CHAID Research Articles

Food security and agricultural challenges in West-African rural communities: a machine learning analysis

ABSTRACT This article investigated household-level food security for Ghana, Liberia, and Senegal. Different agroclimatic, ecological, social, and farming conditions in West Africa were represented. Using data-driven Random Forest and Chi-Square Automatic Interaction Detection (CHAID) decision tree methodology, this study classified 644 Ghanaian, 323 Liberian, and 510 Senegalese households for comparison and interpretation on food security. The predictors growing Liberian and Senegalese decision trees imply community support, diverse selling channels outside villages, resolving the dispute over farmland, and increasing community-level investment for food availability and access demonstrate household food security. Predictor importance on food security for Ghana highlighted the role of independent producers and food suppliers toward stability. Household food security or insecurity was distinguished by location-specific and gender-led households in Liberia and Senegal. Practically, the results presented a need to step-up agricultural education and extension based on an empirical field survey and its interpretations. The results can add considerations to the role of farming households as independent and individual suppliers and consumers to long-standing dimensions of food security, i.e., food availability, access, and stability.

Employing biochemical biomarkers for building decision tree models to predict bipolar disorder from major depressive disorder

BackgroundConventional biochemical parameters may have predictive values for use in clinical identification between bipolar disorder (BD) and major depressive disorder (MDD). MethodsThis study enrolled 2470 hospitalized patients with BD (n = 1333) or MDD (n = 1137) at reproductive age from 2009 to 2018 in China. We extracted 8 parameters, uric acid (UA), direct bilirubin (DBIL), indirect bilirubin (IDBIL), lactic dehydrogenase (LDH), free triiodothyronine (FT3), thyroid-stimulating hormone (TSH), high-density lipoprotein (HDL) and prealbumin of male, patients and 12 parameters, UA, DBIL, IBIL, LDH, FT3, TSH, glutamic-pyruvic transaminase (GPT), white blood cell (WBC), alkaline phosphatase (ALP), fasting blood glucose (FBG), triglyceride and low-density lipoprotein (LDL) of female patients. Backward stepwise multivariate regression analysis and the Chi-Square Automatic Interaction Detection (CHAID) segmentation analysis via SPSS Decision Tree were implemented to define the discrimination of BD and MDD. ResultsDBIL was extracted as the first splitting variable, with LDH and IBIL as the second, TSH and prealbumin as the third in the model of male patients (p-value < .05). For the model of female patients, DBIL was also extracted as the first splitting variable, with UA, LDH, and IBIL as the second, triglyceride and FT3 as the third (p-value < .05). The predictive accuracies of the Decision Tree and multiple logistic regression models were similar (74.9% vs 76.9% in males; 74.4% vs 79.5% in females). ConclusionsThis study suggests the value of the Decision Tree models, which employ biochemical parameters as diagnostic predictors for BD and MDD. The CHAID Decision Tree identified that patients with concomitantly increased LDH, IBIL, and decreased DBIL could be in the group that showed the highest risk of being diagnosed as BD.

Prone Apprehension Relocation Test significantly correlates with radiological instability scores of the hip.

ABSTRACTRecently, there was a debate about whether borderline dysplastic hips should be treated surgically with hip arthroscopy or periacetabular osteotomy (PAO). Current studies recommend a classification into stable and unstable hips. Therefore, radiological scores have been described in recent years. Likewise, a new clinical stability test with the Prone Apprehension Relocation Test (PART) has been described. However, there has been no correlation between the modern radiological scores and the PART. We prospectively studied a consecutive group of patients who presented to our clinic. The PART and radiological scores were assessed in these patients. We divided the patients into a PART-positive and a PART-negative group and analyzed the associated clinical and radiological findings. Out of 126 patients (126 hips) included, 36 hips (29%) were evaluated as PART positive. There were significantly more females in the PART positive group (P = 0.005). Comparing the PART groups, significant differences (P < 0.0001) were found for the lateral center edge angle (LCEA), Femoro-Epiphyseal Acetabular Roof (FEAR) index, Gothic arch angle (GAA), anterior wall index (AWI), the occurrence of the upsloping lateral sourcil (ULS) and signs of acetabular retroversion. The correlation analysis showed an association between LCEA, FEAR index, GAA, AWI, ULS and the PART. A chi-square automatic interaction detection algorithm revealed that the strongest predictor of positive PART was the GAA. In conclusion, a high correlation between the PART and known radiological instability parameters was found. Consequently, a combination of clinical instability testing and radiological instability parameters should be applied to detect unstable hips.

