• The Korean Journal of Applied Statistics
• /
• v.23 no.5
• /
• pp.967-975
• /
• 2010

Since the least squares estimation is not appropriate when multicollinearity exists among the regressors of the linear regression model, the principal components regression is used to deal with the multicollinearity problem. This article suggests a new procedure for the selection of suitable principal components. The procedure is based on the condition index instead of the eigenvalue. The principal components corresponding to the indices are removed from the model if any condition indices are larger than the upper limit of the cutoff value. On the other hand, the corresponding principal components are included if any condition indices are smaller than the lower limit. The forward inclusion method is employed to select proper principal components if any condition indices are between the upper limit and the lower limit. The limits are obtained from the linear model which is constructed on the basis of the conjoint analysis. The procedure is evaluated by Monte Carlo simulation in terms of the mean square error of estimator. The simulation results indicate that the proposed procedure is superior to the existing methods.

• The Korean Journal of Applied Statistics
• /
• v.30 no.2
• /
• pp.301-309
• /
• 2017

It is often possible to test for differences in population means when two or more samples are extracted from each N population. However, it is not possible to test for the mean difference if one sample is extracted from each population since a sample mean does not exist. But, by dividing a group of samples extracted one by one into two groups and generating a sample mean, we can identify a heterogeneity that may exist within the group by comparing the differences of the groups' mean. Therefore, we propose a minimum combination t-test method that can test the mean difference by the number of combinations that can be divided into two groups. In this paper, we proposed a method to test differences between means to check heterogeneity in a group of extracted samples. We verified the performance of the method by simulation study and obtained the results through real data analysis.

• The Journal of Korean Society for School & Community Health Education
• /
• v.21 no.3
• /
• pp.35-47
• /
• 2020

Objectives: This study is a descriptive research study conducted to grasp the Prediction factors of the sexual violence experience of college students. Methods: A convenience sampling was performed for 500 students from one college located in Gyeongsangbuk-do, who agreed to the purpose of this study. Data collection was conducted from October 5, 2015, to October 23, 2015, by filling out the self-report questionnaire. Among the 450 subjects excluding those with missing values, a questionnaire of dating violence experience was applied to 317 college students who answered that they had a friend of the opposite sex, and variables and prediction factors related to dating violence experiences were identified. The statistical methods used were descriptive statistics, x²-test, t-test, Pearson's correlation coefficient and binary logistic regression analysis. Results: As a result of the study, the experience of sexual behavior before entering college (𝑥²=6.52, p=.011), experience of sexual violence damage before entering college(p=.045), the experience of sexual assault before entering college (p=.007) and experience of school violence damage(p=.002) were variables related to the sexual violence experience of college students. School violence victimization (OR=4.831, p=.007) and controlling dating partners (OR=1.349, p<.001) were predictors of dating sexual violence. Dating sexual violence experience group were compared to dating sexual violence non-experience group, the relative degree of controlling dating partners was high (t=4.25, p<.001) and had a traditional gender role attitude (t=2.94, p=.004). and there was a positive correlation (r=.358, p<.001) between controlling dating partners and gender role attitude. Conclusions: In order to prevent sexual violence on dating among college students, it is expected that more effective health education results will emerge if the contents of the school-age school violence victimization experience and the control of dating partners, which are predicted factors of sexual violence on dating, are included in the sexual violence prevention program.

• Weed & Turfgrass Science
• /
• v.7 no.2
• /
• pp.130-139
• /
• 2018

The colors of packaging of herbicides and non-selective herbicides on the market in Korea are defined as brown and red, respectively, according to the notification of RDA. The present study aimed to understand consumer's electroencephalographic (EEG) response when looking at brown and red colors of herbicide and non-selective herbicide packaging papers. The EEG cap was placed on the scalp of each participant (men and women, 10 to 20 years old) and white (control) - brown - white - red colors were sequentially displayed for 5 seconds using the computer monitor. The EEG was measured and statistical analysis was performed using SPSS. For the brown color of the herbicide, men showed a decrease in concentration and a distracting response due to a decrease in the ratio of mid beta to theta (RMT) and the spectral edge of frequency (SEF90). In women, an increase in the ratio of SMR to theta (RSMT) and the spectral edge frequency 50% of the alpha (ASEF) was observed in different brain regions and these EEG changes may enhance the relaxation, stabilization and awakening states of the brain. For the red color of the non-selective herbicide, ASEF increased psychological stability in men. In women, a decrease in absolute high beta (AHB) may associate with a decrease in attention state of the brain. Overall data of the present study clearly revealed that the colors of two herbicides showed significantly different EEG response and gender difference.

