Stata Tip 87: Interpretation of Interactions in Nonlinear Models

Buis, Maarten L.

doi:10.1177/1536867x1001000211

Cited by 407 publications

(226 citation statements)

References 6 publications

Supporting

Mentioning

222

Contrasting

Unclassified

Order By: Relevance

“…Control variables of gender, race, family income, and maternal education were modeled as categorical variables; the remaining continuous predictors were mean-centered first. All analyses were conducted in Stata statistical software using the zinb command, and the results were further probed and plotted using the margins and marginsplot command (Buis, 2010;Jann, 2013;StataCorp., 2013). The results were plotted following the standard procedures for probing significant interactions in regression-based models (Bauer & Curran, 2005;Buis, 2010;Curran, Bauer, & Willoughby, 2006;Jann, 2013;Preacher, Curran, & Bauer, 2006).…”

Section: Discussionmentioning

confidence: 99%

Person-Environment Interactions and Adolescent Substance Use: The Role of Sensation Seeking and Perceived Neighborhood Risk

Andreas

Watson

2016

Journal of Child & Adolescent Substance Abuse

View full text Add to dashboard Cite

Aims: This longitudinal study investigated person-environment interactions and adolescent substance use by examining whether, and if so how, early temperamental quality of sensation seeking interacted with current neighborhood characteristics to shape underage smoking, drinking, and marijuana use. Sample: Participants consisted of a community-representative sample of 352 adolescents. Methods: Early temperament was assessed when adolescents were between seven and 13 years of age, and current neighborhood characteristics (as perceived by participants and their mothers) and current adolescent substance use (self-reported smoking, drinking, and marijuana use in the past 30 days) were assessed approximately three years later. Results: The results from a zeroinflated negative binomial (ZINB) regression revealed a potent negative effect of the perceived neighborhood risk, such that riskier neighborhoods were associated with both a greater probability for any substance use among adolescents, and a greater frequency of substance use among those who were engaging in these behaviors. High sensation seeking was a risk factor only for frequency of substance use among adolescent extant users, but not for the likelihood of any use. In addition, a significant interaction between sensation seeking and neighborhood risk revealed that adolescent sensation seekers engaged in smoking, drinking, and marijuana use primarily under conditions of elevated neighborhood risk and increased opportunity for such behaviors. Conclusion: Underage substance use was affected by synergistic effects between personal and community risk factors.

show abstract

Section: Discussionmentioning

confidence: 99%

Person-Environment Interactions and Adolescent Substance Use: The Role of Sensation Seeking and Perceived Neighborhood Risk

Andreas

Watson

2016

Journal of Child & Adolescent Substance Abuse

View full text Add to dashboard Cite

show abstract

“…The proportion of zeroes is lower in Lambeth than the London-wide average because Lambeth is a high-incidence borough. 33 Following Buis (2010), given that both interacted variables are binary, the average interaction effect on each margin can be calculated by: first, using the Tobit estimates to produce the conditional expected value of admissions for the four Lambeth × policy period (PP qy or P qy ) cells (e.g, Lambeth = 0, PP qy = 0; Lambeth = 1, PP qy = 0; Lambeth = 0, PP qy = 1; Lambeth = 0, PP qy = 0); and, second, taking the double difference of those conditional expected admission rates. The average interaction effect for the intensive margin in the postpolicy period is therefore equal to the following:…”

Section: A2 Tobit Estimatesmentioning

confidence: 97%

Policing cannabis and drug related hospital admissions: Evidence from administrative records

Kelly

Rasul

2014

Journal of Public Economics

View full text Add to dashboard Cite

a b s t r a c tWe evaluate the impact of a policing experiment that depenalized the possession of small quantities of cannabis in the London borough of Lambeth, on hospital admissions related to illicit drug use. To do so, we exploit administrative records on individual hospital admissions classified by ICD-10 diagnosis codes. These records allow the construction of a quarterly panel data set for London boroughs running from 1997 to 2009 to estimate the short and long run impacts of the depenalization policy unilaterally introduced in Lambeth between 2001 and 2002. We find that the depenalization of cannabis had significant longer term impacts on hospital admissions related to the use of hard drugs, raising hospital admission rates for men by between 40 and 100% of their pre-policy baseline levels. The impacts are concentrated among men in younger age cohorts. The dynamic impacts across cohorts vary in profile with some cohorts experiencing hospitalization rates remaining above pre-intervention levels three to four years after the depenalization policy is introduced. We combine these estimated impacts on hospitalization rates with estimates on how the policy impacted the severity of hospital admissions to provide a lower bound estimate of the public health cost of the depenalization policy.Crown

show abstract

“…In the probit model, however, the estimated coefficients show the multiplicative effect, or a ratio change of the dependent variable for each unit change in explanatory variable (Buis 2010). Therefore, for analysis the estimated coefficients needed to be transformed into marginal effects.…”

Section: Data Analysesmentioning

confidence: 99%

Willingness to pay for marine-based tourism in the Ponta do Ouro Partial Marine Reserve, Mozambique

Daly

Fraser

Snowball

2015

African Journal of Marine Science

View full text Add to dashboard Cite

Marine and coastal ecosystems face widespread degradation largely because market failure hides the economic value of the goods and services they provide. Marine protected areas (MPAs) can serve as structures that ensure the continued functioning of marine and coastal ecosystem goods and services. Yet, to be effective and sustainable, MPAs must be able to prove their economic worth and generate revenue. User-fee systems are used commonly to partially finance multi-use MPAs. This study applies contingent valuation as a method of economic valuation within an MPA in southern Mozambique. Using a payment card and questionnaire from November 2012 to April 2013, this study determined the willingness to pay (WTP) of three user groups for access to the Ponta do Ouro Partial Marine Reserve. The study also investigated the potential for the reserve to increase revenues for conservation through the implementation of a user fee for marine-based activities. Probit and ordinary least squares (OLS) regressions were used to determine the effects of various independent variables on WTP. The OLS model found income, African residency and environmental awareness to be significant factors that influenced visitors' WTP for access to the reserve. The mean WTP was R 43.75 (South African rands) per person per day. Using data supplied by the reserve management, conservative estimated annual revenues based on the implementation of this fee amount ranged between R 1.46 million and R 3.3 million.

show abstract

Stata Tip 87: Interpretation of Interactions in Nonlinear Models

Cited by 407 publications

References 6 publications

Person-Environment Interactions and Adolescent Substance Use: The Role of Sensation Seeking and Perceived Neighborhood Risk

Person-Environment Interactions and Adolescent Substance Use: The Role of Sensation Seeking and Perceived Neighborhood Risk

Policing cannabis and drug related hospital admissions: Evidence from administrative records

Willingness to pay for marine-based tourism in the Ponta do Ouro Partial Marine Reserve, Mozambique

Contact Info

Product

Resources

About