Related Covariates Research Articles

Sexual coercion and health-risk behaviors among urban Chinese high school students

Objective To determine the association between health-risk behaviors and a history of sexual coercion among urban Chinese high school students. Design A cross-sectional study was performed among 109,754 high school students who participated in the 2005 Chinese Youth Risk Behavior Survey. Data were analyzed for 5,215 students who had experienced sexual intercourse (1,483 girls, 3,732 boys). Multivariate logistic regression was used to determine the relationship between sexual coercion and the related covariates, and data were stratified by gender. Results Of those students who had had sexual intercourse, 40.9% of the females and 29.6% of the males experienced sexual coercion (p<0.01). When analyses controlled for demographic characteristics, in the study sample, that is, students who had sexual intercourse, drug use (odds ratios [OR], 2.44), attempted suicide (OR, 2.30), physical abuse (OR, 1.74), binge drinking (OR, 1.62), verbal abuse (OR, 1.29), experience of being drunk (OR, 0.68), and smoking of cigarettes (OR, 0.52) were related to a history of sexual coercion. Patterns of health-risk behaviors also differed among female and male students who had experienced sexual coercion. Conclusions Sexual coercion is associated with health-risk behaviors. Initiatives to reduce the harm associated with sexual coercion among high school students are needed.

Coherence Structure and Its Application in Mortality Forecasting

Interaction terms expressed as products of covariates may prove useful in regression problems. A graphical method is presented for identifying potentially useful interaction and related covariates in modeling mortality time series using lagged coherence, a nonlinear extension of the squared coherence, and its residual coherence offshoot. The identified covariates are tested for their significance within a regression model using quasi-likelihood and log link.

Plant functional types in Earth system models: past experiences and future directions for application of dynamic vegetation models in high-latitude ecosystems.

Earth system models describe the physical, chemical and biological processes that govern our global climate. While it is difficult to single out one component as being more important than another in these sophisticated models, terrestrial vegetation is a critical player in the biogeochemical and biophysical dynamics of the Earth system. There is much debate, however, as to how plant diversity and function should be represented in these models. Plant functional types (PFTs) have been adopted by modellers to represent broad groupings of plant species that share similar characteristics (e.g. growth form) and roles (e.g. photosynthetic pathway) in ecosystem function. In this review, the PFT concept is traced from its origin in the early 1800s to its current use in regional and global dynamic vegetation models (DVMs). Special attention is given to the representation and parameterization of PFTs and to validation and benchmarking of predicted patterns of vegetation distribution in high-latitude ecosystems. These ecosystems are sensitive to changing climate and thus provide a useful test case for model-based simulations of past, current and future distribution of vegetation. Models that incorporate the PFT concept predict many of the emerging patterns of vegetation change in tundra and boreal forests, given known processes of tree mortality, treeline migration and shrub expansion. However, representation of above- and especially below-ground traits for specific PFTs continues to be problematic. Potential solutions include developing trait databases and replacing fixed parameters for PFTs with formulations based on trait co-variance and empirical trait-environment relationships. Surprisingly, despite being important to land-atmosphere interactions of carbon, water and energy, PFTs such as moss and lichen are largely absent from DVMs. Close collaboration among those involved in modelling with the disciplines of taxonomy, biogeography, ecology and remote sensing will be required if we are to overcome these and other shortcomings.

Imputing Forest Structure Attributes from Stand Inventory and Remotely Sensed Data in Western Oregon, USA

Imputation is commonly used to assign reference stand observations to target stands based on covariate relationships to remotely sensed data to assign inventory attributes across the entire landscape. However, most remotely sensed data are collected at higher resolution than the stand inventory data often used by operational foresters. Our primary goal was to compare various aggregation strategies for modeling and mapping forest attributes summarized from stand inventory data, using predictor variables derived from either light detection and ranging (LiDAR) or Landsat and a US Geological Survey (USGS) digital terrain model (DTM). We found that LiDAR metrics produced more accurate models than models using Landsat/USGS-DTM predictors. Calculating stand-level means of all predictors or all responses proved most accurate for developing imputation models or validating imputed maps, respectively. Developing models or validating maps at the unaggregated scale of individual stand subplots proved to be very inaccurate, presumably due to poor geolocation accuracies. However, using a sample of pixels within stands proved only slightly less accurate than using all available pixels. Furthermore, bootstrap tests of similarity between imputations and observations showed no evidence of bias regardless of aggregation strategy. We conclude that pixels sampled from within stands provide sufficient information for modeling or validating stand attributes of interest to foresters.

