Search

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.

Non-réponse versus erreur de mesure – Doit-on insister auprès des répondants réticents ? : Pour augmenter les taux de réponse, les enquêteurs intensifient leurs efforts pour amener les personnes échantillonnées parmi les répondants. La question est de savoir si les répondants « réticents » fournissent des réponses de qualité moindre que les répondants « impatients ». Nous définissons les répondants impatients comme des personnes qui répondent au premier envoi d'une enquête postale, et les répondants réticents comme des personnes qui répondent plus tardivement. Nous avons utilisé une analyse multitrait-multiméthode (MTMM) qui permet la séparation statistique de la variance de fond ou de trait, la variance de méthode et la variance d'erreur. Les résultats montrent que la structure de mesure ne diffère pas entre les répondants impatients et réticents. Il n'y avait également aucune différence systématique dans les estimations de la fiabilité et la validité des deux groupes. To increase response rates, survey researchers intensify their efforts to bring sampled persons into the respondent pool. The question is whether or not “reluctant” survey respondents provide answers of lower quality than “eager” respondents. We define eager respondents as persons who respond to the first round of a mail survey, and reluctant respondents as persons who respond in later rounds. We used a multitrait-multimethod (MTMM) design, which allows statistical separation of substantive or trait variance, method variance, and error variance. The results show that the measurement structure does not differ between eager and reluctant respondents. There was also no systematic difference in the reliability and validity estimates for both groups.

To increase response rates, survey researchers intensify their efforts to bring sampled persons into the respondent pool. The question is whether or not “reluctant” survey respondents provide answers of lower quality than “eager” respondents. We define eager respondents as persons who respond to the first round of a mail survey, and reluctant respondents as persons who respond in later rounds. We used a multitrait-multimethod (MTMM) design, which allows statistical separation of substantive or trait variance, method variance, and error variance. The results show that the measurement structure does not differ between eager and reluctant respondents. There was also no systematic difference in the reliability and validity estimates for both groups.

The analysis of measurement invariance of latent constructs is important in research across groups, or across time. By establishing whether factor loadings, intercepts and residual variances are equivalent in a factor model that measures a latent concept, we can assure that comparisons that are made on the latent variable are valid across groups or time. Establishing measurement invariance involves running a set of increasingly constrained structural equation models, and testing whether differences between these models are significant. This paper provides a step-by-step guide to analysing measurement invariance.

Meuleman and Billiet (2009) have carried out a simulation study aimed at the question how many countries are needed for accurate multilevel SEM estimation in comparative studies. The authors concluded that a sample of 50 to 100 countries is needed for accurate estimation. Recently, Bayesian estimation methods have been introduced in structural equation modeling which should work well with much lower sample sizes. The current study reanalyzes the simulation of Meuleman and Billiet using Bayesian estimation to find the lowest number of countries needed when conducting multilevel SEM. The main result of our simulations is that a sample of about 20 countries is sufficient for accurate Bayesian estimation, which makes multilevel SEM practicable for the number of countries commonly available in large scale comparative surveys., Survey Research Methods, Vol 6, No 2 (2012)

A daily diary method was used to examine the daily dynamics of adolescent conflict and perceived relationship satisfaction with mothers, fathers, and best friends among a sample of 72 Dutch adolescents (M = 15.59 years). Multilevel analyses revealed that perceived relationship satisfaction with mothers, fathers, and best friends was lower on days on which conflict occurred with mothers, fathers, and best friends than on days on which no conflict occurred. More specifically, perceived relationship satisfaction was highest in a particular relationship on days when no conflict occurred, second highest on days on which constructive conflict occurred, and lowest on days on which unconstructive conflict occurred. Whereas in adolescents’ relationships with their parents, conflict and perceived relationship satisfaction were not found to be related to each other one day later, conflict with their best friends—and especially unconstructive conflict—was found to be related to higher perceived relationship satisfaction one day later.

Recently, the leading position of telephone surveys as the major mode of data collection has been challenged. Telephone surveys suffer from a growing nonresponse, partly due to the general nonresponse trend for all surveys and partly due to changes in society and technology influencing contactability and willingness to answer. One way to counteract the increasing nonresponse is the use of an advance letter. In mail and face-to-face surveys, advance letters have been proven effective. Based on the proven effectiveness in face-to-face and mail surveys, survey handbooks advise the use of advance letters in telephone surveys. This study reviews the evidence for this advice and presents a quantitative summary of empirical studies on the effectiveness of advance letters in raising the response rate for telephone surveys. The major conclusion is that advance letters are also an effective tool in telephone surveys, with an average increase in response rate (RR1) from 58 percent (no letter) to 66 percent (advance letter), and an average increase in cooperation rate (COOP1) from 64 percent (no letter) to 75 percent (advance letter).

This study examined the effects of daily support exchanges in couples facing multiple sclerosis (MS). Two issues were examined: the imbalance between received and provided support, and the extent to which reciprocal exchanges of received and provided support are associated with end-of-day well-being (positive and negative mood and selfesteem). Guided by equity theory, we expected that one-sided support provision or receipt would be harmful for well-being for both patients and partners. We argued that these negative outcomes could be offset by reciprocating support, that is, when both partners receive and provide support. Sixty-one patients and their partners filled out questionnaires on demographics and disease-related characteristics and subsequently completed computerized daily diaries for 14 days. At the end of each day, both partners completed diaries on end-of-day mood, self-esteem, received and provided emotional and instrumental support, and several control variables (daily hassles and MS-related symptoms for patients). Reciprocity in instrumental support transactions was associated with higher levels of self-esteem among both patients and partners. However, the other results all showed independent effects of support received and provided. Patients’ well-being was related to providing emotional support and instrumental support, whereas partners’ well-being was related to receiving emotional support from patients.

This article examines strategies to reduce nonresponse rates among ethnic minorities. The authors review nonresponse rates and data collection strategies among ethnic minorities with respect to response rates and response bias in six European countries. The national statistical institutes of these six countries use different definitions of ethnic minorities. This is why the definitions of ethnic minorities and their impact on the number of members of ethnic minorities in the six countries are compared. Nonresponse rates are usually higher among ethnic minorities than among the native population. Dissecting the nonresponse phenomenon shows that contact rates among ethnic minorities are lower, nonresponse due to an inability to produce the required information is higher, and cooperation rates are higher among ethnic minorities than among the native population. Increasing the response rates among ethnic minorities should focus on enhancing the contact rate and reducing the number of nonrespondents who are unable to produce the required information. Keywords: nonresponse reduction; tailored strategies; ethnic minorities

The goal of meta-analysis is to integrate the research results of a number of studies on a specific topic. Characteristic for meta-analysis is that in general only the summary statistics of the studies are used and not the original data. When the published research results to be integrated are longitudinal, multilevel analysis can be used for the meta-analysis. We will demonstrate this with an example of longitudinal data on the mental development of infants. We distinguish four levels in the data. The highest level (4) is the publication, in which the results of one or more studies are published. The third level consists of the separate studies. At this level we have knowledge about the degree of prematurity of the group of infants in the specific study. The second level are the repeated measures. We have data about the test age, the mental development, the corresponding standard deviations, and the sample sizes. The lowest level is needed for the specification of the meta-analysis model. Both the way in which the multilevel model has to be specified (the Mln-program is used) as the results will be presented and interpreted.