Datum MEL & Data Analytics Consulting

03/05/2026

✅ Linear Regression Basics 📈🤖

👉 This is the most important and beginner-friendly algorithm in Machine Learning.

🔹 1. What is Linear Regression?

Linear Regression is used to predict a continuous value.

👉 Example:
✔ Predict salary
✔ Predict house price
✔ Predict sales

🔥 2. Basic Idea

👉 It finds a straight line that best fits the data.

Equation:
y = mx + c
Where:
✔ y → Output (target)
✔ x → Input (feature)
✔ m → Slope
✔ c → Intercept

🔹 3. Example

👉 Predict Salary based on Experience

Experience Salary
1 year 20k
2 years 30k
3 years 40k

👉 Model learns pattern → predicts future salary.

🔹 4. Simple Implementation (Python)
from sklearn.linear_model import LinearRegression

# Sample data
X = [[1], [2], [3]]
y = [20000, 30000, 40000]

model = LinearRegression()
model.fit(X, y)

# Prediction
print(model.predict([[4]]))

👉 Output: ∼50000 (approx)

🔹 5. Important Terms ⭐

✔ Feature (X) → Input
✔ Target (y) → Output
✔ Model → Learns relationship
✔ Prediction → Output from model

🔹 6. Assumptions of Linear Regression

✔ Linear relationship
✔ No extreme outliers
✔ Independent features

🔹 7. Why Linear Regression is Important?

✔ Easy to understand
✔ Used in real-world predictions
✔ Foundation for advanced ML

🎯 Today’s Goal

✔ Understand regression concept
✔ Learn equation (y = mx + c)
✔ Implement simple model

👉 Linear Regression = First step into ML modeling 🚀

💬 Tap ❤️ for more!

