Sampling in research on interpersonal aggression

The aim of this study was to investigate the usefulness of convenience samples in research on interpersonal aggression among adults. It was hypothesised that convenience sampled targets of aggression differs from targets in general with regards to both demographic characteristics and degree of aggression exposed to. A convenience sample comprising support-seeking targets of workplace bullying was compared with a representative sample of Norwegian targets of bullying. The results showed that the two samples differed significantly on all demographic variables investigated, except gender. A far higher percentage of the convenience sample had blown the whistle on illegal, immoral or illegitimate practice at their workplace, whereas they also reported significantly more frequent and more intense exposure to aggression. The findings confirm that convenience samples have low external validity when generalising to the general population. Such samples should therefore mainly be used to investigate tendencies in, and the phenomenology of, interpersonal aggression, in studies where generalisability is not the principal objective.     

The Generalizability of Personality Effects in Politics

A burgeoning line of research examining the relation between personality traits and political variables relies extensively on convenience samples. However, our understanding of the extent to which using convenience samples challenges the generalizability of these findings to target populations remains limited. We address this question by testing whether associations between personality and political characteristics observed in representative samples diverged from those observed in the sub‐populations most commonly studied in convenience samples, namely, students and Internet users. We leverage 10 high‐quality representative datasets to compare the representative samples with the two subsamples. We did not find any systematic differences in the relationship between personality traits and a broad range of political variables. Instead, results from the subsamples generalized well to those observed in the broader and more diverse representative sample.     

Children of Lesbian and Gay Parents

ABSTRACT— Does parental sexual orientation affect child development, and if so, how? Studies using convenience samples, studies using samples drawn from known populations, and studies based on samples that are representative of larger populations all converge on similar conclusions. More than two decades of research has failed to reveal important differences in the adjustment or development of children or adolescents reared by same-sex couples compared to those reared by other-sex couples. Results of the research suggest that qualities of family relationships are more tightly linked with child outcomes than is parental sexual orientation.     

Mock Jury Research: Where Do We Go from Here?

This paper reviews the four types of validity that make up Cook and Campbell's traditional approach for social science research in general and psychological research in particular: internal validity, statistical conclusion validity, external validity, and construct validity. The most important generalizability threat to the validity of jury research is not likely a selection main effect (i.e., the effect of relying solely on undergraduate mock jurors) but is more likely the interaction of sample with construct validity factors. Researchers who try to capture the trial process with experimental paradigms may find that undergraduate mock jurors react differently to those efforts than do more representative community samples. We illustrate these issues with the seven papers that make up this volume, and conclude by endorsing Diamond's call for a two-stage research process in which findings with samples of convenience gradually add more realistic trial processes and representative samples to confirm the initial findings and increase the research program's credibility.     

Height, Gender, and Authority Status at Work: Analyses for a National Sample of Canadian Workers

A small number of previous studies using convenience samples from outside Canada, and mostly for males, show positive relationships between physical height and holding a position of authority as a manager or supervisor. The present study employs Multiple Classification Analysis to assess the generality of these patterns to a representative sample of full-time Canadian workers (2,210 males and 1,815 females) using seven alternative measures of authority status. The results for male workers, after controls, generally show significant positive relationships between height and authority status. The controlled analyses for female workers, however, do not. Additional analyses for males show height to be a comparatively strong predictor relative to other social background predictors of authority status. Alternative interpretations of the patterns of findings are discussed.     

Cross-sectional age differences in personality: Evidence from nationally representative samples from Switzerland and the United States

Identifying reliable patterns of age differences in personality can help clarify the nature of adult personality development. Previous studies have been limited because many have relied on convenience samples. In this study, we examined age differences in personality in two nationally representative samples, one from Switzerland and one from the United States. The results indicated that Agreeableness and Conscientiousness were positively associated with age, whereas Extraversion was negatively associated with age. However, the magnitude of age differences for Extraversion was much smaller than for Agreeableness and Conscientiousness. Openness showed a more complex trend such that 30- to 34-year-olds scored lower on Openness than younger age groups, whereas older groups scored somewhere in between. Inconsistent age differences were observed for Neuroticism.     

Public attitudes toward stuttering in Turkey: probability versus convenience sampling

Purpose: A Turkish translation of the Public Opinion Survey of Human Attributes-Stuttering (POSHA-S) was used to compare probability versus convenience sampling to measure public attitudes toward stuttering. Method: A convenience sample of adults in Eski?ehir, Turkey was compared with two replicates of a school-based, probability cluster sampling scheme. Results: The two replicates of the probability sampling scheme yielded similar demographic samples, both of which were different from the convenience sample. Components of subscores on the POSHA-S were significantly different in more than half of the comparisons between convenience and probability samples, indicating important differences in public attitudes. Conclusions. If POSHA-S users intend to generalize to specific geographic areas, results of this study indicate that probability sampling is a better research strategy than convenience sampling.Educational objectives: The reader will be able to: (1) discuss the difference between convenience sampling and probability sampling; (2) describe a school-based probability sampling scheme; and (3) describe differences in POSHA-S results from convenience sampling versus probability sampling.     

