Z-TEST / Z-STATISTIC: used to test hypotheses about. µ when the population standard deviation is unknown



Similar documents
One-sample test of proportions

Hypothesis testing. Null and alternative hypotheses

1. C. The formula for the confidence interval for a population mean is: x t, which was

Confidence Intervals for One Mean

Inference on Proportion. Chapter 8 Tests of Statistical Hypotheses. Sampling Distribution of Sample Proportion. Confidence Interval

PSYCHOLOGICAL STATISTICS

Confidence Intervals

Exam 3. Instructor: Cynthia Rudin TA: Dimitrios Bisias. November 22, 2011

Case Study. Normal and t Distributions. Density Plot. Normal Distributions

Overview. Learning Objectives. Point Estimate. Estimation. Estimating the Value of a Parameter Using Confidence Intervals

5: Introduction to Estimation

I. Chi-squared Distributions

Center, Spread, and Shape in Inference: Claims, Caveats, and Insights

Practice Problems for Test 3

Confidence Intervals. CI for a population mean (σ is known and n > 30 or the variable is normally distributed in the.

CHAPTER 7: Central Limit Theorem: CLT for Averages (Means)

Determining the sample size

Chapter 14 Nonparametric Statistics

Lesson 15 ANOVA (analysis of variance)

Mann-Whitney U 2 Sample Test (a.k.a. Wilcoxon Rank Sum Test)

The following example will help us understand The Sampling Distribution of the Mean. C1 C2 C3 C4 C5 50 miles 84 miles 38 miles 120 miles 48 miles

Statistical inference: example 1. Inferential Statistics

Multi-server Optimal Bandwidth Monitoring for QoS based Multimedia Delivery Anup Basu, Irene Cheng and Yinzhe Yu

Unit 8: Inference for Proportions. Chapters 8 & 9 in IPS

Chapter 7 - Sampling Distributions. 1 Introduction. What is statistics? It consist of three major areas:

STA 2023 Practice Questions Exam 2 Chapter 7- sec 9.2. Case parameter estimator standard error Estimate of standard error

University of California, Los Angeles Department of Statistics. Distributions related to the normal distribution

Lesson 17 Pearson s Correlation Coefficient

Chapter 7: Confidence Interval and Sample Size

Normal Distribution.

Incremental calculation of weighted mean and variance

This document contains a collection of formulas and constants useful for SPC chart construction. It assumes you are already familiar with SPC.

Sampling Distribution And Central Limit Theorem

GCSE STATISTICS. 4) How to calculate the range: The difference between the biggest number and the smallest number.


Measures of Spread and Boxplots Discrete Math, Section 9.4

Confidence Intervals for Linear Regression Slope

Math C067 Sampling Distributions

OMG! Excessive Texting Tied to Risky Teen Behaviors

Confidence intervals and hypothesis tests

Output Analysis (2, Chapters 10 &11 Law)

1 Computing the Standard Deviation of Sample Means

Analyzing Longitudinal Data from Complex Surveys Using SUDAAN

MEI Structured Mathematics. Module Summary Sheets. Statistics 2 (Version B: reference to new book)

Repeating Decimals are decimal numbers that have number(s) after the decimal point that repeat in a pattern.

THE REGRESSION MODEL IN MATRIX FORM. For simple linear regression, meaning one predictor, the model is. for i = 1, 2, 3,, n

Properties of MLE: consistency, asymptotic normality. Fisher information.

Topic 5: Confidence Intervals (Chapter 9)

Research Method (I) --Knowledge on Sampling (Simple Random Sampling)

, a Wishart distribution with n -1 degrees of freedom and scale matrix.

Chapter 7 Methods of Finding Estimators

Maximum Likelihood Estimators.

Quadrat Sampling in Population Ecology

THE TWO-VARIABLE LINEAR REGRESSION MODEL

1 Correlation and Regression Analysis

A Test of Normality. 1 n S 2 3. n 1. Now introduce two new statistics. The sample skewness is defined as:

Chapter 6: Variance, the law of large numbers and the Monte-Carlo method

Hypothesis testing using complex survey data

Hypergeometric Distributions

Definition. A variable X that takes on values X 1, X 2, X 3,...X k with respective frequencies f 1, f 2, f 3,...f k has mean

INVESTMENT PERFORMANCE COUNCIL (IPC)

CHAPTER 3 DIGITAL CODING OF SIGNALS

Chapter 5 Unit 1. IET 350 Engineering Economics. Learning Objectives Chapter 5. Learning Objectives Unit 1. Annual Amount and Gradient Functions

Non-life insurance mathematics. Nils F. Haavardsson, University of Oslo and DNB Skadeforsikring

