’t imply that results apply to anyone other than those who responded.

2/27 INCREASING RESPONSE RATE

Survey topic is interesting to respondents (saliency)

Personalize communications related to survey

KISS: Keep It Short and SimpleFollow-up Trust, respect, like researchers

TYPE OF SURVEY TO CHOOSE?

“It depends”… What do you want to know How complex or sensitive is information

Characteristics of respondents

Time lineResources

You determine survey is best and most appropriate way to collect information

Take time to plan the survey

TYPE OF SURVEY TO CHOOSE?

1. Who to involve in conducting survey - - engage them

2. What information to collect Use of information?

3. Identify respondents Sampling strategy

4. Select survey distribution: telephone, mail, hand-out, email, web-based

5. Data analysis – what will product/final report include

PLANNING A SURVEY

SURVEY PLANNING CONTINUED…

6. Develop questionnaire

7. Pilot test questionnaire 8. Communication strategy to get

support for survey9. Budget, timeline, and

management process Available resources?

•Questions must be:•Screened•Tested•Revised

•Until researcher has confidence they will be clear to respondents

 

Constructing Survey Questions

• Closed-ended or fixed choice questions offer respondents explicit responses from which to choose

•  • Responses must be exhaustive (all respondents can find an appropriate response)

•  • Responses must be exclusive (all respondents will find only one appropriate response)

•  • The exceptions to these rules are “Check all that apply” question

• When in doubt, use “Other, please explain

SINGLE QUESTIONS

OPEN-ENDED QUESTIONS

•Open-ended questions lack response choices•Allow respondents to fill in own answers

•Preferable when full range of responses cannot be anticipated•Allows clear answers when questions involve complex concepts

 

INDEXES AND SCALES

• Index: Composite measure based on sum or average of responses of several questions that measure same concept•  Many pre-existing indexes exist•Such as the CES-D, commitment, dependency• These have been pretested and promote comparability

*** Indexes are usually calculated by summing or averaging responses with each question counting equally

INDEX: EXAMPLE

• Four questions to measure job-related depression, each with the response choices of "yes" or "no":

1. "When I think about myself and my job, I feel downhearted and blue."2. "When I’m at work, I often get tired for no reason."3. "When I’m at work, I often find myself restless and can’t keep still."4. "When at work, I am more irritable than usual."

SCALES

Indicates differences in intensity among the indicators of a variable.

In a hierarchy in which one answer effectively indicates answers on other questions, this is a scale.

The most commonly used scale is the Likert Scale.

Bogardus Social Distance ScaleA technique for measuring willingness of

people to participate in social relations with other kinds of people.

Interested in the extent to which U.S. Christians are willing to associate with, say, Muslims.

1. Are you willing to live in the same country as Muslims? 2. Are you willing to live in the same community as Muslims? 3. Are you willing to live in the same neighborhood as Muslims? 4. Are you willing to live next door to a Muslim? 5. Are you willing to let your child marry a Muslim?