Assessing inter- and intra-individual cognitive variability in patients at risk for cognitive impairment: the case of minimal hepatic encephalopathy

Recent evidence reveals that inter- and intra-individual variability significantly affects cognitive performance in a number of neuropsychological pathologies. We applied a flexible family of statistical models to elucidate the contribution of inter- and intra-individual variables on cognitive functioning in healthy volunteers and patients at risk for hepatic encephalopathy (HE). Sixty-five volunteers (32 patients with cirrhosis and 33 healthy volunteers) were assessed by means of the Inhibitory Control Task (ICT). A Generalized Additive Model for Location, Scale and Shape (GAMLSS) was fitted for jointly modeling the mean and the intra-variability of Reaction Times (RTs) as a function of socio-demographic and task related covariates. Furthermore, a Generalized Linear Mixed Model (GLMM) was fitted for modeling accuracy. When controlling for the covariates, patients without minimal hepatic encephalopathy (MHE) did not differ from patients with MHE in the low-demanding condition, both in terms of RTs and accuracy. Moreover, they showed a significant decline in accuracy compared to the control group. Compared to patients with MHE, patients without MHE showed faster RTs and higher accuracy only in the high-demanding condition. The results revealed that the application of GAMLSS and GLMM models are able to capture subtle cognitive alterations, previously not detected, in patients' subclinical pathologies.

Possible determinants and spatial patterns of anaemia among young children in Nigeria: a Bayesian semi-parametric modelling

Anaemia is a global public health problem affecting both developing and developed countries with major consequences for human health and socioeconomic development. This paper examines the possible relationship between Hb concentration and severity of anaemia with individual and household characteristics of children aged 6-59 months in Nigeria; and explores possible geographical variations of these outcome variables. Data on Hb concentration and severity of anaemia in children aged 6-59 months that participated in the 2010 Nigeria Malaria Indicator Survey were analysed. A semi-parametric model using a hierarchical Bayesian approach was adopted to examine the putative relationship of covariates of different types and possible spatial variation. Gaussian, binary and ordinal outcome variables were considered in modelling. Spatial analyses reveal a distinct North-South divide in Hb concentration of the children analysed and that states in Northern Nigeria possess a higher risk of anaemia. Other important risk factors include the household wealth index, sex of the child, whether or not the child had fever or malaria in the 2 weeks preceding the survey, and children under 24 months of age. There is a need for state level implementation of specific programmes that target vulnerable children as this can help in reversing the existing patterns.

Modeling destructive earthquake casualties based on a comparative study for Turkey

The statistical modeling of destructive earthquakes is an indispensable tool for extracting information for prevention and risk reduction casualties after destructive earthquakes in a seismic region. The linear regression (LR) model can reveal the relation between casualty rate and related covariates based on earthquake catalog. However, if some covariates affect the casualty rate parametrically and some of them nonparametrically, the LR model may entail serious bias and loss of power when estimating or making inference about the effect of parameters. We suggest that semi-parametric beta regression (SBR), semi-parametric additive regression (SAR), and beta regression (BR) models could provide a more suitable description than the LR model to analyze the observed casualties after destructive earthquakes. We support this argument using destructive earthquakes occurred in Turkey between 1900 and 2012 having surface wave magnitudes five or more. The LR, SAR, BR, and SBR models are compared within the context of this data. The data strongly support that the SBR and SAR models can lead to more precise results than the BR and LR models. Furthermore, the SBR is the best model for the earthquake data since the beta distribution provides a flexible model that can be used to analyze the data involving proportions or rates. The results from this model suggest that the casualty rate depends on energy, damaged buildings, and the number of aftershocks of a destructive earthquake.

Are age‐related differences in the consequence of work injury greater when occupational physical demands are high?

To examine if age differences in the consequences of work injury are exacerbated when occupational physical demands are higher. A secondary analysis of workers' compensation claims in British Columbia (N = 373,672). Regression models examined the relationship between age and health care expenditures, days of wage replacement and the occurrence of long-term-disability following a work-related injury in occupations with lower and higher physical demands. Models were adjusted for individual and injury related covariates. Older age and higher occupational physical demands were associated with worse work-injury outcomes. The relationship between age and each outcome was not exacerbated when occupational physical demands were higher compared to when they were lower. Counter to our hypotheses age differences in health care expenditures were smaller among women in more demanding occupations. In this study, we found no evidence that the relationship between age and the consequences of work injury is exacerbated when physical occupational demands are high.

