Mobile phone surveys: The Slovenian case study

In 2004, the number of mobile phone subscriptions in Slovenia reached the total number of inhabitants. Consequently, the fixed telephone coverage has started to decline; almost 10% of households are now available only over the mobile phone. With this, Slovenia positions itself as a typical EU country and can serve as a case study for issues related to mobile phone interview surveys. The paper addresses the general context of mobile phone usage and the calculations of mobile phone coverage rates. It also discusses the non-coverage problems related to mobile and mobile-only households. It is shown, that even with a relatively small non-coverage the corresponding estimates can be considerably biased, as in the case of the unemployment rate in the Slovenian Labour Force Survey. There are severe methodological problems with mobile phone interview surveys. In particular, a pilot mobile phone survey confirmed the disadvantages of costs, frames and response rates, at least when compared to fixed telephone surveys. In addition, the response rates are dramatically lower for less intensive mobile phone users. The comparisons of respondents in mobile phone surveys with the corresponding sub-samples in Labour Force Surveys and in fixed telephone surveys revealed some specific discrepancies in the socio-demographic structure. Due to non-coverage, the respondents in the mobile phone survey tend to be younger, higher educated, from larger households and are represented by a larger share of males. The non-response mechanism can additionally reinforce these effects (e.g, gender), however it can also cancel them (e.g., age, household size). Special complexity arises from the diverse effects of the non-response components (refusal vs. non-contact). According to their attitudes towards mobile phone use, mobile phone users compose three distinct segments (intensive pragmatic and emotional users, less intensive users) that may behave differentially also during the mobile phone survey process.

Response trends in a national random digit dial survey

This paper reviews response trends over 24 consecutive quarters of a National Random Digit Dial telephone survey. Trends for response rates and refusal rates are studied as well as the components of response rate, namely, contact, cooperation and completion rates. In addition other rates, including answering machine, busy and no answer are studied. While refusal rates declined over the six year period, contact and cooperation rates significantly declined causing response rates to decline. Answering machine rates and busy rates also showed a significant increase over time. Finally, correlation’s among the variables of interest are presented. The response rate is negatively correlated with the busy rate, the answering machine rate and the no answer rate. Implications of the above trends are discussed.

Measuring information accessibility and predicting response-effects: The validity of response-certainties and response-latencies

Respondents’ reports about the frequency of everyday behavior are often found to differ considerably when either low- or high-frequency response scales are used to record the answers. It has been hypothesized that the susceptibility to this type of response effect is determined by the cognitive accessibility of the respective target information in respondents’ memories. The first aim of the present paper is to test this hypothesis using two alternative, individual level indicators of the cognitive accessibility of information. These measures are the subjects’ self-reported response certainty and the time needed to answer the question under consideration. A second issue is how response certainties and response latencies should be transformed prior to data analysis in order to maximize their predictive power for response effects. Accordingly, the ability of untransformed measures to predict scale effects is compared with that of logarithmic, square-root and reciprocally transformed versions. The empirical results show that untransformed response certainties and response latencies are equally valid predictors of whether and to what extent subjects’ answers are affected by the presentation of response options. A square-root transformation is found to have no effect on both measures, whereas a logarithmic transformation slightly improves the validity of response certainties. In contrast, a reciprocal transformation proves to have a substantially positive effect on both measures and improves their ability to predict the reliability of respondents’ survey reports.

Optimal multimedia combination for students with dyslexia

Many contemporary online learning resources use more than one sensory modality (audio, visual) or presentation mode (pictures, text). Multimedia learning material is often recommended for and adopted by students with dyslexia. However, there are not many empirical studies that support the benefits of multimedia for this group. Due to the variety of difficulties that these students may experience, multimedia may not be the most appropriate type of online content for them. Our study aimed to identify the most appropriate combination of multimedia content for them from among three commonly-used types of online learning resources. An experiment was conducted in which 27 students with dyslexia learned about a topic under one of three conditions: a scroll-down website with onscreen text and static pictures, video lecture and subtitled video lecture. No significant differences in learning outcomes were found between the groups. However, the effect sizes indicated the superiority of the website over the other two conditions. Because of the small sample size, further research is needed.

blockmodeling

This paper presents the R package blockmodeling which is primarily meant as an implementation of generalized blockmodeling (more broadly blockmodeling) for valued networks where the values of the ties are assumed to be measured on at least interval scale. Blockmodeling is one of the most commonly used approaches in the analysis of (social) networks, which deals with the analysis of relationships or connections, between the units studied (e.g., peoples, organizations, journals etc.). The R package blockmodeling implements several approaches for the generalized blockmodeling of binary and valued networks. Generalized blockmodeling is commonly used to cluster nodes in a network with regard to the structure of their links. The theoretical foundations of generalized blockmodeling for binary and valued networks are summarized in the paper while the use of the R package blockmodeling is illustrated by applying it to an empirical dataset.

