| 研究生: |
陳綉真 Chen, Zhen |
|---|---|
| 論文名稱: |
隨機應答取樣之研究 |
| 指導教授: |
李隆安
Li, Long-An |
| 學位類別: |
碩士
Master |
| 系所名稱: |
理學院 - 應用數學系 Department of Mathematical Sciences |
| 論文出版年: | 1993 |
| 畢業學年度: | 81 |
| 語文別: | 中文 |
| 論文頁數: | 64 |
| 中文關鍵詞: | 隨機應答 、參數設計 、數學 、不相關問題模式 、應用數學 |
| 相關次數: | 點閱:219 下載:0 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
在隨機應答的技巧之中,自華納(Warner,1965) 提出華納模式後,即有多位學者以華納的觀點為籃圖,展開一系列的縱橫發展,並插枝分葉出各式技巧來(Chandhuri & Makerjee,1988)。
本文在第一、二章內,將分別介紹其中最典型的兩種方式:華納模式(Warner Model)與不相關問題模式(Unrelated-Question Model)。在第一章中並圖示固定的參數設計p 之下,比較華納模式的隨機應答方式與一般的直接應答方式效率之優劣,隨受訪者的誠實度變化之情形。在第二章中,除介紹不相關問題模式的技巧與方法外,也圖示比較華納模式與不相關問題模式取樣結果的估計值有效性之優劣。在第三章中,則闡述保護受訪者隱私的立場與獲得有效估計的目的如何互相衝突、矛盾,而定下參數設計的準則,以求取二者之間的平衡。最後一章則因Greenberg(1977) 所定義的冒險函數未能由所給的數學式完全描述,故在此章提出一些較合理的修正。
自序
謝辭
摘要﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒vi
專有名詞暨簡記符號﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒vii
第零章 導論﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒1
一、 引言﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒1
二、 本文架構﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒2
第一章 華納模式﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒4
1-1:華納模式的技巧與方法﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒4
1-2:p值大小隊估量的影響﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒6
1-3:”誠實回答”假設的修正﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒8
1-4:非MLE.的爭論﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒10
1-5:DR與RR的比較﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒12
1-6:討論﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒ 16
一、兼具MLE.及UE.的估量﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒16
二、樣本置回的考慮﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒17
第二章 不相關命題模式﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒18
2-1: πY為已知的技巧與方法﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒19
2-2: πY為未知的技巧與方法﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒20
2-3: 最佳參數的設計原則﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒21
一、最佳n1n2設計﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒ ﹒﹒﹒22
二、最佳πY設計﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒23
三、最佳p1設計﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒24
四、最佳p2設計﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒24
2-4:華納模式與UQM.的比較﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒25
2-5:兩組UQM.的設計﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒31
2-6:討論﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒33
一、設計πY為已知的方法﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒33
二、有關πY的最佳設計﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒33
第三章 風險函數和有效估計﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒34
3-1: 危及函數﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒35
3-2: 洩漏函數﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒41
一、Anderson觀點﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒41
二、Lanke觀點﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒43
三、Flinger 觀點﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒44
3-3:冒險函數﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒45
第四章 新的冒險函數﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒48
4-1:定值常數的猜測值﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒50
一、”是”的情形﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒50
二、”不是”的情形﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒52
4-2:與洩漏風險有關猜測值﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒54
一、”是”的情形﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒54
二、”不是”的情形﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒56
4-3:與敏感性族群大小有關的猜測值﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒57
一、”是”的情形﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒57
二、”不是”的情形﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒﹒59
[1] 楊文山 (1993)隨機應答在大型實地調查中的應用─『以估計台灣地區賄選為例』,台灣地區社會意向調查資料運用學術研討會;中央研究院中山人文社會科學研究所。
[2] 楊文山與伊慶春 (1992) 『台灣社會意向調查』資料檔簡介,科學發展月刊,第二十卷第二期,頁144-148。
[3] Anderson ,H. (1975a) “Efficiency versus protection in the RR designs for estimating proportions.” Tech. Rep. 9, University of Lund, Sweden.
[4] Aderson ,H. (1975b) “Efficiency versus protection in a general RR model for estimation proportions.” Tech. Rep. 10, University of Lund, Sweden.
[5] Anderson ,H. (1975c) “Effciency versus protection in RR designs.” Mimeo notes, University of Lund , Sweden.
[6] Bourke, P.D. , and Dalenius , T. (1974) “RR models with lying.” Tech. Rep. 71, Institute of Statistics, University of Stockholm, Sweden.
