{"id":127,"date":"2021-05-04T22:19:06","date_gmt":"2021-05-04T22:19:06","guid":{"rendered":"https:\/\/webs.uab.cat\/biologiamarina\/analisis-de-dades\/"},"modified":"2021-05-04T22:19:06","modified_gmt":"2021-05-04T22:19:06","slug":"analisis-de-dades","status":"publish","type":"page","link":"https:\/\/webs.uab.cat\/biologiamarina\/analisis-de-dades\/","title":{"rendered":"An\u00e0lisis de dades"},"content":{"rendered":"<p>El primer que s&#8217;ha de fer \u00e9s estandaritzar les dades recollides al camp o al laboratori.<\/p>\n<p>Per exemple:<\/p>\n<ul>\n<li>posar sempre el mateix n\u00famero de decimals,<\/li>\n<li>definir si els decimals s\u00f3n comes o punts,<\/li>\n<li>en les variables categ\u00f2riques, que les abreviacions de cada categoria sempre siguin iguals (exemple, no posar &#8220;Barcelona&#8221; i &#8220;BCN&#8221;, per categoritzar el mateix).<\/li>\n<\/ul>\n<p>En segon lloc hem de triar el test estad\u00edstic que volem fer per comprovar la nostra hip\u00f2tesi.<\/p>\n<p>&nbsp;<\/p>\n<p><strong>DISTRIBUCI\u00d3 DE LES VARIABLES<\/strong><\/p>\n<p><strong>Segueixen una normal?<\/strong><\/p>\n<p>S&#8217;ha d&#8217;avaluar la manera en la que es distribueixen les dades al llarg de les variables.<\/p>\n<ul>\n<li><strong>Distribuci\u00f3 de Poisson: <\/strong> Tracta freq\u00fc\u00e8ncies. Utilitza valors discrets.<\/li>\n<li><strong>Distribuci\u00f3 binomial: <\/strong>\u00e9s la que t\u00e9 2 resultats possibles \u00e0 Control\/Experiment, Mascle\/Femella&#8230;<\/li>\n<li><strong>Distribuci\u00f3 normal:<\/strong> Utilitza valors continus i es descriu mitjan\u00e7ant la mitjana i la vari\u00e0ncia. Una distribuci\u00f3 normal ideal \u00e9s la que segueix la campana de Gauss.<\/li>\n<\/ul>\n<p><strong>Segueixen una vari\u00e0ncia constant?<\/strong><\/p>\n<p>HOMOSTECE\u00cfTAT<\/p>\n<p>Quan les vari\u00e0ncies s\u00f3n iguals.<\/p>\n<p>HETEROSTECE\u00cfTAT<\/p>\n<p>Quan les vari\u00e0ncies s\u00f3n diferents.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"193\" src=\"https:\/\/grupsderecerca.uab.cat\/biologiamarina\/sites\/grupsderecerca.uab.cat.biologiamarina\/files\/destacats\/homoihetero.png\" width=\"906\" \/><\/p>\n<p>&nbsp;<\/p>\n<p><u><strong>Tipus d\u2019an\u00e0lisi<\/strong><\/u><\/p>\n<ul>\n<li><strong>Estad\u00edstica descriptiva: <\/strong>Quan volem resumir, observar, descriure i\/o explicar com \u00e9s una poblaci\u00f3. \u00c9s on utilitzem: la mitjana, la mediana, percentatges, dispersi\u00f3&#8230;<\/li>\n<li><strong>Estad\u00edstica de difer\u00e8ncies: <\/strong>Quan volem comparar 2 o m\u00e9s categories de dades. \u00c9s on utilitzem: t-test, ANOVA, Xi<sup>2<\/sup>.<\/li>\n<li><strong>Estad\u00edstica de relacions: <\/strong>Quan volem veure si existeix alguna relaci\u00f3 entre les variables, \u00e9s a dir, veure si hi ha un patr\u00f3 en el qual quan una variable canvia l\u2019altre tamb\u00e9 ho faci. \u00c9s on utilitzem: regressi\u00f3, correlaci\u00f3&#8230;<\/li>\n<li><strong>An\u00e0lisi multivariant: <\/strong>Quan volem treballar amb un gran nombre de dades i variables. \u00c9s on fem servir l\u2019an\u00e0lisi multivariant. \u00c9s on utilitzem: PCA.<strong>Tipus de test<\/strong><\/li>\n<\/ul>\n<p><u><strong>Tests param\u00e8trics<\/strong><\/u><\/p>\n<p>S\u2019han de complir les seg\u00fcents condicions:<\/p>\n<ul>\n<li>Les variables han de seguir una distribuci\u00f3 determinada. Si la distribuci\u00f3 que se\u2019ns demana seguir \u00e9s normal, i observem que no ho \u00e9s, primer provem de transformar les variables:<\/li>\n<\/ul>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"113\" src=\"https:\/\/grupsderecerca.uab.cat\/biologiamarina\/sites\/grupsderecerca.uab.cat.biologiamarina\/files\/destacats\/transformacions.png\" width=\"615\" \/><\/p>\n<ul>\n<li>Hi ha d\u2019haver homoscedasticitat<\/li>\n<li>Hi ha d\u2019haver relaci\u00f3 lineal en el cas de l\u2019an\u00e0lisi de relacions<\/li>\n<\/ul>\n<p><em>Exemples tests param\u00e8trics: t-test, ANOVA, Xi<sup>2<\/sup>, R-Pearson&#8230;<\/em><\/p>\n<p><strong>Tests NO param\u00e8trics <\/strong>\u00e0 alternativa si no es compleixen les condicions dels test param\u00e8trics. Indueix errors i \u00e9s millor evitar-la.