Growth among HIV-infected Children Receiving Antiretroviral Therapy in Dar es Salaam, Tanzania

We assembled a prospective cohort of 3144 human immunodeficiency virus (HIV) infected children aged <15 years initiating antiretroviral therapy (ART) in Dar es Salaam, Tanzania. The prospective relationships of baseline covariates with growth were examined using linear regression models. ART led to improvement in mean weight-for-age (WAZ), height/length-for-age (HAZ) and weight-for-length or body mass index (WLZ/BMIZ) scores. However, normal HAZ values were not attained over an average follow-up of 17.2 months. After 6 months of ART, underweight (P < 0.001), low CD4 count or percent (P < 0.001), stavudine containing regimens (P = 0.05) and advanced WHO disease stage (P < 0.001) at ART initiation were associated with better WAZ scores. Age >5 years on the other hand was associated with less increase in WAZ score after 6 months of ART (P < 0.001). These findings suggest that although ART improved the growth of the HIV-infected children in Tanzania, adjunct nutritional interventions may be needed to ensure that the growth of these children is optimized to the greatest extent possible.

Relationship between Individual Characteristics and Social Network: Its Localization Development

After reviewing the effect of individual characteristics on social network, we found that little research has been conducted on:(1) the underlying mechanism of individual characteristics on social network, the moderation effect of context variables and the interaction effect of individual characteristics on social network;(2) the dynamic and covariant relationship between individual characteristics and social network;(3) the unique individual characteristics in Chinese context, especially characteristics about Chinese personalities and "Quan-Zi". Due to the importance of localized personality theory and social network research, future research should focus on localization and covariant relationship between individual characteristics and social network.

Association between leg length-to-height ratio and metabolic syndrome in Chinese children aged 3 to 6 years

ObjectiveThe aim of this study is to investigate the association between leg-length-to-height ratio (LLHR) and metabolic syndrome (MetS) among Chinese children. Methods1236 children (619 obese and 617 nonobese children) aged 3–6years participated in a cross-sectional survey in 2005 in Tianjin, China. Information on body adiposity, metabolic traits, and related covariates was obtained using a standardized protocol. LLHR was calculated as the ratio of leg length to stature. ResultsIn the multivariable logistic regression analyses, compared with those in the lowest quartile, odds ratios (OR) and 95% confidence intervals (CI) of MetS among children in the second through the highest quartiles of LLHR Z-score were 0.89 (95% CI, 0.64–1.25), 0.45 (95% CI, 0.32–0.63), and 0.37 (95% CI, 0.26–0.53), respectively, (Pfor trend<0.0001 across LLHR Z-score quartiles). Compared with children with both higher levels of LLHR and lower levels of adipose indices, the corresponding ORs of MetS for those with both lower levels of LLHR and higher levels of anthropometric indices were 4.51 (95% CI, 3.08–6.62) for BMI Z-score, 3.86 (95% CI, 2.60–5.73) for waist circumference, and 2.75 (95% CI, 1.85–4.10) for waist-to-hip ratio, respectively. ConclusionsGreater LLHR is inversely associated with MetS in Chinese children.

Relationship between blood levels of methyl donor and folate and mild cognitive impairment in Chinese patients with type 2 diabetes: a case-control study

Type 2 diabetes is a risk factor for Alzheimer’s disease and mild cognitive impairment. Folate insufficiency fosters a decline in the sole methyl donor, S-adenosylmethionine, and decreases methylation potential, which is associated with Alzheimer’s disease in non-diabetic patients. However, little is known in diabetic patients. We analyzed plasma levels of S-adenosylmethionine, S-adenosylhomocysteine and serum level of folate in 100 elderly type 2 diabetic patients with and without mild cognitive impairment. S-adenosylmethionine/S-adenosylhomocysteine ratio was used to reflect the methylation potential. Patients with mild cognitive impairment had significantly lower levels of S-adenosylmethionine, folate and S-adenosylmethionine/S-adenosylhomocysteineratios. Furthermore, logistic regression analysis indicated the plasma S-adenosylmethionine, S-adenosylmethionine/S-adenosylhomocysteine ratio and serum folate (OR, 0.96, 0.698, 0.72, respectively; p<0.05) were negatively associated with risk of mild cognitive impairment, even after adjusting for related covariates. In addition, folate level was positively correlated with S-adenosylmethionine and the S-adenosylmethionine/S-adenosylhomocysteine ratio (r = 0.38, 0.46, respectively; p<0.05) among patients within the middle tertile of folate levels (6.3–9.1 µg/L). These findings indicate mild cognitive impairment is associated with lower levels of S-adenosylmethionine, folate and weakened methylation potential; plasma S-adenosylmethionine and methylation potential may be predicted by serum folate within a suitable range of folate concentrations in diabetic patients.