Non-compliance with a nurse\u2019s advice to visit the primary care provider: an exploratory secondary analysis of the TRIAGE-trial

BackgroundDuring the cluster randomised TRIAGE-trial, a nurse advised 13% of low-risk patients presenting at an emergency department in Belgium to visit the adjacent general practitioner cooperative. Patients had the right to refuse this advice. This exploratory study examines the characteristics of refusers by uncovering the determinants of non-compliance and its impact on costs, as charged on the patient’s invoice.MethodsBivariate analyses with logistic regressions and T-tests were used to test the differences in patient characteristics, patient status, timing characteristics, and costs between refusers and non-refusers. A chi-square automatic interaction detection analysis was used to find the predictors of non-compliance.Results23.50% of the patients refused the advice to visit the general practitioner cooperative. This proportion was mainly influenced by the nurse on duty (non-compliance rates per nurse ranging from 2.9% to 52.8%) and the patients’ socio-economic status (receiving increased reimbursement versus not OR 1.37, 95%CI: 0.96 to 1.95). Additionally, non-compliance was associated (at the 0.10 significance level) with being male, not living nearby and certain reasons for encounter. Fewer patients refused when the nurse perceived crowding level as quiet relative to normal, and more patients refused during the evening. The mean cost was significantly higher for patients who refused, which was a result of more extensive examination and higher out-of-pocket expenses at the ED.ConclusionsThe nurse providing the advice to visit the general practitioner cooperative has a central role in the likelihood of patients’ refusal. Interventions to reduce non-compliance should aim at improving nurse-patient communication. Special attention may be required when managing patients with a lower socio-economic status. The overall mean cost was higher for refusers, illustrating the importance of compliance.Trial registrationThe trial was registered on registration number NCT03793972 on 04/01/2019.

Comparison of Montreal cognitive assessment and Mattis dementia rating scale in the preoperative evaluation of subthalamic stimulation in Parkinson's disease.

IntroductionThe preoperative evaluation of Parkinson’s Disease (PD) patients for subthalamic nucleus deep brain stimulation (STN-DBS) includes the assessment of the neuropsychological status of the patient. A widely used preoperative test is the Mattis Dementia rating scale (MDRS). However, the Montreal cognitive assessment (MoCA) has also been proven to be a sensitive, time-sparing tool with high diagnostic validity in PD. We evaluate the utility of the MoCA as a preoperative screening test for PD patients undergoing bilateral STN-DBS.MethodsIn this single-centre, retrospective study, we analysed pre- and postoperative assessments of MoCA, MDRS, Movement disorder society-Unified PD Rating Scale-motor examination, PD Questionnaire-39 and levodopa equivalent daily dose. Longitudinal outcome changes were analysed using paired t-test, Pearson’s correlation coefficient, linear regression and CHAID (chi-square automatic interaction detector) regression tree model.ResultsClinical motor and cognitive scores of 59 patients (61.05±7.73 years, 24 females) were analysed. The MoCA, but not the MDRS, identified significant postoperative cognitive decline in PD patients undergoing STN-DBS. The preoperative MoCA score correlated with postoperative quality of life improvement, whereas the MDRS did not. PD patients with a MoCA score ≤ 23 points had a significant decline of quality of life after DBS surgery compared to patients > 23 points.ConclusionThis study identifies the MoCA as an alternative test within the preoperative evaluation of PD patients for the detection of neuropsychological deficits and prediction of the postoperative improvement of quality of life.

Characteristics of early-career nurse researchers negatively impacted during the COVID-19 pandemic: a cross-sectional study

ObjectivesThis study aimed to identify typical patterns and determinants of early-career nurse researchers (ECNRs: age ≤45 years) who reported that their research activities were negatively impacted during the COVID-19 pandemic,...

Detection of subclinical keratoconus using a novel combined tomographic and biomechanical model based on an automated decision tree

