This Week Homework 2 parts 1 and 2 due by 5pm today. Answer keys posted tonight. Midterm 2 on Thursday, bring 8 ½ by 11 inch note sheet. No calculator.

This Week

• Homework 2 parts 1 and 2 due by 5pm today. Answer keys posted tonight.

• Midterm 2 on Thursday, bring 8 ½ by 11 inch note sheet. No calculator needed (no calculations). About 35 questions.

• Review Session on Wednesday.• Practice problems on Tuesday.

Euro Cup Round 2Exercise (required)

• Use FIREFOX.• Due by 5pm Tuesday.• Complete at least 3 sample mean

problems, one problem from each category.

• Do at least 10 of the Variable Trainer problems.

Hypothesis Tests Only

Population Percentages: Planning the Study

How many observations will I need to get an ‘accurate’ estimate of the population percentage?

For example, suppose I wish to ‘know’ (estimate with 95% confidence) the true percentage of Calpoly students who live with their parents.

If I want to have a margin for error of 2 points, how many students should I (randomly) sample?

cutoffz

n

2

2 )100()(

E

zn cutoff

E

Sample Size Calculator II(Population Percentage)

Round UP this number

Calculate

Clear

Population Percentages: Planning the Study

How many observations will I need to get an ‘accurate’ estimate of the population mean?

If I want to have a margin for error of 2 points, how many observations will I probably need?

Preparing an Experiment:Quantitative Dependent Variable

How many samples should you collect if:

You want a 95% confidence interval on the population mean.

You want it to be accurate within __ points ( E ).

X

EE

cutoffz

n

2)(

E

zn cutoff

E

SampleSample SizeSize CalculatorCalculator

Round UP this number

Calculate

Clear

Example

If the standard deviation of a population is 0.83 inches, how many samples would be needed to obtain a 95% confidence interval whose total width is 0.2 inches wide (the upper bound minus the lower bound equals 0.2)?

cutoffz

n

2)(

E

zn cutoff

E

SampleSample SizeSize CalculatorCalculator

Round UP this number

Calculate

Clear

Power and Effect Size Simulator Power and Effect Size Simulator

True True μμ::

True True σσ::

Sample Size Sample Size nn::

μμ : :HH00

% times reject null:% times reject null:

Significance level (Significance level (αα):):

p p Value:Value:

Experiment has been run times.Experiment has been run times.

Refresh All

Run Experiment

Run Experiment 1000 Times

Reset Counters

Review

The power of your hypothesis test is the probability of rejecting a false null hypothesis.

It is

1. zero if your null hypothesis test is true.

2. unknown if your null hypothesis test is false (because we don’t know the effect size).

3. increasing with the effect size.