Welcome to Data Handling!
Once a psychologist has finished their research, they are often left with a mountain of numbers or pages of notes. How do they make sense of it all? This chapter is all about Data Handling: the tools we use to organize, summarize, and visualize information so we can see what it actually tells us about human behavior.
Don't worry if you aren't a "maths person." For AQA Psychology, you just need to understand the logic behind the numbers and follow some simple step-by-step methods.
1. Types of Data
Before we can analyze data, we need to know what kind of data we have. We categorize data in two main ways.
Qualitative vs. Quantitative Data
Qualitative Data is non-numerical data, usually expressed in words. It’s "rich" and detailed, often coming from open-ended questions in interviews or diaries.
Example: A participant describing how they felt during an experiment.
Quantitative Data is numerical data. It’s "narrow" but easy to compare.
Example: The number of words a participant remembered from a list.
Primary vs. Secondary Data
Primary Data is "first-hand" data collected by the researcher specifically for the study they are currently doing.
Example: You conduct an experiment and record the results yourself.
Secondary Data is "second-hand" data that already exists, collected by someone else for a different purpose.
Example: Using government statistics or results from a study published in the 1980s.
Meta-analysis is a special type of secondary research where a researcher looks at the findings of many different studies on the same topic to produce an overall conclusion.
Quick Review: Coding
In Content Analysis, researchers often use Coding. This is the process of turning qualitative data (like a transcript) into quantitative data by counting how many times certain themes or words appear. It’s like turning "descriptions" into "numbers."
Key Takeaway: Quantitative data is about numbers; Qualitative data is about meaning.
2. Descriptive Statistics: Measures of Central Tendency
These are "averages." They give us a single value that represents the "typical" score in a data set.
The Mean
The arithmetic average. You add up all the scores and divide by the number of scores (\( N \)).
Calculation: \( \text{Mean} = \frac{\sum x}{N} \)
Pros: It is the most sensitive measure because it includes every single score.
Cons: It is easily distorted by "outliers" (extremely high or low scores).
The Median
The middle value when scores are put in order.
Calculation: Arrange from lowest to highest. If \( N \) is odd, it’s the middle one. If \( N \) is even, it's the mean of the two middle scores.
Pros: It isn't affected by extreme outliers.
Cons: It doesn't use all the values in the set, so it's less sensitive than the mean.
The Mode
The most frequently occurring score.
Pros: Very easy to calculate and is the only measure you can use for "nominal" data (categories).
Cons: A set of data might have several modes (bi-modal) or no mode at all, which isn't very helpful.
3. Descriptive Statistics: Measures of Dispersion
These tell us how "spread out" the scores are. Are they all bunched together or widely scattered?
The Range
The difference between the highest and lowest scores.
Calculation: \( (\text{Highest score} - \text{Lowest score}) + 1 \)
Note: We usually add \( 1 \) to account for the fact that scores are rounded to the nearest whole number.
Cons: It only looks at the two most extreme scores and ignores the rest.
Standard Deviation (SD)
A more sophisticated measure that tells us the average distance of each score from the mean.
Calculation: While the formula is complex, for the exam, you need to know that a high SD means the scores are spread out (the mean isn't very representative), and a low SD means the scores are clustered close to the mean.
Pros: Much more precise than the range as it uses every score in the set.
Key Takeaway: Use the Mean and Standard Deviation together for the most accurate picture of your data.
4. Percentages and Fractions
In Psychology, you will often need to calculate percentages to compare groups of different sizes.
Calculation: \( \frac{\text{Score}}{\text{Total}} \times 100 \)
Example: If \( 15 \) out of \( 60 \) participants obeyed, the percentage is \( \frac{15}{60} \times 100 = 25\% \).
5. Levels of Measurement
Before choosing a statistical test, you must identify the "level" of your data. Think of this as a ladder of precision (NOIR).
- Nominal: Data in separate categories. (e.g., "Yes" or "No", "Tall" or "Short").
- Ordinal: Data that can be ordered or ranked. The gaps between ranks aren't equal. (e.g., finishing \( 1st \), \( 2nd \), and \( 3rd \) in a race).
- Interval: Data measured using fixed, public scales with equal units. (e.g., temperature in \( ^{\circ}C \), time in seconds, or standardized IQ tests).
6. Data Display: Graphs and Tables
Visuals help us see patterns instantly. Here are the ones you need to know:
- Tables: Raw data is usually converted into a "Summary Table" showing the mean and SD for each condition.
- Bar Charts: Used for discrete data (categories). The bars do not touch. The y-axis represents the frequency or mean.
- Histograms: Used for continuous data (like time or height). The bars do touch because the data flows on a scale.
- Scattergrams: Used specifically for Correlations. Each dot represents a participant's score on two different co-variables.
7. Distributions
When we plot data on a graph, it usually forms a "shape."
Normal Distribution
A symmetrical, bell-shaped curve. Most people are in the middle, with very few at the extreme ends.
Important: In a perfect normal distribution, the Mean, Median, and Mode all occupy the same midpoint.
Skewed Distributions
Sometimes data is pushed to one side.
Positive Skew: Most scores are low, with a few extreme high scores pulling the "tail" to the right. (e.g., a very difficult exam). The Mean is pulled highest, followed by the Median, with the Mode at the peak.
Negative Skew: Most scores are high, with a few extreme low scores pulling the "tail" to the left. (e.g., a very easy exam). The Mean is pulled lowest.
Mnemonics:
- Positive = Point to the right.
- Negative = Nudge to the left.
8. Correlations
Correlations look for a relationship between two co-variables. Unlike experiments, we aren't saying one causes the other.
- Positive Correlation: As one variable increases, the other increases. (\( / \))
- Negative Correlation: As one variable increases, the other decreases. (\( \backslash \))
- Zero Correlation: No relationship between variables.
Correlation Coefficients
This is a number between \( -1.0 \) and \( +1.0 \) that tells us the strength and direction.
- \( +1.0 \) is a perfect positive correlation.
- \( -1.0 \) is a perfect negative correlation.
- \( 0 \) means no relationship.
- The closer the number is to \( +1 \) or \( -1 \), the stronger the relationship is.
Did you know? A correlation coefficient of \( -0.85 \) is actually stronger than a coefficient of \( +0.40 \). The plus or minus just tells you the direction!
Final Tip for the Exam: Always read the graph axes carefully! Examiners often ask you to interpret what a graph is showing or to identify the skew based on where the "tail" is pointing.