Learn how to perform regression analysis using Tableau and how to interpret the results.
- [Instructor] Let's continue to build on what we've learned so far, now with Tableau. I've opened up our application and connected to our data. If you need help on how to connect to your data, feel free to open up the video on exploratory analysis in Tableau. And I've connected specifically to the exercise files 03_04 and the .csv file that you will find in that directory. Okay, great. We will navigate into our workspace by clicking on the tab that reads Sheet 1. And again, similar to how we have seen in our two previous videos, we're going to plot our data.
So we're going to grab our broadcast measure and we're going to drop that onto our Column shelf. Which you can think of the Column shelf as the same as your X axis. So Broadcast, Columns, and just dropping that right there. And now let's grab our Net Sales and drop that onto our Rows shelf, which you can think of as the same as our Y axis. So let's get out Net Sales and bring that over. Now there's a little trick at this stage of the game where you're going to want to turn off your aggregate measures.
So this will ensure that we can see each data point on our plot. So the way I do that is I come up to Analysis and I deselect Aggregate Measures. All right. So instantly we have our plot. And now we can add our trend line, or what we've been calling our best fit line. And there are two ways that we can do this. One is that Analysis bar again. So if we click on Analysis and scroll down, we have our Trend Lines.
So we can select that there, or directly from the plot, I can do a right click and from here I can select Trend Lines and Show Trend Lines. Now if we hover over this line, we'll see a few different values on the tooltip. And sure enough, there's our good old R-Squared. And there's a P-value as well, which is another new concept for this course. So lets take a moment to discuss it. A P-value is the calculated probability of obtaining a result.
Generally, the lower the better. So in other words, the lower this value is, means the more we can infer that there's a correlation between our independent and our dependent variables. The accepted rule of thumb in statistics is that if the P-value is less than .05, then there is a strong correlation. And if it's greater than that, it cannot be determined that there's a correlation. So that's a big concept, but an important one. So I'll repeat that one more time.
Less than .05 means that there may be a correlation. Greater than .05, no correlation. And the way we can interpret the P-value for our particular analysis here is that there is indeed a correlation right here in our tooltip, with a value of less than .0001. And we're seeing the same R-Squared value that we saw from the previous chapter. And you can explore what other sorts of relationships might exist in your data.
So we can bring in, for example... Let's grab Out-Of-Home, and we're going to drop that in our Columns shelf. Let's do a few others too. Let's grab Print and let's do Social-Media-Volum. So instantaneously you can see the regression line for each of these independent variables. Each of these different marketing channels. So we can begin to hover over here and you can begin to see what some of these outputs look like for these different channels too.
So feel free to spend some time exploring what these different values are. And I'd like to do a little bit of a recap. We, at this stage of the game, over these last three videos, we've covered a lot of ground. We performed a regression analysis in each of our three platforms. We discussed what independent and dependent variables are, and the fact that they are synonymous with the X and the Y on our scatter plots. We discussed coefficience, which is a statistical measure that can be used as a multiplicative property in calculating impact on our dependent variables.
We discussed R-Squared, a statistical measure that indicates how strongly our variables are correlated. And we did the same for P-value. There are some big fundamentals there. And that winds down our chapter on regression analysis so we can move onto discussing predictive analytics.
In this course, discover how to gain valuable insights from large data sets using specific languages and tools. Follow Chris DallaVilla as he walks through how to use R, Python, and Tableau to perform data modeling and assess performance. As Chris dives into these concepts, he shares specific case studies that come directly from his own work with clients. Plus, he shares three essential—and practical—best practices for data-driven marketing that you can use to bolster your organization's marketing performance.
- Installing R, Python, and Tableau
- Navigating the UI for R, Python, and Tableau
- Using R, Python, and Tableau
- Exploratory analysis
- Performing regression analysis
- Performing a cluster analysis
- Performing a conjoint assessment
- Stakeholder alignment