PSYC 2260 Lecture Notes - Lecture 4: Null Hypothesis, Statistical Hypothesis Testing, Research Question
26 Jun 2018
School
Department
Course
Professor
PSYC 2260 Introduction to Research Methods in Psychology
Chapter 4 Introduction to Hypothesis Testing
Hypothesis testing: procedure for deciding whether the outcome of a study (result for a sample)
supports a particular theory or practical innovation (thought to apply to a population).
A hypothesis is not the same thing as a research question
- Research question: “does amount of sleep increase alertness?”
- Hypothesis: those who sleep 8 hours will be more alert than those who sleep 6 hours
A declarative statement ; Makes specific prediction; Is testable
E.g. we need to reject the hypothesis that sleep has no effect
If we reject the “no effect” hypothesis. Then, by default, all we have left is the alternative
Thus testing a hypothesis is about rejecting its opposite; “Opposite of what-you-predict”
P.S critic: “opposite” is a bad wording
- E.g. “drug X will increase Y by 10 point what is the opposite of 10 point?
- Possible alternative: does not support ….is in opposition to….
The Hypothesis-Testing Process
Step1: Restate the Question as a research hypotheses
When stating a hypothesis, actually stating 2 hypotheses
1. The Null Hypothesis (HO)
- There is no different in alertness between those who sleep 8 vs 6 hours
- Ho: µ1 = µ2
2. The alternative hypothesis (HA)
- People who sleep 8 hours will be more alert than sleep 6 hours
- Ha: µ1 > µ2; µ1 ≠µ2
Step2: Determine the Characteristics of the Comparison Distribution
Comparison distribution: distribution used in hypothesis testing. It represents the population
situation if the null hypothesis is true. It is the distribution to which you compare the score
based on your sample’s results.
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Hypothesis testing: procedure for deciding whether the outcome of a study (result for a sample) supports a particular theory or practical innovation (thought to apply to a population). A hypothesis is not the same thing as a research question. Research question: does amount of sleep increase alertness? . Hypothesis: those who sleep 8 hours will be more alert than those who sleep 6 hours. A declarative statement ; makes specific prediction; is testable. E. g. we need to reject the hypothesis that sleep has no effect. Then, by default, all we have left is the alternative. Thus testing a hypothesis is about rejecting its opposite; opposite of what-you-predict . Possible alternative: does not support . is in opposition to . Step1: restate the question as a research hypotheses. When stating a hypothesis, actually stating 2 hypotheses: the null hypothesis (ho) There is no different in alertness between those who sleep 8 vs 6 hours. Ho: 1 = 2: the alternative hypothesis (ha)
