2005 53 1 57 81 c 2005 50 2004 10 22 2005 1 4 11 1 2 2 1 2 1. 1.1 30 106 8569 4 6 7; maeda@ism.ac.jp
58 53 1 2005 Dillman 1978 Total Design Method Mangione 1995 2004 2004 2001 2 60 1.2 2003 9 10 11 2004 3 1987 2001 2
59 1.3 3 4 2 4 2 5 2. 2.1 ML03 2003 9 11 2005 ML00 2001 1 2 2002 ML MaiL 03 00 1 1 1 3 ML03 ML00 1 ML03 2 ML03 2 ML03, ML00 40 KS11 ML03 2 B A B 2
60 53 1 2005 1. 4 ML00 1 ML00, ML03 ML03 KS11 KS11 1 KS11 2 ML00 ML00 ML03 2.2 1 KS11 K, M 11 2004 SB03 ML03, ML00 1 3 KS11 2 K M SB03 KS11 #o.oo 2004
61 3. 2 60 ML03 KS11 1 1 3 81 65 ML03 KS11 ML03 82.6 (= 24/29) KS11 85.7 (= 12/14) 40.0 =12/30 14.3 (= 2/14) 9.4 15.6 3.1 3.2 3.3 3.1 3.1.1 1 KS11 1 3 ML03, ML00 1 20, 2 20 4 2 10 1998 KS11 1 3 ML03, ML00 2 ML03 KS11 > 2 > 1> KS11 1 3 2 ML03 ML00 1 3 3.1.2 1-0 10 6 1 2 4 2 3 2 4 5 2 KS11 1 3 2
62 53 1 2005 1. 3 n =, 2 5 1 0 Wald 12 1 3 2
63 2. KS11 1 3 ML03 ML00 n = ML00 ML03 ML03 ML03 ML00 ML00 ML03 10 KS11 1 3 KS11 1 3 1. 2. ML00
64 53 1 2005 2. 4 1 3. 1 3 4. 5. 3.2 2 3.2.1 10 2004 5 non-contact 1 2 KS11
65 2. 2 11 7.2 KS11 1 3 7.6 8.3 14.0 11.8 20 1 3 28.3 1.1 ML00 0.9 ML03 1.3 ML00 1.5 ML03 3.2.2 KS11 1 3
66 53 1 2005 3. 2 KS11 1 3 2 3 70 11 12 1 1 3 3.2.3 ML00 91.7, ML03 82.9 2002 2004 6
67 4. ML03 3 1 4 ML03 4 6 2 3.3 6 10 2000
68 53 1 2005 2000 11 KS11 3.3.1 3 KS11 1 3 ML03 2004 3 KS11 ML00 error-bar 3. 4 n = /
69 2 KS11 1 3 1. 2. ML03 KS11 1 3 ML03 ML00 ML03 KS11 3.3.2 5 ML03 ML00 KS11 2002 2002 1 OM01 SB03 1. 2. 3. 5. 4
70 53 1 2005 1. 2. 3. 4. A B 4.1 ML03 A B B 4.2 4.3 4.1 2002 2001 4.1.1 6 1 1 2 AB 1 #4.11 3: A B 7 #4.5 A B 3: #4.11 #4.5 B A B #7.1 #2.1 2 B
71 6. #4.5 B 8 4.1.2 4 A B A B B 6 2 B
72 53 1 2005 6 2 2 #7.1 #7.2 2 3 5 A 1 2 #2.1 #9.14 B 4. KS11 ML03 4 K M
73 4.2 A 4 4 2 D.K. 1 2 2 4 4 ML03 KS11 KS11 K M K 25 79 M 21 72 D.K. 0 1 45 n =1, 500 3 2 3 K M M 1 KS11 M 5 KS11 ML03 SB03 3 14 50 1 2 KS11 SB03 SB03 ML03 SB03 KS11 ML03 1995a, 1995b 4.3 ML03 KS11 7 KS11 K M 5 KS11
74 53 1 2005 7 #2.7 ML03 2, 3 KS11 K #5.23 M #5.6 K #2.12c ML03 K #2.10 5. KS11 ML03 SB03 1 2
75 7. ML03 KS11 Sudman and Bradburn 1974 Schwarz and Sudman 1996, Sudman et al. 1996
76 53 1 2005 5. 5.1 2 1 A D.K. Mangione 1995 2 ML00, ML03 D.K. 2002 ML00 ML03 #7.19 3. 10.5 KS11 10.9 1 3, 7.7 2 2004 ML00 ML03 Mangione 1995 5.2 2 1. 2. 3.
