Ordinal Regression Methods Research Articles

Econometric Research of the Mix of Factors Influencing First-Year Students’ Dropout Decision at the Faculty of Agri-Food and Environmental Economics

Students dropping out of higher education is a common issue many universities are facing all around the world. This paper complements the existing literature with a unique perspective on the mix of factors influencing first year students' dropout decision at the Faculty of Agri-Food and Environmental Economics from Bucharest, Romania. The objective of this research was to explore these factors through the lens of an unprecedented crisis caused by the COVID-19 pandemic. The ordinal logistic regression method was used based on data resulted from carrying out an online survey. The results of this econometric research can help decision-makers improve retention, the quality of education and the healthy integration of the first-year economics students in the university environment.

Relationship red distribution width to platelet ratio with fibrosis degrees based on transient elastography in chronic hepatitis B patients

Introduction: Chronic hepatitis B virus infection is a major health problem in developing countries like Indonesia. A chronic HBV infection of approx. 40% leads to liver fibrosis and hepatocellular carcinoma. Furthermore, liver biopsy is the gold standard in diagnosing liver fibrosis, but this examination is invasive and has many complications. Transient elastography has been validated to assess fibrosis in several liver diseases including HBV infection. However, not all hospitals have temporary elastography facilities in such a way that cheaper and more affordable examinations are needed. This study aims to examine the relationship between Red cell distribution width to platelet ratio with the degree of liver fibrosis based on transient elastography in chronic hepatitis B patients.Method: This study was a cross-sectional examination, which involves 48 chronic hepatitis B patients receiving outpatient care from September to November 2020 at dr. Zainoel Abidin Banda Aceh. The routine blood tests were performed using the Sysmex XT 1800i. The red cell distribution width to platelet ratio value is calculated by dividing the Red Distribution Width value by the platelets, and the degree of liver fibrosis was assessed using TE (Fibroscan® 502 Echosens).Result: In the 48 patients, the mean value of Red cell Distribution Width to Platelet Ratio for various degrees of liver fibrosis F0-1, F2, F3 and F4 were 0.051; 0.050; 0.077 and 0.108, respectively. It was reported that red cell distribution width to Platelet Ratio levels also increased according to the increase in liver fibrosis degree and was statistically significant with a p-value of 0.002. Analysis of the relationship between red cell distribution width to platelet ratio with the degree of liver fibrosis using ordinal regression method, while the red cell distribution width to platelet ratio level variable has a fairly large estimate value that is 31098 with a p-value of 0.003. The p-value of the test of less than 0.05 proves that the red cell distribution width to platelet ratio level affects and has a relationship with the degree of liver fibrosis.Conclusion: Red cell distribution width to platelet ratio relates to the degree of liver fibrosis in patients with chronic hepatitis B.

Involvement of women in forest conservation- A comparative analysis of fringe villages in Sundarban Biosphere Reserve.

The participation of women in the environmental movements and their association with forest resources in some areas are notable throughout the world. A comparative study to understand the perception and participation of women in forest protection programmes in the fringe villages of Sundarban Biosphere Reserve is the main focus of this paper. Four villages from two blocks of Sundarban Biosphere Reserve are selected. The entire work is mainly based on primary data. For comparative analysis, Chi-square and Cramer's V values are used. The ordinal regression method is used for relationship analysis. Although there are many differences in socio-economic factors and involvement in forest protection committees, there are no disagreements in the perception of women regarding the significance of forest protection of Sundarban. In fine, the focus is given on the inclusion of women's ideas in the decision-making process for the advancement of Sundarban Biosphere Reserve.

Protein–protein interaction prediction based on ordinal regression and recurrent convolutional neural networks

