Scribd
Upload
3 views

module 3 _lecture notes

Uploaded by

Phil Acuña
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
3 views

module 3 _lecture notes

Uploaded by

Phil Acuña
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Lecture Notes – Module 3

Statistical Significance and Hypothesis Testing

What You’ll Learn About in This Chapter

 Understanding the concept of significance is and why it is important


 Distinguishing between Type I and Type II errors
 Understanding how inferential statistics work
 Selecting the appropriate statistical test for your purposes
 Deciding when the z-test for one sample is appropriate to use
 Computing the observed z value (for illustrative purposes only)
 Interpreting the z value – *a must for every student of development
 Understanding what the z value means
 Understanding what effect size is and interpreting it

Statistical Significance

I. The Concept of Significance—In statistics, a significant result is one that means


the difference between or among the variables is due to some systematic influence
and not due to chance.
A. If Only We Were Perfect
1. In most scientific endeavors, there is some amount of error that cannot be
controlled.
2. Significance level is the risk associated with not being 100% confident
that what you observe in an experiment is due to the treatment or what was
being tested.
3. Statistical significance (here’s the formal definition) is the degree of risk
you are willing to take that you will reject a null hypothesis when it is
actually true.
4. There’s always a chance, no matter how small, that you are wrong.

1
B. The World’s Most Important Table (For This Semester Only)—The null can
either be true or false, but you will never know for sure. You may reject the
null or fail to reject it. That produces four possible outcomes.
C. More About Table 9.1
1. If you fail to reject the null hypothesis, and the null hypothesis is true,
there is no difference between the groups (good job).
2. If you reject the null, and the null hypothesis is false, there are real
differences between the groups (good job).
3. If you reject the null, and the null hypothesis is true, you have committed a
Type I error, where there really is no difference between the groups.
4. If you fail to reject the null, and the null hypothesis is false, you have
committed a Type II error, where there really is a difference between the
groups.
D. Back to Type I Errors—Typically, researchers set a tolerance or alpha level
for how willing they are to reject the null hypothesis when it might actually be
true. Traditionally, this level is set at 5% or 0.05.
E. Significance Versus Meaningfulness—Statistical significance means that the
research was a technical success, and the null hypothesis is not a reasonable
explanation for what was observed. Some expensive, complicated treatment
might produce the tiniest effect. It may be statistically significant, but does
that mean the treatment is a success and worth implementing? Not necessarily.
II. An Introduction to Inferential Statistics
A. Descriptive statistics are used to describe a sample’s characteristics.
B. Inferential statistics are used to infer something about the population based on
the sample’s characteristics.
C. How Inference Works—There are general steps to each research project,
including the following:
1. Select representative samples.
2. Collect the relevant data.
3. Reach a conclusion as to whether or not the difference between the scores
is the result of chance.

2
4. Reach a conclusion that applies to the whole population based on the
finding within the samples.
D. How to Select What Test to Use
1. In Figure 9.1, you will see a chart sheet on how to select which test to use.
2. Start by answering the questions at the top of the flowchart.
3. Proceed down the chart by answering each of the questions until you get
to the end of the chart.
4. Note: Figure 9.1 does not include all the statistical tests in existence.
III. An Introduction to Tests of Significance
A. Tests of significance allow decisions to be made about entire populations
based on known information of samples.
B. How a Test of Significance Works: The Plan
1. A statement of the null hypothesis.
2. Setting the level of risk (or the level of significance or Type I error)
associated with the null hypothesis.
3. Selection of the appropriate test statistic.
4. Computation of the test statistic value.
5. Determination of the value needed for rejection of the null hypothesis
using the appropriate table of critical values for the particular statistic.
6. Comparison of the obtained value to the critical value.
7. If the obtained value is more extreme than the critical value, the null
hypothesis cannot be accepted.
8. If the obtained value does not exceed the critical value, the null hypothesis
is the most attractive explanation.
IV. Here’s the Picture That’s Worth a Thousand Words—The text explains how
people can use the normal curve to visualize how the test of significance works.
A. Be Even More Confident
1. A confidence interval (or CI) is the best estimate of the range of a
population value (or population parameter) that we can come up with the
given sample value (or sample statistic).
2. 95% Confidence Interval = Score + 1.96 (standard deviation)

3
3. 99% Confidence Interval = Score + 2.56 (standard deviation)

4
Part II. Z-test

I. Introduction to the One-Sample z-Test


A. If a researcher wants to compare a sample to an entire population, a one-
sample z-test may be appropriate.
B. The Path to Wisdom and Knowledge—Use the flowchart to determine if a
one-sample z-test is appropriate.
II. Computing the Test Statistic
A. z-Test statistic

B. SEM Formula

III. Eight Famous Steps


A. State the null and research hypotheses.
B. Set the level of risk associated with the null hypothesis.
C. Select the appropriate test statistic.
D. Compute the test statistic value (called the obtained value).
E. Determine the value needed for rejection of the null hypothesis using the
appropriate table of critical values for the particular statistic.
F. Compare the obtained value and the critical value.
G. Make your decision.
IV. So How Do I Interpret z = 2.38, p < 0.05?
A. z represents the test statistic.
B. 2.38 is the obtained value in the example used in this chapter.
C. p < 0.05 indicates that the probability is less than 5% that on any one test of
the null hypothesis, the sample and the population averages differ.

V. Special Effects: Are Those Difference for Real?


A. Effect size is the strength of a relationship between variables.
1. A measure of the magnitude of the treatment.

5
2. So effect sizes can be correlational values or values that estimate
difference.
3. Cohen’s d is the effect size for the one-sample z-test.
4. Cohen’s d is the effect size for the one-sample z-test.
VI. Understanding Effect Size
A. A small effect size range from 0.0 to 0.20.
B. A medium effect size range from 0.20 to 0.50.
C. A large effect size is any value above 0.50.
VII. Summary—The one-sample z-test is the simplest example of an inferential test
and leads the reader to a very common inferential test, known as the t-test.

You might also like