• 제목/요약/키워드: multiple linear analysis

검색결과 1,526건 처리시간 0.029초

위계적 선형모형을 이용한 강의평가 결정요인 분석 (Determinants of student course evaluation using hierarchical linear model)

  • 조장식
    • Journal of the Korean Data and Information Science Society
    • /
    • 제24권6호
    • /
    • pp.1285-1296
    • /
    • 2013
  • 강의평가 결과에 영향을 미치는 특성변수로는 교과목 수준의 다양한 강좌특성 변수들과 수강생 수준의 다양한 인적특성 변수들이 있다. 특정 수강생은 다수의 교과목을 이수하기 때문에 다수의 교과목들은 동일한 수강생 안에 속하게 됨으로써 공유되는 특성이 있게 된다. 즉 강의평가 결과는 교과목 수준의 강좌특성 (1-수준)과 수강생 수준의 인적특성 (2-수준)에 의해 영향을 받는 다층구조 (multilevel)를 가지게 되며, 위계적 자료 특성을 가지는 복수의 분석단위의 구조가 된다. 따라서 전통적인 회귀분석에서와 같이 개별 교과목들이 독립이라는 가정을 할 수 없게 된다. 본 논문에서는 강의평가결과에 영향을 미치는 다층구조의 특성을 가진 변수들의 영향력을 보다 타당하게 분석하기 위한 방법으로 위계선형모형 (HLM; hierarchical linear model)을 이용하였다. 분석결과는 다음과 같다. 먼저 교과목 수준의 특성변수들 중에 강좌규모, 개설학년, 담당교수의 전임여부, 해당 교과목의 총 평균평점, 원어강좌 여부가 통계적으로 유의하게 강의평가 결과에 영향을 미친 것으로 나타났다. 또한 수강생 수준의 인적특성 변수들 중에는 성별, 학과계열, 대입당시 전형방법, 평균평점 등이 유의하게 강의평가 결과에 영향을 미친 것으로 나타났다.

울산지역 대기오염과 알레르기 비염 유병률과의 관계 (An Association Between Air Pollution and the Prevalence of Allergic Rhinitis in the Ulsan Metropolitan Region)

  • 오인보;이지호;심창선;김양호;유철인
    • 한국환경보건학회지
    • /
    • 제36권6호
    • /
    • pp.465-471
    • /
    • 2010
  • This study aims to investigate the relationship between air pollution exposure and the prevalence of allergic rhinitis in a young population in the Ulsan metropolitan region (UMR). Data on physician-diagnosed allergic rhinitis (past 12 months) in 1,449 infants and children aged 1-18 years who lived within 1.5 or 2 km of air quality monitoring sites were collected in a cross-sectional health interview survey conducted between January-February 2006 in the UMR. Comparisons of the spatial distribution of the prevalence rates for allergic rhinitis and annual average pollutant concentrations over the region showed that a relatively high prevalence rate occurred around the coastal industrial area, with high PM10 concentrations. A linear correlation analysis demonstrated a positive correlation relationship between them (R = 0.680, p = 0.04). Multiple linear regression analysis revealed that the combined effect of the PM10 and $SO_2$ variables accounts for approximately 81% of the variance (R-square: 0.81) in the prevalence rate. From the multiple logistic regression analysis after adjustment by age, sex, and air-pollutant factors, the PM10 and $SO_2$ which were mainly from industrialrelated emissions were found to be significantly associated with an increased risk of allergic rhinitis (aOR: 1.76, 95% CI: 1.15-2.70 for PM10 ; aOR: 1.63, 95% CI: 1.12-2.35 for SO2).