Early detection of keratoconus is a crucial factor in monitoring its progression and making the decision to perform refractive surgery. The aim of this study was to use the decision tree technique in the classification and prediction of subclinical keratoconus (SKC). A total of 194 eyes (including 105 normal eyes and 89 with SKC) were included in the double-center retrospective study. Data were separately used for training and validation databases. The baseline variables were derived from tomography and biomechanical imaging. The decision tree models were generated using Chi-square automatic interaction detection (CHAID) and classification and regression tree (CART) algorithms based on the training database. The discriminating rules of the CART model selected metrics of the Belin/Ambrósio deviation (BAD-D), stiffness parameter at first applanation (SPA1), back eccentricity (Becc), and maximum pachymetric progression index in that order; On the other hand, the CHAID model selected BAD-D, deformation amplitude ratio, SPA1, and Becc. Further, the CART model allowed for discrimination between normal and SKC eyes with 92.2% accuracy, which was higher than that of the CHAID model (88.3%), BAD-D (82.0%), Corvis biomechanical index (CBI, 77.3%), and tomographic and biomechanical index (TBI, 78.1%). The discriminating performance of the CART model was validated with 92.4% accuracy, while the CHAID model was validated with 86.4% accuracy in the validation database. Thus, the CART model using tomography and biomechanical imaging was an excellent model for SKC screening and provided easy-to-understand discriminating rules.

Stakeholder engagement and SDGs: the role of social media in the European context

PurposeThe authors aim to investigate which social media posts' characteristics result in higher stakeholder engagement in the European context.Design/methodology/approachThe authors apply chi-square automatic interaction detection (CHAID) method to categorize social media posts based on their engagement levels. The authors conduct analysis on the tweets published in 2019 by companies that draw up and publish a sustainability report on Global Reporting (www.globalreporting.org). The final sample is based on 3,045 tweets from 173 firms of different industries and countries.FindingsThe authors find that tweet characteristics, such as content type, industry type and country of a firm, significantly influence stakeholder engagement rate.Practical implicationsThe findings are useful for practitioners and stakeholders. Firm management should consider the most important Sustainable Development Goals (SDGs) concerning their business activities for engaging stakeholders and contributing to sustainable development.Originality/valueThe authors consider stakeholder theory and contribute to stakeholder engagement and sustainability debate. For this purpose, the authors focus on firms' communication of SDGs through social media. To the best of the authors’ knowledge, this is the first study which considers the SDGs’ communication to determine the most significant SDGs resulting in higher stakeholder engagement.

Developing a Clinical Prediction Rule for Gait Independence at Discharge in Patients with Stroke: A Decision-Tree Algorithm Analysis

ObjectivesTo develop a clinical prediction rule (CPR) for gait independence at discharge in patients with stroke, using the decision-tree algorithm and to investigate the usefulness of CPR at admission to the rehabilitation ward. Materials and MethodsWe included 181 subjects with stroke during the postacute phase. The Chi-squared automatic interaction detection analysis method with 10-fold cross-validation was used to develop two CPRs; CPR 1 using easily obtainable data available at admission; CPR 2 using easily obtainable data available 1 month after admission, for prediction of gait independence at discharge. ResultsThe degree of independence of toileting was extracted as a first node in the development of two CPRs to predict gait independence at discharge. CPR 1 included the presence of delirium. CPR 2 included problem-solving abilities. The accuracy and area under the curve of CPR 1 were 84.5% and 0.911, respectively; those of CPR 2 were 89.0% and 0.958, respectively. ConclusionsToileting independence is a key factor in predicting the gait independence for the discharge of patients with stroke during the postacute phase. Early intervention, during the acute phase, for delirium and cognitive decline, as well as for toileting, increases the possibility of gait independence at discharge.

Socio-economic determinants of land use and land cover change in South-Kivu wetlands, eastern D.R. Congo: Case study of Hogola and Chisheke wetlands

In the South-Kivu province, wetlands are mainly converted into farmlands to ensure food and income security among rural populations. This study aimed at assessing the land use and land cover (LULC) change for the last three decades (1990–2020) in South-Kivu wetlands; mainly the Hogola and Chisheke, in Walungu territory, eastern Democratic Republic of Congo (DRC). Socio-economic determinants of LULC in the two wetlands were also assessed. Google earth (CNI/airbus) and airborne images were used for LULC while socio-economic data were collected through a survey questionnaire from 369 households. The Chi-squared Automatic Interaction Detector (CHAID) model allowed assessing factors determining conversion of wetlands' use. For the last three decades, there were 30 and 40% decreases in the acreage covered by Hogola and Chisheke wetlands, respectively, as a result of rapid farmland expansion and brick making. Farmers perceived wetlands as wastelands and thus unfit for agriculture and brick making activities. These two human activities provided significant monetary benefits to wetland users though the profitability of agriculture was dependent on practiced crops. The conversion of wetlands into farmlands was driven by annual household income, wetland utilization patterns, households' main activity and the seniority in exploiting wetlands. The perception of wetlands as degraded and wastelands, the exploited acreage and the farmer gender had also influenced significantly wetland conversions. On the other hand, perceiving wetlands as wastelands depended on the age of the household head, the inclination to brick making activities, the household's main activity and crop, and the wetland traditional/cultural values in the South-Kivu province. Though generating significantly higher annual incomes than any other land use in target wetlands, the conversion of wetlands for agriculture and brick making, under the current practices, is unsuitable for their sustainable management (use), as recommended to achieve the RAMSAR objectives of “wise use” of wetlands.

