Measurement, Quantification and Analysis

Slides:



Advertisements
Similar presentations
Tests of Hypotheses Based on a Single Sample
Advertisements

“Students” t-test.
Chi-square and F Distributions
Chapter 6 Sampling and Sampling Distributions
PTP 560 Research Methods Week 9 Thomas Ruediger, PT.
CmpE 104 SOFTWARE STATISTICAL TOOLS & METHODS MEASURING & ESTIMATING SOFTWARE SIZE AND RESOURCE & SCHEDULE ESTIMATING.
POINT ESTIMATION AND INTERVAL ESTIMATION
Chap 8-1 Statistics for Business and Economics, 6e © 2007 Pearson Education, Inc. Chapter 8 Estimation: Single Population Statistics for Business and Economics.
Evaluating Hypotheses Chapter 9. Descriptive vs. Inferential Statistics n Descriptive l quantitative descriptions of characteristics.
DATA ANALYSIS I MKT525. Plan of analysis What decision must be made? What are research objectives? What do you have to know to reach those objectives?
Chapter 7 Sampling and Sampling Distributions
Chapter 8 Estimation: Single Population
Inference about a Mean Part II
Measurement, Quantification and Analysis Some Basic Principles.
Standard error of estimate & Confidence interval.
AM Recitation 2/10/11.
1/2555 สมศักดิ์ ศิวดำรงพงศ์
Chapter 11: Estimation Estimation Defined Confidence Levels
Estimation of Statistical Parameters
Lecture 12 Statistical Inference (Estimation) Point and Interval estimation By Aziza Munir.
The Scientific Method Formulation of an H ypothesis P lanning an experiment to objectively test the hypothesis Careful observation and collection of D.
Educational Research: Competencies for Analysis and Application, 9 th edition. Gay, Mills, & Airasian © 2009 Pearson Education, Inc. All rights reserved.
Copyright © 2013, 2009, and 2007, Pearson Education, Inc. Chapter 13 Multiple Regression Section 13.3 Using Multiple Regression to Make Inferences.
Physics 270 – Experimental Physics. Let say we are given a functional relationship between several measured variables Q(x, y, …) x ±  x and x ±  y What.
CHEMISTRY ANALYTICAL CHEMISTRY Fall Lecture 6.
Sampling distributions rule of thumb…. Some important points about sample distributions… If we obtain a sample that meets the rules of thumb, then…
Review - Confidence Interval Most variables used in social science research (e.g., age, officer cynicism) are normally distributed, meaning that their.
Statistical Inference Statistical inference is concerned with the use of sample data to make inferences about unknown population parameters. For example,
Statistics for Political Science Levin and Fox Chapter Seven
NURS 306, Nursing Research Lisa Broughton, MSN, RN, CCRN RESEARCH STATISTICS.
Tom.h.wilson Department of Geology and Geography West Virginia University Morgantown, WV.
Tom.h.wilson Department of Geology and Geography West Virginia University Morgantown, WV.
Variability. The differences between individuals in a population Measured by calculations such as Standard Error, Confidence Interval and Sampling Error.
Variability.
GOVT 201: Statistics for Political Science
Advanced Quantitative Techniques
Statistical analysis.
More on Inference.
Psych 231: Research Methods in Psychology
6. Simple Regression and OLS Estimation
Statistics made simple Dr. Jennifer Capers
INF397C Introduction to Research in Information Studies Spring, Day 12
Chapter 4. Inference about Process Quality
Inference and Tests of Hypotheses
Statistical analysis.
Understanding Results
Sampling Distributions
Week 10 Chapter 16. Confidence Intervals for Proportions
Statistics in Applied Science and Technology
INTRODUCTORY STATISTICS FOR CRIMINAL JUSTICE Test Review: Ch. 7-9
More on Inference.
Chi-square and F Distributions
Sampling Distribution
Sampling Distribution
Geology Geomath Chapter 7 - Statistics tom.h.wilson
Samples and Populations
Statistical Process Control
Data analysis and basic statistics
Interval Estimation and Hypothesis Testing
Psych 231: Research Methods in Psychology
Psych 231: Research Methods in Psychology
Psych 231: Research Methods in Psychology
Intro to Confidence Intervals Introduction to Inference
Psych 231: Research Methods in Psychology
Statistical Inference for the Mean: t-test
MGS 3100 Business Analysis Regression Feb 18, 2016
Chapter 4 (cont.) The Sampling Distribution
Correlation and Simple Linear Regression
Correlation and Simple Linear Regression
Presentation transcript:

Measurement, Quantification and Analysis Some Basic Principles And Some Basic Statistics

Three Major Issues Biological and especially ecological data show high variability in quantitative traits We almost never measure everything in field research; rather we sample from larger populations or data sets Sampling leads to uncertainty about conclusions, so we always must estimate our uncertainty Items 1) – 3) above are why we use Statistics!