BackgroundProtein protein interactions (PPIs) are essential to most of the biological processes. The prediction of PPIs is beneficial to the understanding of protein functions and thus is helpful to pathological analysis, disease diagnosis and drug design etc. As the amount of protein data is growing fast in the post genomic era, high-throughput experimental methods are expensive and time-consuming for the prediction of PPIs. Thus, computational methods have attracted researcher’s attention in recent years. A large number of computational methods have been proposed based on different protein sequence encoders.ResultsNotably, the confidence score of a protein sequence pair could be regarded as a kind of measurement to PPIs. The higher the confidence score for one protein pair is, the more likely the protein pair interacts. Thus in this paper, a deep learning framework, called ordinal regression and recurrent convolutional neural network (OR-RCNN) method, is introduced to predict PPIs from the perspective of confidence score. It mainly contains two parts: the encoder part of protein sequence pair and the prediction part of PPIs by confidence score. In the first part, two recurrent convolutional neural networks (RCNNs) with shared parameters are applied to construct two protein sequence embedding vectors, which can automatically extract robust local features and sequential information from the protein pairs. Based on it, the two embedding vectors are encoded into one novel embedding vector by element-wise multiplication. By taking the ordinal information behind confidence score into consideration, ordinal regression is used to construct multiple sub-classifiers in the second part. The results of multiple sub-classifiers are aggregated to obtain the final confidence score. Following that, the existence of PPIs is determined by the confidence score. We set a threshold theta, and say the interaction exists between the protein pair if its confidence score is bigger than theta.ConclusionsWe applied our method to predict PPIs on data sets S. cerevisiae and Homo sapiens. Through experimental verification, our method outperforms state-of-the-art PPI prediction models.

Dimensionality reduced robust ordinal regression applied to life cycle assessment

Life Cycle Assessment quantifies the multi-dimensional impact of goods and services and can be handled by Multi-Criteria Decision Analysis. In Multi-Criteria Decision Analysis, Robust Ordinal Regression manages all the compatible preference functions at once when assessing a set of alternatives and a group of preferences on reference alternatives. Robust Ordinal Regression is thus a versatile method of reducing the cognitive effort required by decision makers for eliciting their preference structures in Life Cycle Assessment, although it does not directly operate on noisy alternatives and requires Stochastic Multicriteria Acceptability Analysis to deal with such scenarios.We propose integrating a dimensionality reduction technique, Principal Component Analysis, and Robust Ordinal Regression methods, to reduce the problem dimensionality and ensure the actual problem features are considered.A generated dataset, a dataset from literature and a Life Cycle Assessment case study are used to test the effectiveness of the proposed methods.

Port Efficiency and the Financial Performance of Greek Public Ports Before and During the Economic Crisis

ABSTRACT This paper investigates the impact of several financial ratios on the efficiency of ten Greek public ports using the ordinal logistic regression method. The selection of the financial ratios was carried out using the indices related to the three efficiency zones. The results indicate that there are financial ratios that positively affect and others that negatively affect port efficiency. In general, our main findings support the notion that the middle efficiency zone was the most influential of all during the economic crisis. This was confirmed by the ANOVA robustness test. The DEA approach suggests that the ports as a whole were inefficient during the three periods—pre-crisis, in-crisis, and across all sample data periods under investigation. Investment in the ports has been held back during the economic crisis; however, the troika (i.e., the International Monetary Fund, European Commission, and European Central Bank) helped the Greek government to accelerate their economic development during this time. Our results have significant implications for the government. It should, for instance, decide whether to privatise the ten ports partially or fully.

The Effects of Multi-Sociodemographic Characteristics of Construction Sites Personnel on Perceptions of Safety Climate-Influencing Factors: The Construction Industry in Saudi Arabia

The construction industry in Saudi Arabia relies prominently on migrant workers of multi-sociodemographic characteristics with different perceptions of a safety climate. The exploration of the perceptions regarding the safety climate among various groups of migrant workers may help identify effective means of improving safety levels at construction sites in Saudi Arabia. This study aimed to examine the effects of multi-sociodemographic characteristics of construction site personnel on their perceptions of the factors that influence the safety climate at construction sites in Saudi Arabia. Data were collected from 401 construction site workers, employed at ongoing construction project sites in Saudi Arabia, using a designed questionnaire. A generalized, linear model approach was applied, using the single ordinal logistic regression method, to analyze the collected data. The results revealed the significant sets of sociodemographic characteristics and their associated subgroups that had significant effects on the perception of importance assigned to each safety climate-influencing factor. These findings provide a better understanding of the views of construction site personnel on the safety climate and can assist construction industry decision-makers, safety policy designers, government agencies, and stakeholders when designing better-targeted enhancement plans and strategies to improve the safety climate of construction sites, based on the sociodemographic makeup of the personnel at each construction site.