Now here is the important step

5.4 Amortization. Question 1: How do you find the present value of an annuity? Question 2: How is a loan amortized?

Week 3 Conditional probabilities, Bayes formula, WEEK 3 page 1 Expected value of a random variable

TI-83, TI-83 Plus or TI-84 for Non-Business Statistics

A GUIDE TO LEVEL 3 VALUE ADDED IN 2013 SCHOOL AND COLLEGE PERFORMANCE TABLES

A Mathematical Perspective on Gambling

Predictive Modeling Data. in the ACT Electronic Student Record

LECTURE 13: Cross-validation

A Review and Comparison of Methods for Detecting Outliers in Univariate Data Sets

Example: Probability ($1 million in S&P 500 Index will decline by more than 20% within a

Chapter 5: Basic Linear Regression

CHAPTER 11 Financial mathematics

Section 11.3: The Integral Test

NATIONAL SENIOR CERTIFICATE GRADE 12

Central Limit Theorem and Its Applications to Baseball

Overview of some probability distributions.

Discrete Mathematics and Probability Theory Spring 2014 Anant Sahai Note 13

In nite Sequences. Dr. Philippe B. Laval Kennesaw State University. October 9, 2008

The analysis of the Cournot oligopoly model considering the subjective motive in the strategy selection

Using Four Types Of Notches For Comparison Between Chezy s Constant(C) And Manning s Constant (N)

COMPARISON OF THE EFFICIENCY OF S-CONTROL CHART AND EWMA-S 2 CONTROL CHART FOR THE CHANGES IN A PROCESS

How To Calculate A Radom Umber From A Probability Fuctio

A modified Kolmogorov-Smirnov test for normality

Approximating Area under a curve with rectangles. To find the area under a curve we approximate the area using rectangles and then use limits to find

CONTROL CHART BASED ON A MULTIPLICATIVE-BINOMIAL DISTRIBUTION

hp calculators HP 12C Statistics - average and standard deviation Average and standard deviation concepts HP12C average and standard deviation

INVESTMENT PERFORMANCE COUNCIL (IPC) Guidance Statement on Calculation Methodology

Optimal Adaptive Bandwidth Monitoring for QoS Based Retrieval

7. Concepts in Probability, Statistics and Stochastic Modelling

PROCEEDINGS OF THE YEREVAN STATE UNIVERSITY AN ALTERNATIVE MODEL FOR BONUS-MALUS SYSTEM

CHAPTER 3 THE TIME VALUE OF MONEY

Factoring x n 1: cyclotomic and Aurifeuillian polynomials Paul Garrett <garrett@math.umn.edu>

TI-89, TI-92 Plus or Voyage 200 for Non-Business Statistics

THE PROBABLE ERROR OF A MEAN. Introduction

Transcription:

Z-TEST / Z-STATISTIC: used to test hypotheses about µ whe the populatio stadard deviatio is kow ad populatio distributio is ormal or sample size is large T-TEST / T-STATISTIC: used to test hypotheses about µ whe the populatio stadard deviatio is ukow Techically, requires populatio distributios to be ormal, but is robust with departures from ormality Sample size ca be small The oly differece betwee the z- ad t-tests is that the t-statistic estimates stadard error by usig the sample stadard deviatio, while the z-statistic utilizes the populatio stadard deviatio Formula: Oe Sample T-test t = x µ where = s = estimated stadard error of the mea Because we re usig sample data, we have to correct for samplig error. The method for doig this is by usig what s called degrees of freedom 1

Degrees of Freedom degrees of freedom ( ) are defied as the umber of scores i a sample that are free to vary we kow that i order to calculate variace we must kow the mea ( ) this limits the umber of scores that are free to vary df = 1 X s = df ( x i x ) 1 where is the umber of scores i the sample Degrees of Freedom Cot. Picture Example There are five balloos: oe blue, oe red, oe yellow, oe pik, & oe gree. If 5 studets (=5) are each to select oe balloo oly 4 will have a choice of color (df=4). The last perso will get whatever color is left.

