Testing the psychometric properties of the Environmental Attitudes Inventory on undergraduate students in the Arab context: A test-retest approach

(Mis)matched direct and moderating relationships among pro-environmental attitudes, environmental efficacy, and pro-environmental behaviors across and within 11 countries

Pro-environmental behaviors are influenced by individuals' pro-environmental attitudes and environmental efficacy, among many other factors. However, attitude-behavior models are inconsistent on whether and how attitudes, efficacy, and behaviors should match in specificity or generality, and on the moderation effect of efficacy. This study first tests a simple model including direct and moderating relationships between pro-environmental attitudes, environmental efficacy, and pro-environmental behaviors. Then it examines relationships among subscales matched or mismatched in their respective specific or general domain of environmental attitudes (concern, values), environmental efficacy (self, collective), and pro-environmental behaviors (private, public). Secondary data come from an overall sample of 11,000 respondents across 11 countries, with n = 1,000 from each country. Pro-environmental attitudes and efficacy have direct relationships with pro-environmental behavior, but efficacy has little moderation effect. Different combinations of (mis)matched measures produce slightly different results, with the most variance explained, counter to hypotheses, by two mismatched models. Results are generally consistent across countries.

A Scoping Review of Nature, Land, and Environmental Connectedness and Relatedness

The importance of nature and the environment in relation to human health is coalescing, as demonstrated by the increased research that attempts to measure nature connectedness and relatedness. These findings align with constructs of cultural connectedness that assess for land connectedness as part of Indigenous ways of knowing. From an Indigenous worldview, relationships with the environment are critical to wellbeing. The purpose of this comprehensive systematic scoping literature review was two-fold: (1) identify and summarize existing measures of land, nature, and/or environmental connectedness, relatedness, and attitudes and (2) evaluate the psychometric properties of these scales. In total, 1438 articles were retrieved from select databases including PubMed/MEDLINE, PsycINFO, CINAHL (EBSCO), and Academic Search Complete (EBSCO). The final searches and application of the inclusion/exclusion criteria resulted in 57 unique articles and 38 scales categorized as connectedness and relatedness scales (n = 9 scales), attitudinal and values-based scales (n = 16 scales), cultural and spiritually based scales (n = 9 scales), and paradigm-based scales (n = 4 scales) (articles could be placed in multiple categories). Psychometric properties and general outcomes associated with nature-related scales are reported, with implications for future education, research, practice, and policy.

Tools for assessing the psychometric adequacy of latent variables in conservation research

Conservation psychology has a history of measuring variables that cannot be seen (e.g., values, attitudes, norms). Such latent variables are critical drivers of human action and are often measured using responses to survey questions. Tools for establishing the psychometric adequacy of unobservable, latent variables has been a century-long pursuit and challenge for quantitative psychologists and statisticians. Fundamental questions at the heart of this challenge include is what is claimed to be measured (validity) being measured and is measurement consistent (reliability)? We examined common methods used to establish the validity and reliability of psychometric instruments. Through a case study of anglers in Texas, we investigated the protocols and metrics used to evaluate the measurement of latent variables. The indicators we tested (identity, awareness of consequences, ascription of responsibility, and personal norms) validly and reliably assessed latent variables. Our findings also illustrated decision protocols (e.g., discriminant validity, convergent validity, internal consistency) involved in assessing the psychometric adequacy of latent variable indicators. The ability to correctly identify significant relationships among unobserved variables and their influence on human action is directly tied to the adequacy of measurement. In an era of instability and change that threatens social-ecological systems worldwide, the need for accuracy and precision in conservation social science has never been greater. Research that employs flawed measures has potential to lead to erroneous conclusions and undermine conservation and biodiversity protection.

Development and initial validation of the chronic hepatitis B quality of life instrument (CHBQOL) among Chinese patients

PURPOSE

This study developed and tested preliminary measurement properties of a Chinese scale specifically designed to measure HRQOL in patients with chronic hepatitis B (CHBQOL).

METHODS

We conducted 94 individual interviews with CHB patients to solicit items and five hepatitis expert interviews along with three rounds of cognitive interviews to identify problems with relevance and understanding of content. A cross-sectional validation study was then conducted to evaluate measurement properties (n = 578). Factor analysis was used to determine the latent structure of the scale. Reliability was evaluated through Cronbach's alpha coefficients and intra-class correlation coefficients (ICCs). Measurement model adequacy, convergent, discriminant, and known-groups validity were also examined.

RESULTS

A scale of 30 items was drafted. After item reduction, the remaining 23 items were assigned to the CHBQOL Somatic symptoms, Emotional symptoms, Belief and Social stigma domains, which had acceptable goodness of fit (χ²/df = 3.13, GFI = 0.90, AGFI = 0.88, RMSEA = 0.06, SRMR = 0.05). All the CHBQOL domains had satisfactory reliability with Cronbach's α coefficients ranging from 0.73 to 0.91 and ICCs were higher than 0.70 except for Belief domain (ICC = 0.54). Convergent and discriminant validity were acceptable as supported by significant item-domain correlations (0.64-0.90). In general, the correlations between CHBQOL and the SF-36 dimensions met hypotheses. Significant differences were found by mean scores in the subgroups of demographic and clinical variables, supporting the known-groups validity.

CONCLUSIONS

The CHBQOL instrument proved to be an appropriate tool for assessing HRQOL among Chinese CHB patients.