Building social capital in cruise travel via social network sites

ABSTRACT The purpose of this study is to investigate what type of Facebook posts help cruise lines build bridging and bonding social capital. The study applies the Chi-Square Automatic Interaction Detection (CHAID) method to identify which types of posts establish bridging and bonding social capital. The analysis is conducted on an international cruise line’s official Facebook posts posted between 1 January 2018 and 1 January 2020 before the Covid-19 pandemic. The results highlight that media type, embedding passenger motivation, and a ship image help establish both bridging and bonding social capital, while content type helps establish bridging social capital. The paper is original because it helps understand how cruise lines can improve bonding and bridging social capital via social media. The paper also enhances understanding of social capital theory in the travel industry by investigating the relationship between Facebook post types and social capital in cruise shipping.

Long-Term Association Between Disease Activity and Disability in Early Axial Spondyloarthritis: Results From a Prospective Observational Study of Inflammatory Back Pain.

Our primary objective was to study the long-term association between disease activity and disability in axial spondyloarthritis (SpA). Our secondary objective was to define patient profiles according to their level of disability. We analyzed data collected during the first 5 years of follow-up of a large early axial SpA cohort, the Devenir des Spondylarthropathies Indifferénciées Récentes (DESIR) cohort. Multivariable models were built to study the association between the Health Assessment Questionnaire for Ankylosing Spondylitis (HAQ-AS) and the Ankylosing Spondylitis Disease Activity Score with C-reactive protein (ASDAS-CRP), adjusting for potential confounders. Hierarchical multivariable analysis was conducted using the chi-square automatic interaction detector (CHAID) method, to help determine how variables best cluster to explain HAQ-AS. Data from 644 patients and 5,152 visits were analyzed. HAQ-AS was longitudinally, independently, and positively associated with ASDAS-CRP (adjusted B [adjB] 0.205 [95% confidence interval (95% CI) 0.187, 0.222]), the enthesitis score (adjB 0.011 [95% CI 0.008, 0.015]), the Bath Ankylosing Spondylitis Metrology Index (adjB 0.087 [95% CI 0.069, 0.105]), and female sex (adjB 0.172 [95% CI 0.120, 0.225]). The CHAID decision tree revealed ASDAS-CRP as the first variable with discriminative power on HAQ-AS. The cutoffs that separated different patient disability profiles were obtained. Disease activity contributes longitudinally to disability and is hierarchically superior to any other variable in explaining this health domain. Enthesitis and spinal mobility are also key drivers of disability in early axial SpA. ASDAS-CRP cutoffs that separated different patient disability profiles largely mimicked the cutoffs previously defined for ASDAS-CRP disease activity states.

Prediction of Chronic Kidney Disease-A Machine Learning Perspective

Chronic Kidney Disease is one of the most critical illnesses nowadays and proper diagnosis is required as soon as possible. Machine learning technique has become reliable for medical treatment. With the help of a machine learning classifier algorithms, the doctor can detect the disease on time. For this perspective, Chronic Kidney Disease prediction has been discussed in this article. Chronic Kidney Disease dataset has been taken from the UCI repository. Seven classifier algorithms have been applied in this research such as artificial neural network, C5.0, Chi-square Automatic interaction detector, logistic regression, linear support vector machine with penalty L1 &amp; with penalty L2 and random tree. The important feature selection technique was also applied to the dataset. For each classifier, the results have been computed based on the following factors given below: (i) full features, (ii) correlation-based feature selection, (iii) Wrapper method feature selection, (iv)Least absolute shrinkage and selection operator regression, (v) synthetic minority over- sampling technique with least absolute shrinkage and selection operator regression selected features, (vi) synthetic minority oversampling technique with full features. From the results, it is marked that LSVM with penalty L2 is giving the highest accuracy of 98.86% in synthetic minority over-sampling technique with full features. Along with accuracy, precision, recall, F- measure, area under the curve and GINI coefficient have been computed and compared results of various algorithms have been shown in the graph. Least absolute shrinkage and selection operator regression selected features with synthetic minority over-sampling technique gave the best after synthetic minority over-sampling technique with full features. In the synthetic minority over-sampling technique with least absolute shrinkage and selection operator selected features, again linear support vector machine gave the highest accuracy of 98.46%. Along with machine learning models one deep neural network has been applied on the same dataset and it has been noted that deep neural network achieved the highest accuracy of 99.6%.

