Welcome to the World of Observations!

Have you ever sat in a coffee shop or a school canteen and just watched how people behave? If so, you’ve already done a basic observation! In Psychology, observation is a formal research method where scientists watch and record behavior instead of asking people questions. It’s one of the best ways to see what people actually do, rather than what they say they do.

In this chapter, we will explore the different ways psychologists watch behavior and how they make sure their "people-watching" is scientific and fair.

1. The Different Types of Observation

Psychologists categorize observations based on where they happen and how the researcher involved themselves. Don't worry if these terms seem like a lot—they usually come in pairs!

A. Naturalistic vs. Controlled Observations

Naturalistic Observation: This happens in a person’s (or animal’s) everyday environment. The researcher doesn't change anything.
Example: Watching how children play on a school playground.
Strength: High ecological validity because the behavior is natural.
Weakness: Hard to control extraneous variables (like a sudden rainstorm distracting the kids).

Controlled Observation: This happens in a set-up environment, like a lab. The researcher might arrange the room in a specific way.
Example: Bandura’s study on aggression, where children were placed in a specific room with a Bobo doll.
Strength: Easy to replicate (repeat) because the setup is the same every time.
Weakness: The environment might feel "fake," so participants might not act naturally.

B. Overt vs. Covert Observations

Overt Observation: The participants know they are being watched. The researcher is clearly visible.
Weakness: This can lead to demand characteristics—participants might "show off" or act differently because they know they are being judged.

Covert Observation: This is a "secret" observation. The participants do not know they are being watched.
Strength: No demand characteristics; behavior is 100% genuine.
Weakness: This raises ethical issues regarding informed consent and privacy.

C. Participant vs. Non-Participant Observations

Participant Observation: The researcher joins in! They become part of the group they are studying.
Strength: The researcher gets a "first-hand" deep understanding of the behavior.
Weakness: The researcher might lose objectivity (become biased) because they become friends with the participants.

Non-Participant Observation: The researcher stands back and watches from a distance without joining in.
Strength: The researcher remains objective and detached.
Weakness: They might miss out on the "meaning" behind certain behaviors because they aren't involved.

2. Structuring the Observation

How do we turn "watching" into "data"? We have two main choices:

Unstructured Observations: The researcher writes down everything they see.
Best for: New topics where you aren't sure what to look for.
Data type: Mostly qualitative data (rich descriptions).

Structured Observations: The researcher uses a pre-made list called a behavioral checklist or coding scheme.
Example: Instead of writing "the child was mean," the checklist has categories like "Pushed another child," "Shouted," or "Took a toy." The researcher just ticks a box every time it happens.
Data type: Mostly quantitative data (numbers and tallies).

Quick Tip: Operational Definitions

In structured observations, behaviors must be operationalized. This means they are defined so clearly that two different people watching the same thing would agree on what happened. For example, "Aggression" is too vague; "Hitting with a closed fist" is an operational definition.

3. Reliability and Validity in Observations

To make sure an observation is "good science," we look at two main things:

Inter-Observer Reliability

If only one person is watching, they might be biased or miss something. To fix this, psychologists use inter-observer reliability. This is when two or more researchers watch the same behavior using the same checklist. If their scores are very similar (e.g., they both counted 10 hits), the observation is reliable.

Validity

Validity is about whether we are measuring "real" behavior.
1. Objectivity: Recording what is actually there, not what you think is there (subjectivity).
2. Ecological Validity: Is the setting natural?
3. Demand Characteristics: Did the participants change their behavior because they saw the observer?

4. Data Analysis for Observations

Once you have your numbers (quantitative data) from a structured observation, you need to summarize them. In Psychology 9990, you need to know:

  • Measures of Central Tendency:
    - Mean: The average score.
    - Median: The middle score in a list.
    - Mode: The most common behavior seen.
  • Measures of Spread:
    - Range: The difference between the highest and lowest score \(Range = highest - lowest\).
    - Standard Deviation: A measure of how much the scores spread out around the mean.

Visualizing Data: You might draw a bar chart to show how many times different behaviors occurred, or a histogram if the data is continuous.

5. Planning an Observation (Paper 2 Skills)

When you are asked to plan an observation in the exam, remember these key features:

1. Type of Observation: Decide if it’s overt/covert, participant/non-participant, etc.
2. Location: Where is it happening (natural or controlled)?
3. Behavioral Categories: What exactly are you looking for? (Provide examples of operationalized behaviors).
4. Sampling of behavior:
- Event Sampling: Every time the behavior happens, you record it.
- Time Sampling: You only record behavior at specific time intervals (e.g., every 30 seconds).
5. Data Type: Will you collect numbers (quantitative) or words (qualitative)?

Common Mistake to Avoid:

In Paper 2 planning questions for observations, you do not need to provide an aim or a hypothesis unless specifically asked. Focus on the procedure—how exactly will you carry out the study?

Summary: Key Takeaways

  • Naturalistic = Real world; Controlled = Lab/Set-up.
  • Covert = Hidden; Overt = Visible.
  • Participant = Joining in; Non-participant = Watching from the side.
  • Structured = Using a checklist; Unstructured = Taking notes on everything.
  • Inter-observer reliability is essential to prove that the recording was accurate and not just one person's opinion.

Did you know? The Piliavin et al. (subway Samaritans) study used non-participant, covert, structured observation to see how people helped a victim on a train. Because it was covert and naturalistic, it had very high ecological validity!