• Journal of the Korean Data and Information Science Society
• /
• 제18권2호
• /
• pp.525-534
• /
• 2007

It is a common practice in many research areas that multiple range tests for a between subject factor such as Tukey are applied to a within subject factor in a repeated measures design. Tukey procedure, however, sometimes detects no pairs with different means even when the hypothesis of all equal level means is rejected. This study attempts to provide a rationale for the proposition that Tukey is inappropriate post hoc procedure for a within subject factor in which the observations are correlated. We introduce two multiple range tests, Bonferroni and Scheffe, for a within subject factor and show that Bonferroni is more appropriate than Scheffe for pairwise multiple comparisons. Subsequent simulation study indicates that Tukey has significantly less power than Bonferroni in detecting actual difference between means of some pairs when the observations of a within subject factor are highly correlated.

• Communications for Statistical Applications and Methods
• /
• 제4권3호
• /
• pp.775-785
• /
• 1997

Robustness rates for repeated measures pairwise multiple comparison procedures were investigated in a split plot design with one between- and one within-subjects factor using untrimmed and trimmed data. Five factors were manipulated in the study: distribution, sphericity, variance-covariance heteroscedasticity, total sample size, and sample size ratio. The Welsch test (W) and the Welsch test on trimmed data $(W_{RT})$ performed better than the other procedures, but had a liberal tendency. The trimmed difference score Bonferroni Procedure $(B_{DT})$ was a good choice in some conditions.

• Communications for Statistical Applications and Methods
• /
• 제12권2호
• /
• pp.305-312
• /
• 2005

In the cluster analysis, Hotelling's $T^2$ can be used to estimate the unknown number of clusters based on the idea of multiple comparison procedure. Especially, its threshold is obtained according to the probability of committing the type one error. Examples are used to compare Hotelling's $T^2$ with other classical location test statistics such as Sum-of-Squared Error and Wilks' $\Lambda$ The hierarchical clustering is used to reveal the underlying structure of the data. Also related criteria are reviewed in view of both the between variance and the within variance.

• Journal of the Korean Data and Information Science Society
• /
• 제25권5호
• /
• pp.971-986
• /
• 2014

동시에 여러 개의 가설검정 수행시 귀무가설이 참일 경우 귀무가설을 기각할 확률이 커지는 문제가 발생한다. 이러한 다중검정 문제 해결을 위해 여러 연구에서는 가설검정시 필요한 집단별 오류율(FWER; family-wise error rate), 위발견율 (FDR; false discovery rate) 또는 위비발견율 (FNR; false nondiscovery rate) 과 통계량을 고려하여 검정력을 높이고자 하였다. 본 연구에서는 T 통계량, 수정된 T 통계량, 그리고 LPE (local pooled error) 통계량 기반 P값을 이용한 Bonferroni (1960) 방법, Holm (1979) 방법, Benjamini와 Hochberg (1995) 방법과 Benjamini와 Yekutieli (2001) 방법 그리고 Z 통계량 기반 Sun과 Cai (2007) 방법을 고찰하고 모의실험을 통해 다중검정 능력을 비교하였다. 또한 실제 데이터로 애기장대 유전자 발현 데이터에 대해 여러 가지 다중검정법을 통해 유의한 유전자들을 선별하였다.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제47권6호
• /
• pp.438-444
• /
• 2021

Objectives: Loss of the interdental papilla is multi-factorial and creates a multitude of problems. Autogenous connective tissue/biomaterial-based regeneration has been attempted for decades to reconstitute the black space created due to the loss of papilla. The aim of this present study was to regenerate papillary recession defects using an amnion-chorion membrane (ACM) allograft and to evaluate the clinical outcome up to six months postoperatively. Materials and Methods: Twenty patients with 25 Nordland and Tarnow's Class I/II interdental papillary recession defects were treated with ACM and coronal advancement of the gingivo-papillary unit via a semilunar incision on the labial aspect followed by a sulcular incision in the area of interest. A photographic image analysis was carried out using the GNU Image Manipulation software program from the baseline to three and six months postoperatively. The black triangle height (BTH) and the black triangle width (BTW) were calculated using the pixel size and were then converted into millimeters. The mean and standard deviation values were determined at baseline and then again at three and six months postoperatively. The probability values (P<0.05 and P≤0.01) were considered statistically significant and highly significant, respectively. An analysis of variance and post hoc Bonferroni test were carried out to compare the mean values. Results: Our evaluation of the BTH and BTW showed a statistically and highly significant difference from the baseline until both three and six months postoperatively (P=0.01). A post hoc Bonferroni test disclosed a statistically significant variance from the baseline until three and six months postoperatively (P<0.05) and a non-significant difference from three to six months after the procedure (P≥0.05). Conclusion: An ACM allograft in conjunction with a coronally advanced flap could be a suitable minimally invasive alternative for papillary regeneration.

