Kruskal-Wallis test

クラスカル・ワリス検定

Wikipedia preview

出典(authority):フリー百科事典『ウィキペディア（Wikipedia）』「2019/05/24 17:40:24」(JST)

wiki en

[Wiki en表示]

The Kruskal–Wallis test by ranks, Kruskal–Wallis H test (named after William Kruskal and W. Allen Wallis), or one-way ANOVA on ranks is a non-parametric method for testing whether samples originate from the same distribution. It is used for comparing two or more independent samples of equal or different sample sizes. It extends the Mann–Whitney U test, which is used for comparing only two groups. The parametric equivalent of the Kruskal–Wallis test is the one-way analysis of variance (ANOVA).

A significant Kruskal–Wallis test indicates that at least one sample stochastically dominates one other sample. The test does not identify where this stochastic dominance occurs or for how many pairs of groups stochastic dominance obtains. For analyzing the specific sample pairs for stochastic dominance, Dunn's test, pairwise Mann-Whitney tests without Bonferroni correction, or the more powerful but less well known Conover–Iman test are sometimes used.

Since it is a non-parametric method, the Kruskal–Wallis test does not assume a normal distribution of the residuals, unlike the analogous one-way analysis of variance. If the researcher can make the assumptions of an identically shaped and scaled distribution for all groups, except for any difference in medians, then the null hypothesis is that the medians of all groups are equal, and the alternative hypothesis is that at least one population median of one group is different from the population median of at least one other group.

Method

1. Rank all data from all groups together; i.e., rank the data from 1 to N ignoring group membership. Assign any tied values the average of the ranks they would have received had they not been tied.
2. The test statistic is given by:

   $H=(N-1){\frac {\sum _{i=1}^{g}n_{i}({\bar {r}}_{i\cdot }-{\bar {r}})^{2}}{\sum _{i=1}^{g}\sum _{j=1}^{n_{i}}(r_{ij}-{\bar {r}})^{2}}},$ where:

   • $n_{i}$ is the number of observations in group $i$
   • $r_{ij}$ is the rank (among all observations) of observation $j$ from group $i$
   • $N$ is the total number of observations across all groups
   • ${\bar {r}}_{i\cdot }={\frac {\sum _{j=1}^{n_{i}}{r_{ij}}}{n_{i}}}$ is the average rank of all observations in group $i$
   • ${\bar {r}}={\tfrac {1}{2}}(N+1)$ is the average of all the $r_{ij}$.
3. If the data contain no ties the denominator of the expression for $H$ is exactly $(N-1)N(N+1)/12$ and ${\bar {r}}={\tfrac {N+1}{2}}$. Thus

   ${\begin{aligned}H&={\frac {12}{N(N+1)}}\sum _{i=1}^{g}n_{i}\left({\bar {r}}_{i\cdot }-{\frac {N+1}{2}}\right)^{2}\\&={\frac {12}{N(N+1)}}\sum _{i=1}^{g}n_{i}{\bar {r}}_{i\cdot }^{2}-\ 3(N+1)\end{aligned}}$
   The last formula only contains the squares of the average ranks.

4. A correction for ties if using the short-cut formula described in the previous point can be made by dividing $H$ by $1-{\frac {\sum _{i=1}^{G}(t_{i}^{3}-t_{i})}{N^{3}-N}}$, where G is the number of groupings of different tied ranks, and t_i is the number of tied values within group i that are tied at a particular value. This correction usually makes little difference in the value of H unless there are a large number of ties.
5. Finally, the decision to reject or not the null hypothesis is made by comparing $H$ to a critical value $H_{c}$ obtained from a table or a software for a given significance or alpha level. If $H$ is bigger than $H_{c}$, the null hypothesis is rejected. If possible (no ties, sample not too big) one should compare $H$ to the critical value obtained from the exact distribution of $H$. Otherwise, the distribution of H can be approximated by a chi-squared distribution with g-1 degrees of freedom. If some $n_{i}$ values are small (i.e., less than 5) the exact probability distribution of $H$ can be quite different from this chi-squared distribution. If a table of the chi-squared probability distribution is available, the critical value of chi-squared, $\chi _{\alpha :g-1}^{2}$, can be found by entering the table at g − 1 degrees of freedom and looking under the desired significance or alpha level.
6. If the statistic is not significant, then there is no evidence of stochastic dominance between the samples. However, if the test is significant then at least one sample stochastically dominates another sample. Therefore, a researcher might use sample contrasts between individual sample pairs, or post hoc tests using Dunn's test, which (1) properly employs the same rankings as the Kruskal–Wallis test, and (2) properly employs the pooled variance implied by the null hypothesis of the Kruskal–Wallis test in order to determine which of the sample pairs are significantly different. When performing multiple sample contrasts or tests, the Type I error rate tends to become inflated, raising concerns about multiple comparisons.

