Forem: Nashipae Ntung'ani

Schemas and Data Modelling in Power Bi

Nashipae Ntung'ani — Mon, 02 Feb 2026 11:31:20 +0000

POWER BI

It is a tool that is mainly aimed to help organizations and individuals to visualize and organize data

Data modelling

Data modelling is a foundational concept in Power BI that determines how data is structured, related, and analyzed within a report. A well-designed data model improves performance, ensures accurate calculations, and makes reports easier to understand and maintain.

In Power BI, the way you structure your data known as Data Modeling is the single most important factor in determining whether your reports are fast, accurate, and easy to use. While Power Query handles the cleaning, the Data Model (Model View) is where you define how different tables "talk" to each other to create a cohesive story.

1. The Building Blocks: Fact vs. Dimension Tables
Before choosing a schema, you must distinguish between your two types of tables:

Fact Tables (The "What"): These contain the quantitative data or "metrics" you want to analyze. They are usually long, containing thousands or millions of rows of transactions.

Example: In the Kenya crops dataset, the Yield or Sales table is the fact table.

Dimension Tables (The "Who, Where, When"): these provide context to your facts. They contain unique lists of attributes used for filtering and grouping.

Example: the County table (Kenya Crop dataset).

2. Choosing a Schema: Star vs. Snowflake
A "schema" is the visual layout of your tables and their relationships.

*The Star Schema *

-In a Star Schema, the central fact table is connected directly to several dimension tables. Visually, it looks like a star.

Power BI is "engine-optimized" for this structure. It simplifies your DAX calculations and provides the fastest query performance because there is only one "hop" from a filter (Dimension) to a result (Fact).
Always aim for a Star Schema. If your data is "flat" (one big table), use Power Query to break it into one Fact and multiple Dimensions.

The Snowflake Schema
A Snowflake Schema is an extension where dimension tables are "normalized" into further sub-dimension tables.

Example: Instead of one Product table, you have a Product table that links to a Category table, which then links to a Department table.

The Downside: While it saves storage by reducing duplicate text, it creates a "chain" of relationships. This makes your DAX more complex and can slow down your reports as Power BI has to "jump" through multiple tables to get a result.

3. Relationships and Cardinality
Relationships are the paths through which filters flow. In Power BI, we primarily deal with One-to-Many relationships.

One to Many: One patient in the Patients (Dimension) table can have many visits in the Treatments (Fact) table.
Directionality: Always try to keep your cross-filter direction as Single. This ensures that filters flow from the Dimension to the Fact, preventing "circular dependency" errors and keeping performance high.

Reference:
file:///C:/Users/USER/Downloads/Introduction%20to%20Power%20BI,%20Power%20Query%20and%20Dax%20part%201.pdf

Beginner's Guide to Basic Data Analysis in Microsoft Excel

Nashipae Ntung'ani — Sun, 25 Jan 2026 13:45:29 +0000

Introduction

MS Excel is one of the ,most widely used micro soft tools for recording and analyzing information. Whether a person is managing personal finances, tracking school marks, or handling business report, Excel provides a simple way to turn a list of numbers to meaningful insights

The Interface

Before analyzing data, you need to know where things are. The excel screen is a grid of columns(labeled as letter A,B,C)and rows (labeled with numbers 1,2,3).

Cell: The intersection of rows and a column (eg cell A1)
Ribbon:The tool bar at the top where you find all your tools like home, Insert and data
Formula Bar: Located above the grid, this shows the exact text or calculation inside the exact text or calculated inside the cell one ahs selected

Cleaning Data

Excel has several built-in tools to help fix errors:

Removing duplicates: If you have the same information repeated by mistake, go to the Data tab and click remove duplicates
Text cleaning: You use simple formulas to fix messy text:

=PROPER(A2): capitalizes the first letter of every word eg names
=TRIM(A2):Remove extra spaces from beginning or end of your text
3.Conditional Formatting
This tool automatically highlights cell based on rules.
Highlight Cell Rules (Greater Than 80,000,Less Than 80,000)

Organizing with Sorting and Filtering

Once your data is clea, you can organize it to see patterns.

Sorting: This changes the order of your list. for example, one can sort a list of employee salary from highest to lowest to see who the highest earners are. Freeze Top Row:
View tab > Freeze Panes.
Choose Freeze Top Row.
Types of sorting:
- Text: A to Z or Z to A
- Numbers: Smallest to Largest or Largest to Smallest
- Dates: Oldest to Newest or Newest to Oldest
Filtering: This hides data you don't need.

An example: If you only want to see employees from the "IT" department, you can click the filter button on the data tab and uncheck everthing except "IT"
Freezing panes: Keeps your header row visible while you scroll through thousands of row data. To freeze Top Row:

-View tab > Freeze Panes.

-Choose Freeze Top Row.