• 한국시뮬레이션학회논문지
• /
• 제10권1호
• /
• pp.51-62
• /
• 2001

For the productivity improvement, it is necessary to analyze problem of the system and evaluate alternative plans. But, it if hard to experiment and evaluate alternative plans in real existing system. Therefore, it is necessary to synthetically analyze and evaluate alternative plans using simulation and statistical analysis method. In this case study, as a simulation language, Promodel is used to compare and evaluate four alternatives which are considered to improve the productivity of an object system. The object system is constituted of a pretreatment line and a painting line. The purpose of this study is to analyze whether the decrease of the down time in pretreatment line can be effective to the productivity improve comparing with exiting system and other alternatives. The other purpose is to choose an alternative that has the best productivity improvement. The results of the simulation show that the possibility of improvement in productivity by decreasing the down time of pretreatment line. Two-Stage Bonferroni Procedure is used to choose the best alternative.

• The Journal of Advanced Prosthodontics
• /
• 제10권4호
• /
• pp.300-307
• /
• 2018

PURPOSE. To investigate the fatigue and fracture resistance of computer-aided design and computer-aided manufacturing (CAD/CAM) ceramic molar crowns on dental implants and human teeth. MATERIALS AND METHODS. Molar crowns (n=48; n=8/group) were fabricated of a lithium-disilicate-strengthened lithium aluminosilicate glass ceramic (N). Surfaces were polished (P) or glazed (G). Crowns were tested on human teeth (T) and implant-abutment analogues (I) simulating a chairside (C, crown bonded to abutment) or labside (L, screw channel) procedure for implant groups. Polished/glazed lithium disilicate (E) crowns (n=16) served as reference. Combined thermal cycling and mechanical loading (TC: $3000{\times}5^{\circ}C/3000{\times}55^{\circ}C$; ML: $1.2{\time}10^6$ cycles, 50 N) with antagonistic human molars (groups T) and steatite spheres (groups I) was performed under a chewing simulator. TCML crowns were then analyzed for failures (optical microscopy, SEM) and fracture force was determined. Data were statistically analyzed (Kolmogorow-Smirnov, one-way-ANOVA, post-hoc Bonferroni, ${\alpha}=.05$). RESULTS. All crowns survived TCML and showed small traces of wear. In human teeth groups, fracture forces of N crowns varied between $1214{\pm}293N$ (NPT) and $1324{\pm}498N$ (NGT), differing significantly ($P{\leq}.003$) from the polished reference EPT ($2044{\pm}302N$). Fracture forces in implant groups varied between $934{\pm}154N$ (NGI_L) and $1782{\pm}153N$ (NPI_C), providing higher values for the respective chairside crowns. Differences between polishing and glazing were not significant ($P{\geq}.066$) between crowns of identical materials and abutment support. CONCLUSION. Fracture resistance was influenced by the ceramic material, and partly by the tooth or implant situation and the clinical procedure (chairside/labside). Type of surface finish (polishing/glazing) had no significant influence. Clinical survival of the new glass ceramic may be comparable to lithium disilicate.

• 벤처창업연구
• /
• 제9권6호
• /
• pp.183-198
• /
• 2014

본 연구는 신제품 개발 전략 보유 여부 및 정형화된 신제품 개발(NPD)활동 여부가 국내 코스닥 기업들의 신제품 개발 성과에어떠한 영향을 미치는 지를 분석하였다. 본 연구결과 신제품 개발 프로그램을 주도하고 통합 할 수 있는 구체적인 신제품전략을 보유하고 있는 기업이 48.5%, 보유하고 있지 않은 기업이 51.5%로 아직은 신제품개발 프로그램을 보유하고 있지 않은 기업이 다소 많은 것으로 나타났다. 신제품개발 성과달성 결과는 신제품 성공비율이 51.21%, 신제품수익 성공비율은 44.87%로 나타났으나, 1993년 미국의 Kuczmarski & Associates의 77개 다양한 산업들을 대상으로 조사하여 신제품 개발 성과가 '매우 성공적'이거나 '성공적'이라고 평가한 응답자들이 53%, 미국의 제품개발관리협회(PDMA)에서 1995년에 두 번째 best practice 연구를 통하여 신제품 개발 성공률이 59% 였던 점을 비교했을 때 이번 조사 기업은 1995년 미국의 제품개발관리협회에서 조사한 결과의 평균에 다소 못 미치는 것으로 나타났다. 신제품 개발 전략을 보유한 여부에 따라 신제품 개발 성과간의 차이가 있는 지를 검증한 결과 신제품개발 전략을 보유여부에 따라 신제품 성공비율이 차이가 나고 신제품전략을 보유할수록 신제품 성공비율이 높은 것으로 나타났다. 신제품 개발활동은 문서화된 프로세스는 없지만 제품 개발에 따르는 개별업무가 완성되기까지 명확하게 인지된 절차를 따르는 기업이 39.6%로써 미국의 1990년 PDMA 결과 76%인 것과 비교해 볼 때 아직 우리 코스닥 기업들이 미국 기업들 보다는 정형화된 NPD 프로그램을 적게 보유하고 있는 것으로 나타났다, ANOVA 검증 결과 공식적으로 문서화된 프로세스 존재여부에 따라 신제품 개발 성공률의 차이가 있는 것으로 나타났다. Scheffe, Bonferroni 사후검증 결과 공식적으로 문서화된 프로세스는 없지만 개별업무 절차를 보유한 기업들 보다 공식 문서화된 프로세스를 보유하고 정해진 부문에서 일련의 업무를 수행하는 기업들이 5%유의수준에서 18.385 만큼 신제품 개발 성공율이 높은 것으로 나타났다. 본 연구의 기존 연구와의 차별성은 미국의 제품개발관리협회(PDMA)가 미국전역에 걸쳐 연구한 내용을 국내 코스닥기업에 적용했다는 점이다. 또한 국내의 신제품개발관련 기존 연구가 과거 일부 산업에 국한된 연구가 이루어졌으나 본 연구에서는 국내 코스닥기업의 다양한 산업에 적용하였다는 것이다. 본연구의 시사점으로는 코스닥기업협회나 관련 부처에서는 기업들의 생존율을 높이고 경쟁력을 강화하기 위해 신제품개발에 대한 정형화된 프로세스를 교육하고 지도할 필요가 있다는 것이다.