03/07/2023

30/03/2022

```Unless you understand your role in society, you can never make any meaningful contribution.

According to the Greeks there are three types of people on earth,
- the idiots,
- the tribesmen, and
- the citizens.

Studies show only 10% of Africans are citizens.
The remaining 90% are either tribesmen or idiots.

2. When the Greeks used the word *"idiot"*, they did not use it as a curse word.
Idiots are people who just don’t care.
If they write exams, they will cheat.
If they are in government they will steal.
An idiot does not care at all, if he eats bananas he throws the peels anywhere instead of putting them in a trash bin.

According to the Greeks, some societies have more idiots than tribesmen and citizens.

3. The next set of people are tribesmen, these are people that look at everything from the point of view of their tribe.

These are people that believe in you only if you are part of their tribe.
It can be terrible to have a tribesman as a leader, he will alienate the rest.

When the Greeks talk about tribes, it’s not just about ethnicity, they also consider religion as a tribe.
A great percentage of Africans are tribesmen, because they view everything from the point of view of their tribes.
They trust only their tribesmen.

4. The last group are citizens.
These are people that like to do things the right way.
They will respect traffic light rules even if no one is watching them.
They drive within speed limits.
They respect the laws, won’t cheat in exams.
In government they won’t steal.
They are compassionate and give to others to promote their wellbeing.

Citizens often promote projects that benefit everyone.
The Greeks called this group the citizens.

Some countries have more citizens than tribesmen and idiots.
Others have so many idiots.

A tribesman can become a citizen through orientation.

And an idiot can become a citizen by training and constant enforcement of the law.

11/03/2021

Baseline Data
A piece of baseline information is a foundation for M&E; without baseline Data, M&E is not effective.

What is baseline Data?
Baseline information describes the socioeconomic situation/status of a facility/ community/society prior to project implementation.

Why baseline information?
Without baseline information, you can't know whether your project made any changes in the lives of the beneficiaries.
For example: if you are creat awareness on Coronavirus in Community A, before the awareness, you must collect baseline Data to understand the percentage (%) of people who are aware of the virus. If the percentage of those aware is 10% for example, that is your baseline.
During or after the implementation of the Coronavirus awareness, whatever percentage (%) you achieved can be compared with 10% to determine progress made.

Example:
If Baseline=10% and achievement of the implemented project is 70%
We can therefore say, 70%(success) -10%(baseline) to understand the progress by the awareness which is 60%.
Now we got 60% progress with the help of the baseline 10% and our success 70%
We couldn't get 60% progress if we didn't collect our baseline (10%)

10/01/2021

to Build and M&E system

Kusek and Rist [2004] suggest that building an M&E system that responds to the results in an organization’s strategy is a Ten-step process, as summarized below:-

One: Conducting a Readiness Assessment - To determine the capacity and willingness of the government/organization and its development partners to construct a results-based M&E system. This assessment addresses such issues as the presence or absence of champions, the barriers to building a system, who will own it, and who will oppose the M&E system.

Two: Agreeing on Outcomes to Monitor and Evaluate - addresses the key requirement of developing strategic outcomes that then focus and drive resource allocation and activities. These outcomes should be derived from strategic priorities (goals).

Three: Developing Key Indicators to Monitor Outcomes - so the degree to which the outcomes are being achieved can be assessed. Developing indicators is a core activity in building an M&E system and drives all subsequent data collection, analysis, and reporting. Both the political and methodological issues in creating credible and appropriate indicators are not to be underestimated.

Four: Gathering Baseline Data on Indicators - involves describing and measuring the initial conditions being addressed by the outcomes. It is the first measurement of the indicators and defines the starting point.

Five: Planning for Improvements - requires setting realistic targets and recognizes that most outcomes are long-term, complex, and not quickly achieved. It is helpful to establish interim targets that specify how much progress towards an outcome is to be achieved each year (or another time period) and the resources needed. Measuring results against targets can involve both direct and proxy indicators and use of both quantitative and qualitative data.

Six: Monitoring for Results - is the administrative and institutional task of establishing data collection, analysis, and reporting guidelines; designating who will be responsible for activities; establishing quality control processes; establishing timelines and costs; working through roles and responsibilities, and establishing guidelines on transparency and dissemination of the information and analysis. It is emphasized that, in constructing an M&E system, the challenges of ownership, management, maintenance, and credibility need to be addressed clearly.

Seven: Evaluative Information to Support Decision Making - focuses on the contributions that evaluation studies and analyses can make throughout this process to assess results and move towards outcomes. Analysis of program theory, evaluability assessments, process evaluations, outcome and impact evaluations, and evaluation syntheses are among the strategies discussed that can be employed in evaluating a results-based M&E system.

*Step Eight: Analyzing and Reporting Findings - is a crucial step, as it determines what findings are reported to whom, in what format, and at what intervals. This step has to address the existing capacity for producing the information, and focuses on the methodologies for accumulating and assessing information, and preparing analyses and reports.

Nine: Using the Findings - emphasizes that the crux of the system is not simply generating results-based information, but getting the information to the appropriate users in a timely fashion so that they can take the information into account in making decisions. This step also addresses the roles of development partners and civil society in using the information to strengthen accountability, transparency, and resource allocation procedures.

