Agricultural subsidies lead to changes in the use of chemical fertilizer by farmers. Using data from a household survey conducted annually by the Rural Economy Research Center of the Ministry of ...Agriculture of China from 2014 to 2018, Control Function (CF) approach and Heteroskedasticity-based identification strategy were employed to analyze the impact of agricultural subsidies on chemical fertilizer use by rice farmers. After addressing the problem of endogeneity, we found that agricultural subsidies have a significantly negative impact on the use of chemical fertilizer. Precisely, every 100% increase in agricultural subsidies would result in an average decrease of 3.4% in chemical fertilizer use. In addition, results of heterogeneity analysis showed that agricultural subsidies had a stronger negative impact on fertilizer use as rice-planting experience increases. But the ability of rice-planting management and off-farm labor within household could reduce this negative impact. We also found that the effect of arable land productivity conservation subsidies policy was short-lived. Furthermore, based on the frame of structural equation model (SEM), the results of mediation analysis showed that agricultural machine and rice-planting area had partial mediation of 5.3% and 41%, respectively. It implied that agricultural subsidies reduced fertilizer use by promoting the adoption of agricultural techniques and expanding the planting area. In brief, agricultural subsidies had both technical effect and scale effect. But the mediating effect of household income was not significant. This study is intended to assist the concerned authority and agriculture sector to understand the positive role of agricultural subsidies in sustainable production, and provides some feasible policy proposals.
•Higher agricultural subsidies in China reduce chemical fertilizer use.•We study the heterogeneous effects of agricultural subsidies on fertilizer use.•Agricultural subsidies have both technical effect and scale effect, but not income effect.•The effect of agricultural subsidies policy is short-lived.
Inadequate translation from theoretical to statistical models of the greenspace – health relationship may lead to incorrect conclusions about the importance of some pathways, which in turn may reduce ...the effectiveness of public health interventions involving urban greening. In this scoping review we aimed to: (1) summarize the general characteristics of approaches to intervening variable inference (mediation analysis) employed in epidemiological research in the field; (2) identify potential threats to the validity of findings; and (3) propose recommendations for planning, conducting, and reporting mediation analyses.
We conducted a scoping review, searching PubMed, Scopus, and Web of Science for peer-reviewed epidemiological studies published by December 31, 2019. The list of potential studies was continuously updated through other sources until March 2020. Narrative presentation of the results was coupled with descriptive summary of study characteristics.
We found 106 studies, most of which were cross-sectional in design. Most studies only had a spatial measure of greenspace. Mental health/well-being was the most commonly studied outcome, and physical activity and air pollution were the most commonly tested intervening variables. Most studies only conducted single mediation analysis, even when multiple potentially intertwined mediators were measured. The analytical approaches used were causal steps, difference-of-coefficients, product-of-coefficients, counterfactual framework, and structural equation modelling (SEM). Bootstrapping was the most commonly used method to construct the 95% CI of the indirect effect. The product-of-coefficients method and SEM as used to investigate serial mediation components were more likely to yield findings of indirect effect. In some cases, the causal steps approach thwarted tests of indirect effect, even though both links in an indirect effect were supported. In most studies, sensitivity analyses and proper methodological discussion of the modelling approach were missing.
We found a persistent pattern of suboptimal conduct and reporting of mediation analysis in epidemiological studies investigating pathways linking greenspace to health; however, recent years have seen improvements in these respects. Better planning, conduct, and reporting of mediation analyses are warranted.
•We summarized the characteristics of approaches to mediation analysis in the field.•Potential threats to the validity of findings were identified.•A persistent pattern of suboptimal conduct and reporting of mediation analysis was found.•We propose recommendations for better planning, conducting, and reporting mediation analyses.
