A primer on bootstrapping
We are almost never only interested about the characteristics of a sample per se. What we want to know is about the population from which the sample was drawn. That is, we want to make inferences about a population on the basis of statistical estimates derived from a sample.
Bootstrapping involves drawing hundreds or even thousands of random sample subsets from a given full sample and calculates statistics on each sample. In this way, bootstrapping provides a way to obtain more robust estimates of standard errors and confidence intervals for many different statistical estimates including the mean, median, proportion, odds ratio, correlation coefficient or regression coefficient.
Multiple Linear Regression
Multiple linear regression involves attempting to model the relationship between two or more predictor variables and an outcome variable by fitting a linear equation to observed data. It is perhaps one of the most useful statistical tools around.
Principle Components Analysis
The primary aim of PCA is to reduce a larger set of variables into a smaller set of factors (called principal components) that account for most of the variance in the original variables. These slides demonstrated how to perform PCA using SPSS software.
Statistical Power
Statistical power is the probability that a test will correctly identify a real effect; that is, the probability that it will correctly reject the null hypothesis. Most studies are very much underpowered. These slides discuss the concept of statistical power and indicate how to determine the required sample size given a desired level of power.
Logistic Regression
Logistic regression is similar to regular multiple linear regression in that the aim is to produce a model of some system that predicts some outcome variable. The difference is that in the case of logistic regression, the outcome variable is dichotomous (e.g., yes, no; survive, die; success, failure; sale, no sale), whereas in the case of linear regression, the outcome variable is continuous (e.g., weight, depression, sales). These slides describe logistic regression and demonstrate how to perform an analysis using SPSS software.
What Does Statistical Significance Mean?
Researchers are always on the lookout for p-values less than 0.05. But what exactly does this mean and why is it important? You may be surprised to find out the truth about statistical significance!
Chi-Square Goodness of Fit and Test of Independence
The chi-square goodness of fit test and the test of independence are both used where variables are frequencies/counts. These slides show how to conduct both types of chi-square analyses using SPSS software.
Binary data effects
Many outcomes in medical, education, and business research are of a binary nature. They can hold one of two values. A patient can survive or die; a student can pass or fail; a customer can buy or not. Researchers have developed different ways of expressing such binary outcomes. These include relative risk, odds ratio, absolute risk difference, and numbers needed to treat, etc. These slides thoroughly describe the major binary outcome measures, how to interpret them, and how they are related to each other.
Information Intelligence 