Investigating directionality for a media consumption construct : establishing the utility of confirmatory vanishing tetrad analysis

Purpose – When assessing the psychometric properties of measures and estimate relations among latent variables, many studies in the social sciences (including management and marketing) often fail to comprehensively appraise the directionality of indicants. Such failures can lead to model misspecification and inaccurate parameter estimates. The purpose of this paper is to apply a post hoc test called confirmatory vanishing tetrad analysis (CTA hereafter) to a single construct called mass media consumption information exposure, which antecedent studies conceptually posited to be a formative (causative) representation.
Design/methodology/approach – This paper analyses a consumer sample of 585 US respondents and applies the CTA test to a single construct by its inclusion in various matrices within a statistical analysis system-macro that takes into account nonnormal data characteristics. The matrices are derived from Mplus 5 through the estimation of a single-factor congeneric model. The CTA test calculates a test statistic similar to an asymptotic x2 distribution with degrees of freedom equal to the number of nonredundant tetrads tested.
Findings – The preliminary data analyses reveal that the data characteristics are nonnormal which is not uncommon in social research. The CTA results reveal that the reflective (emergent) item orientation cannot be fully ruled out as being the correct model representation. This is in contrast to prior theoretical conceptual work which would strongly support this construct being a formative representation.
Originality/value – Insofar as the authors are aware, there is no paper with a particular focus on how the CTA might not provide sound results with a demonstrated example. The paper makes a valuable contribution by discussing modelling philosophy and a procedure for directionality testing. The authors advocate the implementation of pre and post hoc tests as a key component of standard research practice.

An examination of the mediating role of salt knowledge and beliefs on the relationship between socio-demographic factors and discretionary salt use: a cross-sectional study

Background Discretionary salt use varies according to socio-demographic factors. However, it is unknown whether salt knowledge and beliefs mediate this relationship. This study examined the direct and indirect effect of socio-demographic factors on salt knowledge and discretionary salt use in a sample of 530 Australian adults.

Methods An internet based cross-sectional survey was used to collect data for this study. Participants completed an online questionnaire which assessed their salt knowledge, beliefs and salt use behaviour. Mplus was used to conduct structural equation modelling to estimate direct and indirect effects.

Results The mean age of the participants was 49.2 years, and about a third had tertiary education. Discretionary salt use was inversely related to age (r=-0.11; p<0.05), and declarative salt knowledge (knowledge of factual information) scores (r = -0.17; p<0.01), but was positively correlated with misconceptions about salt (r = 0.09; p<0.05) and beliefs about the taste of salt (r = 0.51; p<0.001). Structural equation modelling showed age, education and gender were indirectly associated with the use of discretionary salt through three mediating pathways; declarative salt knowledge, misconceptions about salt and salt taste beliefs.

Conclusions Inequalities observed between socio-demographic groups in their use of discretionary salt use can potentially be reduced through targeted salt knowledge and awareness campaigns.

Improvements in fundamental movement skill competency mediate the effect of the SCORES intervention on physical activity and cardiorespiratory fitness in children

Numerous studies have identified a positive association between fundamental movement skill (FMS) competency and physical activity in children; however, the causal pathways have not been established. The aim of this study is to determine if changes in FMS competency mediated the effect of the Supporting Children's Outcomes using Rewards, Exercise and Skills (SCORES) intervention on physical activity and cardiorespiratory fitness in children. Eight primary schools (25 classes) and 460 children (aged 8.5 ± 0.6, 54% girls) were randomised to the SCORES intervention or control group for the 12-month study. The outcomes were accelerometer-determined moderate-to-vigorous physical activity (MVPA) and cardiorespiratory fitness. The hypothesised mediators were actual FMS competency and perceived sport competence. Mediation analyses were conducted using multilevel linear analysis in MPlus. From the original sample, 138 (30.0%) and 370 (80.4%) children provided useable physical activity and cardiorespiratory fitness data at post-test assessments. There were significant treatment effects for locomotor skills and overall FMSs. Changes in MVPA were associated with changes in object-control skills, overall FMSs and perceived competence. The overall FMSs had a significant mediating effect on MVPA (AB = 2.09, CI = 0.01-4.55). Overall FMSs (AB = 1.19, CI = 0.002-2.79) and locomotor skills (AB = 0.74, CI = 0.01-1.69) had a significant mediating effect on cardiorespiratory fitness. The results of this study conclude that actual but not perceived movement skill competency mediated the effect of the SCORES intervention on physical activity and cardiorespiratory fitness.