• Korean Journal of Social Welfare
• /
• v.48
• /
• pp.150-178
• /
• 2002

The elderly have higher potential for contracting chronic diseases and suffering from development of a complication. Also, the extended old age period leads the elderly to demand more medical services. All those facts indicate that the elderly need more medical services than any other age groups. Consequently, medical care for the elderly with chronic diseases causes high costs burden. However, there is few studies researching the financial burden of chronic illness of the elderly. This study aimed to 1) understand how much the elderly with chronic diseases pay for medical expenses; 2) find out some specific factors related to health care financial burden; 3) suggest the alternative policies to decrease excessive financial burden of caring for the elderly with chronic illness. National Health and Nutrition Survey, which was surveyed by the Korea Institute for Health and Social Affairs in 1998, was used in this study. 4,707 persons with chronic diseases out of 5385 persons over age 60 were selectively sampled. Using SPSSWIN, correlation analysis, T-test, ANOVA and Regression were used as statistical methods in this study. Stepwise multiple regression was employed to analyze the data with a ratio of health care expenditure to income(financial burden) as a dependent variable. Out of Korean old people, 87% had the chronic diseases and their health care financial burden rate showed the average of 17.9%, which meaned they expended almost 20% income to buy medical services. The variables having a great influence on financial burden were monthly income, activity, limitation and single household of an old person. The excessive financial burden was experienced by people who had more than 4 activity limitations(37.1%) and were in the lowest Income level(32.6%), and single household of an old person(31.4%). The new policies should be considered to 1) reduce the financial burden in these groups and to develop the sliced medical cost system considering the characteristics of chronic illness and income level; and 2) develop the medical management system to care for the elderly with chronic illness.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.11 no.3
• /
• pp.92-101
• /
• 2012

Trip demands and patterns on weekends have been changed significantly over the past decade due to the income growth and the spread of the 5-day workweek in Korea. The increased weekend trips for shopping, leisure activities, entertainment and friendship have exacerbated traffic congestion in major highways or principal arterial roads from Friday afternoon through Sunday. Therefore, it is necessary to focus on travel demand forecasts and transport policies for weekend trips by investigating specific characteristics of the trips. Previous research efforts focus on simple analysis of characteristics of weekend trips and comparison of travel characteristics between weekdays and weekends. The paper analyzes the differences between weekday and weekend trips via statistical analyses to derive multiple types of characteristics of weekend trips, and develops Tobit models to identify key factors that may affect the number of trips, using Seoul city's weekend trip survey data in 2006. The model results show that weekend trips appear differently from weekdays by household or individual characteristics. Age, residence area and type of residence affected the number of trips, regardless of the type of the day, whereas gender, occupation, income, presence of household vehicle showed different impacts on trips between weekdays and weekends.

• Korean Journal of Child Studies
• /
• v.26 no.6
• /
• pp.217-236
• /
• 2005

The purpose of this study was to investigate whether preschoolers' peer interaction type and level differ depending on pair compositon and task type. Preschoolers' peer interaction type and level were compared among three kinds of pair compositions(with a Younger peer, with a same-age peer, and with an older peer) as well as between two types of tasks(tasks involving real objects, and tasks involving pictures). Subjects were seventy-eight four-year-old children recruited from six child-care centers in Seoul and Gyeonggi Province. The preschoolers were randomly assigned to one of the three kinds of pair compositions. Interaction processes of the pairs were recorded by a video camera, and the scenes were transcribed for content analyses. The traniscribed peer interactions were coded according to the categories that the researcher had generated. Statistical methods used for the data analysis were frequencies, percentiles, means, standard deviations, repeated measures ANOVA, paired t-tests, one-way ANOVA. Major findings of this study were as follows: There were significant differences preschoolers' in interaction type depending on the pair compositions and the types of tasks. The preschoolers paired with the older peer used abstract collaborative explanations more frequently than in two other types of pair composition. The preschoolers paired with the younger peer used explanations without collaboration and parallel behaviors more frequently than in two other types of pair composition.

