Réliabilitas (statistika): Béda antarrépisi

Konten dihapus Konten ditambahkan
Uchup19 (obrolan | kontribusi)
mTidak ada ringkasan suntingan
Ilhambot (obrolan | kontribusi)
m Ngarapihkeun éjahan, replaced: nyaeta → nyaéta (3), make → maké (2), dipake → dipaké , ea → éa, kabeh → kabéh (2)
Baris ka-1:
Dina [[statistika]], '''réliabilitas''' hartina runtuyan ukuran nu angger atawa alat keur ngukur. Réliabilitas henteu pakait langsung jeung [[validitas (psikométri)|validitas]]. Réliabilitas nyaéta ukuran nu bisa dipercaya keur ngukur sacara angger, tapi teu perlu kana naon anu diukur. Conto, sanajan loba tes anu bisa diandelkeun, tapi teu sakabehsakabéh tes nembongkeun hasil nu hade keur ngagambarkeun hij pagawaeanpagawaéan.
 
Dina elmu [[experiment|percobaan]], '''reliabilitas''' nyaetanyaéta ukuran tes nu masih keneh angger sanggeus tes dipigawe sababaraha kali kana subyek nu sarua dina kaayan nu ampir sarua oge. Hij percobaan bisa diandelkeun lamun hasilna angger dina unggal ukuran, sarta teu bisa diandelkeun lamun hasilna beda.
 
== Estimasi ==
 
Reliabilitas bisa diestimasi ku sababaraha cara nu bisa dikelompokkeun kana dua tipe nyaetanyaéta: administrasi-tunggal jeung administrasi-multiple. Metoda administrasi-multiple merlukeun dua peniley dina administrasina. Dina metoda ''test-retest'', reliabiliti dianggap minangka ''[[Pearson product-moment correlation coefficient]]'' antara dua ukuran administrasi nu sarua. Dina metoda ''alternate forms'', reliabiliti diitung makemaké ''Pearson product-moment correlation coefficient'' tina dua bentuk nu beda, ilaharna di-administrasi-keun babarengan. Metoda administrasi-tunggal kaasup ''split-half'' sarta ''internal consistency''. Metoda ''split-half'' ngawengku ukuran dua ''halves'' minangka bentuk alternatipna. Ieu estimasi "halves reliability" saterusna dilajuning lakukeun ku cara makemaké ''[[Spearman-Brown prediction formula]]''. Ukuran internal nu ilahar dipakedipaké nyaetanyaéta [[Cronbach's alpha]], nu ilaharna dianggap minangka [[mean]] keur sakabehsakabéh koefisien ''split-half''.
<!--
Each of these estimation methods is sensitive to different sources of error and so might not be expected to be equal. Also, reliability is a property of the ''scores of a measure'' rather than the measure itself and are thus said to be ''sample dependent''. Reliability estimates from one sample might differ from those of a second sample (beyond what might be expected due to sampling variations) if the second sample is drawn from a different population because the true reliability is different in this second population. (This is true of measures of all types--yardsticks might measure houses well yet have poor reliability when used to measure the lengths of insects.)
Baris ka-12:
 
* R(t) = 1 - F(t).
 
* R(t) = e^-ʎ.(t). (where ʎ is the failure rate)
 
Baris 21 ⟶ 20:
: <math>{\rho}_{xx'}=\frac{{\sigma}^2_T}{{\sigma}^2_X}=1-\frac{{{\sigma}^2_E}}{{{\sigma}^2_X}}</math>
 
where <math>{\rho}_{xx'}</math> is the symbol for the reliability of the observed score, ''X''; <math>{\sigma}^2_X</math>, <math>{\sigma}^2_T</math>, and <math>{\sigma}^2_E</math> are the variances on the measured, true and error scores respectively. Unfortunately, there is no way to directly observe or calculate the true score, so a variety of methods are used to estimate the reliability of a test.
 
Some examples of the methods to estimate reliability include test-retest reliability, internal consistency reliability, and parallel-test reliability. Each method comes at the problem of figuring out the source of error in the test somewhat differently.