Distributed Ordinal Regression Over Networks

Many real-world data are labeled with natural orders, i.e., ordinal labels. Examples can be found in a wide variety of fields. Ordinal regression is a problem to predict ordinal labels for given patterns. There are specially developed ordinal regression methods to tackle this type of problems, but they are usually centralized. However, in some scenarios, data are collected distributedly by nodes of a network. For the purpose of privacy protection or due to some practical constraints, it is difficult or impossible to transmit the data to a fusion center for processing. Thus the centralized ordinal regression methods are inapplicable. In this paper, we formulate a distributed generalized ordered logit model for distributed ordinal regression. To estimate parameters in the model, a distributed constrained optimization formulation based on maximum likelihood methods is established. Then, we propose a projected gradient based algorithm to solve the optimization problem. We prove the consensus and the convergence of the proposed distributed algorithm. We also conduct numerical simulations on synthetic and real-world datasets. Simulation results show that the proposed distributed algorithm is comparable to the corresponding centralized algorithm. Even when the data label distribution among nodes is unbalanced, the proposed algorithm still has competitive performance.

Exploring loyalty of generation X in a retail store context

Despite the increased interest in consumers pertained to Generation X, there is still much that is not understood about this particular market segment. In particular, this refers to their retail-related behaviour as well as the satisfaction and loyalty of customers in traditional brick and mortar retail environment. The purpose of this paper is to expand the knowledge base about Generation X’s retail store attitudes and retail behaviour in order to identify their loyalty in a retail store context. For that reason, an empirical research was carried out using a convenience sample of 153 Generation X members from the Dubrovnik- Neretva County during the period from June 1 to October 1, 2016. To analyse the data, ordinal logistic regression and ANOVA methods were used. The results revealed that several variables were found partially statistically significant, i.e. retail service quality, retail store quality, product price, product assessment, retail store format and frequency of purchase. This paper provides a framework for an improved understanding of Generation X as consumers and their overall store experience. Moreover, it contributes to the existing literature by providing new insights into the loyalty of Croatian Generation X consumers.

Accurate Monocular Depth Estimation via Interaction of Hierarchical Features

Monocular depth estimation is a challenging task, which assists in understanding 3D scene geometry from the same 2D scene. The ordinal-regression-method demonstrates superior performance in this issue but naive ordinal inference strategy for inferring the final depth values and naive operations to up-sample to the desired space scale introduce significant discretization errors and object boundary confusion. Firstly, we come up with a novel inference strategy to reduce the discretization errors. And then a specifically designed decoder that completes the fusion of different hierarchical features under guidance and the fusion feature reconstruction. We evaluate on a public monocular depth-estimation benchmark dataset (NYU Depth V2). The experimental results show that the method proposed outperforms other ordinal regression methods.

Non-Uniform Discretization-based Ordinal Regression for Monocular Depth Estimation of an Indoor Drone

At present, the main methods of solving the monocular depth estimation for indoor drones are the simultaneous localization and mapping (SLAM) algorithm and the deep learning algorithm. SLAM requires the construction of a depth map of the unknown environment, which is slow to calculate and generally requires expensive sensors, whereas current deep learning algorithms are mostly based on binary classification or regression. The output of the binary classification model gives the decision algorithm relatively rough control over the unmanned aerial vehicle. The regression model solves the problem of the binary classification, but it carries out the same processing for long and short distances, resulting in a decline in short-range prediction performance. In order to solve the above problems, according to the characteristics of the strong order correlation of the distance value, we propose a non-uniform spacing-increasing discretization-based ordinal regression algorithm (NSIDORA) to solve the monocular depth estimation for indoor drone tasks. According to the security requirements of this task, the distance label of the data set is discretized into three major areas—the dangerous area, decision area, and safety area—and the decision area is discretized based on spacing-increasing discretization. Considering the inconsistency of ordinal regression, a new distance decoder is produced. Experimental evaluation shows that the root-mean-square error (RMSE) of NSIDORA in the decision area is 33.5% lower than that of non-uniform discretization (NUD)-based ordinal regression methods. Although it is higher overall than that of the state-of-the-art two-stream regression algorithm, the RMSE of the NSIDORA in the top 10 categories of the decision area is 21.8% lower than that of the two-stream regression algorithm. The inference speed of NSIDORA is 3.4 times faster than that of two-stream ordinal regression. Furthermore, the effectiveness of the decoder has been proved through ablation experiments.