One can choose to freeze the top column or custom freeze

Basic calculation (Functions)

Calculations are the heart of data analysis.Ever formula in excel must start with an equal sign =. Here are the most common functions used to summarize data:
*Aggregate Function
*

SUM: Adds up all the numbers in a range(e.g., =SUM(E2:E10)adds all salaries in those cells

Average: Find the middle value of a group of numbers. Example:
Average salary of all employees:
=AVERAGE(E2:E877)
MIN/MAX: Find the smallest or the largest number in your list. =MIN(E2:E877) to look for the minimum salary.
=MAX(E2:E877) to look for maximum salary.
Count: Tells you how many cells actually contain a number.
Count number of employees with a numeric salary:
=COUNT(E2:E877)

Conditional Math

These are advanced beginner tools that only calculate if a rule is met
SUMIFS example
Total Salary for Full-Time IT employees
SUMIF(D2:D876,"IT"
COUNTIFS :Count employees who re both full-time and in IT

Advanced Analysis and Logic

logic(IF/OR/AND)- These functions act like a brain for your spreadsheet. For example,=IF(score>7,"pass","fail" automatically grades your data
Lookups (VLOOKUP/XLOOKUP/HLOOKUP)- These act like a digital "search" tool. If you have an employee ID, you can use VLOOKUP to instantly find that person's name or salary from a massive list
Date and Time: Tools like DateIf allows you to calculate exactly how many years an employee has been with the company or how many working days a task took.

PIVOT TABLES

PivotTables are one of the most powerful features in Excel. They allow you to summarize, analyze, and explore large datasets like HR data without writing formulas. It summarizes large amounts of data by grouping, counting, summing, or averaging values.

Example: Total Salary by Department Question
What is the total salary cost per department?
Steps
• Rows → Department
• Values → Salary
Excel defaults to SUM.
Result
Total payroll cost per department.

*Pivot chart
*

Visualizing Data with Charts

The final step is to turn your numbers into a picture so that they are easier to understand.
A chart is a visual representation of data that helps users quickly see comparisons, trends, distributions and proportions.

Column charts: Best for comparing different categories like "total sales by department"
Pie chart- Best for showing parts for a whole, such as the "Gender distribution" in an office.

-** Line chart**- Best for showing trends

A BEGINNER'S GUIDE TO GIT

Nashipae Ntung'ani — Sun, 18 Jan 2026 08:59:17 +0000

Introduction

Git is a free, open source version control system that developers and analyst use for tracking changes,collaborating and managing project history efficiently, allowing them to work offline and merge their work seamlessly.

Getting started

Before getting started with Git you’ll need to configure by setting your email and user name.
This is important for version control system as each Git commit uses this information
An example
git config --global user name "nashipae"
git config --global user. email "nashipaentungani@gmail.com

The Git work plan

The life cycle of file has 3 stages. They include:

Your working directory - This is where you are actively working on the file or editing
Staging area-This is a loading dock for changes you want to save (almost final)
The repository-This is a permanent storage area for your project
An example, GitHub, GitLab

Essential commands

Starting a project
There are 2 ways to configure this

Initialize a new project

git init

Clone an existing project git clone (github project)

Checking status and tracking changes

Git allows you to check the status of a file

git status-see what has been changed in a file
An example

git status 
on branch master 
no commits yet`
changes to be committed:`
use `git restore --staged ..."to unstage
`
new file: index.html

git diff- see detailed changes made to a file
git add (file name)-adding specific file to the staging area
git add.- adding all changes to your staging area

Saving your work

To save changes on Git we use the command
git commit -m "commit message"
To view history on commits for repository you can use the git log command

Pushing your code

To create a remote connection to github we use the command

git remote add origin "Github project"
To push to a main branch, we use the command
git push origin main
To push for the first time
git push -u origin main

Pulling code

git pull origin main

See what has changed

Version control deals with understanding what changes have been made to a file
For this you have to check the history of your commit
git log- see commit history
git log --oneline see compacted history
git blame(file name)find out who made changes
git diff (commit 1) (commit 2)- compare between commit
If you made a mistake during staging you need to go back to your previous changes, we use the command

git rese HEAD*
We have 3 examples of this
git reset --soft (HEAD ~1)
git reset --hard (HEAD ~1)
git reset --mixed(HEAD ~1)

Forem: Nashipae Ntung'ani

Schemas and Data Modelling in Power Bi

POWER BI

Data modelling

Beginner's Guide to Basic Data Analysis in Microsoft Excel

Introduction

The Interface

Cleaning Data

Organizing with Sorting and Filtering

Basic calculation (Functions)

Conditional Math

Advanced Analysis and Logic

PIVOT TABLES

Visualizing Data with Charts

A BEGINNER'S GUIDE TO GIT

Introduction

Getting started

The Git work plan

Essential commands

Checking status and tracking changes

Saving your work

Pushing your code

Pulling code

See what has changed