Intraclass Correlations and Covariate Outcome Correlations for Planning Two- and Three-Level Cluster-Randomized Experiments in Education.

Cluster-randomized experiments that assign intact groups such as schools or school districts to treatment conditions are increasingly common in educational research. Such experiments are inherently multilevel designs whose sensitivity (statistical power and precision of estimates) depends on the variance decomposition across levels. This variance decomposition is usually summarized by the intraclass correlation (ICC) structure and, if covariates are used, the effectiveness of the covariates in explaining variation at each level of the design. This article provides a compilation of school- and district-level ICC values of academic achievement and related covariate effectiveness based on state longitudinal data systems. These values are designed to be used for planning group-randomized experiments in education. The use of these values to compute statistical power and plan two- and three-level group-randomized experiments is illustrated. We fit several hierarchical linear models to state data by grade and subject to estimate ICCs and covariate effectiveness. The total sample size is over 4.8 million students. We then compare our average of state estimates with the national work by Hedges and Hedberg.

Overall and Disease-Specific Outcomes Following T-Cell Replete Haploidentical Transplants Using Post-Transplant Cyclophosphamide: An Analysis Of 115 Patients Transplanted At a Single Center

Many patients with hematologic malignancies who may benefit from the curative potential of allogeneic hematopoietic cell transplantation (allo-HCT) will lack a conventional matched sibling (MRD) or matched unrelated (MUD) donor. An HLA-haploidentical related donor is available for nearly all such patients. Attempts to use haploidentical donors that employ stringent ex-vivo T-cell depletion to abrogate GVHD are associated with poor immune reconstitution and high rates of graft rejection. In contrast, the recent use of T-cell replete grafts with post-transplant cyclophosphamide (haplo-ptCy) has resulted in effective control of alloreactivity with low rates of post-transplant infection, graft-rejection and non-relapse mortality using both non-ablative and ablative regimens (Luznik at al BBMT 2008, Solomon et al BBMT 2012.). We recently demonstrated similar outcomes for 53 haplo-PtCy to contemporaneous MRD and MUD transplants performed at our enter (Bashey et al JCO 2013). However there are limited data with respect to the outcome of specific hematologic malignancies using haplo-ptCy. For this study all consecutive patients who underwent haplo-ptCy as a first allogeneic transplant at our center between Oct 2005 and May 2013 (n=115) were included [median age 50 (20-74); M 57%, F 43%; diagnosis- AML(38), ALL(20), CLL(13), NHL(13) HL(10), CML+MPS (12), MDS (7) others (2); graft-BM(58%),PBSC(42%); prior autotranplant(18%); CIBMTR disease risk score high(35%), intermediate(32%) low(33%). Sorror Comorbidity Index-median =1(0-5)]. Preparative regimens used included non-myeloablative [fludarabine 30 mg/m2/d d -6 to -2; TBI 200cGy d-1, Cy 14.5 mg/kg/d on d-6,-5 and 50mg/kg/d on d+3,+4, n=73 ] and myeloablative [regimen 1- same as non-ablative except busulfan 110-130 mg/m2 /d i.v. on d-7 to-3 instead of TBI n=21; regimen 2- fludarabine 30 mg/m2/d d -7 to -5, TBI 150cGy bid x 8 doses (total dose =1200cGy) and Cy 50mg/kg/d d+3,+4 n=21]. Patient characteristics and outcome data were collected prospectively and obtained from our comprehensive database. Survival and disease-free survival (DFS) were estimated using the Kaplan-Meier method. Cumulative incidence of non-relapse mortality (NRM) was calculated using relapse as a competing risk. GVHD was assessed and documented prospectively by a dedicated GVHD nurse. Median CD34+ and CD3+cell doses infused were 4.21 x 10e6/kg (0.84-8.13) and 5.81 x 10e7/kg (1.4-59.5). median time to ANC > 500/mm3 and platelets > 20,000/mm3 were 16 and 26 d. Median % donor chimerism on d 30 were 100% and 100% for PB CD3+ and CD33+ cells respectively. With a median follow-up for living patients of 20 months, the 12 m and 24 m cumulative incidences (CI) of NRM were 10% and 18% and of relapse/progression were 25% and 25% respectively (Fig.1). CI of acute GVHD gd II-IV and gd III-IV at 6 m were 34% and 13% respectively, and of extensive and severe chronic GVHD at 12 m were 40% and 9% respectively. Estimated 12m and 24m probabilities of survival were 73% and 56% and of DFS were 65% and 57% respectively. On a Cox multivariate analysis assessing patient, disease and transplant related covariates, high risk CIBMTR disease score (HR 1.88, p=0.043) and acute GVHD gd 3-4 HR=2.54, p=0.0145) were the only factors significantly associated with survival. For the most frequently treated diagnoses - AML, ALL and mature lymphoid malignancies (CLL, NHL, HL) - the respective estimated 24 m probabilities of DFS were 61%, 58% and 50% respectively (Fig.2). These data demonstrate that haplo-ptCy transplants represent a safe and effective alternative for most patients who may benefit from allo-HCT but lack a conventional donor. They should be offered as standard of care in such patients. [Display omitted] Disclosures:No relevant conflicts of interest to declare.