Ten: Sustaining the M&E System - recognizes the long-term process involved in ensuring the longevity and utility of an M&E system. Six criteria are seen as crucial to the sustainability of an M&E system: demand, structure, trustworthy and credible information, accountability, incentives, and capacity. Each dimension needs constant attention overtime to ensure the viability of the system.

As noted earlier, building an M&E system does not have to be done according to these 10 steps. One could define a more detailed number of steps or fewer. The issue is to ensure that key strategies and activities are recognized, clustered together in a logical manner, and then completed in an appropriate sequence.

What does a sustainable results-based M&E system look like? How can we make and keep such a system functioning? To answer these questions, it is necessary to understand the components of a results-based M&E system and then to build each component.

31/12/2020

Pillars of Monitoring and Evaluation (M&E) Practice

In this piece, highlighted tips that are capable of transforming the career of an M&E practitioner, irrespective of the levels in which you are operating.

1. Management & Analysis:* Because the job generates a lot of routine program activities and implementation research data, you need to have data management skills especially large databases. You should know how to use MS Excel package at an advanced level for creating pivots and interactive dashboards. Get familiar with Statistical packages like SPSS, Stata, SAS, R, etc and know how to use them in analyzing survey data. What can you start doing immediately, identify one statistical package in addition to MS Excel that you want to master at an expert level? Download PDFs files and YouTube videos and learn it for the next 2 months. Also plan to take basic statistical concept courses, it will help you a great deal.

2. Design:* The Evaluation part of M&E is majorly about research. Ranging from baseline to Mid-term to end-line surveys. That means you must understand how to design both quantitative and qualitative research. What can you do? Simple, take short online courses both free and paid from Coursera, EdX, Udemy, and the University of Washington. You can even purchase a textbook like the types written by Prof. E.A. Bamgboye.

3. Writing & Publication:* M&E practitioners have a lot in common with academics because they know how to develop concept notes, write research protocols, and draft conference abstracts. That’s why you meet a lot of them at various conferences which are most times fully funded, all because they have mastered the art of writing scientific articles from their routine program or evaluation data. Also, you should learn how to develop bulletins and factsheets because they are means of disseminating project achievements. There are free YouTube videos that can help you with this.

4. Presentation:* M&E practice allows one to make presentations on program research findings and facilitate training. That means you need to develop expertise in the use of MS PowerPoint for presentations embellished with info-graphics.

5. Management:* The M&E work presents one with a lot of leadership opportunities because you will be supervising research assistants/data entry clerks and also working with other project staff. So you need to have leadership and coordination skills to be able to make the best of your career. You must develop strong communication skills and be able to motivate your subordinates to deliver on targets.

In , networking, branding, and packaging are key to a successful M&E career. “If you are good and people don’t know that you are good, then you are not good”. Make a good first impression, let people have a perception of competence about you. Save up and attend high premium training at least once a year. When you attend such training, the profile of such highbrow organizations will rub off on your credentials. You will meet people who are movers and shakers of the sector since it is for high premium individuals. You will have the chance to rub minds with high caliber professionals because there will likely be group work sessions that will bring you guys together. Your name will begin to ring bells in high places.

Thanks for taking the time to read the piece.

29/06/2020

Methods that can be used to collect information for monitoring and evaluation purposes

You need to select methods that suit your purposes and your resources. Do not plan to do a comprehensive survey of 100 000 households if you have two weeks and very little money! Use sampling in this case.

: is important concept when using various tools for a monitoring or evaluation process. Sampling is not really a tool in itself, but used with other tools it is very useful. Sampling answers the question: Who do we survey, interview, include in a focus group etc? It is a way of narrowing down the number of possible respondents to make it manageable and affordable. Sometimes it is necessary to be comprehensive. This means getting to every possible household, or school or teacher or clinic etc. In an evaluation, you might well use all the information collected in every case during the monitoring process in an overall analysis. Usually, however, unless numbers are very small, for in-depth exploration you will use a sample. Sampling techniques include:
sampling (In theory random sampling means doing the sampling on a sort of lottery basis where, for example all the names go into a container, are tumbled around and then the required number are drawn out. This sort of random sampling is very difficult to use in the kind of work we are talking about. For practical purposes you are more likely to, for example, select every seventh household or every third person on the list. The idea is that there is no bias in the selection.);
sampling (e.g. every seventh household in the upper income bracket, every third household in the lower income bracket);
sampling (e.g. only those people who have been on the project for at least two years).

25/06/2020

Core Humanitarian Standard on Quality and Accountability

The Core Humanitarian Standard on Quality and Accountability #(CHS) sets out Commitments that organizations and individuals involved in humanitarian response can use to improve the quality and effectiveness of the assistance they provide. It also facilitates greater accountability to communities and people affected by crisis: knowing what humanitarian organizations have committed to will enable them to hold those organizations to account.

The CHS places communities and people affected by crisis at the centre of humanitarian action and promotes respect for their fundamental human rights. CHS describes the essential elements of principled, accountable and high-quality humanitarian action. Humanitarian organizations may use it as a voluntary code with which to align their own internal procedures. It can also be used as a basis for verification of performance, for which a specific framework and associated indicators have been developed to ensure relevance to different contexts and types of organization.