<\/p>\n<p><em>Exemples: U Mann-Whitney, Wilcoxon test, Kruskal Wallis, Friedman Test, S-Spearman, T-Kendall&#8230;<\/em><\/p>\n<p><strong>SELECCI\u00d3 DEL TEST<\/strong><\/p>\n<p>Quan tenim clares les variables que voldrem mesurar podem proposar un test per tal d\u2019acceptar o rebutjar la hip\u00f2tesi.<\/p>\n<p>Hem de saber si les variables s\u00f3n quantitatives o qualitatives i a partir d\u2019aqu\u00ed podem proposar un test.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" alt=\"\" height=\"561\" src=\"https:\/\/grupsderecerca.uab.cat\/biologiamarina\/sites\/grupsderecerca.uab.cat.biologiamarina\/files\/destacats\/dissenytest_0.jpg\" width=\"861\" \/><\/p>\n<p>I recordar si s&#8217;ha d&#8217;utilitzar un test param\u00e8tric o no, en funci\u00f3 de si les variables segueixen una normal, tenen vari\u00e0ncia constant i si hi ha momostece\u00eftat.<\/p>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n<p><strong>QUIN TEST \u00c9S M\u00c9S ADIENT? <\/strong><\/p>\n<p><strong>La clau dicot\u00f2mica de la estad\u00edstica.<\/strong><\/p>\n<p>A continuaci\u00f3 es mostra un clau dicot\u00f2mica per a fer servir en an\u00e0lisis de relacions i difer\u00e8ncies:<\/p>\n<p>1) Vull fer una an\u00e0lisi de relacions&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.2<\/p>\n<p>&nbsp;&nbsp;&nbsp; Vull fer una an\u00e0lisi de difer\u00e8ncies&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..7<\/p>\n<p>2) Les dades segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;3<\/p>\n<p>&nbsp;&nbsp;&nbsp; Les dades no segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.6<\/p>\n<p>3) Les dades s\u00f3n independents&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;4<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp; Les dades no s\u00f3n independents&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..<strong>.S-Spearman o T-Kendall<\/strong><\/p>\n<p>4) Hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.5<\/p>\n<p>&nbsp;&nbsp;&nbsp; No hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;<strong>&#8230;&#8230;&#8230;&#8230;S-Spearman o T-Kendall<\/strong><\/p>\n<p>5) Hi ha relaci\u00f3 lineal entre les variables&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..<strong>R-Pearson<\/strong><\/p>\n<p>&nbsp;&nbsp;&nbsp; No hi ha relaci\u00f3 entre les variables&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..<strong>S-Spearman o T-Kendall<\/strong><\/p>\n<p>6) Transforma les variables i ara s\u00ed hi ha distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.3<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp; Transforma les variables i segueix sense haver distribuci\u00f3 normal&#8230;&#8230;<strong> S-Spearman o T-Kendall<\/strong><\/p>\n<p>7) Es vol comprar 2 variables qualitatives&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;<strong>Xi<sup>2<\/sup><\/strong><\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp; Es vol comparar 1 variable quantitativa i 1 variable qualitativa&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.8<\/p>\n<p>8) La variable quantitativa es de 2 categories&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;9<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp; La variable quantitativa es de m\u00e9s de 2 grups&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..16<\/p>\n<p>9) Les dades s\u00f3n dependents&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.10<\/p>\n<p>&nbsp;&nbsp;&nbsp; &nbsp;Les dades s\u00f3n independents&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;13<\/p>\n<p>10) Les dades segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..12<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp; &nbsp;&nbsp;Les dades no segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;11<\/p>\n<p>11) Transforma les variables i ara s\u00ed hi ha distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;12<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Transforma les variables i segueix sense haver distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..<strong> Wilcoxon-Test<\/strong><\/p>\n<p>12) Hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.