This preview shows pages 1–4. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Statistics Laboratory Grace Dion September 15, 2006 Biology 201 10:00 Introduction: While the collection and analysis of data is the basis of the scientific method, the statistical studies of the test results more often provide the information sought by researchers to prove or disprove hypotheses. Even four centuries ago, prominent scientists such as Bernoulli, Gauss, and Fisher recognized the need to filter experimental data in order to draw inferences. When applying statistics to data, one assesses the kind of data, quantitative or qualitative, present to determine the appropriate statistical tests to run. Once determined, descriptive statistical studies are done which usually relate to a sample or representative section of a larger set and the measurements that will be analyzed are either continuous which includes data that can be fractions or discontinuous where no fraction is possible, such as the number of females present in a study. Once these facts are determined, a researcher must identify whether his data is nominal or able to be categorized, like eye color or ordinal where there are different degrees of the variable such as blonde to dark brown hair color. Numerical measures are often used to summarize the collected data with the most common being the mean, the median, the mode, percentile, range, variance, and standard deviation related to each variable studied. In addition to descriptive statistics, a researcher might use derived variables, data that hasn’t been measured directly but that is calculated from the measurements taken, such as percentages or ratios. In this case, a scientist might employ linear regression to look at the relationship between dependent and independent variables in a test situation. A model is created and a probability of error calculated for the model. This model can then be used to predict what the value of the dependent variable will be when the independent variable changes. Finally, a researcher looks at the probabilities derived from the data he has obtained and what they might predict about other outcomes. (p219). 2 Statistical analysis of data is critical to scientific researchers in testing hypotheses, especially those based on a sampling of a larger population. In our experiments, we employed common statistical methods to analyze collected data in order to become more familiar with the math behind the methods and the value of statistics in experimental research. Methods and Materials: I. Coin Toss For this experiment, we designated Allison Galassie as the coin flipper, while Kelly Martin, Cedric Harville, and I recorded the results of each flip. To perform this, Allison flipped the same standard U.S. quarter 10 times and 100 times, respectively. She performed each of these flips by placing the quarter on the middle finger of her right hand and then thumping the quarter upward with her thumb. Next, she caught the quarter and turned it over onto the top of her left hand. She flipped the coin about two feet in the air. While she was doing this, the rest of her left hand....
View
Full
Document
This note was uploaded on 10/12/2011 for the course BIOLOGY 100 taught by Professor Smith during the Spring '11 term at South Carolina Upstate.
 Spring '11
 Smith
 Biology

Click to edit the document details