Turkey’s Coffee Adventure: An Analysis of Coffee Experience Typologies and Business Type Preferences

The rise in the coffee consumption, as one of the most popular beverages in the world, has opened the door to “coffee shops” in their modern form. The present study analyses the type of Turkish consumers who prefer national and international coffee shops based on their demographic characteristics and motivations. A total of 448 visitors to coffee shops during the Istanbul Coffee Festival filled out the developed survey form. 411 of these were used for data analysis. The data obtained in the research were divided into four groups through a Chi-squared Automatic Interaction Detection (CHAID) Analysis and Correspondence Analysis, depending on the demographic and motivational characteristics of the consumers with experience of national and international coffee shops. Consumers attribute importance to the atmosphere, image, coffee quality and location of coffees shops, while the target market of coffee shops is generally well-educated, consisting of mostly female consumers over the age of 30. The first and the second group preferred the coffee experience of international coffee shops, while the third group preferred only national coffee shops. The respondents in the fourth group preferred the national coffee shops, but also stated a desire to experience coffee in international coffee shops

Impact of Social Capital on Residents' Choice of Medical Institutions: A Study Based on Classification Tree Model and Logistic Regression

To explore the influence of social capital on the local residents' choice of medical institutions and to provide a reference basis for promoting diagnosis and treatment services available at different tiers. A classification tree model was established using the exhaustive chi-square automatic interaction detection (Exhaustive CHAID) method to screen for factors influencing the residents' choice of medical institutions, and a logistic regression model was used to quantitatively analyze the interaction effect of the influencing factors. The classification tree model showed that there were four layers and eight terminal nodes, identifying a total of six influencing factors, including individual social capital, self-reported physical health, education, community social capital, chronic disease prevalence, and self-reported mental health. Logistic regression analysis showed that education (odds ratio [OR]=0.660, 95% confidence interval [CI]: 0.502-0.869), community social capital (OR=0.746, 95% CI: 0.589-0.943), and individual social capital (OR=0.405, 95% CI: 0.287-0.572) (P<0.001) had an impact on residents' choice of medical institution. There was an interaction between individual social capital and self-reported physical health on residents' choice of medical institution (OR=1.872, 95% CI: 1.180-2.969, P<0.05). Interventions in terms of social capital factors should be considered in order to promote the rational use of medical resources.

Operation Risk Structural Analysis and Evaluation System Construction for Multistation Integration Project

Digitalization in the power grid sector will become a strong support in the future for the upgrading of the power industry. And, multistation integration has become a leading project in the process. However, there are some hidden risks to its security with the boom of new technologies. Risk analysis and evolution of the multistation integration have an important role in safe operation. This paper analyzes the relationship among risk factors by using both the structural equation modeling (SEM) and the chi-squared automatic interaction detector (CHAID) approach, which is more objective than subjective judgment. The result shows the following: (1) The operational risks come from 10 risk-derived sources which therefore can be qualitatively divided into 2 risk categories with 11 risk indicators. (2) Concerning the structural analysis, the main risk source of multistation integration comes from endogenous risks. (3) In terms of procedural analysis, cultural risk, information risk, and accidental risk are the most critical external risks, while performance risk and location risk are the most critical internal ones. The results of models provide a reference for engineering management in digitalization in the power grid sector.

