Modélisation de la structure de la covariance sous Windows: Une analyse multitratsmultiméthodea utilisant Amos, Eqs, et Lisrel. Cet article étudie les trois grands logiciels d'analyse de la structure de la covariance sous Windows. Le point essentiel de la comparaison réside en une analyse d'une matrice 5x5 multitraits-multiméthodes. On montre ainsi que la possibilité de différents fonctions d'ajustement et de procédures bootstrap dans les programmes actuels d'analyse de la structure de la covariances permettent une analyse fine de données à problèmes.
The task of survey interviewers includes contacting target persons, gaining their cooperation, & using appropriate interview methods to obtain the most reliable data. However, interviewers are not equally successful in their task. Factors that might differentiate good & mediocre interviewers are examined in this article, focusing on the influence of the interviewers' contact methods on respondents' cooperation rate. Data from the German ALLBUS 2000 General Social Survey are used in multilevel analyses of both contact & cooperation rates to assess both interviewer & target person variables. Results support the hypothesis that interviewer contact methods do influence respondent cooperation, but that respondent variables play an important role in the selection of contact methods. Tables, Appendixes, References. J. Stanton
Explores response quality of children in surveys as related to the cognitive developmental stages of children. The abilities to interpret the question, retrieve information, & synthesize & report that information are assessed as an outcome of question structure & age variables. The study focused on partially labeled response options & vague quantifiers in contrast to completely labeled, clearly worded response options. The effects of cognitive processes, age, & question-&-answer options structure are evaluated. 1 Table, 2 Figures, 1 Appendix, 34 References. L. Collins
The authors of this article, who are also the guest editors for this issue on multilevel analysis, give an overview and brief history of multilevel analysis and present the following four research articles. Multilevel Analysis - Overview. Methods. History.
Household survey nonresponse is a matter of concern in many countries. In one of the first international trend analyses, de Leeuw and de Heer (2002) found that response rates declined over the years, and that countries differed in response rates and nonresponse trends. Their analyses cover longitudinal data on the Labour Force Survey from National Statistical Institutes for the period 1980 to 1997. We added a new data set, covering the period 1998 -2015, and analysed nonresponse data over time and countries. In these analyses we differentiated between voluntary and mandatory surveys. The trends visible in de Leeuw and de Heer (2002) continue with possibly a small deceleration in refusal rates.
This study applies ordinal confirmatory factor analysis for multiple groups to assess equivalence of scale, random errors and systematic (nonrandom) errors of attitudinal questions surveyed on rating scales under different survey modes (Face-to-Face [F2F], Telephone, Paper, and Web). Empirical findings from a large-scale experiment are presented. Consistent with theoretical expectations, interviewer- and self-administered surveys measured all assessed questions on systematically different scales, with different systematic bias, and with differing extents of random error. These measurement effects were absent when comparing Paper with Web or F2F with Telephone. It is concluded that modes impact primarily systematic measurement effects affecting multiple items equally. Interviewer- and self-administered modes should only be combined with great care in mixed-mode surveys that focus on attitudinal constructs. Combining Paper and Web or Telephone and F2F are the viable options. Thereby choosing the self-administered modes appears more efficient, because these modes exhibited higher indicator reliabilities (smaller random error) than the interviewer modes.
With the decrease of landline phones in the last decade, telephone survey methodologists face a new challenge to overcome coverage bias. In this study we investigate coverage error for telephone surveys in Europe over time and compare two situations: classical surveys that rely on landline only with surveys that also include mobile phones. We analyzed Eurobarometer data, which are collected by means of face-to-face interviews and contain information on ownership of landline and mobile phones. We show that for the period 2000-2009, time has a significant effect on both mobile phone penetration and coverage bias. In addition, the countries' development significantly affects the pace of these changes.
A multilevel problem concerns a population with a hierarchical structure. A sample from such a population can be described as a multistage sample. First, a sample of higher level units is drawn (e.g. schools or organizations), and next a sample of the sub‐units from the available units (e.g. pupils in schools or employees in organizations). In such samples, the individual observations are in general not completely independent. Multilevel analysis software accounts for this dependence and in recent years these programs have been widely accepted. Two problems that occur in the practice of multilevel modeling will be discussed. The first problem is the choice of the sample sizes at the different levels. What are sufficient sample sizes for accurate estimation? The second problem is the normality assumption of the level‐2 error distribution. When one wants to conduct tests of significance, the errors need to be normally distributed. What happens when this is not the case? In this paper, simulation studies are used to answer both questions. With respect to the first question, the results show that a small sample size at level two (meaning a sample of 50 or less) leads to biased estimates of the second‐level standard errors. The answer to the second question is that only the standard errors for the random effects at the second level are highly inaccurate if the distributional assumptions concerning the level‐2 errors are not fulfilled. Robust standard errors turn out to be more reliable than the asymptotic standard errors based on maximum likelihood.
The effectiveness of using a "not selling anything" introduction in encouraging people to participate in random sampling household telephone surveys is studied. Twenty-nine split run experiments conducted by multiple Dutch market research firms were performed to ascertain the efficacy of standard & nonsolicitation introductions & whether the latter introductions increased participant gain. The findings revealed that the use of the "not selling anything" introduction increased participant response rates by 2% regardless of the topic of the household telephone survey. Although the data analysis demonstrated that placement of the nonsolicitation technique at the beginning of an introduction provided the best participant response rate, it is revealed that such introduction methods worked equally well for general & special groups. The use of nonsolicitation introductions in carrying out household telephone interviews is also recommended since "not selling anything" introductions are easy & inexpensive to implement. 2 Tables, 31 References. J. W. Parker
