Nonresponse Bias in a Dual Frame Sample of Cell and Landline Numbers

We conducted a dual frame survey of landline and cell phonenumbers in 2004 to evaluate the feasibility of including cellphone numbers in random digit dial telephone surveys in theUnited States. By sampling cell phone numbers, the coveragebias associated with households that have only cell phones iseliminated. However, we discovered two major sources of nonresponsebias in the dual frame sample. In an attempt to reduce thesebiases, we applied several different estimation schemes. Buta comparison to the 2004 Current Population Survey Cell PhoneSupplement showed that none of the estimation schemes substantiallyreduced the nonresponse bias of the estimates. We suggest othermethods that might be used in future surveys that include cellphones and discuss the need for additional data collection andresearch on this issue.     

Italian Preschoolers' Peer‐status Linkages with Sociability and Subtypes of Aggression and Victimization

Little is known about the behavior of preschool children belonging to peer sociometric status groups (popular, average, rejected, neglected, and controversial) in cultural contexts outside North America. This study examined the social interactions of Italian preschoolers. The sample consisted of 266 Italian preschoolers (mean age of 64 months). Physical and relational subtypes of aggression and victimization, as well as sociable behavior, were measured by peer nominations and teacher ratings. Peer nominations of acceptance and rejection (like and dislike nominations) were also collected in order to form the sociometric status groups. Results of confirmatory factor analyses showed that items developed with US preschoolers appeared to work fairly well in identifying behavioral constructs in Italian preschoolers. Findings generally supported previous research with American preschoolers. Sociometrically popular children were highest in sociability, and lowest in physical and relational aggression and victimization. The opposite pattern emerged for rejected status children. Neglected status children were generally not distinguished from average status children. Results also suggested that the enactment of relational aggression promotes greater social impact for some Italian children (controversial children), and this aggression also invites more conflict and victimization. However, the sociability of controversial children appears to buffer them from rejected group status.     

Validation Theory and Research for a Population-Level Measure of Children’s Development,Wellbeing, and School Readiness

This paper delineates general validity and research questions that are underlying an ongoing program of research pertaining to the Early Development Instrument (EDI, Janus and Offord 2007), a population-level measure, on which teachers rate kindergarten children’s developmental outcomes in the social, emotional, physical, cognitive, and communicative domains. It describes the large-scale research projects that are using the EDI to measure children’s developmental outcomes for entire populations (e.g., provinces, cities). Given the uniqueness of the EDI’s use as monitoring tool, or social indicator, for children’s developmental health at an early age, we spell out the opportunities and challenges that these projects provide with regard to validation research. The article sets the stage for the special issue, as it is dedicated to showcase the different aspects of the conceptual, theoretical, and empirical validation research as well as the collaborative community-based projects that are currently being undertaken with respect to the EDI.     

HIV seroprevalence and risk behaviors among clients attending a clinic for the homeless in Miami/Dade County,Florida, 1990–1996

In the early 1990s, HIV seroprevalence was highest at the Miami homeless clinic among 16 homeless sites participating in a nationwide sentinel survey. To examine dynamic seroprevalence patterns in Miami's homeless clients in relation to demographics and risk behaviors over six years, we analyzed data from an unlinked (blinded) serosurvey of clients attending the principal primary care clinic serving Miami's homeless. Data were from 3,797 medical encounters with homeless persons who, on their initial clinic visit within an annual survey period, received routine serologic testing and a risk behavior survey. Overall HIV seroprevalence was 15.9% and infection rates for men (16.4%) and women (14.5%) did not differ. Seroprevalence for blacks (19.9%) was significantly higher than for Hispanics (9.1%) or whites (8.3%) (p < 0.0001). Seroprevalence was 12.6% (35 times the national rate) for clients reporting heterosexual contact as their only risk. Significant increases in seroprevalence, above this heterosexual-contact-only 'baseline', were found for clients disclosing high-risk behaviors: male-to-male sex, drug injection, receiving or giving money/drugs for sex, and sexual contact with a drug injector or HIV-infected partner (p < 0.0001). Seroprevalence declined over six years from 23.2 to 7.2% (p < 0.0001). Significant downward trends were observed for men and women, blacks and Hispanics, men who have sex with men, and clients reporting heterosexual contact. The proportion of clients reporting high-risk behaviors decreased sharply (p < 0.0001). Elevated HIV seroprevalence in Miami's homeless clients was strongly associated with high-risk behaviors. Expansion of HIV prevention and HIV/drug treatment services for homeless persons is strongly recommended.     

Conjugate analysis of multivariate normal data with incomplete observations

The authors discuss prior distributions that are conjugate to the multivariate normal likelihood when some of the observations are incomplete. They present a general class of priors for incorporating information about unidentified parameters in the covariance matrix. They analyze the special case of monotone patterns of missing data, providing an explicit recursive form for the posterior distribution resulting from a conjugate prior distribution. They develop an importance sampling and a Gibbs sampling approach to sample from a general posterior distribution and compare the two methods.