Professional Documents
Culture Documents
CH 5 Student Notes
CH 5 Student Notes
Objectives:
Distinguish between, and discuss the advantages of, observational studies and experiments. Identify and give examples of different types of sampling methods, including a clear definition of a simple random sample. Identify and give examples of sources of bias in sample surveys. Identify and explain the three basic principles of experimental design. Explain what is meant by a completely randomized design. Distinguish between the purposes of randomization and blocking in an experimental design. Use random numbers from a table or technology to select a random sample.
SECTION 5.1: Designing Samples What if, we wanted to find out, on average, how many evenings per week all Americans dine out? Since we cannot put a question to the entire population of the U.S., we can put the question to
Caution:
In order to see the response change, we must actually impose the change.
Observational studies of the effect of one variable on another often fail because the explanatory variable is confounded with lurking variables. Review: Confounding: Simulation provides an alternative method in these circumstances. After producing data, the next logical step is to use formal statistical inference, which answers specific questions with a known degree of confidence. DEFINITIONS: 1. 2. 3. 4. If conclusions based on a sample are to be valid for the entire population, a sound design for selecting the sample is required. The design of a sample refers to Poor sample designs can produce misleading conclusions. 5.
6.
Simple Random Samples The statistical remedy for personal choice bias samples is to allow impersonal chance to choose the sample. A sample chosen by chance allows neither favoritism by the sampler nor selfselection by respondents. Choosing a sample by chance attacks bias by The simplest way to use chance to select a sample is to place names in a hat ( ) and draw out a handful ( ). This is the idea of DEFINITION:
An SRS not only gives each individual an equal chance to be chosen, but it also The idea of an SRS is to choose our sample by drawing names from a hat. In practice, computer software can shoose an SRS from a list of individuals in the population by using a random number generator or by consulting a table of random digits.
A table of random digits is a long string of the digits 0, 1, 2, 3, , 9 with two properties: 1. 2. Table B on the back page of your book is a table of random digits. HOW TO CHOOSE AN SRS: 1. 2.
The use of chance to select the sample is the essential principle of statistical sampling. A probability sample is a sample chosen by chance. We must know what samples are possible and what chance, or probability each possible sample, has. When sampling from large populations, it is common to sample important groups within the population separately, then combine these samples. This is a To select a stratified random sample: 1. 2.
Another common means of restricting random selection is to choose the sample in stages i.e the current population survey uses a multistage sampling design along with opinion polls and other nation samples. There are a few cautions about using sample surveys in particular. When the population consists of human beings accurate information from a sample requires much more than a good sampling design. To begin we need an accurate and complete list of the population. Because such a list is rarely available,
Again,
In addition, the behavior of the respondent of or the interviewer can cause response bias.
Because we deliberately use chance, the results obey the laws of probability that govern chance behavior. (We will study all the laws of probability in Chapter 6.) Results from a survey usually come with a margin of error which we will learn in Chapter 10. Finally,
Treatment: The distinction between explanatory and response variables is important! Factors: Many experiments study the joint effects of several factors. Each treatment is formed by combining a specific value, called a level, of each of the factors. Does regularly taking aspirin help protect people against heart attacks? The Physician's Health Study looked at the effect aspirin. One-half of the 21,966 physicians in the study took aspirin and the other half did not. The subjects: The factors: Treatments: The physicans that did not take the aspirin took a ______________: Why do we use them?
Laboratory experiments in science and engineering often have a simple design with only a single treatment, which is applied to all of the experimental units. The design of such an experiment can be outlined as
With comparative experiments we need to guard against the placebo effect: Especially in medical experiments where there are two groups, one treated with medication and one treated with a placebo,
To limit the effect of the placebo response and also the confounding of variables,
The group of patients who receives a sham treatment is called a _______________ because
1st rule:
2nd rule:
How can we assign experimental units to treatments in a way that is fair to all of the treatments? The statistician's remedy is to rely on chance to make an assignment that does not depend on any characteristic of the experimental units and that does not rely on the judgment of the experimenter in any way. The use of chance can be combined with matching, but the simplest design creates groups by chance alone.
To avoid making quick conclusions about cause and effect from a single study, The 3 principles of good experimental design are: 1. 2. 3. You will often see the "statistically significant" in reports of investigations in many fields of study.
Cautions about experimentation: One type of study is called a double blind experiment:
Many behavioral science experiments use subjects that know they are subjects... Sometimes a matched pairs design is used to compare subject preference between two objects. The two objects taken together are called a ____________. Arrangement of the treatments within the block must be _______ and _________. Tossing a coin, which doesn't seem too scientific, is a great way to accomplish this. The order that the treatments are given can also influence the subject response, so ___________________________________________ also, again by a coin toss. The matched pairs design uses
The idea of blocking is an important additional principle of statistical design of experiments. A wise experimenter will form blocks based on Randomization will
EXPERIMENTAL DESIGN ACTIVITY Surfing Clams of North Carolina Snood Power! Personal SpaceThe Final Frontier Return of the Hungry Turkeys
Snood Power!