[7] Brewer,K. R. W. (1981) “Estimating marijuana asage using RR—Some paradoxical findings.” Australian Journal of Statistics. 23, 139-48.
[8] Chaudhuri,A. , and Rahal , M. (1988)”Randomized Response: The-ory and Techniques.” New York : Marcel Dekker Inc.
[9] Devore , J. L. (1977) ”A note on the RR techniques.” Common Statist. –Theory Methods 6, 1525-1529.
[10] Dowling ,T.A. , and Shachtman , R. (1975) “On the relative effciency or RR models.” J. Amer. Statist. Assoc. 70, 84-87.
[11] Duffy , J. C. , and Waterton ,J. J. (1988) “RR vs. Direct question-ing : estimating the prevalence of alcohol reated problems in a field survey.” Australian Journal of Statistics. 30, 1-14
[12] Flinger,M. A. , Policello, G. E. ,and Singh, J. (1977) “A comparison of two RR survey methods with considerson for the level of respondent protection.” Common. Statist. – Theory Methods 6, 1511-1526.
[13] Folsom , R. E. Greenberg ,B. G. ,Horvitz ,D. G. And Abernathy,J. R. (1973) “The two alternate questions RR model for human surveys.” J. Amer. Statist. Assoc. 68, 525-530.
[14] Fox , J. A. , and Tracy , P. E. (1986) ”RR : A method for sensitive surveys.” Sage University Paper Series on Quantitative Applications in the Social Science 07-058 ,Beverly Hill , CA:Sage.
[15] Greenberg ,B. G. , Abul-Ela ,Abdel-Latif,A. , Simmons , W. R. , and Horvitz,D. G. (1969) “The unrelated uestion RR Model : Theoretical framework.” J. Amer. Statist. Assoc. 64, 520-539.
[16] Greenberg ,B. G. ,Kubler , R. R. , Abernathy,J. R.,and Horvitz ,D. G. (1977) “Respondent hazards in the unrelated question RR Model.” J. Statist. Plann. Inference 1, 53-60.
[17] Horvitz ,D. G. , Shah ,B. V. , and Simmons ,W. R. (1967) “The unrelated question RR Model. “Proccedings of the Social Statistics Section, American Statistical Association, 65-72.
[18] Horvitz ,D. G. ,Greenberg ,B. G. ,and Abernathy,.J. R. (1976) “RR: adata gathering device for sensitive uestions.” Internat. Statist. Reu. 44, 181-196.
[19] I-Cheng,C. , Chow ,L. P. , and Rider , R. V. (1972) “The RR technique as used in Taiwan outcome of pregency study.” Studies in Family Planning. 3 265-269.
[20] Kim, Jong-Ik , and Flueck , J. A. (1978a) “Modifications of the RRT for Sampling without replacement.” Proc. ASA. Sec. Suru. Res. Methods , 346-350.
[21] Krotki ,K. J. ,and Fox ,B. (1974) “The RR technique ,the interview and the self-administrated questionnaire. An empirical comparison of fertility reporting.” Proccedings of the Social Statistics Section , American Statistical Association, 357-371.
[22] Lanke, J. (1975) “On the choice of unrelated question in Simmons’version of RR.” J. Amer. Statist. Assoc. 70, 80-83.
[23] Lanke, J. (1976) “On the degree of protection in randomized inter-views.” Internat. Statist. Reu. 44, 197-203.
[24] Leysieffer, R. W. ,and Warner , S. L. (1976) “Respondent jeopardy and optimal designs in RR models.” J. Amer. Statist. Assoc. 71, 649-656.
[25] Moors , J. J. A. (1971) “Optimization of the unrelated question RR model.” J. Amer. Statist. Assoc. 66, 627-629.
[26] Shimizu , I. M. , and Bonham , G. S. (1978) “RR technique in anational survey.” J. Amer. Statist. Assoc. 73, 35-39.
[27] Tracy , P.E. , and Fox , J. A. (1981) “The validity of RR for sensitive measurement.” American Sociologiacal Reuiew. 46, 187-200.
[28] Umesh , U. N. , and Peterson R. A. (1991) “A critical evaluation of the RR method, applications , validation ,and research Agenda.” Sociological Methods and Research. Bd 20 ,104-138.
[29] Warner , S. L. (1965) “RR: a survey technique for elimination evasive answer bias.” J. Amer. Statist. Assoc. 60. 63-69.
(限達賢圖書館四樓資訊教室A單機使用)