• Communications for Statistical Applications and Methods
• /
• v.16 no.4
• /
• pp.697-705
• /
• 2009

The Pearson chi-squared statistic or the deviance statistic is widely used in assessing the goodness-of-fit of the generalized linear models. But these statistics are not proper in the situation of continuous explanatory variables which results in the sparseness of cell frequencies. We propose a goodness-of-fit test statistic for the cumulative logit models with ordinal responses. We consider the grouping of a dataset based on the ordinal scores obtained by fitting the assumed model. We propose the Pearson chi-squared type test statistic, which is obtained from the cross-classified table formed by the subgroups of ordinal scores and the response categories. Because the limiting distribution of the chi-squared type statistic is intractable we suggest the parametric bootstrap testing procedure to approximate the distribution of the proposed test statistic.

• The Korean Journal of Applied Statistics
• /
• v.23 no.6
• /
• pp.1057-1065
• /
• 2010

In this paper we discuss some cautionary notes in using the Pearson chi-squared test statistic for the goodness-of-fit of the ordinal response model. If a model includes continuous type explanatory variables, the resulting table from the t of a model is not a regular one in the sense that the cell boundaries are not fixed but randomly determined by some other criteria. The chi-squared statistic from this kind of table does not have a limiting chi-square distribution in general and we need to be very cautious of the use of a chi-squared type goodness-of-t test. We also study the limiting distribution of the chi-squared type statistic for testing the goodness-of-t of cumulative logit models with ordinal responses. The regularity conditions necessary to the limiting distribution will be reformulated in the framework of the cumulative logit model by modifying those of Moore and Spruill (1975). Due to the complex limiting distribution, a parametric bootstrap testing procedure is a good alternative and we explained the suggested method through a practical example of an ordinal response dataset.

• Journal of Korean Society for Quality Management
• /
• v.30 no.3
• /
• pp.38-52
• /
• 2002

This paper is concerned with the analysis of structural equation model(SEM) with the ordinal data such as Likert scale. The SEM is misused when the arbitrary scores allocated to the Likert scale are treated as quantitative data. The underlying distribution approaches have been studied to solve this problem, and the partial least squares(PLS) Is also tried. In this paper the quantification methods for the Likert scale are proposed to analyze the SEM. We assume that the Likert scale is an observation of the interval of the continuous underlying distribution, and the respondents have their own patterns in the response of some questions. Normal and beta distributions as the response patterns are considered to quantify the Likert scale. To compare the efficiency of the proposed method the bootstrap simulations are tried.

• IE interfaces
• /
• v.15 no.2
• /
• pp.209-217
• /
• 2002

Quality of Service(QoS) in the telecommunications can be evaluated by analyzing the opinion data which result from the surveyed opinions of respondents and quantify subjective satisfaction on the QoS from the customers' viewpoints. For analyzing the opinion data, MOS(mean opinion score) method and Cumulative Probability Curve method are often used. The methods are based on the scoring method, and therefore, have the intrinsic deficiency due to the assignment of arbitrary scores. In this paper, we propose an analysis method of the opinion data using logit models which can be used to analyze the ordinal categorical data without assigning arbitrary scores to customers' opinion, and develop an analysis procedure considering the usage of procedures provided by SAS(Statistical Analysis System) statistical package. By the proposed method, we can estimate the relationship between customer satisfaction and network performance parameters, and provide guidelines for network planning. In addition, the proposed method is compared with Cumulative Probability Curve method with respect to prediction errors.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.4
• /
• pp.63-71
• /
• 2016

This paper is about applying efficient data mining method which improves the score calculation and proper building performance of credit ranking score system. The main idea of this data mining technique is accomplishing such objectives by applying Correlation based Feature Selection which could also be used to verify the properness of existing rank scores quickly. This study selected 2047 manufacturing companies on KOSPI market during the period of 2009 to 2013, which have their own credit rank scores given by NICE information service agency. Regarding the relevant financial variables, total 80 variables were collected from KIS-Value and DART (Data Analysis, Retrieval and Transfer System). If correlation based feature selection could select more important variables, then required information and cost would be reduced significantly. Through analysis, this study show that the proposed correlation based feature selection method improves selection and classification process of credit rank system so that the accuracy and credibility would be increased while the cost for building system would be decreased.

• Journal of Labour Economics
• /
• v.40 no.2
• /
• pp.1-29
• /
• 2017

We examine the extent to which students' performance relative to peers affects their career choice. Specifically, we analyze the relationship between a student's mathematics ranking in his/her school and the likelihood of choosing Mathematics and Science track in high school. Using a panel dataset of students in Seoul, we measure a student's performance using two variables: absolute performance and relative performance. The former measures a student's performance relative to the entire sample, while the latter measures performance relative to the student's peers in the same school. After controlling for test scores and other characteristics, we find that the students with a poor relative ranking are 11 percentage points less likely to choose the Mathematics and Science track. Relative performance affects girls more greatly than boys. Although relative performance affects a student's self-efficacy and class participation, our accounting exercise suggests that this channel accounts for only 12 percent of the impact, implying that students may respond to the relative ranking mostly due to other factors, such as strategic consideration to perform well in college applications.

