Class Notes (1,019,094)
US (398,522)
UND (504)
PSYC (105)
Lecture 8

PSYC 241 Lecture Notes - Lecture 8: Standard Deviation, Sample Size Determination, Statistical SignificancePremium

5 pages88 viewsSpring 2017

Department
Psychology
Course Code
PSYC 241
Professor
Kristen Leighton
Lecture
8

This preview shows page 1. to view the full 5 pages of the document.
Lesson 8: Introduction to Hypothesis Testing
Hypothesis Testing
Goal: rule out chance (sampling error) as a plausible explanation for results
Hypothesis testing helps determine whether a treatment has an effect
To evaluate research in which:
1. A sample is selected from the population
2. The treatment is administered to the sample
3. The people in the sample are measured after treatment
If people in sample are noticeably different from those in original population, we have
evidence of treatment effect
It is also possible, however, that the difference between sample and population is due
to sampling error
A statistical hypothesis is an assumption about a population parameter
Best way to determine whether hypothesis is true would be to examine entire
population
Impractical --> impossible
Random sample is proxy for population
Use this sample to test hypothesis
Two possible explanations:
1. The difference between sample and population can be explained by sampling
error
No treatment effect
2. The difference is too large to be due to sampling error
Treatment effect
Cannot "prove" things to be true
Much easier to show that something is false
Dilemma for researchers:
We want to support our hypotheses, but the techniques available are better for showing
that something is false
Null Hypothesis
Two types of statistical hypotheses:
1. Null hypothesis: No difference
H0
No difference between sample statistics and population parameters or
No difference between samples
--> no difference between populations
Any minor differences due to chance
Alternative Hypothesis
2. Alternative hypothesis: Difference
H1 or Ha
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Subscribers Only

Only half of the first page are available for preview. Some parts have been intentionally blurred.

Subscribers Only
Difference between sample statistics and population parameter or
Difference between two samples
--> difference between populations
Difference due to some non-random cause
Even though we are using samples to test our hypothesis, the hypotheses are always stated
based on populations
Statistical Notation
H0: µ0 = µ1
No difference in population means
H1: µ0 doesn't equal µ1
Ha: µ0 doesn't equal µ1
Population means are not equal (two-tailed, non-directional)
Non-directional tests are more stringent; more conservative
Directional Tests
If researchers predicts a specific direction for the treatment effect (increase or
decrease), can incorporate the directional prediction into the hypothesis test
The result is called a directional test or a one-tail test
Directional test includes directional prediction in statement of hypotheses and location
of the critical region
For example, if the original population has a mean of 80 and treatment is predicted to
increase the scores, the null hypothesis would state that after treatment:
H0: mean is equal to or < 80 (there is no increase)
Statistical Notation
H1: µ0 < µ1
H1: µ0 > µ1
Ha: µ0 < µ1
Ha: µ0 > µ1
One mean is greater (or less than) the other (one-tailed, directional)
The Null hypothesis, the Alpha Level, the Critical Region, and the Test Statistic (can be written on
calculation sheet)
5 steps in hypothesis testing:
Step 1: State Hypothesis
State the hypotheses; select an α (alpha) level
Null hypothesis, H0, always states that the treatment has no effect
Alpha level establishes a criterion, or "cut-off," for making a decision
about the null hypothesis
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Subscribers Only

Loved by over 2.2 million students

Over 90% improved by at least one letter grade.