Topic

Univariate and bivariate statistics

Lecture

topic 1, Uni- versus bi-variate statistics

Purpose

Practice identifying how many variables and their scale level. Let's do it!

Question

An important first step is to identify given an managerial question or problem how many variables (or columns in your data table) are needed to address the question, and what the scale level(s) is (are) of the variable(s).

For each of the following cases, identify how many variables are involved and what the scale levels are of the variables. You may have to also brainstorm a bit of what data you would need to collect to address the question.

Case 1

The data science team at LinkedIn is asked to find out the proportion of users that have a post-college degree and the proportion of users that only have a high school degree.

Case 2

The MBA office asks you to investigate which section (ES1, ES2, etc.) was more satisfied with Data Science Camp, or whether there were no differences.

Case 3

The president is wondering whether men or women are more likely to vote for the green party.

Case 4

Linda is selling ice creams on the beach. She is wondering whether the sales of ice cream are affected by temperature.

Topic

ANOVA and mean comparisons

Lecture

topic 3, Comparing more than two means

Purpose

Practice interpreting the results of an ANOVA test. Let's do it!

Question

What do you conclude from these analyses?

Topic

Bivariate statistics -- t/Z test

Lecture

topic 2, Comparing two means

Purpose

Practice carrying out the 6 steps of hypothesis testing for two means by hand. Let's do it!

Topic

Practice working with SPSS

Lecture

topic 2, Comparing two means

topic 3, Comparing more than two means

This activity is only useful if you have reviewed the corresponding SPSS How to guide.

Purpose

Use SPSS to test hypotheses about the population. Follow the following managerial/research questions to further practice your SPSS skills. Let's do it!

In this module you learned about comparing means when your analysis requires you to use a quantitative variable and a categorical variable at the same time. SPSS can easily perform such analyses for you. It is always a good idea to write down the 6 steps of hypothesis testing on scratch paper, even when using SPSS. Practice conducting a t-test and an ANOVA with SPSS following two scenarios from the American Express mini-case (module 1, SPSS file 'mini_case_AE_credit_card_web.sav').

Question 1

Suppose you are interested in monthly credit card spending in the grocery category. You are only interested in how much customers spend once they have made the decision to use their primary card for groceries. Such an analysis can be interesting for instance if we believe that the first decision is whether or not to use the primary card to pay, followed by the second decision of how much to spend. Here we would focus on the second step. In other words, in your analyses, you should ignore the zeros (a zero means that presumably the customer did not use their primary card that month to buy groceries). In particular, you are interested in comparing the spendings (on average) of men and women, irrespective of their card brand. What do you conclude? Are the monthly grocery spendings using the primary card, on average, higher for men or women? Or is there no difference?

Question 2

Following the previous question, suppose that you want to know whether there is a difference, on average, between the primary card brands with respect to monthly grocery spending (again, conditional on users having decided to use the card). How would you proceed? What analyses technique should you use? What is your conclusion?

Topic

Five multiple choice practice questions

Lecture

Module 4, all topics

Purpose

Test your knowledge about the subjects of this module. Let's do it!