With two field studies, this research highlights the role of brand community identification (BCI) as a source of customers' repurchase intentions and also reveals how BCI creates a buffering effect, ...protecting against the negative repercussions of service failures for repurchase intentions. The first study builds on social identity theory and investigates BCI as a driver of repurchase intentions; it explains this positive relationship according to single dimensions of customer citizenship behavior. The second study establishes that BCI mitigates the negative effect of service failures on customers' repurchase intentions. These results call attention to the need to build and maintain customers' strong BCI, because such investments encourage favorable customer–customer helping and advocacy that drive repurchase intentions. Moreover, strong BCI may reduce a firm's required service recovery efforts, because customers with strong BCI perceive service failures less negatively.
•The article shows innovative results based on two field studies with active members of a brand community.•The impact of brand community identification on repurchase intentions is explained by customer citizenship behaviors.•In contrast to previous studies, we found direct effects of Brand community identification on repurchase intentions.•Strong brand community identification weakens the effect of service failures on repurchase intentions (buffering effect).
Until now, there has been limited research on the specific relationship between exposure to child maltreatment, the development of low self-control and subsequent delinquency among non-western ...samples from a general theory of crime perspective. To address this research gap, this study explores the role of low self-control as a mediator between exposure to child maltreatment and delinquent behaviour of South Korean children. By using longitudinal data from South Korean children, structural equation modeling was employed to examine this relationship. The results of the analysis indicated that child maltreatment significantly influenced the development of low self-control in children. Also, the findings revealed that low self-control mediated the relationship between child maltreatment and delinquent behaviour. In other words, consistent with the proposition of the general theory of crime, children’s low self-control played a mediating role in the impact of exposure to child maltreatment on their engagement in delinquent behaviour. Based on these findings, the importance of early interventions to prevent child maltreatment and foster the self-control level of children is suggested.
The incorporation of causal inference in mediation analysis has led to theoretical and methodological advancements-effect definitions with causal interpretation, clarification of assumptions required ...for effect identification, and an expanding array of options for effect estimation. However, the literature on these results is fast-growing and complex, which may be confusing to researchers unfamiliar with causal inference or unfamiliar with mediation. The goal of this article is to help ease the understanding and adoption of causal mediation analysis. It starts by highlighting a key difference between the causal inference and traditional approaches to mediation analysis and making a case for the need for explicit causal thinking and the causal inference approach in mediation analysis. It then explains in as-plain-as-possible language existing effect types, paying special attention to motivating these effects with different types of research questions, and using concrete examples for illustration. This presentation differentiates 2 perspectives (or purposes of analysis): the explanatory perspective (aiming to explain the total effect) and the interventional perspective (asking questions about hypothetical interventions on the exposure and mediator, or hypothetically modified exposures). For the latter perspective, the article proposes tapping into a general class of interventional effects that contains as special cases most of the usual effect types-interventional direct and indirect effects, controlled direct effects and also a generalized interventional direct effect type, as well as the total effect and overall effect. This general class allows flexible effect definitions which better match many research questions than the standard interventional direct and indirect effects.
To investigate associations of leptin and adiponectin levels with knee and hand osteoarthritis, and explore whether these mediate the association between adiposity and osteoarthritis.
This is a ...cross-sectional analysis of baseline data from the population-based Netherlands Epidemiology of Obesity study. Adiposity was assessed with body mass index (BMI) and percentage total body fat (%TBF). Osteoarthritis, defined as hand or knee osteoarthritis, was determined using American College of Rheumatology criteria. Fasting serum adipokine levels were measured using immunoassays. Associations between adiposity and osteoarthritis were examined with logistic regression, adjusted for age, sex, ethnicity and education, and additionally for leptin and adiponectin as potential mediators.
