Step-by-step approach
Insights from School Rankings Data
Know more about your data with Power BI
This article is going to be based on a step-by-step approach to extracting insights from your data that might not be “visible” to the eye from your data using Power BI. If you are new to power BI and you haven't read my last article on introduction to Power BI, click here to read it up so that you can have a good understanding of the things I will be writing about in this article.
Using Power BI, you can detect hidden insights through visualization which could have been impossible for you to see merely by going through the raw data. Let us hop right into it, shall we?
About the Data
The data being used is the World University Rankings gotten from Kaggle. This dataset basically consists of different universities around the world and their different rankings based on several features such as name, country, research score, teaching score, international outlook score, and other important features that largely determine the rankings of a university.
The Step-by-step Approach
The first thing is to click on the Power BI icon on your computer and load it. When the app is fully initialized, it will take you to this screen. Power BI is an “easy-to-use” software and that also is one of its strengths. In case you are a beginner, Power BI’s got you covered.
The next step to take is to load your data by selecting the get data tab. In the image above, we are selecting text/CSV because our data is in a CSV format.
What we are doing here is selecting CSV data from the location where it is located on my computer. The university rankings dataset consists of two CSV files which are university rankings and university scores.
This screen pops up immediately after we select the data and we have the option of loading the data the way it is or transforming it before loading it into Power BI. In this case, we would be loading the university rankings data first before transforming it so that it can be joined to university score data and we then do the transformations together
After finally loading the two CSV files, Power BI has been able to identify a relationship between the datasets and it joined them together. We can see this in the image above which shows the relationship view of the Power BI. If you are unfamiliar with the term relationship view, do well to check out my first article on Power BI introduction. Let's move in straight to the transformation of the data.
The next thing to do is to transform the data using the transform function of Power BI. It is during this step that you clean the data, convert types, fill in missing values, etc. The above image shows the transformations I applied to the data in the applied steps tab. I will be showing you how I did some of them
Over here, I discovered that the gender ratio column contained null values represented with “n/a”.I made use of the replace values function indicated with the red arrow in the power query editor to replace everywhere “n/a” appears in the data with the value of zero.
What I did here was to change the column name of the title column from “Title” to “University”.
Let us start creating our dashboard. The first visual we will be creating is a card with the percentage of international students.
A card can be created by clicking on the card object in the visualization pane as shown by the red arrow in the image above and selecting the particular field you want it to display in this scenario, we selected the percentage of international students.
I created another card with the number of students column in the fields pane using the same method I used above
I then created a text field and fill the name “University Rankings” using the text box button as indicated by the arrow in the above image. The text box allows you to enter any text of your choice. You can adjust the font size, change the font style, make it bold, and do other varieties of operations using the dialogue box that comes on the screen when the text box is selected.
The Clustered Column Chart displays vertical bars from numerous data areas (Measures) against a single Metric. It is a component in the visualization pane that allows you to create charts by selecting the desired columns. In this scenario, we wanted to visualize the number of students by location, so we select those two columns from the fields pane and then select the clustered bar column in the visualizations pane to produce the bar chart.
The clustered column chart as indicated in the arrow above is very much like the clustered bar chart. The only difference is their orientations.
I added two more visuals to the plot. The first one is a card from the universities column and I added a slicer as indicated by the arrow icon in the image above.
With the help of the slicers, a user may sort and filter a dense report to see only the data they need. The slicers, in contrast to filters, are visible as a visual on the report and allow a user to pick values while reviewing the report.
This is the final outcome of my visual. I can view the ranking of different schools, their percent of international students, number of students, and their international outlook score by selecting any number on my slicer.
If you see this report to be informative enough, do well to drop some claps. Happy Reading!