Prediction of Chronic Kidney Disease Using Machine Learning

Chronic Kidney Disease is one of the most serious illnesses nowadays, and it is vital to have a good diagnosis as soon as possible. Machine learning has proven to be effective in medical therapy. The doctor can diagnose the ailment early with the use of machine learning classifier algorithms. This article has examined Chronic Kidney Disease prediction from this standpoint. The Chronic Kidney Disease dataset was obtained from the University of California at Irvine's repository. The artificial neural network, C5.0, Chi-square Automatic interaction detector, logistic regression, linear support vector machine with penalty L1 &amp; with penalty L2, and random forest classifier techniques were used in this study. The dataset was also subjected to the significant feature selection technique. The results were computed for each classifier using I full features, (ii) correlation-based feature selection, (iii) Wrapper method feature selection, (iv) Least absolute shrinkage and selection operator regression, (v) synthetic minority over-sampling technique with least absolute shrinkage and selection operator regression selected features, and (vi) synthetic minority over-sampling technique with full features. The results show that in synthetic minority over-sampling technique with full features, LSVM with penalty L2 has the maximum accuracy of 98.86 percent. Along with precision, recall, F-measure, and area, accuracy, precision, recall, and area. The GINI coefficient and beneath the curve have been computed, and the results of various algorithms have been compared in the graph. After synthetic minority over-sampling technique with full features, the least absolute shrinkage and selection operator regression selected features with synthetic minority over-sampling approach produced the best results. Again, the linear support vector machine had the maximum accuracy of 98.46 percent in the synthetic minority over-sampling technique with the least absolute shrinkage and selection operator selected features. Along with machine learning models, a deep neural network was used on the same dataset, and the deep neural network was found to have the greatest accuracy of 99.6%.

Development of direct marketing strategy for banking industry: The use of a Chi-squared Automatic Interaction Detector (CHAID) in deposit subscription classification

A comparison between Chi-squared Automatic Interaction Detector (CHAID) and logistic regression analysis was performed for classification problems on bank direct marketing data. CHAID Performance Comparison and comparison with Logistic Regression (LR) performance were also conducted. Priority performance with two statistical measures was evaluated: classification accuracy and sensitivity in the presence of data containing categorical imbalances. Random over sampling (ROS) was then applied to deal with class balance problems to get better performance of CHAID analysis. Segmentation analysis was also performed using the CHAID approach to improve the performance of the analysis results. CHAID outperforms LR because of its advantages that it can be used to perform segmentation modeling. Direct marketers should pay attention to traits are Duration, Month, Contact, and Housing. To get a higher subscription, the bank must extend the call duration. Based on these results, the banking industry needs to prepare regulations related to human resources, infrastructure, costs, and government support to achieve higher subscriptions.JEL Classification A10; C10; G21

Aberrant salience relationship with first rank symptoms

BackgroundAberrant salience is the incorrect assignment of salience, significance, or value to different innocuous stimuli that might precede the onset of psychotic symptoms. The present study aimed to perform a preliminary evaluation of potentially different correlations between the Aberrant Salience Inventory (ASI) score and dimensional or categorical diagnostic approaches.Methods168 adult outpatients with a current psychiatric diagnosis were consecutively enrolled. Patients were evaluated using different psychometric scales. ASI was used to evaluate aberrant salience, and to evaluate the association between ASI scores and first rank symptoms (FRS), and/or with a psychiatric diagnosis. Principal dichotomic clusters of ASI were identified using the Chi-square automatic interaction detection (CHAID) method.ResultsCurrent (16.76 ± 6.02 vs 13.37 ± 5.76; p = 0.001), lifetime (15.74 ± 6.08 vs 13.16 ± 5.74; p = 0.005) and past (15.75 ± 6.01 vs 13.33 ± 5.80; p = 0.009) FRS were the main clusters dichotomizing ASI. The average ASI score did not significantly differ among patients with different diagnoses.ConclusionsASI could be used as a tool to identify psychopathological dimensions, rather than the categorical diagnoses, in the schizophrenic spectrum.

Prediction of body weight from morphological traits of South African non-descript indigenous goats of Lepelle-Nkumbi Local Municipality using different data mining algorithm.

Body weight is a vital trait which can assist farmers on selecting animals to use during breeding season. Therefore, the study was conducted to develop the best model to predict body weight from morphological traits through classification and regression tree (CART) and chi-square automatic interaction detector (CHAID) and to determine the relationship between body weight and some morphological traits. A total of 700 South African non-descript indigenous goats (female = 417 and male = 283) between the age of 1 and 5years old were used in the study. Body weight and some morphological traits viz. body length (BL), heart girth (HG), withers height (WH), rump height (RH), and rump length (RL) were measured in the study. CART, CHAID, and Pearson's correlation were used for data analysis. CART and CHAID algorithms indicated that predictor factors such as BL, HG, age, and villages had statistical influence on body weight of goats. The study suggests that BL can be used to estimate body weight of South African non-descript indigenous goats. Goodness of fit test suggests that CHAID is a suitable algorithm for prediction of body weight of South African non-descript indigenous goats. Correlation findings indicated that BW had positive highly statistical correlation (P < 0.01) with BL in male and female goats with correlation values of r = 0.65 and r = 0.65, respectively. Findings suggest that improving BL of South African non-descript indigenous goats might improve body weight.