Psychology (9685) | Research methods 1

Research Methods 1: Your Guide to Becoming a Psychological Investigator

Welcome to Research Methods! This might feel like a separate subject, but understanding how psychologists gather evidence is the single most important skill you'll learn. Everything you study—from memory models to theories of conformity—rests on the foundation of solid research.

Don't worry if terms like 'variables' or 'distributions' seem daunting. We will break them down into simple, manageable steps. By the end of this chapter, you’ll not only know the terminology but understand why these methods are used to answer big questions about human behaviour.

Key Takeaway from the Introduction:

Research methods are the 'tools' psychologists use. If you understand the tools, you can evaluate the quality of the findings!

---

1. Core Research Methods: The Ways We Investigate

Psychology uses four main approaches to gather data. The method chosen depends entirely on the question the researcher is trying to answer.

1.1 The Experimental Method

The gold standard for determining cause and effect. The researcher manipulates one variable and measures the effect on another, while controlling all other factors.

Types of Experiments

• Laboratory Experiment:
  

  Conducted in a highly controlled environment (e.g., a lab or classroom specifically set up for the task). This allows the researcher to control extraneous variables effectively.

  Advantage: High control = high internal validity (we are sure the IV caused the DV).
  Disadvantage: Artificial setting = low ecological validity (results might not reflect real life).

• Field Experiment:
  

  Conducted in a natural, everyday setting (e.g., a park, street, or school). The IV is still manipulated by the researcher.

  Advantage: Natural setting = higher ecological validity.
  Disadvantage: Less control = increased risk of extraneous variables messing up the results.

• Natural Experiment:
  

  The Independent Variable (IV) occurs naturally; the researcher simply records the effect on the Dependent Variable (DV). The researcher does not manipulate the IV (it happens anyway, e.g., a natural disaster, a law change, or having a biological condition like being a split-brain patient).

  Advantage: Allows investigation of variables that would be unethical or impractical to manipulate.
  Disadvantage: Lack of control over who is studied and the conditions.

1.2 Observation Techniques

Observation involves watching and recording behaviour. This is useful when self-report might be biased, or when we want to see behaviour as it happens.

Types of Observational Design

• Naturalistic vs. Controlled:
  

  Naturalistic observation is watching behavior in the environment where it would normally occur (e.g., watching playground behaviour). Controlled observation takes place under structured conditions, often in a lab setting, where variables can be managed.

• Overt vs. Covert:
  

  Overt observation: Participants know they are being watched. (Risk of changing behaviour).
  Covert observation: Participants are unaware they are being watched. (Ethical concerns, but high validity).

• Participant vs. Non-participant:
  

  Participant observation: The researcher becomes part of the group they are studying (e.g., joining a cult). Risk of losing objectivity.
  Non-participant observation: The researcher remains separate, observing from a distance (e.g., through a two-way mirror).

1.3 Self-Report Techniques

These methods involve asking people directly about their feelings, thoughts, or actions.

• Questionnaires:
  

  A set of written questions used to assess thoughts or feelings. They can use open questions (detailed, descriptive answers, resulting in qualitative data) or closed questions (fixed choices, easy to count, resulting in quantitative data).

• Interviews:
  

  Face-to-face or telephone interaction.
  Structured Interviews: Questions are pre-set and asked in a fixed order (like a questionnaire read aloud).
  Unstructured Interviews: Works more like a conversation; the researcher has a general aim but the questions develop as the interview progresses.

1.4 Correlations

Correlations analyze the relationship between co-variables (two or more measurable traits or characteristics).

Crucial Point: A correlation tells us if two things are linked, and how strongly, but it does not show cause and effect.

• Positive Correlation: As one variable increases, the other also increases (e.g., hours spent studying and exam score).
• Negative Correlation: As one variable increases, the other decreases (e.g., hours spent watching TV and exam score).
• Zero Correlation: No relationship exists between the co-variables.

---

2. Scientific Processes: Planning and Executing Research

2.1 Aims and Hypotheses

Before any research begins, a psychologist must define their purpose.

• Aims: A general statement of what the researcher intends to investigate (e.g., To investigate whether coffee consumption affects reaction time.)
• Hypotheses: A precise, testable statement predicting the outcome of the research.