Social Support and its Predictors Among Iranian Cancer Survivors

  • Faghani, Safieh;Rahmani, Azad;Parizad, Naser;Mohajjel-Aghdam, Ali-Reza;Hassankhani, Hadi;Mohammadpoorasl, Asghar
    • Asian Pacific Journal of Cancer Prevention
    • /
    • 제15권22호
    • /
    • pp.9767-9771
    • /
    • 2014
  • Background: Social support is an important factor in psycho-social well-being of cancer survivors. There is little information about level of social support and its predictors among cancer survivors in Iran or other Middle Eastern countries. The aims of present study were to determine the social support and its prediction factors among Iranian cancer survivors. Materials and Methods: In this descriptive-correlational study 187 cancer patients in one educational center and one private oncology office in northwest of Iran participated using a convenient sampling method. The data collection tool consisted of a researcher-prepared checklist and the Multidimensional Scale of Perceived Social Support Assessment (MSPSS). Data analysis was performed using SPSS statistical software with descriptive statistics and multiple linear regression analysis. Results: The total score of MSPSS was 68 from a possible score between 7 and 84. Participants believed that they received a high level of support from their family members and significant others. Multiple linear regression analysis showed that single and depressed cancer survivors and participants with lower levels of physical activity believed that they received lower levels of social support. Conclusions: Iranian cancer survivors receive high levels of social support and family members are the most important source of this support. In planning any supportive care program for Iranian cancer survivors this strength should be considered. Especially, single and depressed and patients with lower levels of physical activity need more attention.

The Effect of Violent Experience on Burnout among Some Dental Hygienists

  • Jeon, Eun-Jeong;Han, Mi Ah;Park, Jong;Choi, Seong Woo
    • 치위생과학회지
    • /
    • 제17권5호
    • /
    • pp.413-422
    • /
    • 2017
  • This study investigated the effect of violent experience on burnout among some dental hygienists. The study subjects were 242 dental hygienists working at dental clinics. Data were collected by self-administered questionnaire including information such as demographics, work-related characteristics, working environment, experience of violence, and burnout. The violence was classified as verbal violence, physical threat, or physical violence committed by dentists, patients, or caregivers. Descriptive analysis, t-test, ANOVA, correlation, and multiple linear regression analysis were performed to examine the factors associated with burnout. The levels of verbal violence, physical threat, and physical violence by dentists were $0.53{\pm}1.26$, $1.12{\pm}2.70$, and $0.04{\pm}0.42$, respectively. The levels of verbal violence, physical threat, and physical violence by patients and caregivers were $1.50{\pm}1.89$, $1.41{\pm}2.24$, and $0.24{\pm}1.38$, respectively. The score of burnout was $3.13{\pm}0.43$. Total violence, verbal violence, and physical violence by dentists were positively correlated with burnout. Total violence, verbal violence, and physical threat by patients and caregivers were positively correlated with burnout. In multiple linear regression analysis, the level of physical violence by dentists was positively associated with burnout of dental hygienists (${\beta}=0.95$, p=0.032). The levels of total physical violence (${\beta}=0.28$, p=0.002), verbal violence (${\beta}=0.15$, p<0.001), and physical threat (${\beta}=0.19$, p=0.009) by the patients or caregivers were positively associated with burnout of dental hygienists. This study examined the association between violence and burnout among dental hygienists. The level of violence showed positive correlation with burnout. Environment improvement to protect employees from violence and for management of employees who experienced workplace violence are needed to reduce the burnout.

다중선형 회귀분석에 의한 LiDAR 자료의 필터링 자동화 기법 (An Filtering Automatic Technique of LiDAR Data by Multiple Linear Regression Analysis)

  • 최승필;조지현;김준성
    • 대한공간정보학회지
    • /
    • 제19권4호
    • /
    • pp.109-118
    • /
    • 2011
  • 본 연구는 지면 데이터�V을 이용하여 다중선형 회귀분석에 의한 평면방정식을 도출하여 전역필터링 한 것을 기준으로 전체 데이터�V을 이용하여 도출된 평면방정식으로 전역필터링 한 것과 가상격자별로 평면방정식을 도출하여 지역필터링을 수행한 결과를 분석하여 정확도를 평가하였다. 그 결과 지면 데이터�V을 이용한 전역필터링의 평균정확도를 기준으로 전체 데이터�V을 이용한 전역필터링의 정확도는 약 2~3%정도 떨어지고, 가상격자를 이용한 지역필터링의 정확도는 약 2~4% 떨어지는 것으로 나타났다. 특히 가상격자가 3~4cm일 때 기준자료와 약 2%의 정확도의 차이가 나타낸 것으로 보아 가상격자 사이즈를 라이다 스캔간격의 3~4배 크기로 지정하여 필터링 하는 것이 바람직 할 것으로 판단된다. 따라서 필터링의 적용방법에 따라 평균정확도가 차이가 발생하였으며, 향후 보다 다양한 실제지형을 선정하여 필터링의 정확도에 대한 연구가 필요할 것으로 생각된다.

