2 Accuracy and Precision Accuracy How close a measurement is to the actual or “true value” high accuracy true value low accuracy true value 3.

Slides:



Advertisements
Similar presentations
Confidence Intervals Objectives: Students should know how to calculate a standard error, given a sample mean, standard deviation, and sample size Students.
Advertisements

Parameter Estimation Chapter 8 Homework: 1-7, 9, 10 Focus: when  is known (use z table)
Sampling: Final and Initial Sample Size Determination
Estimation of Sample Size
Objectives Look at Central Limit Theorem Sampling distribution of the mean.
Confidence Intervals for Proportions
EPIDEMIOLOGY AND BIOSTATISTICS DEPT Esimating Population Value with Hypothesis Testing.
Parameter Estimation Chapter 8 Homework: 1-7, 9, 10.
Intro to Statistics for the Behavioral Sciences PSYC 1900 Lecture 9: Hypothesis Tests for Means: One Sample.
Fall 2006 – Fundamentals of Business Statistics 1 Business Statistics: A Decision-Making Approach 6 th Edition Chapter 7 Estimating Population Values.
Chapter 7 Estimating Population Values
Chapter 7 Probability and Samples: The Distribution of Sample Means
DATA QUALITY and ANALYSIS Strategy for Monitoring Post-fire Rehabilitation Treatments Troy Wirth and David Pyke USGS – Biological Resources Division Forest.
QM-1/2011/Estimation Page 1 Quantitative Methods Estimation.
Standard error of estimate & Confidence interval.
Leon-Guerrero and Frankfort-Nachmias,
2 Accuracy and Precision Accuracy How close a measurement is to the actual or “true value” high accuracy true value low accuracy true value 3.
1 Psych 5500/6500 Statistics and Parameters Fall, 2008.
Estimation Statistics with Confidence. Estimation Before we collect our sample, we know:  -3z -2z -1z 0z 1z 2z 3z Repeated sampling sample means would.
Chapter 11: Estimation Estimation Defined Confidence Levels
Jan 17,  Hypothesis, Null hypothesis Research question Null is the hypothesis of “no relationship”  Normal Distribution Bell curve Standard normal.
A Sampling Distribution
Populations, Samples, Standard errors, confidence intervals Dr. Omar Al Jadaan.
Sampling and Confidence Interval
Estimation of Statistical Parameters
ESTIMATION. STATISTICAL INFERENCE It is the procedure where inference about a population is made on the basis of the results obtained from a sample drawn.
Topics: Statistics & Experimental Design The Human Visual System Color Science Light Sources: Radiometry/Photometry Geometric Optics Tone-transfer Function.
Timberlake LecturePLUS1 Chapter 1 Measurements Accuracy and Precision.
1 Measurement in Chemistry In chemistry we  do experiments  measure quantities  use numbers to report measurements.
Accuracy and Precision
Understanding and Presenting Your Data OR What to Do with All Those Numbers You’re Recording.
Physics 270 – Experimental Physics. Standard Deviation of the Mean (Standard Error) When we report the average value of n measurements, the uncertainty.
1 Things That May Affect Estimates from the American Community Survey.
Standard Error and Confidence Intervals Martin Bland Professor of Health Statistics University of York
1 Estimation From Sample Data Chapter 08. Chapter 8 - Learning Objectives Explain the difference between a point and an interval estimate. Construct and.
Statistical estimation, confidence intervals
CONFIDENCE INTERVAL It is the interval or range of values which most likely encompasses the true population value. It is the extent that a particular.
6.1 Inference for a Single Proportion  Statistical confidence  Confidence intervals  How confidence intervals behave.
Medical Statistics as a science
How confident are we in the estimation of mean/proportion we have calculated?
CHEMISTRY ANALYTICAL CHEMISTRY Fall Lecture 6.
CHAPTER-6 Sampling error and confidence intervals.
Introduction to Inference: Confidence Intervals and Hypothesis Testing Presentation 4 First Part.
Chapter 8 Parameter Estimates and Hypothesis Testing.
Chapter 10: Confidence Intervals
1 UNIT ONE Topic: Accuracy and Precision. 2 Accuracy How close a measurement is to the actual or true value good accuracy true value poor accuracy true.
Stats Lunch: Day 3 The Basis of Hypothesis Testing w/ Parametric Statistics.
Statistical Analysis II Lan Kong Associate Professor Division of Biostatistics and Bioinformatics Department of Public Health Sciences December 15, 2015.
1 Sampling Distribution of Arithmetic Mean Dr. T. T. Kachwala.
Statistical significance using Confidence Intervals
Statistics Nik Bobrovitz BHSc, MSc PhD Student University of Oxford December 2015
1 Probability and Statistics Confidence Intervals.
Course: Research in Biomedicine and Health III Seminar 5: Critical assessment of evidence.
SAMPLING DISTRIBUTION OF MEANS & PROPORTIONS. SAMPLING AND SAMPLING VARIATION Sample Knowledge of students No. of red blood cells in a person Length of.
SAMPLING DISTRIBUTION OF MEANS & PROPORTIONS. SAMPLING AND SAMPLING VARIATION Sample Knowledge of students No. of red blood cells in a person Length of.
Confidence Intervals Dr. Amjad El-Shanti MD, PMH,Dr PH University of Palestine 2016.
April Center for Open Fostering openness, integrity, and reproducibility of scientific research.
Dr.Theingi Community Medicine
CHAPTER 6: SAMPLING, SAMPLING DISTRIBUTIONS, AND ESTIMATION Leon-Guerrero and Frankfort-Nachmias, Essentials of Statistics for a Diverse Society.
Variability. The differences between individuals in a population Measured by calculations such as Standard Error, Confidence Interval and Sampling Error.
Variability.
And distribution of sample means
STANDARD ERROR OF SAMPLE
ESTIMATION.
Meta-analysis statistical models: Fixed-effect vs. random-effects
Using Statistics in Biology
Using Statistics in Biology
Statistics in Biology.
GENERALIZATION OF RESULTS OF A SAMPLE OVER POPULATION
Statistics in Biology: Standard Error of the Mean & Error Bars
Presentation transcript:

2 Accuracy and Precision

Accuracy How close a measurement is to the actual or “true value” high accuracy true value low accuracy true value 3

Precision How well several measurements agree with each other high precision low precision 4

Precision or the reproducibility of a set of measurements. A precise sample estimate will have a very small random error of estimation. Precision

Accuracy

Precision

Accuracy and Precision What can you say about the accuracy and precision in each of the following: High precision, low accuracy High precision, high accuracy 8

Question: How will each of the following affect accuracy and precision? 1. A meter stick that is missing the first centimeter. 2. A scale that has a zero point that is really five pounds above zero. 9

Solution How will each of the following affect accuracy and precision? 1. The shortened meter stick will produce measurements that have poor accuracy, but good precision is possible. 2. The poorly calibrated scale will give a weight that is not accurate, but good precision is possible. 10

A blood sample was taken from a patient and four different assays were used to measure blood glucose. The true value of the blood glucose was known to be 4.5 mmol/L. State whether the accuracy and the precision of each assay is high or low: Assay A: 4,6; 4,6; 4,8; 4,5; 4,5; 4,4; Assay B: 4,3; 3,5; 5,3; 4,6; 5,5; 3,7 Assay C: 3,5; 3,6; 3,3; 3,5; 3,4; 3,5 Assay D: 8,5; 6,4; 5,3; 7,6; 4,8; 9,3 11

Literature incosistency? High accuracy and low precision or both – low accuracy and low precision?

How confident are we in the estimation of mean/proportion we have calculated?

14

15

Measures of precision: 1. Standard error of mean, SEM Standard error of proportion, SE(p) 2. Confidence interval for mean Confidence interval for proportion

Standard error of mean, SEM Number of patients Standard deviation, SD SEM is smaller (estimate is more precise): the larger is N (number of patients) the smaller is SD (dispersion of data)

19

95% confidence interval for mean, 95% CI  Together with SEM, 95% CI is also the measure of precision  Unlike SEM, 95% CI also estimates accuracy of the result ie. 95% is accurate that interval includes true (population) mean)

95% confidence interval for mean If we draw a 100 samples from our population we would find the true population value within 95% confidence interval in 95 samples samples

Critical values for 90%, 95% and 99% level of confidence 90% CI => mean ± 1.65 SEM 95% CI => mean ± 1.96 SEM 99% CI => mean ± 2.58 SEM Level of Confidence - Critical Value 0.75, or 75% , or 80% , or 85% , or 90% , or 95% , or 98% , or 99% 2.58

Example 1 The average systolic BP before treatment in study A, of a group of 100 hypertensive patients, was 170 mmHg. After treatment with the new drug the mean BP dropped by 20 mmHg. If the 95% CI is 15–25, this means: 23 we can be 95% confident that the true effect of treatment is to lower the BP by 15–25 mmHg.

Example 2 In study B 50 patients were treated with the same drug, also reducing their mean BP by 20 mmHg, but with a wider 95% CI of -5 to +45. This CI includes zero (no change). This means: 24 there is more than a 5% chance that there was no true change in BP, and that the drug was actually ineffective..

Watch out for... The size of a CI is related to the sample size of the study. Larger studies usually have a narrower CI. 25

Example 3 – Meta analysis Fig. Plot of 5 studies of a new antihypertensive drug. 1.Which study showed the greatest change? 2.Did all the studies show change in favour of the intervention? 3.Were the changes statistically significant?

Proportion 1. Standard error of proportion, SE(p) SE( p) = √(p(1 – p)/n) 2. Confidence interval for proportion

The standard deviation describes the variability of a sample; does not describe the sample The standard error of the mean (SEM) does not describe the sample but uncertainty describes the uncertainty of how the sample mean represents the population mean.

Krebs NF, Westcott JE, Culbertson DL et. al. Comparison of complementary feeding strategies to meet zinc requirements of older breastfed infants. Am J Clin Nutr. 2012; 96:30-35 “Mean (±SEM) total absorbed zinc amounts were 0.80 ± 0.08, 0.71 ± 0.09, and 0.52 ± 0.05 mg/d for the: meat, iron-and-zinc-fortified infant cereal, and whole-grain, iron-only-fortified infant cereal groups of infants.” SEMCI Meat Fe&Zn Fe

Common mistake in the literature Misuse of standard error of the mean (SEM) when reporting variability of a sample. A critical evaluation of four anaesthesia journals. P. Nagele* British Journal of Anaesthesia 90 (4): (2001)

SD CI Standard deviation tells us about the variability (spread) in a sample. The CI tells us the range in which the true value (the mean if the sample were infinitely large) is likely to be.

What does a small standard error tell us about the sample estimate of the mean? That it is highly variable That the population standard deviation may be small That the sample size is probably small That it is imprecise

What will tend to make the standard error larger? A small variance A large standard deviation Imprecise data Inaccurate data 33