Types of Hypotheses

• Directional Hypothesis (One-tailed): Predicts the direction of the results.
  Example: "People who drink coffee will have faster reaction times than those who do not."
• Non-Directional Hypothesis (Two-tailed): Predicts that a difference or relationship exists, but doesn't state which way.
  Example: "There will be a difference in reaction times between people who drink coffee and those who do not."

2.2 Sampling: Who are we studying?

We cannot study everyone (the population). We use a smaller group, the sample, to represent the population.

Population and Sample

The Population is the entire group you are interested in (e.g., all 16-18 year old students globally). The Sample is the group of participants actually selected from that population.

Why is sampling technique important? If your sample is biased (unrepresentative), you cannot generalise your findings back to the whole population.

Key Sampling Techniques

• Random Sampling: Every member of the population has an equal chance of being selected (like names out of a hat). This is often the best technique for avoiding bias, but it can be very difficult to achieve in practice.
• Opportunity Sampling: Selecting anyone who is available and willing to take part at the time of the study (e.g., asking people walking past in the street). This is the easiest and quickest method but usually the most prone to sampling bias.

2.3 Pilot Studies

A pilot study is a small-scale trial run of the actual investigation.

Aim of Piloting: To check if the procedure works, if the instructions are clear, if the task is too difficult, or if the materials are appropriate. It helps the researcher save time and money by identifying flaws before the main study.

2.4 Variables and Operationalisation

In an experiment, we work with specific types of factors (variables):

• Independent Variable (IV): The variable the researcher manipulates or changes.
• Dependent Variable (DV): The variable the researcher measures. The DV is expected to change because of the IV manipulation.
• Extraneous Variables: Any other variables that might affect the DV, but are not the IV. We try to control these!

Operationalisation: This means clearly defining variables in terms of how they can be measured. You must make abstract concepts concrete and measurable.

Example: If your IV is 'tiredness', you must operationalise it. Instead of saying 'tired vs awake', you operationalise it as: "Participants who have slept less than 4 hours (tired group) vs. participants who have slept 8 hours (awake group)."

2.5 Experimental Designs

How do we allocate the participants to the different conditions?

• Independent Groups Design: Two separate groups of participants are used; each group experiences only one condition of the IV.
  Example: Group A gets coffee, Group B gets water.
• Repeated Measures Design: The same participants take part in all conditions of the IV.
  Example: All participants first take the test after coffee, and then take the test again after water.
• Matched Pairs Design: Two separate groups are used, but participants are paired up based on a key characteristic (e.g., IQ, age, memory score) so that each condition has equivalent participants.

2.6 Control Techniques

• Random Allocation: Used in the Independent Groups design. This ensures that participants are randomly assigned to conditions, distributing participant variables (like individual differences in ability) evenly across groups.
• Counterbalancing: Used in the Repeated Measures design (as explained above).

2.7 Demand Characteristics

This is a potential problem in any study. Demand characteristics occur when participants guess the purpose of the study and change their behavior to meet what they think the researcher expects.

Analogy: When you are being observed in a class, you behave better than usual! You are responding to the 'demand' of the situation.

2.8 Specific Design Features (Non-Experimental)

• Observational Design (Behavioural Categories): When observing, researchers must break down the stream of continuous behaviour into measurable units called behavioural categories (e.g., 'hitting', 'sharing toy', 'sitting alone'). These must be objective and clearly defined.
• Questionnaires/Interviews (Open vs. Closed Questions): Remember, open questions yield detailed, qualitative data, while closed questions yield specific, quantifiable (numerical) data.

---

3. Ethics in Psychological Research

Psychological research must protect the participants’ physical and psychological well-being. Psychologists follow a strict ethical code. If a study violates these, it may be deemed unacceptable.

Memory Trick: Think of the 7 Key Ethics as C-C-D-D-P-R-P.