Loneliness and mental health in a representative sample of community-dwelling Spanish older adults

Research seems to support loneliness as a risk factor for mental health problems in the elderly. Most studies analyzing the effects of loneliness on older adults' mental health have relied on convenience samples. In this study, the prevalence and predictors of feelings of loneliness were studied in a representative sample of 272 community-dwelling Spanish older adults. The potential of feelings of loneliness to significantly contribute to the explanation of mental health of the elderly was also explored. The percentage of people reporting feelings of loneliness was 23.1%. Being a woman, being older, living alone, having fewer economic resources, having lower perceived health, and being dissatisfied with the frequency of contact with relatives and friends were found to be significant predictors of feelings of loneliness. Loneliness contributed significantly to the explanation of mental health, even when other significant variables were statistically controlled. The results of this study suggest that loneliness is a relevant factor for the analysis and understanding of mental health in the elderly.     

Individual phase course and role change in neurosis patients in open intention-centered dynamic psychotherapy groups

Sociograms are used to demonstrate the phase-typical occurrence of changes in individual roles among a sample comprising 79 neurotics who had received in-patient treatment in intention-centred dynamic open groups. This change depends on age and sex. The results can, on the basis of the four fundamental principles be used for controlling developments, be interpreted as indicating a development, but require closer definition by the use for more representative samples and variables.     

Dimensional analysis of sense of coherence using structural equation modelling

The sense of coherence (SOC) concept has been criticized on conceptual and psychometric grounds. The aim of this study was to develop the concept using structural equation modelling. The short 13‐item version of the SOC questionnaire was used and data were collected from two different samples. Each sample consisted of about 2000 individuals, selected to be representative of the Swedish population. A model with a general factor and two subordinate specific factors yielded a satisfactory solution in both samples. The two factors were labelled Social Comprehension and Commitment and Unpleasant Emotions and Inner Tension respectively. It was concluded that the 13‐item SOC scale could to a certain degree be regarded as a measure of negative affectivity and that the suggested model makes the SOC concept more theoretically coherent. Copyright © 1999 John Wiley & Sons, Ltd.     

Can treatment trial samples be representative?

A catchment area-based sample of patients recruited for an eating disorder treatment trial was compared with patients from the same geographical area seen in the 12 months before and after the trial. The three samples were very similar. The research sample was representative of the usual clinic sample from which it had been selected and thus the results could be extrapolated with some confidence to other similar clinical settings. It is concluded that whilst treatment trials, by their very nature, have explicit and implicit inclusion and exclusion criteria with appropriate designs they can be usefully representative.     

Is “Generation Me” Really More Narcissistic Than Previous Generations?

ABSTRACT In this commentary, we identify several methodological and conceptual issues that undermine Twenge, Konrath, Foster, Campbell, and Bushman's (this issue) claim that narcissism levels have been rising over the past few decades. Specifically, we discuss (a) the limitations of convenience samples for making inferences about generational differences, (b) our failure to replicate other cross-temporal meta-analytic findings using data from a nationally representative sample, and (c) issues surrounding the interpretation of the Narcissistic Personality Inventory. It is important to consider these issues given the extensive media coverage of Twenge et al.'s claim that today's youth are particularly narcissistic, which has wide-ranging implications for how this generation of young adults views itself and is viewed by society at large.     

A Swedish validation of the Berlin Numeracy Test

Recent research has highlighted the importance of considering an individual's level of numeracy, that is their numerical abilities, in a vast variety of judgment and decision making tasks. To accurately evaluate the influence of numeracy requires good and valid measures of the construct. In the present study we validate a Swedish version of the Berlin Numeracy Test (Cokely, Galesic, Schulz, Ghazal & Garcia‐Retamero, 2012). The validation was carried out on both a student sample and a sample representative of the Swedish population. The Swedish BNT showed sound psychometrical properties in both samples. Further, in both samples the BNT had satisfactory convergent and discriminant validity when correlating with other measures of numeracy, while not being significantly related to measures of personality. With respect to predictive validity the results indicated divergent patterns in the two samples. In the student sample, participants scoring highest on the BNT outperformed those in the other three levels, which did not differ in performance. In contrast, in the population sample participants scoring lowest on the BNT performed worse than those in the other three levels, which did not differ in performance. Taken together, however, the results suggest that the Swedish version of the BNT should be considered a valid measure of numeracy in both Swedish student and population representative samples.     

Approaches to the Empirical Study of Convenience Theory for White-Collar Crime

Convenience theory suggests that members of the elite in society commit financial crime in their professional roles when alternative actions require too much effort. Convenience is a relative concept where white-collar crime is chosen over legitimate actions when there is a strong economical motive, ample organizational opportunities, and acceptance of deviant behavior. To study convenience theory, four investigations are presented in this article: statistical sample of white-collar criminals, autobiographies by white-collar criminals, internal investigations of white-collar crime, and student elicitation on white-collar crime. The strongest relationship within convenience theory seems to be the effect from willingness to commit crime based on deviant behavior on organizational opportunity to commit white-collar crime.     

