Lab 2 new

The smaller the sample, the bigger the standard error.

• True

• False

Which of the following statements is false?

• A positive correlation coefficient means that if the value of y is large, x is unlikely to have a value lower than the mean of x.

• T-tests and correlations assume that variables are normally distributed.

• You can plot a QQ-Plot in Stata with the command qqplot.

• Examples of test statistics are Pearson's r, regressions, and differences in means.

Why do we construct a null hypothesis (H0)?

• To test if X (our assumed effect) is true.

• To start with the assumption that X (our assumed effect) is not true.

• To be able to reject the alternative hypothesis H1.

A p-value is the probability that H1 is true or false.

• True.

• False

What do we learn from a p-value?

(several answers correct!)

• The smaller the p-value, the stronger the evidence against H0.

• The smaller the p-value, the less statistically significant is our estimate.

• The probability of observing something as extreme as our estimate under H1.

• The smaller the p-value, the lower the probability that we accept H0.

The standard error is used to calculate the confidence interval

• True

• False

Which interpretation of confidence intervals is correct?

Under repeated sampling, the 95% confidence interval...

• ... indicates that a specific interval has a 95% probability of containing the true effect size.

• ... indicates the probability that the true effect size is included in the interval.

Testing for a significance level of 0.05, based on this t-test, we can say...

(several answers correct)

• The hypothesis that there is no difference between the Nicotine and Placebo groups can be rejected.

• The t-test only shows a one-sided test.

• The confidence interval of the difference between the Nicotine and Placebo groups includes 0.