Exact probability tables

A large amount of computing resources is required to compute exact probabilities for the Kruskal–Wallis test. Existing software only provides exact probabilities for sample sizes less than about 30 participants. These software programs rely on asymptotic approximation for larger sample sizes.

Exact probability values for larger sample sizes are available. Spurrier (2003) published exact probability tables for samples as large as 45 participants. Meyer and Seaman (2006) produced exact probability distributions for samples as large as 105 participants.

Exact distribution of $H$

Choi et al. make a review of two methods that had been developed to compute the exact distribution of $H$, propose a new one, and compare the exact distribution to its chi-squared approximation.

See also

• Friedman test
• Jonckheere's trend test

Further reading

• Daniel, Wayne W. (1990). "Kruskal–Wallis one-way analysis of variance by ranks". Applied Nonparametric Statistics (2nd ed.). Boston: PWS-Kent. pp. 226–234. ISBN 0-534-91976-6.

External links

• An online version of the test

UpToDate Contents

全文を閲覧するには購読必要です。 To read the full text you will need to subscribe.

• 1. オーストラリア、ニュージーランド、南西太平洋（オセアニア）への旅行者における健康被害の可能性potential health hazards in travelers to australia new zealand and the southwestern pacific oceania [show details]
  … is endemic in some parts of Oceania, including Wallis and Fortuna Islands and Tonga. Infection is usually acquired during the butchering of wild pigs . Papua New Guinea has several venomous snakes …
• 2. デングウイルス感染症（デング熱）：疫学dengue virus infection epidemiology [show details]
  … 2531 paired maternal-umbilical cord blood samples were tested for dengue-specific immunoglobulin (Ig)M to determine… including American Samoa, Cook Islands, French Polynesia, New Caledonia, and Tonga .…
• 3. クリプトコッカス・ガッティ感染症：臨床的特徴および診断cryptococcus gattii infection clinical features and diagnosis [show details]
  …False-positive tests are also uncommon . Agreement between the two most commonly used commercial test formats, latex agglutination and enzyme-linked immunoassays (EIAs), is excellent (92 to 98 percent) . Test utility… thought to be largely restricted to tropical and subtropical regions, including Australia and Papua New Guinea. It is now clear that sporadic cases occur in various regions around the world. The clinical …
• 4. シガテラ中毒ciguatera fish poisoning [show details]
  … Fish can be tested using a mouse bioassay and an IgG immunoassay, but these tests are costly and time consuming and are not widely used .… toxicity although it can also be seen in patients with neurotoxic shellfish poisoning. In the South Pacific, paresis may occur in up to 10 percent of patients . Reversible cerebellar dysfunction also can …
• 5. 鼻疽の疫学、臨床症状、および診断epidemiology clinical manifestations and diagnosis of melioidosis [show details]
  … clinical specimen Serologic testing alone is not a reliable method of diagnosis. Indirect hemagglutination test (IHAT) and other serologic tests are available,… Cases are also reported from Papua New Guinea and New Caledonia .…

English Journal

• Interleukin-10 responses from acute exercise in healthy subjects: A systematic review.

• Cabral-Santos C, de Lima Junior EA, Fernandes IMDC, Pinto RZ, Rosa-Neto JC, Bishop NC, Lira FS.
• Journal of cellular physiology. 2019 Jul;234(7)9956-9965.
• Interleukin 10 (IL-10) is a cytokine that plays a critical role with potent anti-inflammatory properties when produced during exercise, limiting host immune response to pathogens and preventing tissue damage. The purpose of this systematic review was to assess the response of IL-10 after acute exerc
• PMID 30536945

• Load-bearing capacities of ultra-thin occlusal veneers bonded to dentin.

• Maeder M, Pasic P, Ender A, Özcan M, Benic GI, Ioannidis A.
• Journal of the mechanical behavior of biomedical materials. 2019 Jul;95()165-171.
• To test whether the load-bearing capacity of occlusal veneers made of ceramic or hybrid materials bonded to dentin does differ from those of porcelain-fused-to metal or lithium disilicate glass ceramic crowns. In 80 human molars, occlusal tooth substance was removed so that the defects extended into
• PMID 31009900

• Graphene Oxide/Silver Nanoparticle Coating Produced by Electrophoretic Deposition Improved the Mechanical and Tribological Properties of NiTi Alloy for Biomedical Applications.