Computer Model Calibration Using the Ensemble Kalman Filter

Computer model calibration is the process of determining input parameter settings to a computational model that are consistent with physical observations. This is often quite challenging due to the computational demands of running the model. In this article, we use the ensemble Kalman filter (EnKF) for computer model calibration. The EnKF has proven effective in quantifying uncertainty in data assimilation problems such as weather forecasting and ocean modeling. We find that the EnKF can be directly adapted to Bayesian computer model calibration. It is motivated by the mean and covariance relationship between the model inputs and outputs, producing an approximate posterior ensemble of the calibration parameters. While this approach may not fully capture effects due to nonlinearities in the computer model response, its computational efficiency makes it a viable choice for exploratory analyses, design problems, or problems with large numbers of model runs, inputs, and outputs.

Fear of Violent Consequences and Condom Use Among Women Attending an STD Clinic

Intimate partner violence (IPV) has been associated with risk for HIV infection. This cross-sectional study tested the hypothesis that fear of violent consequences when negotiating condom use mediated the relation between IPV and condom use. Participants (n = 478) were recruited between March 1, 2004 and June 30, 2006 from a public clinic that treats sexually transmitted diseases in upstate New York as part of a randomized controlled trial. They completed an audio, computer-assisted, self-administered questionnaire with items on risky sexual behavior, intimate relationships, and related covariates and confounding variables. Seventeen percent of the sample reported IPV in the past three months. Recent IPV was associated with fear of violent consequences to requests for condom use, and such fear was associated with inconsistent condom use. Women who reported IPV also reported greater difficulties in negotiating safer sex behaviors with their abusers. So fear of violent consequences appeared to hinder their ability to protect themselves against HIV infection. The results were consistent with fear of violent consequences mediating the relationship between IPV and condom use. Health care providers involved in HIV prevention and sexual risk reduction interventions need to address IPV and, more specifically, fear of IPV when negotiating safer sex as part of their services for providing more comprehensive care to the women they serve.

Dispersion Pattern of PGE and Ag in Chromiferous Ultramafic Suite of Rocks in Sukinda Valley, India