• The Journal of the Korean dental association
• /
• v.53 no.9
• /
• pp.628-643
• /
• 2015

Purpose: The aim of this retrospective study was to compare marginal bone loss and survival rates of double short implants(multiple implant) which had been installed and restored in severely atrophic maxillary molar site without a grafting procedure. Material and Method: The subjects were patients (90 patients, 180 implants) who had been installed double short implants in severely atrophic maxillary single molar site without bone augmentation procedure from 2006 to 2014 in dental clinic in Chuncheon city. Following data were collected from dental records and radiographic panoramic views: patient's age, gender, smoking status, implant site, timing of implant installation, residual ridge height. The correlation between those factors and survival rate and marginal bone loss were analyzed. Statistical analysis was performed using Chi-square test, Student's t- test and ANOVA. Result: Eleven implants in 6 patients failed and the cumulative survival rate was 93.9%. No significant differences were found in relation to the following factors: patient's age, gender, implant site, timing of implant installation (P> .05). There were significant differences in smoking status and residual ridge height(P< .05). The average follow-up time was $45{\pm}14.7months$. The mean marginal bone loss of survived 169 implants was $0.08{\pm}0.59mm$. Conclusion: Despite the short term outcomes, the survival rate of double short implants was comparable to normal length implants. This study demonstrated that placement of double short implants without the use of bone grafting procedure for severely atrophic posterior maxilla is a simple and predictable treatment procedure.

• Journal of Technologic Dentistry
• /
• v.33 no.4
• /
• pp.495-503
• /
• 2011

Purpose: The purpose of this study was to obtain data that can be utilized in the planning of oral health education plan, by surveying the correlation between oral health knowledge, attitude, and behavior among high grade students elementary school in some regions of the metropolitan. Methods: The subject of this study through the survey of the recognition of the oral health, a questionnaire survey was conducted for 548 male and female students who were in the 4th, 5th, 6th grade of elementary school in some regions of the metropolitan. Statistical analysis was conducted using the SPSS 12.0 with t-test, ANOVA, and correlation. Results: Knowledge of oral health was very high. Correct attitudes and positive behavior were indicated. There was a significant difference in oral health knowledge according to school year and parents' dental condition(P<.05). The student one recording oral health behavior degree where the oral health knowledge is high appears highly, considers statistically the difference which it was visible. Positive correlation was significantly observed among oral health knowledge, attitudes, and behavior. Especially, attitudes and toothbrush method were highly correlated(r=.853). Conclusion: Oral health program through the elementary school students and their parents for the development of proper oral health care education programs continue to be made should be considered.

• Journal of Technologic Dentistry
• /
• v.37 no.1
• /
• pp.9-15
• /
• 2015

Purpose: The purpose of this study was to compare the accuracy of the maxillary three-unit fixed dental prosthesis (FDPs) made using two CAD/CAM milling machines : DCM Group(Dentaim CAD/CAM milling machine), WCM Group(Wieland CAD/CAM milling machine). Methods: Each of 10 duplicate models was scanned by blue light scanner(Identica, Medit, Korea), and the three-unit FDPs (STL file) was designed using DelcamCAD. A total of 20 three-unit FDPs was fabricated, comprising 2 groups of 10 specimens each (shrinkage ratio is 1:1). The first three-unit FDPs STL file was used as a CAD reference model (CRM). Obtained STL files by scanning the inner surface of three-unit FDPs were convened into the point clouds-ASC II files. Discrepancies between the point clouds and CRM were measured by superimposition software. Statistical methods to analyze the data were used non-parametric method. The mean (SD) values were compared by a Mann-Whitney U-test. Type one error rate was set at 0.05. Results: WCM group had small discrepancies with $2.17{\mu}m$ of mean value compared to $4.44{\mu}m$ in DCM group. The accuracy values between the two groups showed a sratistically significant difference (Table 2, p<.05). Conclusion: The accuracy of the three-unit fixed dental prosthesis(FDPs) made of two CAD/CAM milling machines were statistically different. Accuracy with which the prosthesis made of WCM group was superior.