• The Korean Journal of Pain
• /
• 제28권2호
• /
• pp.88-95
• /
• 2015

Background: Patients suffering from non-cardiac chest pain (NCCP) can interpret their chest pain wrongly despite having received a correct diagnosis. The objective of this study was to compare the efficacy of the relaxation method with metaphor therapy for reducing irrational beliefs and pain severity in patients with NCCP. Methods: Using a randomized controlled trial, 33 participants were randomly divided into a relaxation training group (n= 13), a metaphor therapy group (n = 10), and a control group (n = 10), and were studied for 4 weeks. The two tools used in this research were the Brief Pain Inventory (BPI) index for determining the degree of pain and the short version of the Jones Irrational Belief Test. Metaphor therapy and a relaxation technique based on ${\ddot{O}}$st's treatment were used as the interventions. The collected data were analyzed with a multivariate analysis of covariance (MANCOVA), a Chi-square test, and the Bonferroni procedure of post-hoc analysis. Results: The relaxation training method was significantly more effective than both metaphor therapy and the lack of treatment in reducing the patients' beliefs of hopelessness in the face of changes and emotional irresponsibility, as well as the pain severity. Metaphor therapy was not effective on any of these factors. In fact, the results did not support the effectiveness of metaphor therapy. Conclusions: Regarding the effectiveness of the relaxation method as compared with metaphor therapy and the lack of treatment in the control group, this study suggests that relaxation should be paid greater attention as a method for improving the status of patients. In addition, more studies are needed to determine the effectiveness of metaphor therapy in this area.

• 한국전문물리치료학회지
• /
• 제20권3호
• /
• pp.81-88
• /
• 2013

Active straight-leg raise (ASLR) is a physical evaluation procedure to test lumbar spine stability. Several previous studies have reported various methods to control the activation of abdominal muscles during ASLR. We investigated the effects of three different hip positions in frontal plane on abdominal muscles to increase or decrease the difficulty level of lumbar spine stability exercise during ASLR in pain free subjects. Eleven young and healthy subjects voluntarily participated in this study (6 men, 5 women; mean age=$24.0{\pm}1.2$ years, height=$160.0{\pm}7.3cm$, weight=$55.0{\pm}10.6kg$, body mass index=$21.5{\pm}2.3kg/m^2$). The subjects had three trials on each ASLR with hip $10^{\circ}$ adduction, neutral hip, and hip $30^{\circ}$ abduction. Separate repeated-measures analysis of variance (ANOVA) and the post hoc Bonferroni tests (with ${\alpha}$=.05/3=.017) were performed for each muscle among the three different hip positions in frontal plane (ASLR with hip $10^{\circ}$ adduction, neutral hip, and hip $30^{\circ}$ abduction). The ipsilateral external oblique (EO), contralateral EO, ipsilateral internal oblique/transverse abdominis (IO/TrA), and contralateral IO/TrA were significantly greater in ASLR with hip $30^{\circ}$ abduction compared with ASLR with hip $10^{\circ}$ adduction. Also, the ipsilateral EO, contralateral EO, and ipsilateral IO/TrA were significantly greater in ASLR with hip $30^{\circ}$ abduction compared with ASLR with neutral hip. These results suggest that ASLR with hip $30^{\circ}$ abduction and neutral would be useful method to strengthen the EO and IO/TrA. And, ASLR with hip $10^{\circ}$ adduction would be effective in early stages of lumbar stabilization program due to low activation of EO and IO/TrA during maintaining of ASLR position with low load.