QUESTIONNAIRE DESIGN Zikmund, W. G

Presentation on theme: "QUESTIONNAIRE DESIGN Zikmund, W. G"— Presentation transcript:

1 QUESTIONNAIRE DESIGN Zikmund, W. G
報告學生： 丁詩芸 指導老師： 任維廉 教授 2019/4/12

2 自我介紹 姓名：丁詩芸 年級：運管系大四 興趣：打排球、跟家裡狗狗玩 2019/4/12

3 Outline What should be asked? Phrasing questions.
The art of asking questions. Question sequence. What is the best layout? Pretest. When it comes to global research. 2019/4/12

4 1. What should be asked ? A survey is only as good as the questions it asks. 2019/4/12

5 Key criteria Questionnaire Relevancy Questionnaire Accuracy
no unnecessary information all needed information is obtained Questionnaire Accuracy reliable and valid use simple, conversational words may help 2019/4/12

6 2. Phrasing Questions Open-Ended Response
Permit free responses What things do you like most about your job? What names of local banks can you think of offhand? Fixed-Alternative Questions choose the one closest alternative Simple-dichotomy question Determinant-choice question Frequency-determination question Attitude rating scales Checklist question 2019/4/12

7 When conducting exploratory research.
Advantage Disadvantage Open-Ended Response When conducting exploratory research. Learn more words people spontaneously give Find unanticipated reaction Costly Take much time Difficult to categorize Interviewer bias Fixed-Alternative Questions Cheap Less interviewer skill Take less time Easier to answer Easy to coding Researcher bias No adequate answer 2019/4/12

8 Phrasing questions for different surveys
Mail and telephone questions must be less complex. Telephone and personal interviews should be written in a “conversational” manner. 2019/4/12

9 2019/4/12 Personal interview Telephone interview
Self-administered survey Advantages 易獲得合作 可伺機調整問題發掘更多資訊 可利用視覺輔助工具 可事先篩選不適合的respondent 較人訪低成本 較不受地理條件限制 較節省人力 降低人訪誤差比率 快速完成調查 可利用電腦隨機(或近乎隨機)抽樣 後續追蹤程序簡便 有較佳的機會接近較高位階人士 可利用獎品鼓勵參與 成本低廉 可匿名完成問卷 容許充裕時間完成問卷 Disadvantages 高成本 訪員須事先受過良好訓練 完成時間長 後續追蹤號時費力 並非所有respondent皆有辦法觸及 有些受訪者不願與陌生人談話 易造成訪員誤差 較人訪的回收率低 訪問時間受限 無法訪問無電話者或不列電話號碼者 訪問的完整性較人訪低 無法用視覺輔助工具 低回收率 無法適度挖掘資訊 問卷長度受限 須正確郵寄地址 對議題有興趣者回答比例高 2019/4/12

10 How satisfied are you with your community. 1. Very satisfied 2
How satisfied are you with your community? 1. Very satisfied 2. Quite satisfied 3. Somewhat satisfied 4. Slightly satisfied 5. Neither satisfied nor dissatisfied 6. Slightly dissatisfied 7. Somewhat dissatisfied 8. Quite dissatisfied 9. Very Dissatisfied (for mail interviews) 1. Very satisfied 2. Somewhat satisfied 3. Neither satisfied nor dissatisfied 4. Somewhat dissatisfied 5. Very Dissatisfied (for telephone interviews) 2019/4/12

11 3. The art of asking questions.
Avoid complexity Not all people have the vocabulary as college students. Avoid jargon Avoid leading and loaded questions Leading questions- A question that suggest or implies certain answers. Split-ballot technique- Two alternative phrasings of the same questions would yield more accuracy. Loaded questions- A question that suggests social-desirability answers or is emotionally charged. ? ? ? 2019/4/12

12 Avoid double-barreled items Avoid making assumptions
Avoid ambiguity Be as specific as possible Avoid double-barreled items Contains two issues at once Avoid making assumptions Express attitudes on subjects which people do not ordinarily think about is meaningless. Avoid burdensome questions Take too much effort from respondents 2019/4/12

13 Mutually exclusive Under $15,000 ___ One more category
$15,000-$30,000___ $30,000-$55,000___ $55,000-$70,000___ Over $70,000___ One more category Bias could be caused by people who do not want to be classified in an extreme category . People with incomes of $30,000 will be in? 2019/4/12

14 4. Question sequence Opening questions To be logical
Interesting, easy to answer, build confidence To be logical Embarrassing questions at the middle or end. Asking at the beginning may irritate respondents affect willingness finish the survey. Funnel technique General questions before specific questions Filter question When someone is not qualified to answer 2019/4/12

15 5. What is the best layout? Appearance and lay out can
influence cooperation and data quality. Never be over crowded Openers 2019/4/12

16 6. Pretest 7. When it comes to global research.
The first pretest should be conducted using personal interviews Select a group that is not too divergent from the actual respondents. Provide sampling procedure and the return rate 7. When it comes to global research. Back translation 2019/4/12

17 Thanks for your listening!
Q & A 2019/4/12