101A_different_designsv3_Fall_08

101A_different_designsv3_Fall_08 - Statistics 101 A Fall...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Fall 2008 Professor Esfandiari Why do I want you to know about different designs? In this lecture my main objective is to convey why it is important for you to know about different designs. Even though we may not have time to discuss the mathematical underpinning and the data analysis related to each of these designs, it is important for you to become introduced to different kinds of designs. This knowledge will not only help you in designing studies, it will also help you develop a better understanding of the research that results from different kind interventions. I will try to achieve this objective through giving examples with fictitious data. Why do we design experiments: When we want to examine the effect of a number of independent or explanatory variables on an outcome variable, we do this through designing of an experiment and development of a statistical model. Example 1: suppose that a physician wants to examine the effect of a new medication (medication vs. placebo) and exercise (aerobic exercise vs. no exercise) on lowering the level of cholesterol. The outcome/dependent variable is lowering the level of cholesterol and the explanatory or independent variables are medication and exercise. Example 2: A researcher wants to examine the effect of a new fertilizer (chemical A and B, vs. chemical A. vs. no chemical) on the growth of tomatoes. He thinks that the type of soil might also impact the growth of tomatoes. In order to control for the effect of soil he divides the land into three categories (very fertile, fertile, and relatively fertile) and he makes it a factor in the study. So, he now has two independent and one dependent variable. Dividing the land into three categories in terms of fertility and making it a variable in the study is called “blocking” and it is used as a mean of dealing with a potential source of error. Example 3: A language specialist wants to examine the effect of teaching two methods of French (novels vs. grammar) on learning this language. However she thinks that the prior knowledge of French vocabulary and having traveled to France might influence the outcome variable. In order to control for these two sources of possible error, she first gives her students a test on French vocabulary. Based on the data obtained, she ranks them in pairs and randomly assigns the two students with the highest scores to novel and grammar, then, she randomly assigns the next pair with the highest grades to novel and grammar, and continues to do so until all the subject are assigned to the two groups. Thus, the independent variables are method of teaching (intervention), and having or not having traveled to France. The dependent variable is learning French which is measured through a test. Prior knowledge of French was controlled by how the subjects were assigned to the control and experimental group. 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/22/2011 for the course STAT 101 taught by Professor Esfandi during the Spring '11 term at UCLA.

Page1 / 9

101A_different_designsv3_Fall_08 - Statistics 101 A Fall...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online