The particular t-distributio to use depeds o the umber of degrees of freedom(df) there are i the calculatio Degrees of freedom (df) df for the t-test are related to sample size For sigle-sample t-tests, df= -1 df cout how may observatios are free to vary i calculatig the statistic of iterest For the sigle-sample t-test, the limit is determied by how may observatios ca vary i calculatig s i t obt = x µ s z obt = x µ σ The z-test assumes that: the umerator varies from oe sample to aother the deomiator is costat Thus, the samplig distributio of z derives from the samplig distributio of the mea Z-test vs. T-test t obt = x µ s The z-test assumes that: the umerator varies from oe sample to aother the deomiator varies from oe sample to aother Therefore the samplig distributio is broader tha it otherwise would be The samplig distributio chages with It approaches ormal as icreases 3

Characteristics of the t-distributio: The t-distributio is a family of distributios -- a slightly differet distributio for each sample size (degrees of freedom) It is flatter ad more spread out tha the ormal z-distributio As sample size icreases, the t-distributio approaches a ormal distributio Itroductio to the t-statistic 3.5 3 t-dist., df=5.5 Normal Distributio, df= t-dist., df=0 1.5 1 t-dist., df=1 0.5 0-3 - -1 0 1 3 Whe df are large the curve approximates the ormal distributio. This is because as is icreased the estimated stadard error will ot fluctuate as much betwee samples. 4

Note that the t-statistic is aalogous to the z- statistic, except that both the sample mea ad the sample s.d. must be calculated Because there is a differet distributio for each df, we eed a differet table for each df Rather tha actually havig separate tables for each t-distributio, Table D i the text provides the critical values from the tables for df= 1 to df= 10 As df icreases, the t-distributio becomes icreasigly ormal For df=, the t-distributio is Procedures i doig a t-test 1. Determie H 0 ad H 1. Set the criterio Look up t crit, which depeds o alpha ad df 3. Collect sample data, calculate x ad s 4. Calculate the test statistic t obt = x µ s 5. Reject H 0 if t obt is more extreme tha t crit 5

Example: A populatio of heights has a µ=68. What is the probability of selectig a sample of size =5 that has a mea of 70 or greater ad a s=4? We hypothesized about a populatio of heights with a mea of 68 iches. However, we do ot kow the populatio stadard deviatio. This tells us we must use a t-test istead of a z-test Step 1: State the hypotheses H 0 : µ=68 H 1 : µ 68 6

Step : Set the criterio oe-tail test or two-tail test? α=? df = -1 =? See table for critical t-value Step 3: Collect sample data, calculate x ad s From the example we kow the sample mea is 70, with a stadard deviatio (s) of 4. Step 4: Calculate the test statistic Calculate the estimated stadard error of the mea = s = 4 5 = 0.8 Calculate the t-statistic for the sample x µ t = t = 70 68 0.8 =.5 7

Step 5: Reject H 0 if t obt is more extreme tha t crit The critical value for a oe-tailed t-test with df=4 ad α=.05 is 1.711 Will we reject or fail to reject the ull hypothesis? 4 3 1 0 t crit =1.711 0.05 Example: A researcher is iterested i determiig whether or ot review sessios affect exam performace. The idepedet variable, a review sessio, is admiistered to a sample of studets (=9) i a attempt to determie if this has a effect o the depedet variable, exam performace. Based o iformatio gathered i previous semesters, I kow that the populatio mea for a give exam is 4. The sample mea is 5, with a stadard deviatio (s) of 4. 8

We hypothesized about a populatio mea for studets who get a review based o the iformatio from the populatio who did t get a review (µ=4). However, we do ot kow the populatio stadard deviatio. This tells us we must use a t-test istead of a z- test Step 1: State the hypotheses H 0 : µ=4 H 1 : µ 4 Step : Set the criterio oe-tail test or two-tail test? α=? df = -1 =? See table for critical t-value Step 3: Collect sample data, calculate x ad s From the example we kow the sample mea is 5, with a stadard deviatio (s) of 4. 9

Step 4: Calculate the test statistic Calculate the estimated stadard error of the mea s 4 4 = = = = 1.33 9 3 Calculate the t-statistic for the sample x µ t = 6 4 t = = 1.33 1.33 = 1.503 Step 5: Reject H 0 if t obt is more extreme tha t crit The critical value for a oe-tailed t-test with df=8 ad α=.05 is 1.86 Will we reject or fail to reject the ull hypothesis? 4 3 1 0 0.05 t crit =-.101 α t + α crit =.101 0.05 10

Assumptios of the t-test: Idepedet Observatios: Each perso s score i the sample is ot affected by other scores; if, for example, subjects cheated from oe aother o the exam, the idepedece assumptio would be violated Normality: The populatio sampled must be ormally distributed Need to kow oly the populatio mea Need sample mea ad stadard deviatio Cofidece Itervals Ofte, oe s iterest is ot i testig a hypothesis, but i estimatig a populatio mea or proportio This caot be doe precisely, but oly to some extet Thus, oe estimates a iterval, ot a poit value The iterval cotais the true value with a probability The wider the iterval, the greater the probability that it cotais the true value Thus there is a precisio/cofidece trade-off The itervals are called cofidece itervals(ci) Typical CIs cotai the true value with probability.95 (95% CI) ad with probability.99 (99% CI) CI is calculated with either t or z, as appropriate 11