The wild turkey, Meleagris gallopavo, is an example of a sexually dimorphic species exhibiting an array of extravagant behavioral and morphological [characteristics] that serve no obvious function other than to attract mates. One of these characteristics is a distensible process at the base of the upper mandible known as a snood. From the female perspective, better father turkeys will mean that the next generateion of turkeys will have a greater chance of survival; a big snood, it is suggested, is taken by the female to indicate good genes. This, from a biological standpoint, it makes sense to select mates from males with bigger snood. It is possible that the snood length might also be regarded as a measure of tough turkey maleness by male turkeys. From the male turkey perspective, it might not be a good idea to ruffle the feathers of a big snooded male because he might ruffle your feathers in response. Your task is to design an experiment to address this question: does the male turkey regard other males snood length as a measure of dont-tread-on-me capability? An experiment is envisioned as follows. One male turkey decoy is to be placed in a small arena with a pile of birdseed nearby. (Using a decoy will
give direct control over some possible confounding variables. For example, a live turkey might provide a different aggressive display for different situations, or be at different levels of hunger during the experiment.) A hungry live male turkey would then be placed in the arena. It is reasoned that if male turkeys regard the snood as indicative of a powerful turkey they will be less fearful of competing for food against a short snooded turkey. You have budgeted for 60 male turkey subjects for your experiment and can order the animals to conform to your size specifications. For example, a 5 lb turkey would be small, a 20 lb turkey would be large.
Section 5.3: Simulating Experiments There are three methods for answering questions like "What is the likelihood that...?", "What are the chances that...?", or "What are the chances that this will happen before that?"
We can: 1) Try to estimate the likelihood of a result of interest by actually carrying out the experiment many times and calculating the result's relative frequency. This is slow, possibly costly, and often impractical. 2) Develop a probability model and use it to calculate a theoretical answer. This requires that we know something about the rules of probability and may not be feasible. 3) Start with a model that reflects the truth about the experiment, and develop a procedure for imitating, or simulating, a number of repetitions of the experiment. This is quicker, cheaper, and relatively easy using technology.
The imitation of chance behavior, based on a model that accurately reflects the experiment under consideration, is called a ________________________. Old statistical technology includes equipment like two sided fair coins, and dice of any number of independent, non-repeated numbers, using a random number table.
Flipping and tossing are really mathematical concepts. Coin tosses are independent because the result of one toss has no effect or influence over the next coin toss. Shooting 10 free throws in basketball and observing the sexes of 10 children have similar models and are simulated the same way. A model is based on opinion and past experience.
Assigning digits can be done in different ways but some are more efficient and preferred. New technology includes assigning outcomes to random digits using a calculator or computer
software. These machines will calculate as many repetitions as requested quickly and without complaining.
Using the TI-83 we can use the command randInt (found under MATH/PRB/5) to generate lists of random integers at will. The command randInt (0, 9, 8) will generate lists of 8 random numbers between 0 and 9 with successive enter commands.
The command randInt (0, 99, 12) generates lists of ____ random numbers between ________. How could we simulate the outcomes of rolling a 6 sided die 15 times?
Repeating the steps for carrying out a simulation (most often carried out with random numbers) are:
1. 2. 3. 4. 5. 6.
Experiment Pepsi vs. Coke In this activity, we want to determine if students can tell the difference between two popular colas. The students will be given 3 cups of cola. Two will be of one type and the other will be of the other type. The goal is not to be able to determine which is Coke and which is Pepsi but to be able to tell which one cup contains a cola different from the other two. We are interested in determining the proportion of students that can tell a difference in the colas. What are some things to keep in mind in conducting this study in order to validate the results? How will we account for these? Each student will take the taste test. The student administering the test will record the result (correct or incorrect in identifying the different cola) Results: We are interested in the proportion of correct choices. Record the number of correct and incorrect choices in the table below. Correct Incorrect
If a student were blindly guessing, what proportion would he/she have guessed correctly?
Did our class do better than blindly guessing? Was our proportion significantly higher than blindly guessing to say that some students can distinguish between the flavors of two popular colas?
(over)
Simulate 40 classes blindly guessing using the calculator. Clearly describe the procedure. And carry out the simulation. Proportion Proportion Trial Trial correct correct 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
What proportion of these simulations did as well or better than our class did in choosing the different cola?
Now use this number to describe whether the proportion of correct choices by our class was significantly higher than blindly guessing to say that some students can distinguish between the flavors of two popular colas.
Ch 5 review