지역사회 스케일링경험률에 영향을 미치는 요인: 지역사회건강조사 자료이용 (Factors associated with community scaling rate: Using community health survey data)

  • 김지민;하주원;김지수;정연호;김동석;이가영;장영은;김남희
    • 한국치위생학회지
    • /
    • 제15권6호
    • /
    • pp.1053-1061
    • /
    • 2015
  • Objectives: The purpose of the study is to investigate the influencing factors of community scaling rate using community health survey data. Methods: The data were extracted from 2013 Community Health Survey, Ministry of education, Korea Dental Association, Statistics Korea, Health Insurance Review and Assessment Service, and Ministry of the Interior. The resource factors of independent variables were analysed by Geographical Information System(GIS) using Map Wizard for Excel 17.0. The data were analyzed by descriptive analysis, pearson correlation and multiple linear regression analysis(p<0.05). Results: Seocho-gu in Seoul had the highest annual scaling rate(55.5%) and Goheung-gun had the lowest rate(11%) showing 44.5 percent gap. The influencing factors of scaling included the number of dental hygienists(r=0.316), dentists(r=0.332), dental hospitals(r=0.470), high school graduation rate(r=0.757) and equivalence scales household income(r=0.764)(p<0.05). Multiple linear regression analysis showed that community scaling rate was closely associated with community education level and monthly income(p<0.05). Conclusions: Community scaling rate was closely related to the community education and income level. It is necessary to provide the equal distribution of the oral health service to the community society.

Widely-Linear Beamforming and RF Impairment Suppression in Massive Antenna Arrays

  • Hakkarainen, Aki;Werner, Janis;Dandekar, Kapil R.;Valkama, Mikko
    • Journal of Communications and Networks
    • /
    • 제15권4호
    • /
    • pp.383-397
    • /
    • 2013
  • In this paper, the sensitivity of massive antenna arrays and digital beamforming to radio frequency (RF) chain in-phase quadrature-phase (I/Q) imbalance is studied and analyzed. The analysis shows that massive antenna arrays are increasingly sensitive to such RF chain imperfections, corrupting heavily the radiation pattern and beamforming capabilities. Motivated by this, novel RF-aware digital beamforming methods are then developed for automatically suppressing the unwanted effects of the RF I/Q imbalance without separate calibration loops in all individual receiver branches. More specifically, the paper covers closed-form analysis for signal processing properties as well as the associated radiation and beamforming properties of massive antenna arrays under both systematic and random RF I/Q imbalances. All analysis and derivations in this paper assume ideal signals to be circular. The well-known minimum variance distortionless response (MVDR) beamformer and a widely-linear (WL) extension of it, called WL-MVDR, are analyzed in detail from the RF imperfection perspective, in terms of interference attenuation and beamsteering. The optimum RF-aware WL-MVDR beamforming solution is formulated and shown to efficiently suppress the RF imperfections. Based on the obtained results, the developed solutions and in particular the RF-aware WL-MVDR method can provide efficient beamsteering and interference suppressing characteristics, despite of the imperfections in the RF circuits. This is seen critical especially in the massive antenna array context where the cost-efficiency of individual RF chains is emphasized.