In 6408 participants (56% women, median age 56 years), prevalence of osteoarthritis was 22% (10% isolated knee and 8% isolated hand osteoarthritis). Leptin levels were positively associated with osteoarthritis, while adiponectin levels were not. Leptin partially mediated the association of adiposity with osteoarthritis (OR 1.40 (95%CI 1.30; 1.52) attenuated to 1.38 (1.24; 1.54) per 5 units BMI and OR 1.25 (1.17; 1.35) to 1.20 (1.10; 1.32) per 5 units %TBF, representing 4% and 17% mediation, respectively). Larger proportion mediation by leptin was found in knee (13%/27%) than in hand osteoarthritis (9%/18%). Sex-stratified analyses generally showed stronger associations between adiposity, leptin and osteoarthritis in women than in men.
Serum leptin levels were associated with osteoarthritis, and partially mediated the association between adiposity and osteoarthritis, while adiponectin levels were not associated with osteoarthritis. These findings provide evidence for systemic effects of adipose tissue in osteoarthritis.
Purpose
Indirect or mediated effects constitute a type of relationship between constructs that often occurs in partial least squares (PLS) path modeling. Over the past few years, the methods for ...testing mediation have become more sophisticated. However, many researchers continue to use outdated methods to test mediating effects in PLS, which can lead to erroneous results. One reason for the use of outdated methods or even the lack of their use altogether is that no systematic tutorials on PLS exist that draw on the newest statistical findings. The paper aims to discuss these issues.
Design/methodology/approach
This study illustrates the state-of-the-art use of mediation analysis in the context of PLS-structural equation modeling (SEM).
Findings
This study facilitates the adoption of modern procedures in PLS-SEM by challenging the conventional approach to mediation analysis and providing more accurate alternatives. In addition, the authors propose a decision tree and classification of mediation effects.
Originality/value
The recommended approach offers a wide range of testing options (e.g. multiple mediators) that go beyond simple mediation analysis alternatives, helping researchers discuss their studies in a more accurate way.
Mediation analysis is crucial for diagnosing indirect causal relations in many scientific fields. However, mediation analysis of nominal variables requires examining and comparing multiple total ...effects and their corresponding direct/indirect causal effects derived from mediation models. This process is tedious and challenging to achieve with classical analysis tools such as Excel tables. In this study, we worked closely with experts from two scientific domains to design MediVizor, a visualization system that enables experts to conduct visual mediation analysis of nominal variables. The visualization design allows users to browse and compare multiple total effects together with the direct/indirect effects that compose them. The design also allows users to examine to what extent the positive and negative direct/indirect effects contribute to and reduce the total effects, respectively. We conducted two case studies separately with the experts from the two domains, sports and communication science, and a user study with common users to evaluate the system and design. The positive feedback from experts and common users demonstrates the effectiveness and generalizability of the system.
The fields of machine learning and causal inference have developed many concepts, tools, and theory that are potentially useful for each other. Through exploring the possibility of extracting causal ...interpretations from black-box machine-trained models, we briefly review the languages and concepts in causal inference that may be interesting to machine learning researchers. We start with the curious observation that Friedman's partial dependence plot has exactly the same formula as Pearl's back-door adjustment and discuss three requirements to make causal interpretations: a model with good predictive performance, some domain knowledge in the form of a causal diagram and suitable visualization tools. We provide several illustrative examples and find some interesting and potentially causal relations using visualization tools for black-box models.
Abstract
STUDY QUESTION
Could the direct contribution of genetic variants to the pathophysiology of uterine fibroids and the contribution mediated by age at menarche be different?
SUMMARY ANSWER
Age ...at menarche plays a mediation role in the genetic influence on uterine fibroids, and four causal genetic mechanisms underlying the age at menarche-mediated effects of common genetic loci on uterine fibroid development were identified.
WHAT IS KNOWN ALREADY
Uterine fibroids are common benign tumors developing from uterine smooth muscle. Genome-wide association studies (GWASs) have identified over 30 genetic loci associated with uterine fibroids in different ethnic populations. Several genetic variations in or nearby these identified loci were also associated with early age at menarche, one of the major risk factors of uterine fibroids. Although the results of GWASs reveal how genetic variations affect uterine fibroids, the genetic mechanism of uterine fibroids mediated by age at menarche remains elusive.
