Introduction to Inferential Statistics: What you need to know before you take regression

This course was designed for people who do not have much prior experience with statsitics or they are not totally comfortable with basic statistical concepts like hypothesis testing, p-values, standard errors, distributions, correlation and etc.  It is designed to provide a solid foundation for people who are starting for practically 0 and would like to work up to running multiple regression (which is the most commonly used statistical technique in the social sciences).  This one week course offered in the first week will cover the basics people need to take regression in the second week.

Day 1:  First we discuss basic distributions with a strong emphasis on the normal distribution (otherwise known as the bell curve) but considering others as well.  We cover the types of variables and how central tendency (mean, median, mode) and variability (range, variance and standard deviation) can be measured as these are the essential foundational building blocks of statistical inference.

Day 2: We continue by understanding how standardized scores (such as IQ, SAT, GRE and, most importantly, z-scores) can get us one step closer to comparable statistical estimates used for inference. Then we turn back to the normal distribution to see how the concepts of probability, distributions and standardized scores come together highlighting the power of both the normal distribution and what it can tell us.

Day 3: We explore why it is not even crucial to have normally distributed variables when doing statistical inference.  With our newly gained knowledge we learn to leverage statistical properties of group means and learn how to formulate (and even test) simple hypotheses using statistical inference.

Day 4: We leverage the discoveries of an old brewery employee who also revolutionized the world of statistics by allowing us to move from simple hypothesis testing of situations with too many assumptions (z test) to be realistic to testing hypotheses commonly found in the social sciences (t test and ANOVA).  We explore both group comparisons, changes over time and comparisons to some gold standard value.

Day 5: We move to testing relationships between variables using correlations and we introduce bivariate regression which provides the foundation of multiple regression analysis. If time allows we explore another class of statistics which make less assumptions about distributions and learn how to answer different types of research questions with this different type of analytical technique (the Chi-Square test).

NONE – REALLY!  If you remember how to add, subtract, divide and multiply using both negative and positive numbers, you are ready to take this course. There may be an occasional square and square root that, in this class, we are happy to calculate with a calculator.