The Impact of Service Quality on Customer Satisfaction in Commercial Bank of Ethiopia, Mekelle Branch

This paper examined the effect of service quality on customer satisfaction in Commercial Bank of Ethiopia, Mekelle branch. The five hypotheses were tested by using statistical techniques. The study is explanatory type of research. The sample size is 300 and is chosen on convenient basis. The primary source of data was collected using questionnaire. Data were analyzed by using SPSS version 21. Ordinal regression method was used to model the relationship between the customer satisfaction and service quality dimensions. The explanatory variables include tangibility, reliability, responsiveness, assurance and empathy; and the response variable is customer satisfaction. Customer satisfaction was measured on ordered, categorical, and five-point Likert scale: strongly disagree, disagree, neutral, agree and strongly agree. The model fitting statistics and the reliability of the model assumption: parallel lines tests, overall goodness of fit information were tested. The findings indicated that the explanatory variables were significantly associated with the customer satisfaction. Keywords: Service quality dimensions and Customer satisfaction. DOI: 10.7176/JMCR/72-02 Publication date: October 31 st 2020

Impact of the COVID-19 Epidemic on Lifestyle Behaviors and Their Association With Subjective Well-Being Among the General Population in Mainland China: Cross-Sectional Study.

BackgroundThe world is experiencing an unprecedented challenge due to the coronavirus disease (COVID-19) pandemic. However, it is unclear whether people’s lifestyles will change as a result.ObjectiveThe aim of this study is to explore perceived lifestyle changes after the outbreak of COVID-19 and their association with subjective well-being (SWB) among the general population in Mainland China.MethodsAn online survey was conducted in May 2020. Lifestyle behaviors including leisure-time physical exercise, leisure-time screen time, and dietary intake were self-reported. SWB was measured using the General Wellbeing Schedule (GWS). Other covariates including sociodemographic factors, self-rated physical health, perceived social support, and loneliness were also assessed by a structured questionnaire. A multivariate ordinal regression method was used to analyze the association between SWB and lifestyle behaviors as well as perceived lifestyle changes.ResultsA total of 1033 participants aged between 18 and 60 years were included in this study. The mean GWS score was 71.7 points. About 70% of the respondents reported spending more time looking at screens, whereas about 30% reported an increased frequency of vegetable and fruit intake after the outbreak of COVID-19. Inactive physical exercise (odds ratio [OR] 1.16, 95% CI 1.02-1.48), infrequent vegetable intake (OR 1.45, 95% CI 1.10-1.90), infrequent fruit intake (OR 1.31, 95% CI 1.01-1.70), and often skipping breakfast (OR 1.43, 95% CI 1.08-1.91) were associated with lower SWB after adjusting for sociodemographic factors, self-rated physical health, perceived social support, and loneliness. Moreover, participants who perceived a decrease in the frequency of vegetable, fruit, and breakfast intake were more likely to report lower SWB.ConclusionsThe COVID-19 pandemic may have positive and negative impacts on different aspects of lifestyle behaviors. Both unhealthy lifestyle behaviors and negative lifestyle changes were associated with lower SWB. These findings provide scientific evidence that can inform lifestyle guidelines and public mental health interventions during the COVID-19 outbreak.

Social deprivation among socio-economic contrasted french areas: Using item response theory analysis to assess differential item functioning of the EPICES questionnaire in stroke patients.

BackgroundMultiple approaches have been proposed to measure low socio-economic status. In France the concept of precariousness, akin to social deprivation, was developed and is widely used. EPICES is a short questionnaire that was developed to measure this concept. This study aimed to evaluate Differential Item Functioning (DIF) in the EPICES questionnaire between contrasted areas: mainland France, French West Indies (FWI) and French Guiana (FG).MethodsThe population was taken from the INDIA study, which aimed to evaluate the impact of social inequalities on stroke characteristics and prognosis. Eligible people were patients referred to neurology or emergency departments for a suspicion of stroke. We assessed the DIF using hybrid ordinal logistic regression method, derived from item response theory.ResultsWe analysed 1 553 stroke patients, including 768 from FWI (49.5%), 289 from FG (18.6%) and 496 from mainland (31.9%). We identified five items with a moderate to large DIF in area comparisons: “meeting with a social worker”, “complementary health insurance”, “home-owning”, “financial difficulties” and “sport activities”. Correlation between EPICES score and the latent variable was strong (r = 0.84).ConclusionThis is the first attempt to assess the DIF of the EPICES score between different French populations. We found several items with DIF, which can be explained by individual interpretation or local context. However, the DIFs did not lead to a large difference between the latent variable and the EPICES score, which indicates that it can be used to assess precariousness and social deprivation between contrasted areas.