• Consent: Participants must agree to take part, based on full information about the study's nature and aims (informed consent).
• Confidentiality: Personal data must be protected (often done by assigning participants a number instead of using their name).
• Deception: Researchers should avoid misleading participants. Mild deception is sometimes unavoidable, but must be justified.
• Debrief: A crucial conversation after the study where the true aim is revealed, and any deception is addressed.
• Protection from Harm: Participants must not be exposed to greater risk than they would encounter in daily life (physical or psychological).
• Right to Withdraw: Participants must be told they can leave the study at any time, and can even withdraw their data afterwards.
• Privacy: Participants' right to control the flow of information about themselves. Observing people in a public place is usually acceptable, but observing them in private spaces is not.

---

4. Data Handling and Analysis

Once data is collected, we need to sort it, describe it, and present it clearly.

4.1 Quantitative and Qualitative Data

• Quantitative Data: Numerical data (numbers, scores, measurements).
  Strengths: Easy to analyze statistically; objective.
  Weaknesses: Lacks depth and detail.
• Qualitative Data: Non-numerical data (words, descriptions, interviews, transcripts).
  Strengths: Provides rich detail; high external validity.
  Weaknesses: Difficult to analyze and compare; subjective interpretations needed.

4.2 Primary and Secondary Data

• Primary Data: Data collected directly by the researcher specifically for the current investigation (e.g., running your own experiment).
• Secondary Data: Data that already exists and was collected by someone else (e.g., government statistics, previous studies).
• Meta-analysis: A special type of secondary data analysis where a researcher statistically combines the results from multiple existing studies addressing a similar hypothesis.

4.3 Descriptive Statistics

Descriptive statistics summarize the main features of the dataset.

Measures of Central Tendency (The Average Score)

These tell us about the typical or middle value in a dataset.

1. Mean: The arithmetic average (add all scores and divide by the number of scores).
   Advantage: Uses all the data, making it the most sensitive measure.
   Disadvantage: Easily distorted by extreme scores (outliers).
2. Median: The middle score when the data is arranged in order.
   Advantage: Not affected by extreme scores.
3. Mode: The most frequently occurring score.
   Advantage: Useful for nominal (category) data.
   Disadvantage: Can have zero modes or multiple modes.

Calculation Example (Mean): To calculate the mean of scores 2, 4, 6, 8:
\( \text{Mean} = \frac{2 + 4 + 6 + 8}{4} = \frac{20}{4} = 5 \)

Measures of Dispersion (The Spread of Scores)

These tell us how spread out the data is.

1. Range: The difference between the highest and lowest scores (plus one, often, to account for rounding).
   Advantage: Easy to calculate.
   Disadvantage: Only uses two scores, so is easily distorted by a single extreme score.
2. Standard Deviation (SD): Measures the average distance of each data point from the mean.
   Concept: A low SD means scores are tightly clustered around the mean (high consistency). A high SD means scores are widely spread out (low consistency).
   Advantage: Much more precise than the range as it uses all data points.

Fractions and Percentages

Students must be able to calculate and use basic fractions and percentages to represent parts of the total data set.

Example: If 15 out of 20 people agreed, the fraction is \( \frac{15}{20} \), and the percentage is \( \frac{15}{20} \times 100 = 75\% \).

Correlations (Review)

When analyzing correlational data, we determine if the relationship is positive, negative, or zero.

4.4 Presentation and Display of Quantitative Data

Visualizing data makes it easier to understand patterns.

• Tables: Present raw scores and summary data clearly (often include measures of central tendency).
• Bar Charts: Used when data is discrete (separate categories, like comparing means of Independent Groups). Bars do not touch.
• Line Graphs: Used when data is continuous (e.g., changes over time, or repeated measures data).
• Scattergrams: Used to plot correlations. Each point represents two scores (co-variables) from one participant. The pattern of the points indicates the type and strength of the correlation.

4.5 Distributions

How the scores are spread across the range.

• Normal Distribution: The classic 'bell curve'. It is symmetrical, with the highest frequency in the middle. Crucially, the Mean, Median, and Mode are all at the same central point.
• Skewed Distributions: Asymmetrical distributions where the data bunches up towards one end.
  Positive Skew: Scores are clustered on the left, with the 'tail' pointing to the right (positive direction). The mean is pulled right.
  Negative Skew: Scores are clustered on the right, with the 'tail' pointing to the left (negative direction). The mean is pulled left.