Variability

All natural processes are variable, Whether continuous or discreet Continuous data All natural processes are variable, Whether continuous or discreet Discreet data Plus, better sampling effort better describes distributions

In many processes, we observe characteristic distributions 37.5% A a AA Aa aa (p + q)4 where p = q = 0.5 When (p + q) is raised to the 20th power, binomial is indistinguishable from the normal. Binomial – Few interacting factors Normal – Many interacting factors 2 factors: One way to get AA or aa, 2 ways to get Aa (p + q)2 4 factors: One way to have AAAA or aaaa, 4 ways to get AAAa or aaaA, and 6 ways to get AAaa

Major point: Biological processes are variable, but they often tend to vary in predictable of consistent patters.

Sampling and Estimation

To calculate the average in a sample: A characteristic of field biology is the attempt to estimate parameters from highly variable populations of uncertain “true” value. To calculate the average in a sample: Mean = Sum of all observations/number of observation To estimate the variability of the observations: Variance = Sum of (individual observation – Mean of observations)2 _____________________________________________ Number of Individual Observations - 1 -1 Or to express this in the same units as the Mean: Standard deviation = Square Root of the Variance

All natural processes are variable, Whether continuous or discreet What happens when we estimate means? Select 5 observations at random. Then 10. Then 25. Probability Better sampled populations yield better distributions Larger sample sizes yield better estimates Means will also be variable, and will have a characteristic distribution If you sample N means, they will have a variance as well

To estimate the variability of the means: Divide the standard deviation (the square root of the variance) by the square root of the sample size (why? Variability of the means is dependent upon sample size – decreases with large samples.) Recall, To estimate the variability of the observations: Variance = Sum of (individual observation – Mean of observations)2 _____________________________________________ Number of Individual Observations – 1 Divide the square root of the variance, the standard deviation, by the square root of the sample size. The bigger the sample size, the less variable the means This is the Standard Error, which is used to calculate a Confidence Interval

If you want, for example, a 95% confidence that a mean measured was from this theoretical distribution, you want to multiply the standard error by a constant the grabs 95% of the distribution. Note that plus or minus 2 SE gives you 4.26% (2.13 x 2), so it needs to be a little smaller, 1.96 to be precise.

Uncertainty

Confidence intervals represent a level of confidence about the true value of the mean. It is the mean, with the expected variability (standard error) combined with a constant derived from a theoretical distribution that defines how much uncertainty is acceptable. Lets pick a constant that circumscribes 95% of the variability. In other words, if you sample repeated with a given sample size, a 95 % CI means that in 95% of the samples you collect, you will have the value of the true mean. The mean is either in or not! No matter how well we sample, we will “miss-estimate” the population parameter a certain percentage. What level of error are we willing to accept? With a 95 % limit, 5 % of the time. In theory, the tails are limitless, so we must set a criterion. Decision rule – 5 % error. Minimize this with replication

Importance of Replication? One sample: Wrong 5% or 1/20 of the times you sample Two replicated samples: Wrong 1/20 x 1/20 or 1/400 Three replicated samples: Wrong 1/20 x 1/20 x 1/20 or 1/8,000

What confidence do we want? What error will we accept? One things we do frequently in science is compare things. For example, if one population bigger than another, which population are we sampling from? A B What kinds of errors can we make?

Fundamental Principles Have clearly defined hypotheses Measure carefully Sample intensively – large sample sizes reduce Beta-Error Replicate – Replication reduces Alpha-Error

Samples of Data Sets from Previous Projects that required Quantification and Statistical Analysis

Sum of Squares df Mean Square F Sig. Forearm (mm) Between Groups 4053.985 2 2026.993 569.784 .000 Within Groups 152.971 43 3.557 Total 4206.957 45 Foot (mm) 254.274 127.137 55.693 98.161 2.283 352.435

Table 2: Chi-Square Tests   Table 2: Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 15.181a 6 .019 Likelihood Ratio 16.437 .012 N of Valid Cases 150 2 cells (16.7%) have expected count less than 5. The minimum expected count is 3.87. Table 2: Chi-Square Tests: The table above shows the Chi-Square value and level of significance.