Italians’ public opinion on road roundabouts: A web based survey

Roundabouts have been shown to provide significant safety and operational benefits, but a better understanding of public perceptions of roundabouts is critical in order to assess the contribution of the driving behavior of the users on their safety. Therefore, research on user behavior and preferences is a helpful tool in improving road safety and accident prevention. The aim of this study is to increase knowledge about Italian users’ public opinion on road roundabouts. The authors estimated cross-national differences, and assessed correlations with personal variables, such as age, gender, travel mode most frequently used, perception of roundabout safety, etc. In order to collecting data, the instrument used is a national web based survey, the total sample comprised 1.728 participants. The data collected were analyzed by both Ordinal regression method and Multiple Correspondence Analysis.The results of this study show that the overall opinion that the respondents expressed against roundabouts is favorable. There are no substantial differences between the judgment of male drivers and that of females. Although for some categories of users the roundabouts are not considered safe and have no advantages in terms of functionality, their general opinion is always positive. The categories of respondents who expressed "Favorable" opinion are: women, those who travel daily with the private car or with public transport and who is a pedestrian. The categories of respondents who expressed "Very favorable" opinion are: men, and those who use the motorcycle as their main means of transport.Moreover, this study showed that web-based surveys provide an efficient means of collecting detailed public opinion data.

APLIKASI REGRESI ORDINAL PADA FAKTOR YANG BERPENGARUH TERHADAP AKTIFITAS ENZIM CHOLINESTERASE DARAH (Studi di Dusun Binangun Desa Bumiaji Kecamatan Bumiaji Kota Batu)

The case of pesticide contamination in Batu City in 2017 wasrelatively high, with the percentage of 82.7%. The case of pesticide contamination could be detected by examining the cholinesterase enzyme activity in blood.The Bumiaji villagers (Batu City) mostly work as a pesticide spray farmers. The study aims to analyze the factors that affect cholinesterase enzyme’s activities on the blood of the farmers who spray pesticide. The study was observational with cross-sectional design and used ordinal regression method. The research sampleswere 81 farmers selected through accidental sampling method. The independent variables of this study were the use of Personal Protective Equipment (PPE) and farmers’work duration. The dependent variable of this study was the cholinesterase enzyme’s activities in the farmers’ blood. The results of the study were that most farmers experienced cholinesterase enzyme activity changes, misused PPE (80.2%), and worked for less than 8 hours each day (92.6%). Based on the ordinal regression analysis, the variables which affected to the cholinesterase enzyme activity of the farmers’ blood was the use of PPE with a p-value of 0.00 <α (0.05). Based on the regression model, the farmers who properly used PPE had 0.12 higher chance to avoid the change of cholinesterase enzyme activity in the blood. It is necessary to held a disseminationabout the use of personal protective equipment to the farmers, as a preventive effort by the district health office, local village office, and the management of farmer groups in Binangun sub-village, Bumiaji Village, Bumiaji Subdistrict, Batu City.

APLIKASI REGRESI ORDINAL PADA FAKTOR YANG BERPENGARUH TERHADAP AKTIFITAS ENZIM CHOLINESTERASE DARAH (Studi di Dusun Binangun Desa Bumiaji Kecamatan Bumiaji Kota Batu)

The case of pesticide contamination in Batu City in 2017 wasrelatively high, with the percentage of 82.7%. The case of pesticide contamination could be detected by examining the cholinesterase enzyme activity in blood.The Bumiaji villagers (Batu City) mostly work as a pesticide spray farmers. The study aims to analyze the factors that affect cholinesterase enzyme’s activities on the blood of the farmers who spray pesticide. The study was observational with cross-sectional design and used ordinal regression method. The research sampleswere 81 farmers selected through accidental sampling method. The independent variables of this study were the use of Personal Protective Equipment (PPE) and farmers’work duration. The dependent variable of this study was the cholinesterase enzyme’s activities in the farmers’ blood. The results of the study were that most farmers experienced cholinesterase enzyme activity changes, misused PPE (80.2%), and worked for less than 8 hours each day (92.6%). Based on the ordinal regression analysis, the variables which affected to the cholinesterase enzyme activity of the farmers’ blood was the use of PPE with a p-value of 0.00 <α (0.05). Based on the regression model, the farmers who properly used PPE had 0.12 higher chance to avoid the change of cholinesterase enzyme activity in the blood. It is necessary to held a disseminationabout the use of personal protective equipment to the farmers, as a preventive effort by the district health office, local village office, and the management of farmer groups in Binangun sub-village, Bumiaji Village, Bumiaji Subdistrict, Batu City.

