Chapter 17: INFERENTIAL stats Flashcards
What tells us that the alternative hypothesis MIGHT be true? There is no relationship between variables (any relationship is due to chance)
Rejection of the null hypothesis (we seek to reject the null hypothesis)
*this examines whether the sample mean falls in a small range of the bell-shaped curve
As a researcher, you are always seeking to ______ the null hypothesis.
REJECT
2 other types of hypotheses
- Directional 2. Non-directional
Which type of hypothesis? *one group mean is expected to be higher than another; greater, less than, increase, decrease
directional
Which type of hypothesis? *a difference between the means or a change of any kind is implied; different from another but not in any specific direction
non-directional
What type of error?: Null hypothesis is incorrectly rejected
Type I error
ex. type I error might allow an ineffective drug to come on the market
(further explanation: hypothesis says that drug is effective, null hypothesis says drug is ineffective –> you reject the null hypothesis…so you are saying the drug is effective…when in reality it isn’t)
Signifies probability of incorrectly rejecting a true null hypothesis; controlling the risk of committing a Type I error
level of significance: increasing this also increases the chance of making a Type I error
reduces the risk of type II error
Power Analysis
Power is typically set at ________
.80
4 components to _____ _______ 1. significance criterion 2. sample size 3. effect size 4. power
power analysis
Viewing a normal curve, right tailed, small shaded region with calculated value and critical value established =
rejection region (you reject the null hypothesis)
means for drawing conclusions about a population, given data from a sample; provide framework for making objective judgements about the reliability of sample estimates
inferential statistics
null hypothesis (ex. drug is ineffective) is wrongly accepted
Type II error
ex. might prevent an effective drug from coming on the market
(Further description: your hypothesis was that the drug was effective, null hypothesis was that the drug was ineffective –> Type II error says that you wrongly accepted the null hypothesis so you said that the drug was ineffective…when in reality, it was effective)
T/F: statistically significant does not mean results are significant. It means that obtained results are not likely to have been result of chance, at a specified level of probability
True
In the small shaded regions of a two-tailed test, do you accept or reject the null hypothesis?
reject the null hypothesis
