A Beginner's Guide to Graphing Data

Main ideas / lessons

• Graphs, charts, and plots are interchangeable terms for visually displaying data.
• Students often make mistakes in graphing because graphing practice is emphasized in elementary school, but less attention is given in later schooling.
• Graphs help convert large lists of numbers into understandable patterns, trends, and relationships—e.g., the Keeling Curve, which shows rising atmospheric CO₂.
• There are five major graph types, and choosing the correct one is crucial:
  • Data comes first, then decide which graph type best represents it.

Five major types of graphs (and when to use each)

1. Line graph

   • Use when: showing change over time
   • How it looks: points connected by a line
   • Axes example: time on the x-axis, quantity (e.g., money) on the y-axis

2. Scatter plot

   • Use when: showing correlation between two variables
   • Key concept: independent vs. dependent variables
     • Independent variable → x-axis
     • Dependent variable → y-axis
   • How it looks: dots not connected; used to reveal relationships

3. Bar graph

   • Use when: comparing groups (two or more groups)
   • How it looks: separate bars with space between columns
   • Common practice mentioned: bars often represent the mean/average

4. Histogram

   • Use when: showing the distribution of data (frequency within ranges)
   • How it looks: bars/blocks that touch each other (no gaps)
   • Axes idea: value ranges on x-axis, number/frequency on y-axis

5. Pie chart

   • Use when: representing parts of a whole
   • How it looks: segments showing percentages
   • Example context: distribution across categories (e.g., rodent families)

Graphing methodology / step-by-step instructions (especially for scatter plots)

Choosing the right graph using examples

• Pie chart: parts of a whole across months (e.g., ice cream sold by month as a fraction/percentage)
• Line graph: over time (e.g., average high temperatures over months)
• Bar graph: compare specific groups (e.g., three months that start with “J” and compare sales)

Elements of a good scatter plot (instructional breakdown)

1. Set up axes

   • Place the independent variable on the x-axis
   • Place the dependent variable on the y-axis

2. Label each axis with units

   • Include units in parentheses (e.g., “average high temperature (°F)”)

3. Write a descriptive title

   • The title should summarize what the graph shows
   • Avoid vague titles like only “ice cream”
   • It should communicate the relationship and typically include context/date if relevant

4. Plot the data points

5. Draw a best-fit line

   • Acts like an “average” trend line
   • Guideline: roughly half the points above and half below
   • Rule: the best-fit line should not extend beyond the data
   • Avoid arrows suggesting extrapolation—interpolate within the data range

Additional quality rules highlighted

• Use appropriate, evenly spaced scaling
  • Major grid steps should have consistent intervals that you can interpret
• Avoid including zero unless it is actually part of the data range
  • “If your data doesn’t have 0, it shouldn’t be on your graph.”

Common mistakes demonstrated (with an example “bad graph”)

• Bad title

  • Too vague; should specify variables and what each axis represents (e.g., relationship between fertilizer and plant height)

• Axis variables swapped

  • Fertilizer is the independent variable but isn’t placed on the x-axis

• Missing units

  • Units should be included in parentheses for both axes

• Scatter plot misuse

  • Dots in a scatter plot should not be connected with lines

• Non-linear / non-uniform axis scaling

  • Tick spacing is uneven (example: inconsistent increments like 5, 7, 11, 15, … where equal spacing should represent equal values)

• Insufficient grid/tick labeling

  • Only one unlabeled/meaningless grid value doesn’t allow accurate reading; include enough tick marks/values to measure properly (at least two numbers on gridlines)

• Best-fit line extends beyond data

  • The trend line should stay within the data range; extending it implies improper extrapolation

Speakers / sources featured

• Mr. Andersen (main speaker/creator of the video)
• Charles David Keeling (historical figure referenced for collecting atmospheric CO₂ data that led to the Keeling Curve)