77 4. 5. 5.3 2 1 4.3 1 A., D.K. A.1 D.K. D.K. ML03 9 ML00 D.K. ML03, ML00 +D.K. 3 2
78 53 1 2005 1 D.K. 2 D.K. 100 A.2 3 Poststratification Thompson 2002 Gelman and Carlin 2002 2, ML03 10 6 KS11 SB03 20 3 1 2 4 3 4 2 6 3 48 24 h h N h N = P N h W h = N h /N h ˆp h ˆp PST : ˆp PST = X h W h ˆp h. 2 1, 2 4 B. ML03 A B 2 AB Split-Ballot 1982 A B AB A B 33 5 2 χ 2 /df 33 0.88 Pearson χ 2 A B 2 1 #4.5 4.1
79 1 ML00 2001 ML03 3 00 2 KS11 3 4 2 5 2 6 2004 7 #9.6 4: ML03 A B ML03 45.5, KS11 20.9 8 NHK 1996, Schuman and Presser 1996 9 Dillman, D. A. 1978. Mail and Telephone Surveys: The Total Design Method, John Wiley & Sons, New York. Gelman, A. and Carlin, J. B. 2002. Poststratification and weighting adjustments, Survey Nonresponse eds. R. M. Groves, D. A. Dillman, J. L. Eltinge and R. J. A. Little, Wiley, New York. 2004 2001 1 50 109 126
80 53 1 2005 2002 2000 1 3 No. 14. 2005 2 2003 1 3 2000 5 48 1 147 178 2001 2000 No. 87 Mangione, T. W. 1995. Mail Surveys: Improving the Quality, Sage Publications, Thousand Oaks. T. W. 1999 NHK 1996 1995a 1 ESTRELA 1995 10 43 51 1995b 2 ESTRELA 1995 11 34 42 2004 11 2003 No. 92 Schuman, H. and Presser, S. 1996. Questions and Answers in Attitude Surveys, Sage Publications, Thousand Oaks. Schwarz, N. and Sudman, S. 1996. Answering Questions: Methodologies for Determining Cognitive and Communicative Processes in Survey Research, Jossey-Bass, San Francisco. Sudman, S. and Bradburn, N. M. 1974. Response Effects in Surveys, National Opinion Research Center, Chicago. Sudman, S., Bradburn, N. M. and Schwarz, N. 1996. Thinking about Answers, Jossey-Bass, San Fransisco. 1987 Thompson, S. K. 2002. Sampling, 2nd ed., Wiley, New York. 1982
Proceedings of the Institute of Statistical Mathematics Vol. 53, No. 1, 57 81 (2005) 81 A Study on the Characteristics of Mail Survey: A Comparison with Face-to-face Interviewing Tadahiko Maeda The Institute of Statistical Mathematics This paper discusses the basic characteristics of mail surveys in comparison with those of a different survey mode: face-to-face interviewing. We conducted a mail survey, to compare it with the eleventh nationwide survey of the Japanese national character study, which was carried out by face-to-face interviewing in almost the same survey period. Additional information from a mail survey carried out three years before, and from another omnibus survey by face-to-face interviewing were also considered. First, we scrutinized the response rates of different demographic groups in each survey, and compared them. The most distincitve result was the lowest response rate in the younger male group. This tendency was more or less common in the two survey modes. Different response rates among groups resulted in a bias of the achieved sample from the population value with regard to some basic demographic variables. While the sex and age distribution of the sample achieved in mail surveys was more similar to the population distribution than that achieved in the face-to-face mode, educational background seemed to be more biased toward higher education in the mail surveys. Second, we discuss the difference between the results of the two survey modes with regard to two groups of survey items. In the first group, items in the questionnair have neutral or vague response options. Based on the results of some experimental items in the questionnaire, it is shown that, as a feature of self-administratoin mode, neutral response options of mail questionnaire attracts by far the larger portion of respondents, and that this difference from the face-to-face interviewing cannot be decreased by merely dealing with the wording of the question. In the second group, though items do not have neutral options, we still found remarkable differences between some of the items in the two survey modes. We could not find the reasons for these differences by contemplating the item content. Based on these results, we point out some characteristics of mail surveys. Key words: Japanese National Character Survey, mail survey, face-to-face interviewing, response rate, mode effect, demographic variables.