5
Apa itu perancangan percobaan? Merupakan suatu uji atau sederetan uji dimana suatu proses atau sistem mengakibatkan terjadinya perubahan yang cukup berarti dari variabel input, yang dapat diamati melalui respon yang muncul. Perencanaan (planning) suatu percobaan untuk memperoleh informasi yang relevan dengan tujuan dari penelitian Langkah-langkah dalam menyusun rancangan percobaan : Rumuskan masalah penelitian Pilih faktor-faktor dan taraf-taraf Tentukan variabel respon Pilih rancangan percobaan Laksanakan percobaan Analisis data Kesimpulan dan rekomendasi Ada tiga prinsip dasar yang perlu diperhatikan dalam merancang suatu percobaan, yaitu: 1. Pengacakan (Randomization) 2. Ulangan (Replication) 3. Pengendalian Lingkungan (Local control) Pengacakan: setiap unit percobaan memiliki peluang yang sama untuk diberikan suatu perlakuan. Menghindari galat sistematik Meningkatkan validitas kesimpulan (pemenuhan asumsi kebebasan) Caranya: lotere, tabel bilangan acak, komputer Ulangan: Penerapan perlakuan terhadap beberapa unit percobaan. Untuk menduga galat percobaan Untuk menduga standard error rataan perlakuan Untuk meningkatkan presisi kesimpulan Berapa jumlah ulangan ? Minimal 3 Minimal db-galat 15

Dasar Dasar Rancob

Embed Size (px)

DESCRIPTION

jjjjj

Citation preview

Page 1: Dasar Dasar Rancob

Apa itu perancangan percobaan?– Merupakan suatu uji atau sederetan uji dimana suatu proses atau sistem

mengakibatkan terjadinya perubahan yang cukup berarti dari variabel input, yang dapat diamati melalui respon yang muncul.

– Perencanaan (planning) suatu percobaan untuk memperoleh informasi yang relevan dengan tujuan dari penelitian

Langkah-langkah dalam menyusun rancangan percobaan :

Rumuskan masalah penelitian Pilih faktor-faktor dan taraf-taraf Tentukan variabel respon Pilih rancangan percobaan Laksanakan percobaan Analisis data Kesimpulan dan rekomendasi

Ada tiga prinsip dasar yang perlu diperhatikan dalam merancang suatu percobaan, yaitu:1. Pengacakan (Randomization)2. Ulangan (Replication)3. Pengendalian Lingkungan (Local control)

Pengacakan: setiap unit percobaan memiliki peluang yang sama untuk diberikan suatu perlakuan.

– Menghindari galat sistematik – Meningkatkan validitas kesimpulan (pemenuhan asumsi kebebasan)– Caranya: lotere, tabel bilangan acak, komputer

Ulangan: Penerapan perlakuan terhadap beberapa unit percobaan.– Untuk menduga galat percobaan – Untuk menduga standard error rataan perlakuan – Untuk meningkatkan presisi kesimpulan

Berapa jumlah ulangan ?– Minimal 3– Minimal db-galat 15– Gunakan formula yang ada

Beberapa istilah: Perlakuan:

– Suatu metode/prosedur yang diterapkan terhadap unit percobaan – Merupakan taraf-taraf dari suatu faktor atau kombinasi taraf dari beberapa faktor

Page 2: Dasar Dasar Rancob

Unit percobaan – Unit terkecil dalam percobaan yang diberikan perlakuan

Unit Pengamatan – Unit terkecil tempat dilakukan pengamatan respon percobaan

Faktor (kualitatif & kuantitatif)– Peubah bebas penyusun perlakuan, dimana nilai-nilainya dapat bersifat kualitatif

maupun kuantitatif Taraf

– Nilai-nilai dari faktor-faktor yang dilibatkan dalam percobaan Interaksi

– Perubahan pengaruh dari suatu faktor pada berbagai taraf faktor yang lain Model acak

– Model yang dibangun oleh peubah bebas-peubah bebas yang bersifat acak Model tetap

– Model yang dibangun oleh peubah bebas-peubah bebas yang bersifat tetap Model Campuran