<strong>Paired t-test<\/strong><\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; No hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;<strong>Wilcoxon-Test<\/strong><\/p>\n<p>13) Les dades segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..15<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Les dades no segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;14<\/p>\n<p>14) Transforma les variables i ara s\u00ed hi ha distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;15<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Transforma les variables i segueix sense haver distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;.<strong>U-Mann Whitney <\/strong><\/p>\n<p>15) Hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;<strong>Unpaired t-test<\/strong><\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; No hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.<strong>U-Mann Whitney<\/strong><\/p>\n<p>16) Les dades s\u00f3n dependents&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..17<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Les dades s\u00f3n independents&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..20<\/p>\n<p>17) Les dades segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..19<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Les dades no segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;18<\/p>\n<p>18) Transforma les variables i ara s\u00ed hi ha distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;19<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Transforma les variables i segueix sense haver distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..<strong>.Friedman Test<\/strong><\/p>\n<p>19) Hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;.<strong>Paired ANOVA<\/strong><\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; No hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;<strong>Friedman Test<\/strong><\/p>\n<p>20) Les dades segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..22<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Les dades no segueixen una distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;21<\/p>\n<p>21) Transforma les variables i ara s\u00ed hi ha distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;22<\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Transforma les variables i segueix sense haver distribuci\u00f3 normal&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;..<strong>Kruskal-Wallis <\/strong><\/p>\n<p>22) Hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;<strong>One-way ANOVA<\/strong><\/p>\n<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; No hi ha homoscedasticitat&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;&#8230;<strong>Kruskal-Wallis<\/strong><\/p>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>El primer que s&#8217;ha de fer \u00e9s estandaritzar les dades recollides al camp o al laboratori. Per exemple: posar sempre el mateix n\u00famero de decimals, definir si els decimals s\u00f3n comes o punts, en les variables categ\u00f2riques, que les abreviacions de cada categoria sempre siguin iguals (exemple, no posar &#8220;Barcelona&#8221; i &#8220;BCN&#8221;, per categoritzar el [&hellip;]<\/p>\n","protected":false},"author":20,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-127","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/webs.uab.cat\/biologiamarina\/wp-json\/wp\/v2\/pages\/127","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/webs.uab.cat\/biologiamarina\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/webs.uab.cat\/biologiamarina\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/webs.uab.cat\/biologiamarina\/wp-json\/wp\/v2\/users\/20"}],"replies":[{"embeddable":true,"href":"https:\/\/webs.uab.cat\/biologiamarina\/wp-json\/wp\/v2\/comments?post=127"}],"version-history":[{"count":0,"href":"https:\/\/webs.uab.cat\/biologiamarina\/wp-json\/wp\/v2\/pages\/127\/revisions"}],"wp:attachment":[{"href":"https:\/\/webs.uab.cat\/biologiamarina\/wp-json\/wp\/v2\/media?parent=127"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}