A scattergram is a visual graph that displays the relationship between two numerical variables using plotted points to reveal patterns or correlations.
What Is a Scattergram?
A scattergram, also known as a scatterplot or scatter diagram, is a type of graph used in social science research to show how two quantitative variables relate to each other. Each point on the graph represents one case or observation, with one variable plotted along the horizontal axis (x-axis) and the other plotted on the vertical axis (y-axis).
Scattergrams help researchers visualize relationships, trends, and patterns. If the points tend to move together in a certain direction, this suggests a correlation between the two variables. A scattergram is one of the most straightforward tools for exploring bivariate relationships and is often used before running statistical analyses like correlation or regression.
Why Scattergrams Matter in Social Science
In social science research, scattergrams serve as an important exploratory tool. They help researchers visually inspect whether a relationship exists between two variables and what form that relationship might take. For example, a scattergram could reveal a positive relationship between years of education and income, a negative relationship between age and social media use, or no clear relationship at all.
Scattergrams are particularly useful because they:
- Provide a quick, intuitive view of data relationships.
- Help detect patterns that statistical summaries may miss.
- Reveal outliers that might skew results or suggest errors.
- Show the strength and direction of a relationship.
They are often one of the first steps in data analysis, giving researchers clues about how to proceed with more formal statistical tests.
Key Components of a Scattergram
Axes
The x-axis (horizontal) usually displays the independent variable—the variable the researcher believes influences or predicts the other. The y-axis (vertical) shows the dependent variable—the outcome being studied.
For example, if a sociologist is investigating how education level affects income, education would be plotted on the x-axis and income on the y-axis.
Data Points
Each dot on the scattergram represents one observation. For instance, in a study of 100 individuals, the graph would contain 100 points. Each point is placed based on the values of both variables for that individual.
Trend or Line of Best Fit
Researchers often draw a trend line (or regression line) to summarize the general direction of the points. This line helps highlight whether the relationship is positive, negative, or nonexistent, and whether it appears linear or nonlinear.
Types of Relationships in a Scattergram
Scattergrams can show various kinds of relationships between two variables. Understanding these patterns is key to interpreting what the graph is telling us.
Positive Relationship
In a positive (or direct) relationship, as values on the x-axis increase, values on the y-axis also increase. Points tend to rise from the bottom left to the top right of the graph.
Example: A psychologist might find a positive relationship between hours of sleep and reported energy levels.
Negative Relationship
In a negative (or inverse) relationship, as one variable increases, the other decreases. Points slope downward from the top left to the bottom right.
Example: A criminologist might observe a negative relationship between age and number of arrests.
No Relationship
If the points on a scattergram appear randomly scattered without a clear pattern, it suggests there is little or no relationship between the variables.
Example: An education researcher might find no relationship between student shoe size and test scores.
Curvilinear Relationship
Sometimes the relationship is not a straight line but a curve. For example, moderate stress may improve performance, while too much stress reduces it. This creates an inverted U-shape on the graph.
Example: In sociology, a study might find that civic engagement increases with age up to a point, then declines in older populations.
When to Use a Scattergram
Scattergrams are used when both variables are numerical (interval or ratio scale) and when the researcher wants to explore or describe the relationship between them. They are especially useful in the following situations:
- Preparing for a correlation or regression analysis.
- Checking for linear or nonlinear patterns.
- Identifying outliers or clusters in data.
- Teaching or presenting research findings in a visual way.
Example from Social Science Research
Imagine a political scientist wants to study the relationship between voter turnout and income. They survey 200 individuals and collect data on income and whether they voted in the last election.
They assign voting as a percentage (0% for those who didn’t vote, 100% for those who did), plot income on the x-axis, and voting percentage on the y-axis. The scattergram shows a general upward trend, suggesting that higher income is associated with greater voter turnout.
This visual helps confirm the relationship before conducting a correlation test or building a predictive model.
Advantages of Scattergrams
- Easy to understand: They show patterns at a glance, even for audiences with limited statistical training.
- Flexible: Scattergrams can reveal linear, nonlinear, or no relationships.
- Detects outliers: Unusual points stand out clearly.
- Pre-analysis tool: They guide researchers on which statistical tests are appropriate.
Limitations of Scattergrams
While scattergrams are powerful visual tools, they do have some limitations:
- Only show two variables: They don’t capture the effect of other factors.
- Cannot prove causation: A scattergram shows correlation, not cause and effect.
- Cluttered with large datasets: Too many points can make patterns harder to see.
- Sensitive to scale: The appearance can change based on axis scaling.
Researchers should interpret scattergrams alongside other analysis tools to draw reliable conclusions.
Best Practices for Creating Scattergrams
Label Axes Clearly
Always include labels for both axes to show what each variable represents. Also, include units of measurement if applicable.
Use Consistent Scales
Choose scales that accurately reflect the data range. Inconsistent scales can distort the relationship or make it harder to interpret.
Include a Title
Provide a clear title that describes what the scattergram shows. For example: “Relationship Between Weekly Exercise and Reported Happiness.”
Consider Adding a Trend Line
Adding a line of best fit (regression line) can make it easier to see the overall direction and strength of the relationship.
Check for Outliers
Identify any data points that fall far outside the general pattern. These can be errors or interesting exceptions worth further investigation.
Examples Across Social Science Fields
Sociology
A sociologist might use a scattergram to examine the relationship between social media use and reported loneliness. The x-axis would show hours on social media per day, and the y-axis would reflect loneliness scores.
Psychology
A psychologist studying stress and performance might create a scattergram showing test performance scores on the y-axis and reported stress levels on the x-axis.
Education
In education research, a scattergram could show the relationship between number of books read per semester and GPA.
Political Science
A political scientist might use a scattergram to look at the relationship between political knowledge (measured by quiz scores) and civic participation.
Criminology
A criminologist might examine the relationship between neighborhood income levels and crime rates, plotting average household income on one axis and crime incidents on the other.
Conclusion
Scattergrams are essential visual tools in social science research. By plotting two variables against each other, they help researchers detect patterns, identify outliers, and prepare for deeper statistical analysis. While simple in design, scattergrams offer powerful insights into how social variables interact. When used alongside other research tools, they contribute to a clearer, more complete understanding of the data.
Meta description: Scattergram is a graph that uses plotted points to show how two numeric variables relate, helping researchers explore patterns and correlations.
Glossary Return to Doc's Research Glossary
Last Modified: 03/27/2025