* CAPI 200 CFPS 8. 4% curbstoning post quasi-experiment design * 77004 Mick Couper curbstoning curbstoning 68
200 CFPS / Couper et al. 997 69
202. 2 200 CFPS CFPS CFPS CFPS 200 CFPS 2 70
I II 200 CFPS 2 2 Computer Assistant Personal Interviewing CAPI 2 7
202. 2 Baker et al. 995 Couper et al. 997 PSID 968 - HRS 992 - Add Health 994 - BHPS 99 - Understanding Society 2009 - SOEP 984 - CAPI CFPS CFPS 2 3 2 CFPS 6000 2 3 CFPS 2009 CAPI 200 4 5 200 9 5 5 72
200 9 5 438 84385 CFPS 50% 794 2 8. 8% 3 946 22522 8. 4% 946 65 38 52% 65 50% 3 0% 2 46 30% 9% 65 65 Ⅱ 5. 73. 08 3. 05 46. 75 946 20%. 02 40% 2. 04 60% 4. 07 70% 7. 09 80% 9. 3 90% 2. 4. 2 2 3 4965 4343 493 3208 8776 CFPS missing data 50% 73
202. 2 II 0 II F = 75. 66 p < 0. 00 2 II 6. 8%. 5% 2. 5 57 36 354. 068 08 630 048. 000 53 0 8563 38 946 22522 Binary Logistic Regression 3 2 2 65 35% 57 36 3% 2 6% - 2 % 5% 24 CFPS 438 3% % 34% 25% 7% 74
3 Binary Logistic Regression Ⅰ Ⅱ. 385 **. 39 **. 428 3. 740. 597. 489. 432. 630. 899 *. 94 **. 324 2. 495 /. 937 *. 09 **. 388 2. 770 8-24 25-34 -. 274 35 -. 398 0 = =. 034 0 = =. 329. 02 ***. 02 ***. 003. 02 -. 495 *** -. 42 ***. 304. 244-2LL 388. 375 39. 228 Nagelkerke R 2. 68. 57 64. 0 63. 0 * p < 0. 05 **p < 0. 0 ***p < 0. 00 95% 3. 740 / 2. 770 2. 495 CFPS 75
202. 2 CFPS 2 I t GLM repeated measures analysis of variance 2005 8-38 3 4 t = 6. 34 p < 0. 000 0. 255 0. 020 0. 098 0. 08 2 3 200 CFPS CFPS 76
4 I t 95% t -. 57. 266. 026. 06. 207 6. 34 07. 000 CFPS 5 I 8-24 25-34 35 I 6 Box's M = 2. 428 F =. 982 p > 0. 05 7 F = 28. p < 0. 000 F = 4. 53 p < 0. 05 77
202. 2 5 I Tests of Within-Subjects Effects Ⅲ F Eta. 326. 326 37. 620. 000. 260 -. 326. 326 37. 620. 000. 260 -. 326. 326 37. 620. 000. 260. 326. 326 37. 620. 000. 260 3. 77 07. 035-3. 77 07. 035-3. 77 07. 035 3. 77 07. 035 - Greenhouse - Geisser - Huynh - Feldt Lower - bound Sphericity Assumed - - 6 Ⅰ 8-24. 308. 222 43 25-34. 220. 205 52 35. 24. 82 3. 255. 22 08 8-24. 062. 68 43 25-34. 33. 22 52 35. 077. 24 3. 098. 88 08 78
7 I Tests of Within-Subjects Effects Ⅲ F Eta. 929. 929 28.. 000. 2 *. 300 2. 50 4. 53. 03. 079 3. 472 05. 033 - - 8-24 25 35 25-34 25-34 46. 2% 7. 7% 3. 5% 79
202. 2 25-34 8-24 CFPS CFPS 25-34 8-24 8. 8% 8. 4% 52% CAPI 200 CFPS 65 24 80
CAPI Groves 989 coverage non-response sampling error measurement or response error CAPI CAPI 2005 Baker Reginald P. Norman M. Bradburn & Robert A. Johnson 995 Computer - assisted Personal Interviewing An Experimental Evaluation of Data Quality and Costs. Journal of Official Statistics 4. Couper Mick P. Sue Ellen Hansen & Sally A. Sadosky 997 Evaluating Interviewer Performance in A CAPI Survey. In Lars Lyberg Paul Biemer Martin Collins Edith De Leeuw Cathryn Dippo Norbert Schwarz & Dennis Trewin eds. Survey Measurement and Process Quality. New York John Wiley and Sons. Groves Robert M. 989 Survey Errors and Survey Costs. New York John Wiley and Sons. 8
Abstract Based on data from An Evaluation Project on Urban Minimum Livinghood Guarantee MLG in China this paper adopts a comprehensive method of both quantitative and qualitative studies to explore welfare dependency in China. The descriptive analysis shows that the beneficiaries are generally satisfied with the MLG scheme and they are willing to work. Through Binary-Probit regression analysis this study finds that the beneficiaries' working willingness and decisions are mainly affected by factors such as working capability health conditions gender and age. Particularly respondents' dependency levels are highly related with their working capability. Qualitative study shows that MLG plays a primary role in securing recipients' lives and respondents do have a strong sense of seeking jobs and improving their lives. Given the empirical findings the authors conclude that there is no significant welfare dependency effect caused by the current urban MLG scheme. However from a forward-looking perspective the possibility of welfare dependency could by no means be neglected. A Study on the Quality Control of Social Survey Intervention in curbstolen behaviors and its effects Yan Jie Qiu Zeqi Ren Liying Ding Hua & Sun Yan 68 Abstract Based on a post quasi-experimental design this paper examines curb-stolen behaviors among field interviewers and the effects of behavior intervention using the data and paradata of CFPS Chinese Family Panel Studies in 200. The study shows that curb-stolen behaviors have directly affected the quality of survey data and their occurrences are significantly related with interviewers' occupations. These behaviors could be successfully detected by the CAPI Computed-Assisted Personal Interviewing technique and effectively reduced by timely intervention. The effects of intervention however vary with interviewers' ages but are indifferent to their gender education levels and occupations. An Ethnography without History Reflections on Malinowski and the successors Zhang Limei & Hu Hongbao 82 Abstract Malinowski is well known as the founder of modern ethnography. Nowadays ethnography in the strict sense still follows the basic guidelines proposed by Malinowski and the face-to-face interaction on the spot is still the core feature of ethnographic method despite the post-modern reflections of anthropologists' self-identity and professional practice and the successors' modifications of the functionalist anthropology. With increasing studies on civilized societies in anthropology documents have become an important complement for fieldwork. In the authors' view however there is only the use of historical documents in fieldwork but no so-called doing fieldwork in the past. Historical ethnography is just literature and its on- 245