Data collection

Habari Mwanafunzi! Welcome to the Heart of Research!

Ever argued with your friends about which spot in town has the best nyama choma? How would you *prove* it? Would you ask people? Would you count how many people visit each spot? Or would you read online reviews? Congratulations, you've just been thinking like a researcher! Today, we're diving into the exciting world of Data Collection, the engine that powers every research project. It’s how we move from a simple question to a solid answer.

Think of yourself as a detective. Your research question is the mystery, and data collection is you gathering clues to solve it. Let's get our detective gear on!

First Things First: What Kind of Clues Are We Looking For?

Before you start collecting, you need to know what you're looking for. In research, our "clues" or data come in two main forms: Primary and Secondary.

        Your Research Project
                |
     +----------+-----------+
     |                      |
     V                      V
  Primary Data           Secondary Data
(You collect it          (Someone else
 yourself, fresh!)         collected it)
     |                      |
   Surveys,               Books, Journals,
   Interviews,            KNBS Reports,
   Observations           Company Data

• Primary Data: This is the data you collect yourself, for your specific research purpose. It’s fresh from the source! Think of it as cooking a meal from scratch with ingredients you picked yourself.
• Secondary Data: This is data that has already been collected by someone else. It's like using a pre-made sauce for your cooking – it saves time, but might not be exactly what you wanted. A great Kenyan example is using data from the Kenya National Bureau of Statistics (KNBS) for a project on population trends.

The Toolkit: Your Primary Data Collection Methods

When you decide to collect primary data, you have a fantastic toolkit of methods. The one you choose depends on your research question. Let's open the toolbox!

Scenario: Let's imagine our research project is: "To understand the transportation challenges faced by university students in Nairobi." We'll use this to explore the methods.

1. Questionnaires & Surveys

These are lists of questions you give to people to answer. They can be on paper, or online (like a Google Form). They are excellent for getting information from many people quickly.

• Strengths: You can reach a large number of people (e.g., hundreds of students) easily. They are often anonymous, so people might be more honest.
• Weaknesses: People might not return them (low response rate), or they might not understand a question and there's no one to ask.
• Our Scenario: We could create a survey asking students about their daily commute time, cost, preferred mode of transport (matatu, bus, walking), and to rate their satisfaction on a scale of 1-5.

Image Suggestion: An engaging, brightly lit photo of a diverse group of young Kenyan researchers in a university common area. One is holding a tablet showing a survey form, while others are engaging with fellow students who are filling out paper questionnaires. The atmosphere is collaborative and positive.

2. Interviews

This is where you have a conversation with someone to get in-depth information. It’s more personal than a survey.

• Structured: You ask every person the exact same questions in the same order. It's like a verbal survey.
• Semi-Structured: You have a guide with key questions, but you can ask follow-up questions to explore interesting answers. This is very popular!
• Unstructured: It's more like a free-flowing conversation about a topic.
• Our Scenario: We could conduct semi-structured interviews with 10-15 students to hear their personal stories about their commute. We could ask things like, "Tell me about your journey to campus yesterday," and "What is the most stressful part of your commute?" This gives us rich, detailed stories (qualitative data!).

3. Observations

Sometimes, the best way to learn is to watch! Observation is systematically watching and recording behaviour in a natural setting.

• Strengths: You see what people *actually do*, not just what they *say* they do.
• Weaknesses: The presence of a researcher can sometimes change people's behaviour (the Hawthorne effect). It can also be time-consuming.
• Our Scenario: We could go to a major matatu terminus near the university (like Archives or Kencom) during rush hour and observe how students queue, how long they wait, and how crowded the matatus are.

4. Focus Group Discussions (FGDs)

A Focus Group is a guided discussion with a small group of people (usually 6-10). It’s great for exploring shared experiences and capturing group dynamics.

• Strengths: The group setting can spark ideas and conversations that wouldn't happen in a one-on-one interview.
• Weaknesses: Sometimes one or two loud people can dominate the conversation.
• Our Scenario: We could organize an FGD with a group of students who all use Route 125. We can ask them to discuss their shared experiences, safety concerns, and ideas for improvement.

Image Suggestion: A medium shot of a researcher facilitating a Focus Group Discussion. A diverse group of 8 Kenyan university students are seated in a circle, engaged in a lively but respectful discussion. The setting is a simple classroom. The mood is thoughtful and communicative.

Let's Talk Numbers: Sampling

It's impossible to survey every single student in Nairobi! So, we take a sample – a smaller group that represents the larger population. But how big should your sample be? There are many complex formulas, but a simple one to start with is Slovin's Formula.

It helps you calculate a sample size (n) from a known population size (N) with a certain margin of error (e), usually 5% (or 0.05).

Formula:
    n = N / (1 + N * e^2)

Where:
    n = Sample size
    N = Total population size
    e = Margin of error (e.g., 0.05)

Example Calculation:
Let's say the total population of university students in Nairobi is 50,000 (N).
We want a 5% margin of error (e = 0.05).

Step 1: Calculate e^2
    e^2 = 0.05 * 0.05 = 0.0025

Step 2: Calculate N * e^2
    50,000 * 0.0025 = 125

Step 3: Add 1 to the result from Step 2
    1 + 125 = 126

Step 4: Divide the total population (N) by the result from Step 3
    n = 50,000 / 126
    n ≈ 396.8

Step 5: Round up to the nearest whole number.
    n = 397

So, you would need to survey about 397 students to represent the whole population with a 5% margin of error.

Putting It All Together: Choosing Your Method

So, which method is best? The answer is always: It depends on your research question!

• If you want to know "how many" students use matatus, a survey (quantitative) is perfect.
• If you want to know "why" students find matatus stressful, interviews or FGDs (qualitative) are your best bet.

Often, the best research uses a mix of methods! This is called mixed-methods research. You could start with a survey to get the numbers, then conduct interviews to understand the stories behind those numbers.

   Research Question: "What are the transport challenges for students?"
            |
            +---------------------------------+
            |                                 |
            V                                 V
   Quantitative Approach                  Qualitative Approach
  (The "What" & "How Many")             (The "Why" & "How")
            |                                 |
      SURVEY                               INTERVIEWS / FGDs
      - Frequencies                         - Rich narratives
      - Percentages                         - Personal stories
      - Averages                            - Deep understanding

You've done it! You now have the foundational knowledge to be a great data detective. Remember, data collection is all about choosing the right tool for the job to help you uncover the truth. Go forth and gather those clues!

Sasa wewe ni mtafiti! (Now you are a researcher!)