STUDY DESIGN, SIZE, DURATION
In this study, we conducted a genome-wide causal mediation analysis in two cohorts covering a total of 69 552 females of Han Chinese descent from the Taiwan Biobank (TWB). TWB is an ongoing community- and hospital-based cohort aiming to enroll 200 000 individuals from the general Taiwanese population between 30 and 70 years old. It has been enrolling Taiwanese study participants since 2012 and has extensive phenotypic data collected from 148 291 individuals as of May 2021.
PARTICIPANTS/MATERIALS, SETTING, METHODS
We recruited individuals in two cohorts, with 13 899 females in TWB1 and 55 653 females in TWB2. The two sets of individuals are almost distinct, with only 730 individuals enrolled in both cohorts. Over 99% of the participants are Han Chinese. Approximately 21% of participants developed uterine fibroids. DNA samples from both cohorts were genotyped using two different customized chips (TWB1 and TWB2 arrays). After quality control and genotype imputation, 646 973 TWB1 single-nucleotide polymorphisms (SNPs) and 686 439 TWB2 SNPs were assessed in our analysis. There were 99 939 SNPs which overlapped between the TWB1 and TWB2 arrays, 547 034 TWB1 array-specific SNPs and 586 500 TWB2 array-specific SNPs. We performed GWASs for screening potential risk SNPs for age at menarche and for uterine fibroids. We subsequently identified causal mediation effects of risk SNPs on uterine fibroids mediated by age at menarche.
MAIN RESULTS AND THE ROLE OF CHANCE
In addition to known loci at LIN28B associated with age at menarche and loci at WNT4 associated with uterine fibroids, we identified 162 SNPs in 77 transcripts that were associated with menarche-mediated causal effects on uterine fibroids via four different causal genetic mechanisms: a both-harmful group with 52 SNPs, a both-protective group with 34 SNPs, a mediator-harmful group with 22 SNPs and a mediator-protective group with 54 SNPs. Among these SNPs, rs809302 in SLK significantly increased the risk of developing uterine fibroids by 3.92% through a mechanism other than age at menarche (P < 10−10), and rs371721345 in HLA-DOB was associated with a 2.70% decreased risk (P < 10−10) in the occurrence of uterine fibroids, mediated by age at menarche. These findings provide insights into the mechanism underlying the effect of genetic loci on uterine fibroids mediated by age at menarche.
LIMITATIONS, REASONS FOR CAUTION
A potential issue is that the present study relied upon self-reported age at menarche and uterine fibroid information. Due to the experimental design, the consistency between self-reports and medical records for uterine fibroids in Taiwan cannot be checked. Fortunately, the literature support that self-reporting even years later remains a practical means for collecting data on menarche and uterine fibroids. We found that the impact of under-reporting of uterine fibroids is less in our study. In addition, the rate of reporting a diagnosis of uterine fibroids was within the rates of medical diagnosis based on national health insurance data. Future work investigating the consistency between self-reports and medical records in Taiwan can remedy this issue.
WIDER IMPLICATIONS OF THE FINDINGS
This study is the first to investigate whether and to what extent age at menarche mediates the causal effects of genetic variants on uterine fibroids by using genome-wide causal mediation analysis. By treating age at menarche as a mediator, this report provides an insight into the genetic risk factors for developing uterine fibroids. Thus, this article represents a step forward in deciphering the role of intermediated risk factors in the genetic mechanism of disease.
STUDY FUNDING/COMPETING INTEREST(S)
This work was supported by the China Medical University, Taiwan (CMU110-ASIA-13 and CMU107-Z-04), the Ministry of Science and Technology, Taiwan (MOST 110-2314-B-039-058) and the International Joint Usage/Research Center, the Institute of Medical Science, the University of Tokyo, Japan (K2104). The authors have no competing interests.
TRIAL REGISTRATION NUMBER
N/A.
PDF
