Learning Objectives

By the end of this tutorial, learners will be able to:

• Decide when to use Python or R instead of Excel alone.
• Use AI assistants to generate and explain Python or R scripts.
• Clean M&E datasets using reproducible code.
• Calculate indicators, achievement rates, and performance status.
• Create descriptive statistics, summary tables, and charts.
• Use AI to debug code and improve documentation.
• Validate AI-generated code before using results in a report.
• Document AI-assisted analysis transparently.

Why Use Python or R for M&E?

Excel is useful for everyday monitoring tasks, but Python and R become valuable when analysis needs to be repeated, audited, scaled, or documented. A script can show exactly how the data was cleaned, which formulas were used, which records were excluded, and how charts were created.

This makes Python and R especially useful for evaluations, donor reporting, longitudinal monitoring, survey analysis, data quality checks, and reproducible evidence workflows.

Tools Used in This Tutorial

The AI-Assisted Coding Workflow for M&E

1. Start with the Evaluation Question

Before asking AI to write code, define the evaluation or monitoring question. Good analysis starts with the question, not with the tool.

2. Prepare the Dataset and Data Dictionary

AI-generated code is only useful if the dataset is clearly structured. Before writing code, prepare the data and define what each column means.

3. Ask AI to Create a Python Analysis Script

Python is useful for cleaning datasets, calculating indicators, creating automated summaries, and exporting results. AI assistants can help write the first draft of the script.

Example Python Code

import pandas as pd
import numpy as np
df = pd.read_excel("monitoring_data.xlsx")
df.columns = (
    df.columns
    .str.strip()
    .str.lower()
    .str.replace(" ", "_")
)
missing_summary = df.isna().sum().reset_index()
missing_summary.columns = ["column", "missing_values"]
df["achievement_rate"] = np.where(
    (df["target"].isna()) | (df["target"] == 0),
    np.nan,
    df["actual"] / df["target"]
)
df["performance_status"] = np.select(
    [
        df["achievement_rate"].isna(),
        df["achievement_rate"] >= 1.0,
        df["achievement_rate"] >= 0.8
    ],
    ["No target", "Achieved", "On track"],
    default="Behind target"
)
district_summary = (
    df.groupby("district", as_index=False)
    .agg(
        total_actual=("actual", "sum"),
        total_target=("target", "sum")
    )
)
district_summary["overall_achievement_rate"] = (
    district_summary["total_actual"] / district_summary["total_target"]
)
df.to_excel("cleaned_monitoring_data.xlsx", index=False)
district_summary.to_excel("district_summary.xlsx", index=False)
missing_summary.to_excel("missing_values_summary.xlsx", index=False)

4. Ask AI to Create an R Analysis Script

R is widely used for statistics, survey analysis, and data visualization. It is especially useful when M&E teams need clear analytical scripts and publication-ready charts.

Example R Code

library(tidyverse)
library(readxl)
library(writexl)
library(janitor)
df <- read_excel("monitoring_data.xlsx") |>
  clean_names()
df_clean <- df |>
  mutate(
    achievement_rate = if_else(
      is.na(target) | target == 0,
      NA_real_,
      actual / target
    ),
    performance_status = case_when(
      is.na(achievement_rate) ~ "No target",
      achievement_rate >= 1 ~ "Achieved",
      achievement_rate >= 0.8 ~ "On track",
      TRUE ~ "Behind target"
    )
  )
district_summary <- df_clean |>
  group_by(district) |>
  summarise(
    total_actual = sum(actual, na.rm = TRUE),
    total_target = sum(target, na.rm = TRUE),
    overall_achievement_rate = total_actual / total_target,
    .groups = "drop"
  )
write_xlsx(df_clean, "cleaned_monitoring_data.xlsx")
write_xlsx(district_summary, "district_summary.xlsx")

5. Use AI to Check Data Quality

AI assistants can help create scripts for data quality review. This is useful when datasets are large or when the same checks need to be repeated every reporting period.

6. Use AI for Descriptive Statistics

Descriptive statistics help evaluators understand the distribution of results before making claims. AI can help generate scripts for counts, means, medians, minimums, maximums, and grouped summaries.

7. Use AI to Create Charts

Python and R can generate reproducible charts. This means the same code can be reused when new data arrives, reducing manual chart editing and improving consistency.

Chart Validation Checklist

• Does the chart use the correct dataset?
• Is the right indicator selected?
• Is the reporting period correct?
• Are percentages calculated correctly?
• Are missing values handled transparently?
• Does the chart title avoid unsupported conclusions?
• Is the chart readable for non-technical audiences?

8. Use AI to Automate Repeated Reporting Tasks

Python and R are useful when the same analysis must be repeated every month, quarter, or reporting cycle. AI can help create scripts that refresh outputs when new data is added.

9. Use AI to Debug Code

One of the most practical uses of AI assistants is debugging. When a script fails, learners can paste the error message and ask AI to explain what went wrong.

10. Validate AI-Generated Code Before Reporting

AI-generated code can be useful, but it can also contain incorrect assumptions. Before using the output in a report, validate the script and the results.

11. Use AI Responsibly with Sensitive M&E Data

M&E datasets may include sensitive information about beneficiaries, households, children, income, health, disability, migration status, protection risks, or exact location. AI-assisted coding must respect data protection rules.

Before Using AI, Ask:

• Does the dataset include names, phone numbers, ID numbers, addresses, or GPS points?
• Does it include sensitive demographic or protection information?
• Can the task be done with anonymized or sample data?
• Does organizational policy allow this type of AI-assisted analysis?
• Are donor, safeguarding, and consent requirements respected?
• Is the code stored in an approved environment?
• Who has access to the data, scripts, and outputs?

Practical Exercise for Learners

You are an M&E officer reviewing a monthly monitoring dataset from implementing partners. The dataset includes district, month, indicator, target, actual, partner, and beneficiary group.

Task 1: Ask AI to Plan the Analysis

Task 2: Ask AI to Generate Code

Task 3: Ask AI to Review the Code

FAQ: Python, R, and AI Assistants for M&E

Documenting AI Use in the Methodology

Learners should practice documenting when and how AI was used. This improves transparency and helps teams explain the role of AI in the analytical process.

Final Takeaway

Python, R, and AI assistants can help M&E professionals move from manual spreadsheet work to more reproducible, transparent, and automated analysis workflows. They can support data cleaning, statistical summaries, visualization, debugging, documentation, and repeated reporting tasks.

But AI-generated code is not automatically correct. The evaluator remains responsible for understanding the analysis, validating calculations, checking assumptions, protecting sensitive data, and ensuring that every conclusion is supported by evidence.

Continue Learning with EvalCommunity Academy

Explore practical resources for evaluators, M&E officers, development professionals, and humanitarian practitioners who want to use AI responsibly in real evaluation workflows.

Explore EvalCommunity Academy Courses