Exam 2

Measure of central tendency: What is a typical value for data set

Measure of dispersion: How different are the data points from one another

• Mean: central tendency; interval data; this is the average

• Median: central tendency; ordinal and interval data; this is the value (data point) in the middle

• Mode: central tendency; nominal, ordinal, and interval data; this is the most frequent value

• Range: dispersion; ordinal data but mostly interval data; the span of the data (largest value minus the smallest value)

• Standard deviation: dispersion; interval data; how much does the typical data point differ from the mean

Under what conditions would one want to use the median instead of the mean? You use the median over the mean when you have extreme data points.

• measure of association - Describe the statistical strength of the relationship between two or more variables

• correlation; tells you the strength of the relationship between two variables

  • different levels of correlation: weak, strong, or no correlation; positive or negative 

• test of significance - Could your results simply be a function of chance? The results that occurred in the sample, would they likely occur in the population? Do I accept my hypothesis or reject it?

• null hypothesis - the negation of the substantive hypothesis

• substantive hypothesis - the hypothesis you think is correct

• criterion of significance level

  • what is - allows the researcher to evaluate the p-value (aka predetermined level) and is typically set to .05

  • how arrived at - determine the cost of accepting the substantive hypothesis and being wrong compared to the cost of accepting the null hypothesis and being wrong

  • how to compare criterion of significance with p-value  - If the p-value is at or less than the criterion of significance, then accept the substantive hypothesis. If the p-value is greater than the criterion of significance, then accept the null hypothesis.

• factors that affect p-values

  • The larger the sample size the smaller the p-value 

  • The greater the correlation among variables, the smaller the p-value (the greater the difference between the subgroups, the smaller the p-value)

  • The type of test - the different types of tests of significance 

• statistical v practical significance - Statistical significance is not a functioning chance, aka I would find similar results if I interviewed the entire population. Practical significance - Whether the difference is big or small is a function of interpretation

• Type I error - incorrectly accept the substantive hypothesis (accept the substantive hypothesis and later find out you were wrong)

• Type II error - incorrectly accept the null hypothesis (accept the null hypothesis and later find out you are wrong)

• Trade-off between Type I v Type II error - If I want to reduce the possibility of a type 1 error, decrease the criterion and make it less than .05. Lowering the criterion increases your possibility of making a type 2 error. To decrease the possibility of a type 2 error, increase the criterion, which increases your chance of making a type 1 error.