지체장애인의 인공지능 스피커 사용 의도에 영향을 미치는 요인에 관한 연구 (A Study on the Factors Affecting the Intention to Use Artificial Intelligence Speakers of the People with Physical Disability)

  • 박혜현;이선민
    • 한국콘텐츠학회논문지
    • /
    • 제21권1호
    • /
    • pp.572-578
    • /
    • 2021
  • 본 연구는 지체장애인을 대상으로 인공지능 스피커에 대한 인지적 요소와 감정적 요소가 인공지능 스피커 사용 의도에 미치는 영향을 검증하는 것을 목적으로 하였다. 연구방법은 지체장애인을 대상으로 온라인 설문 조사를 실시하였다. 인공지능 스피커에 대한 인지도와 필요도, 인지된 친밀함과 즐거움, 사용 의도를 파악하였으며, 각 변인이 지체장애인의 인공지능 스피커 사용 의도에 미치는 영향력을 확인하기 위해 다중회귀분석(Multiple linear regression analysis)을 실시하였다. 연구 결과, 지체장애인의 인공지능 스피커에 대한 인지된 즐거움은 사용 의도에 유의한 정적 영향을 나타내었다. 그러나 지체장애인의 인공지능 스피커에 대한 인지도와 필요도, 인지된 친밀함은 인공지능 스피커 사용 의도에 통계적으로 유의미한 영향을 나타내지 않는 것으로 분석되었다. 본 연구의 결과는 장애인의 인공지능 스피커 사용 의도 향상을 위해 즐거움 요소를 강화하는 것이 필요함을 시사하며, 장애인을 위한 인공지능 제품과 맞춤형 서비스를 개발하기 위한 기초자료를 제공하는 점에서 의의가 있다.

Multi-material topology optimization for crack problems based on eXtended isogeometric analysis

  • Banh, Thanh T.;Lee, Jaehong;Kang, Joowon;Lee, Dongkyu
    • Steel and Composite Structures
    • /
    • 제37권6호
    • /
    • pp.663-678
    • /
    • 2020
  • This paper proposes a novel topology optimization method generating multiple materials for external linear plane crack structures based on the combination of IsoGeometric Analysis (IGA) and eXtended Finite Element Method (X-FEM). A so-called eXtended IsoGeometric Analysis (X-IGA) is derived for a mechanical description of a strong discontinuity state's continuous boundaries through the inherited special properties of X-FEM. In X-IGA, control points and patches play the same role with nodes and sub-domains in the finite element method. While being similar to X-FEM, enrichment functions are added to finite element approximation without any mesh generation. The geometry of structures based on basic functions of Non-Uniform Rational B-Splines (NURBS) provides accurate and reliable results. Moreover, the basis function to define the geometry becomes a systematic p-refinement to control the field approximation order without altering the geometry or its parameterization. The accuracy of analytical solutions of X-IGA for the crack problem, which is superior to a conventional X-FEM, guarantees the reliability of the optimal multi-material retrofitting against external cracks through using topology optimization. Topology optimization is applied to the minimal compliance design of two-dimensional plane linear cracked structures retrofitted by multiple distinct materials to prevent the propagation of the present crack pattern. The alternating active-phase algorithm with optimality criteria-based algorithms is employed to update design variables of element densities. Numerical results under different lengths, positions, and angles of given cracks verify the proposed method's efficiency and feasibility in using X-IGA compared to a conventional X-FEM.

주·야간 교통사고의 특성 및 사고모형 비교분석 -청주시 4지 신호교차로를 중심으로 - (Comparative Analysis on the Characteristics and Models of Traffic Accidents by Day and Nighttime in the Case of Cheongju 4-legged ignalized Intersections)

  • 유두선;오상진;김태영;박병호
    • 대한토목학회논문집
    • /
    • 제28권2D호
    • /
    • pp.181-189
    • /
    • 2008
  • 본 연구는 주간과 야간의 교통사고 특성과 사고모형을 비교 분석하는데 그 목적이 있다. 이를 위해 본 연구에서는 청주시 4지 신호교차로를 중심으로 주 야간 교통사고의 차이를 검증하고, 주야간별 다중선형, 다중비선형 과 포아송 및 음이항회귀모형을 개발하는데 중점을 두고 있다. 분석된 주요 연구결과는 다음과 같다. 첫째, 사고특성 분석을 통해 주간과 야간 사고의 차이점이 파악되었다. 둘째, 통계적으로 설명력이 높은 주 야간 총 12개의 사고모형이 개발되었다. 마지막으로 공통변수와 특정변수를 활용하여 주야간 사고모형의 차이가 비교 분석되었다.