Learning with diagrams makes it easier to understand! | Marketing and Multiple Regression Analysis – Part 1
table of contents
Learning with diagrams makes it easier to understand! | Marketing and Multiple Regression Analysis - Part 1
Essential skills for marketersPromotion AnalysisLet's try using multiple regression analysis. Over the course of five sessions, we will learn the basics of multiple regression analysis. First, we will start by using diagrams to get a rough understanding of multiple regression analysis.
Multiple regression analysis is relatively easy to understand, and is a versatile analytical method that can be used with a certain level of understanding. However, if you try to learn it on your own, you may get stuck before you can appreciate its value because you cannot understand various technical terms and formulas. Let's put aside the difficult parts and use diagrams to understand multiple regression analysis simply. We will not perform an actual analysis. It is OK if you understand the concept and analytical steps.
Free downloads of related materials
A guide to multiple regression analysis in Excel that empowers marketers
~ Understand the correlation between marketing measures and business results ~
First of all, what is multiple regression analysis?
First, let's define data analysis. Data analysis is "the technology to estimate unknown information based on a finite amount of data." To put it more simply, it is "finding new information from a lot of data and using it as a hint to what you want to know."
Multiple regression analysis is an analysis that can measure the correlation and the degree of influence when there are "multiple factors" for "one result." For example, "sales of a new product" (one result) is related to multiple factors such as "TV commercials," "online ads," and "SNS campaigns." Do the results and each factor really have an influence? If so, it is possible to estimate the degree of influence. This is why multiple regression analysis is suitable for analyzing promotions.
Analysis procedure
Next, let's look at the analysis procedure. There are three steps: choose a result, list the elements, and create combinations of results and elements.
Step 1: Choose an outcome
First, decide what you want to know from this analysis as the outcome (objective variable). In practice, this would be something you want to "increase" (or decrease) through some kind of measure. For example, when thinking of the outcome of a promotion project, it's easy to imagine "the number of product purchases." You can also set a variety of outcomes depending on the type of analysis you want to perform, such as "number of website visits," "number of members," or "number of email newsletter subscriptions."
Step 2: List candidate elements
Next, make a list of items that are likely to be related to the results as candidate factors. If you have an idea of the likely influencing factors from the start, there is no need to identify too many. On the other hand, if you don't have a clear idea, you will often get better analysis results by coming up with a wide range of candidates and narrowing them down as you analyze. When the results are "number of product purchases": The most likely factor to come to mind is "advertising costs," but there are several other possible factors that could apply, such as "development costs," "price reductions by competitors," and "campaigns by competitors." Once you have come up with candidate factors and decided on their priorities, you can move on to preparing the data.
Step 3: Create the optimal combination of candidate elements
After following these steps, you can run the analysis program for the first time. Multiple regression analysis can be done in Excel, but please understand that behind the scenes, a model (a combination of factors that explain the results) is being created by selecting and discarding candidates for the picked factors. Indicators for evaluating the created model include the coefficient of determination, t-value, and p-value. More details on these indicators will be provided later.
Next time, let's understand seven statistical terms
Multiple regression analysis. It may seem difficult to explain in words, but it's easier to understand when illustrated. Please use your own work as an example and write down the results and elements according to the analysis procedure. You will be able to get a better understanding of what multiple regression analysis is. In the next issue, we will introduce statistical terms necessary for multiple regression analysis.
Marketing and Statistics Article Summary
Seven statistical terms to know | Marketing and multiple regression analysis - Part 7
There are many terms that appear in statistics. Here are seven statistical terms that you will need to know to understand them all.
Read on to understand. 10 common mistakes in multiple regression analysis | Marketing and multiple regression analysis - Part 3
What should you do in this situation? Here are 10 common cases where multiple regression analysis tends to fail, along with solutions.
Free download of case study
A collection of on-off integrated analysis cases useful for formulating promotion strategies
We conducted an integrated analysis that took into account not only online advertising but also offline advertising to determine which advertising measures have the greatest impact on results such as sales, and compiled the findings and results into a collection of case studies.