AbstractPotential chromite ore deposits of India are situated in Sukinda, Odisha, which may also be considered as a potential resource for platinum group elements (PGEs). This paper reports on PGE geochemistry in twenty six samples covering chromite ores, chromitites and associated ultramafic rocks of the Sukinda ultramafic complex. Platinum group element contents range from 213 to 487 ppb in the chromite ore body, from 63 to 538 ppb in rocks that have chromite dendrites or dissemination and from 38 to 389 ppb in associated olivine–peridotite, serpentinite, pyroxenite and brecciated rocks. The PGEs are divided into two sub‐groups: IPGE (Ir, Os, and Ru) and PPGE (Pd, Pt, and Rh) based on their chemical behaviour. The IPGE and PPGE in these three litho‐members show a contrasting relationship e.g. average IPGE content decreases from chromite to chromitite and associated rocks while PPGE increases in the same order. Appreciable Ag in chromitite (270–842 ppb) is recorded. Positive correlation between IPGE with Cr2O3 and with Al2O3 is observed while these are negatively correlated with MgO. Covariant relationships between Au and Mg in rocks devoid of chromite and between Ag and Fe in chromitite sample are observed. Chromite in all seams and some chromitite samples exhibit an IPGE‐enriched chondrite normalized pattern while PPGE are highly fractionated and show a steep negative slope, thereby indicating that PGE in the parental melt fractionates and IPGE‐compatible elements prefer to settle with chromite. The rocks devoid of chromite and rocks containing accessory chromite exhibit a nearly flat pattern in chondrite‐normalized PGE plots and this suggests a limited fractionation of PGE in these rocks. Variation in the distribution pattern of PGE and Ag in three typical litho‐members of the Sukinda Valley may be related to multiple intrusion of ultramafic magma, containing variable volume percentage of chromite.

The role of diffusion of responsibility, responsibility, and attitude in willingness to donate to a world hunger-related charity

The role of diffusion of responsibility, responsibility, and attitude in willingness to donate to a world hunger-related charity was examined. Attitude towards helping others and attitude towards charitable organizations were the two attitude measures considered. In accordance with earlier studies on diffusion of responsibility, participation in the study was either as a member of a group or alone. Group versus non-group was established using priming techniques in a virtual environment. The intervention consisted of the viewing of a video from the World Food Programme website. Findings determined that the video exerted a significant effect on participants whether in the virtual group or alone, and on all variables with the exception of attitude towards charitable organizations, suggesting that such a video is an effective tool for fundraisers. It was also determined that, although responsibility was not diffused, it significantly impacted willingness to donate, with personal responsibility producing a more robust effect over social responsibility. Attitude towards helping others significantly impacted willingness to donate and also exhibited a covariate relationship with personal responsibility. Both of these findings suggest that a more personal, attitudinal type of responsibility is what promotes helping behaviors, rather than social responsibility that arises from social norms. These findings are supportive of the theory on prosocial behaviors and confirm the contributory role of responsibility when engaging in prosocial behaviors.

Latitude or biogeographic breaks? Determinants of phenotypic (co)variation in fitness-related traits in Betaeus truncatus along the Chilean coast

Ectothermal organisms distributed along environmental gradients in a wide geographical distribution display extensive phenotypic variation. This is particularly pervasive along latitudinal clines, which are linked to gradual changes in environmental factors. Widespread species may also be distributed among biogeographic breaks, which in contrast to smooth clines, often show abrupt changes in phenotypic traits. In species with widespread latitudinal distribution that also encompass important biogeographical breaks, it is not clear which of those factors prevails on shaping the phenotypic variation or if some traits are particularly more sensitive to one or the other. To evaluate this, we measured 4 fitness-related traits in 6 populations of the intertidal snapping shrimp Betaeustruncatus, as its distribution along Chile expands over 40° in latitude and three major biogeographical provinces. Here, we statistically evaluated the role of both, latitude and biogeographic breaks, on mean population values of fitness-related traits but also on the variances and covariances (i.e., P-matrix) between them. Overall, our results (1) indicate that latitude is more important than breaks in shaping the phenotypic variation of most of these fitness-related traits, (2) show that the differences in the variance–covariance relationship among traits between the extremes of the gradient arises from gradual increases in variance and rather sharp changes in covariance at mid-latitudes and (3) show that at present, it is difficult to unambiguously determine whether natural selection or plasticity is responsible for the observed pattern in means, variances and covariances and only further work might disentangle these possibilities.

Developing Choropleth Maps of Parameter Results for Quantile Regression

Local regression methods can provide specific information about individual observations (places) in spatial analysis that is often useful in understanding non-stationary covariate relationships. Geographically weighted regression (GWR) is one widely used local regression model whose parameter estimates are mapped as continuous or discrete surfaces. A less frequently used local regression method used in spatial analysis is quantile regression (QR). One drawback to the use of QR is its restricted visualization possibilities; currently, regression parameter summaries for QR are visualized only as line graphs that plot parameter estimates against quantile levels. This article presents a method that also permits the spatial display of these values in a choropleth map so that QR can be added to the cartographic repertoire of exploratory data models.