statistics: 4.3 HYPOTHESIS TESTING

what questions does inferntial statistics serve to answer? 2 pts

1. is there any relationship between the variables

2. are there differences between groups or conditions

research hypothesis vs statistical hypothesis? 2 pts

1. research hypothesis- involves making a substantive scientific claim that can be tested (scientific claims)

2. statistical hypothesis- involves making a statement about the probability distribution of a population value, must be mathematically precise and have a clear correspondence with the research hypothesis, and are specific claims about the population that are expressed formally/mathematically

from research hypothesis to statistical hypothesis? 6 pts

1. state the research hypothesis

2. identify the variables

3. define how the variables will be measured

4. choose the appropriate statistical test

5. translate the prediction into statistical language

6. write the statistical hypothesis

7. create the null hypothesis and test if it is true or false depending on the data obtained from samples

steps of hypothesis formulation? 2 pts

1. a researcher has some theory about the world and wants to determine whether or not the data actually supports that theory

2. the researcher must then create a null hypothesis (after the alt one)

why do we have to create a null hypothesis? 1 pt

because we do not directly rest our hypothesis in stats; instead we try to reject the null hypothesis

null hypothesis vs alternative hypothesis? 2 pts

1. null- states there is no effect of not variable on another variable

2. alt- states that there is an effect of one variable on another variable

how to create a statistical hypothesis? 2 pts

there are two scenarios:

1. values change but you are unsure about how they change (direction of effect is unknown) → run a nondirectional/two tailed test to predict a change in scores or run a directional/one tailed test to predict the direction of scores

2. variable change and you predict how they change (direction); whether they will increase or decrease

hypothesis testing: decisions and consequences? 2 pts

1. type 1 error- rejecting a true null hypothesis (false positive)

2. type 2 error- accepting a null hypothesis (false negative)

what is a test procedure? 2 pts

a rule based on sample data for deciding whether to reject or retain the null hypothesis and requires:

1. a test statistic

2. a rejection region (alpha or significance level)

what is a test statistic? 1 pt

a number calculated from a statistical test of a hypothesis that tells us how far a sample result is from what the null hypothesis predicts

what is the purpose of a test statistic? 2 pts

1. to show how closely our sample data match the distribution under the null hypothesis of that statistical test

2. used to calculate the p value of your results which helps us decide whether to reject a null hypothesis or not

what is a rejection region? 1 pt

the set of all test statistic value for which the null hypothesis will be rejected; values of the test statistic that we consider to be extreme and inconsistent with the null hypothesis

how do we choose the rejection region? 4 pts

1. we choose a region that minimizes the probability of error in our decision

2. in general we decide to allow for a 5% chance of error so we choose a rejection region that leaves behind 5% of probability

3. a statistic in our sample data is chosen to be calculared using statistical software which tells us the the value for the test statisitic and the probability of finding that value if the null hypothesis were true

4. with that info we can make a decision of rejecting or retaining the null hypothesis

what is alpha value? 1 pt

the probability that we assign to the rejection region, the amount of error that we are willing to tolerate in our decisions; strictiness varies

what is p-value? 4 pts

1. the probability of our results if the null hypothesis is true; how likely our data is under the hypothesis that there is no effect

2. the smaller the p value the less likely your data could have occured under if the null hypothesis were true

3. in order to reject the null hypothesis our p-value should be smaller than or equal to our alpha level e.g. alpha=5% p-value=</=0.5

4. if the p-value is greater than the alpha value then we acept the null hypothesis

what is the 5% rule in p-values? 2 pts

1. f the p-value is less than 0.05 (5%), reject the null hypothesis

2. The 5% level is called the significance level that tells us how much risk a researcher is willing to accept for rejecting a null hypothesis