• Rokaya D, Srimaneepong V, Qin J, Siraleartmukul K, Siriwongrungson V.
• Journal of nanoscience and nanotechnology. 2019 Jul;19(7)3804-3810.
• The aim of this study was to evaluate the mechanical and tribological properties of graphene oxide/silver nanoparticle (GO/AgNP) coated medical grade nickel-titanium (NiTi) alloy. The alloy substrates were coated using electrophoretic deposition at 30 V for 1, 5, and 10 min and were characterized by
• PMID 30764937

Japanese Journal

• 足部足関節評価質問票SAFE-Q を用いた慢性足関節不安定症を有するスポーツ選手における足部・足関節QOL の検討

• 兼岩 淳平,田中 亜吏紗,羽生 瑚幸,下村 咲喜,工藤 慎太郎
• 森ノ宮医療大学紀要 = Bulletin of Morinomiya University of Medical Sciences (13), 37-44, 2019-03-20
• … SAFE-Qについて、Kruskal-Wallis検定を用いてCAIの有無や程度、競技種目を比較した。 …
• NAID 120006586084

• 中学生の保護者を対象とした柔道授業に関する調査

• 與儀 幸朝
• 鹿児島大学教育学部研究紀要. 教育科学編 = Bulletin of the Faculty of Education, Kagoshima University. Studies in education (70), 45-53, 2019-03-11
• … The analysis of the survey results compared grades by way of the Kruskal-Wallis one-way analysis of variance test.Consideration of physique and differences in physical fitness were inspected and researched by way of the Chi-squared test and Spearman's correlation between two variables.The results of the study revealed that parents' anxiety over the safety of first grader's' who learn judo was significantly higher than that of second and third grader's' parents. …
• NAID 120006605157

• 地域包括ケアシステムにおける訪問看護ステーションの経営状況と事業所特性及び地域特性,経営管理との関連 : 全国と群馬県の比較

• 飯田 苗恵,鈴木 美雪,塩ノ谷 朱美,坪井 りえ,佐々木 馨子,大澤 真奈美,齋藤 基,柿沼 直美,山路 聡子,神山 智子,北爪 明子,牛込 久美子,依田 裕子,Saito Motoi,柿沼 直美,Kakinuma Naomi,山路 聡子,Yamazi Satoko,神山 智子,Kamiyama Tomoko,北爪 明子,Kitazume Akiko,牛込 久美子,Ushigome Kumiko,依田 裕子,Yoda Yuko
• 群馬県立県民健康科学大学紀要 14, 19-34, 2019-03
• … 明らかにし，比較検討する．方法：全国2,000人及び群馬県163人のST管理者を対象とし，自記式質問紙調査を実施した．有効回答は全国473人，群馬県55人で，経営状況と各項目との関連はχ2検定及びKruskal-Wallis検定を，さらに経営状況を目的変数とするロジスティック回帰分析を行った．結果：全国では，看護師常勤・看護師非常勤・事務職員非常勤の従業者数，1人1日平均訪問回数，北海道・東北地域 …
• NAID 120006592284

Related Links

• Kruskal-Wallis検定 - Study channel

この例の場合，水準の数は3つなので，脳梗塞の発症回数が初発，2回目，3回目の人たちの間で，発症時年齢に差があるかどうかを知りたい時にはKruskal-Wallis検定を使って水準間のいずれかに差があるかを検定します。

• Kruskal-Wallis Test in SAS | Statistical Methods

Introduction A Kruskal-Wallis test is typically performed when an analyst would like to test for differences between three or more treatments or conditions. However, unlike a one-way ANOVA, the response variable of interest is not ...

• Kruskal–Wallis test - Handbook of Biological Statistics

Summary Use the Kruskal–Wallis test when you have one nominal variable and one ranked variable. It tests whether the mean ranks are the same in all the groups. When to use it The most common use of the Kruskal–Wallis test is ...

---

★リンクテーブル★

リンク元	「クラスカル・ワリス検定」
関連記事	「testing」「test」

「クラスカル・ワリス検定」

　　[★]

英

Kruskal-Wallis test, Kruskal-Wallis H test

「testing」

　　[★]

• n.

• 試験

関

assessment、data quality、exam、examination、examine、experimental design、matched group、measurement、research design、scoring method、test、trial

　 　

「test」

　　[★] テスト