Extent and limits of the matching concept in monkeys (Cebus apella)

In Experiment 1, 8 monkeys, experimentally naive with regard to visual stimuli, were trained on identity matching with a two-sample set based on two-dimensional stimuli. On a subsequent test employing two new samples, 4 of the 8 applied the matching rule to the new sample stimuli (as defined by our transfer criterion), and 3 showed substantial savings in learning to match the new samples. Two of these 3 transferred the matching rule when given a second test with two new samples, and the third showed immediate and complete transfer when tested with a third pair of new stimuli. These results indicate a much stronger representation of the matching concept in monkeys than in pigeons, even when the conditions of assessment are reasonably comparable. In Experiment 2, however, 4 monkeys from Experiment 1 failed to transfer the matching rule to steady versus flashing green samples, indicating that the matching concept did not immediately extend beyond the general class of visual stimuli with which it was developed. These and related results in the literature suggest that representation of the matching concept in animals varies along a specificity-abstractness dimension, reflecting the degree to which the concept is tied to the conditions and context of its development.     

Sampling in developmental science: Situations,shortcomings, solutions,and standards

Sampling is a key feature of every study in developmental science. Although sampling has far-reaching implications, too little attention is paid to sampling. Here, we describe, discuss, and evaluate four prominent sampling strategies in developmental science: population-based probability sampling, convenience sampling, quota sampling, and homogeneous sampling. We then judge these sampling strategies by five criteria: whether they yield representative and generalizable estimates of a study’s target population, whether they yield representative and generalizable estimates of subsamples within a study’s target population, the recruitment efforts and costs they entail, whether they yield sufficient power to detect subsample differences, and whether they introduce “noise” related to variation in subsamples and whether that “noise” can be accounted for statistically. We use sample composition of gender, ethnicity, and socioeconomic status to illustrate and assess the four sampling strategies. Finally, we tally the use of the four sampling strategies in five prominent developmental science journals and make recommendations about best practices for sample selection and reporting.     

Pathological Video-Game Use Among Youth Ages 8 to 18: A National Study

ABSTRACT— Researchers have studied whether some youth are "addicted" to video games, but previous studies have been based on regional convenience samples. Using a national sample, this study gathered information about video-gaming habits and parental involvement in gaming, to determine the percentage of youth who meet clinical-style criteria for pathological gaming. A Harris poll surveyed a randomly selected sample of 1,178 American youth ages 8 to 18. About 8% of video-game players in this sample exhibited pathological patterns of play. Several indicators documented convergent and divergent validity of the results: Pathological gamers spent twice as much time playing as nonpathological gamers and received poorer grades in school; pathological gaming also showed comorbidity with attention problems. Pathological status significantly predicted poorer school performance even after controlling for sex, age, and weekly amount of video-game play. These results confirm that pathological gaming can be measured reliably, that the construct demonstrates validity, and that it is not simply isomorphic with a high amount of play.     

Utility of a CPT in Diagnosing ADHD among a Representative Sample of High-Risk Children: A Cautionary Study

Continuous performance tests (CPTs) are widely used in the assessment and study of attention deficit hyperactivity disorder (ADHD). Although CPTs have reliably found differences between children with ADHD and normal controls, discriminating between children with ADHD and children with subclinical levels of behavioral or cognitive problems is a more clinically relevant and difficult endeavor. Additionally, most studies use convenience samples from clinical care settings that may not represent the ADHD population as a whole. The current study assessed the utility of a clinically used CPT, the Test of Variables of Attention (TOVA), in distinguishing between children with ADHD and children with subclinical levels of attention/behavior problems. Participants constituted a representative sample of elementary school students at high risk for ADHD, including 116 children with ADHD and 51 subclinical controls. Results found no significant differences between the ADHD and subclinical group on CPT variables, and CPT performance did not reliably predict group membership. Implications of the findings are discussed.     

Utility of a CPT in diagnosing ADHD among a representative sample of high-risk children: a cautionary study.

Continuous performance tests (CPTs) are widely used in the assessment and study of attention deficit hyperactivity disorder (ADHD). Although CPTs have reliably found differences between children with ADHD and normal controls, discriminating between children with ADHD and children with subclinical levels of behavioral or cognitive problems is a more clinically relevant and difficult endeavor. Additionally, most studies use convenience samples from clinical care settings that may not represent the ADHD population as a whole. The current study assessed the utility of a clinically used CPT, the Test of Variables of Attention (TOVA), in distinguishing between children with ADHD and children with subclinical levels of attention/behavior problems. Participants constituted a representative sample of elementary school students at high risk for ADHD, including 116 children with ADHD and 51 subclinical controls. Results found no significant differences between the ADHD and subclinical group on CPT variables, and CPT performance did not reliably predict group membership. Implications of the findings are discussed.