Quality guidelines for mixed methods research in intervention studies

Among researchers' many investigations of the use of mixed methods in intervention studies, more recent discussions especially concern the roles played in such studies by qualitative research, intervention phases, procedures, and integration (Gallo and Lee, 2016; Woolcock, 2018; O'Cathain, 2018; Creswell and Plano Clark, 2018). One can find the basic procedures to follow while realizing a mixed methods experimental design (Creswell and Plano Clark, 2018), practical guidance (O'Cathain, 2018) for using qualitative research with a randomized control trial (RCT), and a mixed methods appraisal tool for appraising the methodological quality of RCTs, non-randomized studies, and mixed methods – MMAT (Hong et al., 2018). However, no model exists to assess the quality of mixed methods research in intervention studies, particularly experimental and quasi-experimental research in complex interventions. Our aim is to develop such a theoretical model. Today, the number of interventions relying on mixed methods methodology is growing exponentially. A theoretical model is called for to help assess the quality of mixed methods research in intervention studies, and in this respect our aim is to: (1) provide an overview of guidelines, recommendations, models, and quality criteria for mixed methods research; (2) overview the guidelines for intervention studies; (3) give a summary of guidelines and models for mixed methods research in such studies; (4) evaluate the mentioned guidelines, models, and quality criteria; (5) identify and describe the key elements of these guidelines, models, and quality criteria; and (6) develop a theoretical model for assessing the quality of mixed methods research designs used in intervention studies.

Limiting spectral distribution of large dimensional random matrices of linear processes

The limiting spectral distribution (LSD) of large sample radom matrices is derived under dependence conditions. We consider the matrices \(X_{N}T_{N}X_{N}^{\prime}\) , where \(X_{N}\) is a matrix (\(N \times n(N)\)) where the column vectors are modeled as linear processes, and \(T_{N}\) is a real symmetric matrix whose LSD exists. Under some conditions we show that, the LSD of \(X_{N}T_{N}X_{N}^{\prime}\) exists almost surely, as \(N \rightarrow \infty\) and \(n(N)/N \rightarrow c > 0\). Numerical simulations are also provided with the intention to study the convergence of the empirical density estimator of the spectral density of \(X_{N}T_{N}X_{N}^{\prime}\).

Estimating Bayes factors from minimal summary statistics in repeated measures analysis of variance designs

In this paper, I develop a formula for estimating Bayes factors directly from minimal summary statistics produced in repeated measures analysis of variance designs. The formula, which requires knowing only the F-statistic, the number of subjects, and the number of repeated measurements per subject, is based on the BIC approximation of the Bayes factor, a common default method for Bayesian computation with linear models. In addition to providing computational examples, I report a simulation study in which I demonstrate that the formula compares favorably to a recently developed, more complex method that accounts for correlation between repeated measurements. The minimal BIC method provides a simple way for researchers to estimate Bayes factors from a minimal set of summary statistics, giving users a powerful index for estimating the evidential value of not only their own data, but also the data reported in published studies.

Recent decline in life expectancy at birth in Slovenia

For many decades, life expectancy at birth (e0) in Slovenia has been increasing at a very rapid pace. However, in 2015, e0 declined slightly; it recovered in 2016, but fell again in 2017 for women. In the same period, a pause in declining mortality was observed in numerous developed countries worldwide. It is too early to provide a thorough analysis and firm conclusions, but we shed some light on the topic by decomposing the observed decline in Slovenia by age and cause of death. In particular, using a life table model and life expectancy decomposition technique, we analyse what cause of death for what age group contributed the most to this decline in life expectancy at birth. We show that the main reason for the recent drop in life expectancy at birth in Slovenia was higher mortality due to external causes for men of all ages and due to neoplasms for women above 60 years and men above 50 years.

Home team advantage in the English Premier League

The home team advantage in association football is a well known phenomenon. The aim of this paper is to offer a different view on the home team advantage. Usually, in association football, every two teams—team A and team B—play each other twice in a season. Once as a home team and once as a visiting, or away team. This gives us two results between teams A and B which are combined together to evaluate whether team A, against its opponent B, recorded a result at its home ground—in the comparison to the away ground—that is better, even, or worse. This leads to a random variable with three possible outcomes, i.e. with trinomial distribution. The combination and comparison of home and away results of the same two teams is the key to eliminate problems with different squad strengths of teams in a league. The bayesian approach is used to determine point and interval estimates of unknown parameters of the source trinomial distribution, i.e. the probability that the result at home will be better, even, or worse. Moreover, it is possible to test a hypothesis that the home team advantage for a selected team is statistically significant.