Stats Test 3 Chapter 8

Front 1

Point Estimate

Back 1

A summary statistic from a sample that is just one number as an estimate of the population parameter.

Front 2

Interval Estimates

Back 2

It conveys the range of the sample statistics we could expect if we conducted repeated hypothesis tests using samples from the same population

"casting a net in hopes to capture the whole dartboard"

Front 3

Confidence Interval

Back 3

A calculated interval estimate that surrounds the point estimate

Front 4

Confidence Interval Steps

Back 4

Step 1: Draw the curve and mark the sample mean in the middle
Step 2: Indicate the bounds of the confidence interval on both ends, and color in the approximate percentages under each segment of the curve
Step 3: Look up z- value for upper and lower ends of the confidence interval
Step 4: Convert the z-values to raw scores for both the upper and lower ends of the curve
Step 5: Check your work
Upper - middle= same
Lower- middle= same

Front 5

Confidence Interval Step 4 Calculations....

Back 5

Front 6

Effect Size Definition

Back 6

Amount 2 populations do not overlap
larger effect size- more significant(most likely)
sometimes small can too

Front 7

Things that influence effect size

Back 7

Front 8

Formula for effect size

Back 8

Front 9

Summary for Cohen's Effect Size Conventions for Mean Differences

Back 9

Front 10

P rep

Back 10

Step 1: look up the percentage using the z- statistic
Step 2: if two tailed, double the percentage
Step 3: Divide by 100

Front 11

Statistical Power

Back 11

Probability that the study will produce a statistically significant result if the research hypothesis is true
Probability you will not make a type 2 error
helps plan future studies, helps evaluating results of studies

Front 12

Calculating Statistical Power

Back 12

Percent above the mean

Front 13

What Influences Power

Back 13

Smaller standard deviations give you more power than larger standard deviations
Larger the sample size more power
Less stringent the significance level (.05 less stringent than .01 ) the greater the power
One tailed test have more power than two tailed test for a result in the predicted direction
One tailed is less extreme cut off score than a two tailed test
ZERO power for a result in the opposite direction using a one tailed test

Front 14

Importance of Power in Interpreting Results

Back 14