Hypothesis Testing

Question 1

null hypothesis (H0)

Answer 1

the hypothesis to be tested (usually of “no difference”, “no effect”, or “no association between a risk factor and disease”).

Question 2

alternative hypothesis (H1)

Answer 2

the hypothesis that contradicts the null hypothesis (usually the research hypothesis of interest).

• Two-sided–Used when we are interested in any deviation from the null hypothesis (used more often than one-sided)
• One-sided–Used when we have prior evidence that leads us to be interested in a deviation from the null hypothesis in one direction

Question 3

significance level (α)

Answer 3

• Objective cutpoint used for choosing between H0 and H1
• Represents the probability of choosing H1 when H0 is really true (i.e. the probability of rejecting a true null hypothesis)
• α = 0.05 implies that there is 5% error. That is, 5% of the time you conclude there is an association when there really isn’t one (false positive)
• If not stated (and often it is not), assume α = 0.05

Question 4

p-value

Answer 4

Probability of obtaining the observed sample estimate (or a more extreme estimate) by chance alone if the null hypothesis is true. The p-value is not the probability the H0 is true (a common error).

Question 5

General guidelines for assessing p-values

Answer 5

p ≤ .001 very highly significant
.001 < p ≤ .01 highly significant
.01 < p ≤ .05 significant
p > .05 non-significant (observed difference is compatible with chance)
.05 < p ≤ .10 borderline significance

Question 6

Paired Samples

Answer 6

two groups of subjects individually matched on certain factors

Question 7

ANOVA (ANalysis Of VAriance)

Answer 7

Extension of t-test to comparison of more than two groups

Question 8

2 x 2 table

Answer 8

a format for displaying data that is classified by two different variables, each of which has only two possible outcomes

Question 9

1-sided

Answer 9

There is an association in a specific direction (positive/negative) between variable A and variable B

Question 10

2-sided

Answer 10

There is an association in no specific direction (positive/negative) between variable A and variable B

Question 11

Chi-Square Test

Answer 11

chi-squared test is used to compare a categorical outcome variable between groups

Question 12

t-test

Answer 12

t-test is used to compare a continuous outcome variable between groups.

Question 13

Estimation of Proportions (Probabilities)

Answer 13

In experiments with two outcomes (e.g. success or failure), we are interested in estimating the population proportion (probability) of success where # of successes/# of subjects = the EoP (Probabilities).