– Model yang dibangun oleh peubah bebas-peubah bebas yang bersifat acak dan tetap

Klasifikasi Rancangan Percobaan Rancangan Perlakuan

Berkaitan dengan kondisi-kondisi apa yang akan diberikan terhadap unit-unit percobaan Contoh: Faktor tunggal, faktorial, split-plot, dll

Rancangan Lingkungan Berkaitan dengan bagaimana perlakuan-perlakuan itu diterapkan pada unit-unit percobaan Contoh: RAL, RAKL, RBSL

Rancangan Pengukuran Berkaitan dengan bagaimana respon unit percobaan diukurTreatments are the different procedures we want to compare. These couldbe different kinds or amounts of fertilizer in agronomy, different longdistancerate structures in marketing, or different temperatures in a reactorvessel in chemical engineering.Experimental units are the things to which we apply the treatments. Thesecould be plots of land receiving fertilizer, groups of customers receivingdifferent rate structures, or batches of feedstock processing at differenttemperatures.Responses are outcomes that we observe after applying a treatment to anexperimental unit. That is, the response is what we measure to judgewhat happened in the experiment; we often have more than one response.Responses for the above examples might be nitrogen contentor biomass of corn plants, profit by customer group, or yield and qualityof the product per ton of raw material.Randomization is the use of a known, understood probabilistic mechanismfor the assignment of treatments to units. Other aspects of an experimentcan also be randomized: for example, the order in which unitsare evaluated for their responses.

Page 3: Dasar Dasar Rancob

Experimental Error is the random variation present in all experimental results.Different experimental units will give different responses to thesame treatment, and it is often true that applying the same treatmentover and over again to the same unit will result in different responsesin different trials. Experimental error does not refer to conducting thewrong experiment or dropping test tubes.Measurement units (or response units) are the actual objects on which theresponse is measured. These may differ from the experimental units.For example, consider the effect of different fertilizers on the nitrogencontent of corn plants. Different field plots are the experimental units,but the measurement units might be a subset of the corn plants on thefield plot, or a sample of leaves, stalks, and roots from the field plot.Blinding occurs when the evaluators of a response do not know which treatmentwas given to which unit. Blinding helps prevent bias in the evaluation,even unconscious bias from well-intentioned evaluators. Doubleblinding occurs when both the evaluators of the response and the (humansubject) experimental units do not know the assignment of treatmentsto units. Blinding the subjects can also prevent bias, becausesubject responses can change when subjects have expectations for certaintreatments.Control has several different uses in design. First, an experiment is controlledbecause we as experimenters assign treatments to experimentalunits. Otherwise, we would have an observational study.Second, a control treatment is a “standard” treatment that is used as abaseline or basis of comparison for the other treatments. This controltreatment might be the treatment in common use, or it might be a nulltreatment (no treatment at all). For example, a study of new pain killingdrugs could use a standard pain killer as a control treatment, or a studyon the efficacy of fertilizer could give some fields no fertilizer at all.This would control for average soil fertility or weather conditions.Placebo is a null treatment that is usedwhen the act of applying a treatment—any treatment—has an effect. Placebos are often used with humansubjects, because people often respond to any treatment: for example,reduction in headache pain when given a sugar pill. Blinding is importantwhen placebos are used with human subjects. Placebos are alsouseful for nonhuman subjects. The apparatus for spraying a field witha pesticide may compact the soil. Thus we drive the apparatus over thefield, without actually spraying, as a placebo treatment.Factors combine to form treatments. For example, the baking treatment fora cake involves a given time at a given temperature. The treatment isthe combination of time and temperature, but we can vary the time andtemperature separately. Thus we speak of a time factor and a temperaturefactor. Individual settings for each factor are called levels of thefactor.Confounding occurs when the effect of one factor or treatment cannot bedistinguished from that of another factor or treatment. The two factorsor treatments are said to be confounded. Except in very special circumstances,confounding should be avoided. Consider planting cornvariety A in Minnesota and corn variety B in Iowa. In this experiment,we cannot distinguish location effects from variety effects—the varietyfactor and the location factor are confounded.