• The Korean Journal of Food And Nutrition
• /
• v.26 no.4
• /
• pp.783-796
• /
• 2013

This study investigated dietary patterns among the elderly over 75 years old living in Jeollanam-do area in May 2012. Although structured interviews were conducted with 236 consenting subjects, only 194 who completed the ADL, IADL, and K-MMSE tests were used for statistical data analysis. Using ADL, IADL, K-MMSE scores, cluster analysis was first performed and resulted in two groups: IFG (Insufficiently Functioning Group) and SFG (Sufficiently Functioning Group). Chi-square tests for nominal scales, Mann-Whitney tests for ordinal scales, and ANOVAs and t-tests for interval and ratio scales were conducted to compare two groups. More than 70% of IFG were illiterates compared to 28.1% of SFG. 'Excessive eating', 'appetite', 'digestion', and 'balanced diet' did not differ between groups. SFG more frequently had snacks and ate out and were more likely to take health supplements than IFG. Among the 100 major food items, consumption frequencies of several foods differed between groups. Study implications and limitations were discussed.

• Korean Journal of Agricultural Science
• /
• v.50 no.4
• /
• pp.927-940
• /
• 2023

Online shopping has been undergoing significant developments in the South Asian region in the last decade. Using a representative sample of Sri Lankan consumers, this study explored online food purchasing behavior in Sri Lanka, a developing nation and island in South Asia. Data were collected from 562 respondents from all nine provinces in Sri Lanka using an online survey. Consumer attitudes were evaluated using factor analysis, and factor scores were added as explanatory variables to the final model. An ordered logistic regression model was used to examine the impact of consumer demographics, economic variables, and consumer attitudes on online food purchases. Online food purchasing intensity was categorized into four groups that suited ordinal rankings: zero for never, low for rarely, medium for occasionally, and high for regularly. Results indicated that age, income, education, and living in urban areas affect the online food purchasing behavior of Sri Lankan consumers. In addition, trust, convenience, and attitudes toward price were powerful drivers of online food purchasing. The findings have a number of significant managerial ramifications for creating strategies to promote online food purchases in developing South Asian nations like Sri Lanka. Moreover, promoting online shopping could be a potential solution for traffic congestion, ultimately helping to mitigate the negative externalities associated with it, such as carbon emissions and air pollution.

• Journal of Korean Academy of Nursing
• /
• v.16 no.2
• /
• pp.13-28
• /
• 1986

The main purpose of th study was to evaluate he validity of Korean Pain Measurement Tool composed of pain terms. The specific purposes of this study were 1. to examine whether pain intensities of pain terms are congruent with those classified in three previous studies. 2. to evaluate the relative intensity of each term by panel of judges. 3. to explore the difference of ranks of pain terms according to the sex, education, and ages. One hundred and sixty normal individuals were selected by 2$\times$2$\times$4 sampling design. Sex (male, female), education (high school, college), and age (20s, 30s, 40s, 50s) were matched. Each individual was asked to rate the ranks of 3~8 pain terms in each subclass. The data measured by ordinal scale were transformed to the interval scale to compare with the pain intensities gained from the previous study. The pain ranks different from previous results were finally rearranged or cancelled through the consultation of 4 panel of judges and sunmed up to 91 pain terms in the scale. As a result, the ranks of pain terms within each of eleven subclasses among the twenty subclasses completely were congruent with the Previous pain ranks, while the ranks of nine subclasses were different from the previous pain ranks. In addition, there was significant relation between sex and pain ranks in skin punctuate pressure pain and cavity pressure. (sp : $\chi$$^2$=5.18 ø=0.26； cp : $\chi$$^2$=5.83 ø=0.24) In conclusion, seven terms from subclasses of inflammatory repeated pain, traction pressure pain, fatigue-related pain, fear-related pain, dull pain, and pulsation. related pain were cancelled. The ranks of four terms in subclasses of incisive Pressure pain and constrictive pressure pain were tentatively rearranged. Ranks of two terms in the tract pain were left as shown in the third study. As a result, six terms must be studied repeatedly for obtaining exact scores from ratio scale.

• The Korean Journal of Pain
• /
• v.34 no.4
• /
• pp.417-426
• /
• 2021

Background: Non-cardiac chest pain (NCCP) is a common patient complaint imposing great costs on the healthcare system. It is associated with psychological factors such as depression. The aim of the present study is determining depression predictors in NCCP patients. Methods: The participants of this cross-sectional study were 361 NCCP patients. Patients filled out questionnaires concerning their sociodemographic, lifestyle, and clinical factors (severity of pain, type D personality, somatization, cardiac anxiety, fear of body sensations, and depression). Results: Based on multiple ordinal logistic regression, lack of physical activity (odds ratio [OR], 1.78; 95% confidence interval [CI], 1.09-2.87), sleep quality (OR, 2.98; 95% CI, 1.15-7.69), being a smoker (OR, 1.33; 95% CI, 2.41-4.03), present pain intensity (OR, 1.08; 95% CI, 1.05-1.11), type D personality (OR, 2.43; 95% CI, 1.47-4.03), and somatization (OR, 1.22; 95% CI, 1.15-1.3) were significant predictors of depression in NCCP patients. Additionally, multiple linear regression showed that being unmarried (β = 1.51, P = 0.008), lack of physical activity (β = 1.22, P = 0.015), sleep quality (β = 2.26, P = 0.022), present pain intensity (β = 0.07, P = 0.045), type D personality (β = 1.87, P < 0.001), somatization (β = 0.45, P < 0.001), and fear of bodily sensation (β = 0.04, P = 0.032) increased significantly depression scores in NCCP patients. Conclusions: Physicians should consider the predictors of depression in NCCP patients which can lead to receiving effective psychological consultations and reducing the costs and ineffectual referrals to medical centers.