Ordinal regression algorithms for the analysis of convective situations over Madrid-Barajas airport

Abstract In this paper we tackle a problem of convective situations analysis at Adolfo-Suarez Madrid-Barajas International Airport (Spain), based on Ordinal Regression algorithms. The diagnosis of convective clouds is key in a large airport like Barajas, since these meteorological events are associated with strong winds and local precipitation, which may affect air and land operations at the airport. In this work, we deal with a 12-h time horizon in the analysis of convective clouds, using as input variables data from a radiosonde station and also from numerical weather models. The information about the objective variable (convective clouds presence at the airport) has been obtained from the Madrid-Barajas METAR and SPECI aeronautical reports. We treat the problem as an ordinal regression task, where there exist a natural order among the classes. Moreover, the classification problem is highly imbalanced, since there are very few convective clouds events compared to clear days. Thus, a process of oversampling is applied to the database in order to obtain a better balance of the samples for this specific problem. An important number of ordinal regression methods are then tested in the experimental part of the work, showing that the best approach for this problem is the SVORIM algorithm, based on the Support Vector Machine strategy, but adapted for ordinal regression problems. The SVORIM algorithm shows a good accuracy in the case of thunderstorms and Cumulonimbus clouds, which represent a real hazard for the airport operations.

Economic valuation of critical land

The purposes of this research are to analyze and to identify the factors influencing the farmers’ Willingness to Pay (WTP) in reducing the impact of the critical land. The research method used to calculate the farmers’ WTP on the land restoration is the Willingness to Pay (WTP) Method, and The Ordinal Logistic Regression Method is used to analyze WTP’s influencing factors. The result showed that the farmers’ average of maximum WTP was IDR 21,196. This means that the farmers’ average of maximum WTP is lower than the average cost incurred by the farmers for the land restoration activity that was IDR 58,000. This indicated the low of farmers’ awareness in the efforts of doing critical land restoration. The independent variable with significant influence is the OWN (status of the land ownership) variable. The other variables that are positive and significant are income, age, education, long stay, and family numbers. The significant variables with negative impact are marital status, occupation, and land restoration activity. In general, farmers thought that the activities of repairing the degraded land were the role of the government. So that the efforts of farmers in recovering land are almost non-existent.

Using Ordinal Logistic Regression Analysis in Evaluating Teachers’ Performance Level of High Schools (12th grades) in Kurdistan Regional Government

The Ministry of Education – Kurdistan Regional Government (MOE-KRG) currently depends only on the students’ grades in the evaluation of the teachers’ performance level in all subjects as low, medium, or high performance. Relying just on one variable to determine teacher’s performance is not fair and this problem must be resolved statistically through finding new proposed statistical model. Therefore, this study tries to find some variables that are available in MOE-KRG for all high schools in order to use them in the proposed model to predict teachers’ performance level instead of the old one. This study aims to predict teacher’s performance level of high schools (12th grades) in KRG who teach Kurdish subject, and also analyze the effects of variables that have impact on the evaluation of teachers’ performance level depending on the data that are available in the MOE-KRG. In this paper Ordinal Logistic Regression (OLR) method is used to find a proposed model for evaluating teachers’ performance in Kurdish subject using the data of all scientific high schools in KRG (646 high schools). The teachers' status was analyzed by selecting nine variables related to the high schools: sector (governmental or non-governmental), geographic location, type of education, status of school (exemplary, non-exemplary), gender of student, year of school establishment, number of classes, number of teachers and student’s average marks in Kurdish subject. It is concluded that four of these variables (student’s average in Kurdish subject, number of classes, geographical location and status of school) have significant effects on teachers' evaluation in 12th class of scientific high schools, and the overall percentage of correct classification is about 87%, it that means the Ordinal logistic regression model has an ability to predict teacher’s performance level very well.